Daily Papers

Recent progress in Large Vision-Language Models (LVLMs) has enabled promising applications in medical tasks, such as report generation and visual question answering. However, existing benchmarks focus mainly on the final diagnostic answer, offering limited insight into whether models engage in clinically meaningful reasoning. To address this, we present CheXStruct and CXReasonBench, a structured pipeline and benchmark built on the publicly available MIMIC-CXR-JPG dataset. CheXStruct automatically derives a sequence of intermediate reasoning steps directly from chest X-rays, such as segmenting anatomical regions, deriving anatomical landmarks and diagnostic measurements, computing diagnostic indices, and applying clinical thresholds. CXReasonBench leverages this pipeline to evaluate whether models can perform clinically valid reasoning steps and to what extent they can learn from structured guidance, enabling fine-grained and transparent assessment of diagnostic reasoning. The benchmark comprises 18,988 QA pairs across 12 diagnostic tasks and 1,200 cases, each paired with up to 4 visual inputs, and supports multi-path, multi-stage evaluation including visual grounding via anatomical region selection and diagnostic measurements. Even the strongest of 10 evaluated LVLMs struggle with structured reasoning and generalization, often failing to link abstract knowledge with anatomically grounded visual interpretation. The code is available at https://github.com/ttumyche/CXReasonBench

Explainable AI (XAI) in medical histopathology is essential for enhancing the interpretability and clinical trustworthiness of deep learning models in cancer diagnosis. However, the black-box nature of these models often limits their clinical adoption. We introduce GRAPHITE (Graph-based Interpretable Tissue Examination), a post-hoc explainable framework designed for breast cancer tissue microarray (TMA) analysis. GRAPHITE employs a multiscale approach, extracting patches at various magnification levels, constructing an hierarchical graph, and utilising graph attention networks (GAT) with scalewise attention (SAN) to capture scale-dependent features. We trained the model on 140 tumour TMA cores and four benign whole slide images from which 140 benign samples were created, and tested it on 53 pathologist-annotated TMA samples. GRAPHITE outperformed traditional XAI methods, achieving a mean average precision (mAP) of 0.56, an area under the receiver operating characteristic curve (AUROC) of 0.94, and a threshold robustness (ThR) of 0.70, indicating that the model maintains high performance across a wide range of thresholds. In clinical utility, GRAPHITE achieved the highest area under the decision curve (AUDC) of 4.17e+5, indicating reliable decision support across thresholds. These results highlight GRAPHITE's potential as a clinically valuable tool in computational pathology, providing interpretable visualisations that align with the pathologists' diagnostic reasoning and support precision medicine.

Artificial intelligence is revealing what medicine never intended to encode. Deep vision models, trained on chest X-rays, can now detect not only disease but also invisible traces of social inequality. In this study, we show that state-of-the-art architectures (DenseNet121, SwinV2-B, MedMamba) can predict a patient's health insurance type, a strong proxy for socioeconomic status, from normal chest X-rays with significant accuracy (AUC around 0.70 on MIMIC-CXR-JPG, 0.68 on CheXpert). The signal was unlikely contributed by demographic features by our machine learning study combining age, race, and sex labels to predict health insurance types; it also remains detectable when the model is trained exclusively on a single racial group. Patch-based occlusion reveals that the signal is diffuse rather than localized, embedded in the upper and mid-thoracic regions. This suggests that deep networks may be internalizing subtle traces of clinical environments, equipment differences, or care pathways; learning socioeconomic segregation itself. These findings challenge the assumption that medical images are neutral biological data. By uncovering how models perceive and exploit these hidden social signatures, this work reframes fairness in medical AI: the goal is no longer only to balance datasets or adjust thresholds, but to interrogate and disentangle the social fingerprints embedded in clinical data itself.

Clinical decision-support systems are expert systems whose recommendations clinicians act on directly, yet they are usually cleared on one aggregate accuracy number from a held-out test set. That number says nothing about input reliability under encoding shifts, subgroup gaps, threshold sensitivity, or operational feasibility. We present RISED, a pre-deployment evaluation framework operationalising five dimensions (Reliability, Inclusivity, Sensitivity, Equity, Deployability) through BCa bootstrap 95% confidence intervals, literature-grounded thresholds, and Holm-Bonferroni-corrected PASS / FAIL / INCONCLUSIVE verdicts; Equity is a proxy-dependence diagnostic rather than a gating test. Applied to seven cohorts spanning 35 years (n from 303 to 99,492), RISED surfaces failures invisible to AUROC: on Diabetes 130, Reliability passes by three orders of magnitude (PSS = 0.0004) while Inclusivity (AUC parity gap = 0.262) and Sensitivity (max threshold-flip rate 49.1%) fail decisively; both NHIS cohorts reproduce this. NHANES 2021-2023, with a complete feature profile, achieves INCONCLUSIVE verdicts; BRFSS 2024 produces the suite's most severe Sensitivity failure (max threshold-flip rate 64.2%) after instrument rotation removed hypertension and cholesterol. The pattern recurs on credit- and income-prediction cohorts, confirming domain-agnosticity; a multi-model check shows the failures are data-driven, not model-specific. RISED ships as an open-source Python package complementing TRIPOD+AI, FUTURE-AI, and Fairlearn with the structured numerical evidence those standards require but do not prescribe.

Predictive artificial intelligence (AI) offers an opportunity to improve clinical practice and patient outcomes, but risks perpetuating biases if fairness is inadequately addressed. However, the definition of "fairness" remains unclear. We conducted a scoping review to identify and critically appraise fairness metrics for clinical predictive AI. We defined a "fairness metric" as a measure quantifying whether a model discriminates (societally) against individuals or groups defined by sensitive attributes. We searched five databases (2014-2024), screening 820 records, to include 41 studies, and extracted 62 fairness metrics. Metrics were classified by performance-dependency, model output level, and base performance metric, revealing a fragmented landscape with limited clinical validation and overreliance on threshold-dependent measures. Eighteen metrics were explicitly developed for healthcare, including only one clinical utility metric. Our findings highlight conceptual challenges in defining and quantifying fairness and identify gaps in uncertainty quantification, intersectionality, and real-world applicability. Future work should prioritise clinically meaningful metrics.

Respiratory acoustic foundation models (FMs) are benchmarked exclusively on smartphone recordings, yet clinical deployment increasingly targets body-coupled (BC) wearables whose sensors attenuate high-frequency content through tissue and bone, leaving FM reliability uncharacterised. We introduce BCoughBench, evaluating five FMs (OPERA-CT/CE/GT, HeAR, M2D+Resp) on nine classification tasks (AUROC, sensitivity at 95% specificity, Expected Calibration Error) and three age regression tasks (MAE vs. a mean-predictor baseline) across five EBEN-simulated BC sensor conditions on five labeled cough datasets. Mean AUROC declines from 0.785 (smartphone) to 0.689-0.723, degrading most under temple vibration pickup (Δ = -0.096) and least under the soft in-ear (Δ = -0.062). No FM meets the clinical sensitivity threshold (Se@Sp95 geq 0.20) on most disease tasks under any BC sensor. Sex classification on the CIDRZ cohort collapses (AUROC 0.954 to 0.596-0.628, Δ = -0.341) while COVID detection is nearly unaffected (Δ = -0.004). Age regression is robust, improving under the forehead accelerometer on CoughVID (MAE 9.61 to 8.97 yr); HeAR leads on regression and demographic tasks, M2D+Resp on disease and characteristic tasks. BCoughBench provides a reproducible framework for FM evaluation under wearable conditions.

We use a Stochastic Hybrid Automaton (SHA) model of prostate cancer evolution under intermittent androgen suppression (IAS) to study a threshold-based policy for therapy design. IAS is currently one of the most widely used treatments for advanced prostate cancer. Patients undergoing IAS are submitted to cycles of treatment (in the form of androgen deprivation) and off-treatment periods in an alternating manner. One of the main challenges in IAS is to optimally design a therapy scheme, i.e., to determine when to discontinue and recommence androgen suppression. The level of prostate specific antigen (PSA) in a patient's serum is frequently monitored to determine when the patient will be taken off therapy and when therapy will resume. The threshold-based policy we propose is parameterized by lower and upper PSA threshold values and is associated with a cost metric that combines clinically relevant measures of therapy success. Using Infinitesimal Perturbation Analysis (IPA), we derive unbiased gradient estimators of this cost metric with respect to the controllable PSA threshold values based on actual data and show how these estimators can be used to adaptively adjust controllable parameters so as to improve therapy outcomes based on the cost metric defined.

Intermittent Androgen Suppression (IAS) is a treatment strategy for delaying or even preventing time to relapse of advanced prostate cancer. IAS consists of alternating cycles of therapy (in the form of androgen suppression) and off-treatment periods. The level of prostate specific antigen (PSA) in a patient's serum is frequently monitored to determine when the patient will be taken off therapy and when therapy will resume. In spite of extensive recent clinical experience with IAS, the design of an ideal protocol for any given patient remains one of the main challenges associated with effectively implementing this therapy. We use a threshold-based policy for optimal IAS therapy design that is parameterized by lower and upper PSA threshold values and is associated with a cost metric that combines clinically relevant measures of therapy success. We apply Infinitesimal Perturbation Analysis (IPA) to a Stochastic Hybrid Automaton (SHA) model of prostate cancer evolution under IAS and derive unbiased estimators of the cost metric gradient with respect to various model and therapy parameters. These estimators are subsequently used for system analysis. By evaluating sensitivity estimates with respect to several model parameters, we identify critical parameters and demonstrate that relaxing the optimality condition in favor of increased robustness to modeling errors provides an alternative objective to therapy design for at least some patients.

Objective: To propose and retrospectively validate an integrated framework addressing three barriers to clinical translation of readmission prediction: lack of explainability, absence of deployment reliability infrastructure, and inadequate demographic fairness evaluation. Materials and Methods: We constructed a cohort of 415231 adult admissions from the MIMIC-IV database (30-day readmission prevalence 18.0%), split 70/15/15. Logistic regression, XGBoost, and LightGBM models were trained on 26 features. SHAP provided per-patient explanations. Fairness was evaluated across 16 subgroups using AUC-ROC, false negative rate (FNR), and positive predictive value (PPV). Calibration was assessed using Brier scores and calibration curves. Results: XGBoost achieved AUC-ROC 0.696 (95% CI 0.691-0.701), outperforming or matching the LACE baseline (AUC 0.60-0.68). LightGBM achieved best calibration (Brier 0.146). Prior admissions were the dominant predictor. All subgroups met equity thresholds (delta AUC <= 0.05, delta FNR <= 0.10). Conclusion: This framework delivers competitive performance, clinically actionable explanations, and strong demographic equity. Code is publicly available at https://github.com/Tomisin92/readmission-prediction.

Multimodal large language models (LLMs) are increasingly explored as automated evaluators in clinical settings, yet their scoring behavior on ordinal clinical scales remains poorly understood. We benchmark three frontier LLM families against supervised deep learning models for scoring Clock Drawing Test (CDT) images on two public datasets using the Shulman rubric. While fully fine-tuned Vision Transformers achieve the best calibration (MAE 0.52, within-1 accuracy 91%), zero-shot LLMs remain competitive on tolerance-based agreement (GPT-5 MAE 0.67, within-1 accuracy 92%) despite higher absolute error. However, per-score analysis reveals that all three LLM families exhibit a pronounced central tendency effect (systematic endpoint compression): predictions are systematically compressed toward the middle of the scale, with over-prediction at the low end (score 0 to 1) and under-prediction at the high end (score 5 to 4). This effect disproportionately affects the clinically critical extremes where accurate scoring most impacts screening decisions for cognitive impairment. Targeted ablations show that neither few-shot exemplars spanning the full score range nor removing clinical terminology from the prompt eliminates the effect. Our findings extend the LLM-as-a-judge bias literature from NLP evaluation to clinical assessment, and highlight the need for calibration-aware evaluation and post-hoc calibration before deploying LLM-based raters in high-stakes screening workflows.

Objective. Clinical AI documentation systems require evaluation methodologies that are clinically valid, economically viable, and sensitive to iterative changes. Methods requiring expert review per scoring instance are too slow and expensive for safe, iterative deployment. We present a case-specific, clinician-authored rubric methodology for clinical AI evaluation and examine whether LLM-generated rubrics can approximate clinician agreement. Materials and Methods. Twenty clinicians authored 1,646 rubrics for 823 clinical cases (736 real-world, 87 synthetic) across primary care, psychiatry, oncology, and behavioral health. Each rubric was validated by confirming that an LLM-based scoring agent consistently scored clinician-preferred outputs higher than rejected ones. Seven versions of an EHR-embedded AI agent for clinicians were evaluated across all cases. Results. Clinician-authored rubrics discriminated effectively between high- and low-quality outputs (median score gap: 82.9%) with high scoring stability (median range: 0.00%). Median scores improved from 84% to 95%. In later experiments, clinician-LLM ranking agreement (tau: 0.42-0.46) matched or exceeded clinician-clinician agreement (tau: 0.38-0.43), attributable to both ceiling compression and LLM rubric improvement. Discussion. This convergence supports incorporating LLM rubrics alongside clinician-authored ones. At roughly 1,000 times lower cost, LLM rubrics enable substantially greater evaluation coverage, while continued clinical authorship grounds evaluation in expert judgment. Ceiling compression poses a methodological challenge for future inter-rater agreement studies. Conclusion. Case-specific rubrics offer a path for clinical AI evaluation that preserves expert judgment while enabling automation at three orders lower cost. Clinician-authored rubrics establish the baseline against which LLM rubrics are validated.

A compartmental deterministic model that allows (1) immunity from two stages of infection and carriage, and (2) disease induced death, is used in studying the dynamics of meningitis epidemic process in a closed population. It allows for difference in the transmission rate of infection to a susceptible by a carrier and an infective. It is generalized to allow a proportion ({\phi}) of those susceptibles infected to progress directly to infectives in stage I. Both models are used in this study. The threshold conditions for the spread of carrier and infectives in stage I are derived for the two models. Sensitivity analysis is performed on the reproductive number derived from the next generation matrix. The case-carrier ratio profile for various parameters and threshold values are shown. So also are the graphs of the total number ever infected as influenced by {\epsilon} and {\phi}. The infection transmission rate (eta), the odds in favor of a carrier, over an infective, in transmitting an infection to a susceptible ({\epsilon}) and the carrier conversion rate ({\phi}) to an infective in stage I, are identified as key parameters that should be subject of attention for any control intervention strategy. The case-carrier ratio profiles provide evidence of a critical case-carrier ratio attained before the number of reported cases grows to an epidemic level. They also provide visual evidence of epidemiological context, in this case, epidemic incidence (in later part of dry season) and endemic incidence (during rainy season). Results from total proportion ever infected suggest that the model, in which {\phi}=0 obtained, can adequately represent, in essence, the generalized model for this study.

We study the problem of detecting the correlation between two Gaussian databases XinR^{ntimes d} and Y^{ntimes d}, each composed of n users with d features. This problem is relevant in the analysis of social media, computational biology, etc. We formulate this as a hypothesis testing problem: under the null hypothesis, these two databases are statistically independent. Under the alternative, however, there exists an unknown permutation sigma over the set of n users (or, row permutation), such that X is rho-correlated with Y^sigma, a permuted version of Y. We determine sharp thresholds at which optimal testing exhibits a phase transition, depending on the asymptotic regime of n and d. Specifically, we prove that if rho^2dto0, as dtoinfty, then weak detection (performing slightly better than random guessing) is statistically impossible, irrespectively of the value of n. This compliments the performance of a simple test that thresholds the sum all entries of X^TY. Furthermore, when d is fixed, we prove that strong detection (vanishing error probability) is impossible for any rho<rho^star, where rho^star is an explicit function of d, while weak detection is again impossible as long as rho^2dto0. These results close significant gaps in current recent related studies.

Clinical LLMs are often scaled by increasing model size, context length, retrieval complexity, or inference-time compute, with the implicit expectation that higher accuracy implies safer behavior. This assumption is incomplete in medicine, where a few confident, high-risk, or evidence-contradicting errors can matter more than average benchmark performance. We introduce SaFE-Scale, a framework for measuring how clinical LLM safety changes across model scale, evidence quality, retrieval strategy, context exposure, and inference-time compute. To instantiate this framework, we introduce RadSaFE-200, a Radiology Safety-Focused Evaluation benchmark of 200 multiple-choice questions with clinician-defined clean evidence, conflict evidence, and option-level labels for high-risk error, unsafe answer, and evidence contradiction. We evaluated 34 locally deployed LLMs across six deployment conditions: closed-book prompting (zero-shot), clean evidence, conflict evidence, standard RAG, agentic RAG, and max-context prompting. Clean evidence produced the strongest improvement, increasing mean accuracy from 73.5% to 94.1%, while reducing high-risk error from 12.0% to 2.6%, contradiction from 12.7% to 2.3%, and dangerous overconfidence from 8.0% to 1.6%. Standard RAG and agentic RAG did not reproduce this safety profile: agentic RAG improved accuracy over standard RAG and reduced contradiction, but high-risk error and dangerous overconfidence remained elevated. Max-context prompting increased latency without closing the safety gap, and additional inference-time compute produced only limited gains. Worst-case analysis showed that clinically consequential errors concentrated in a small subset of questions. Clinical LLM safety is therefore not a passive consequence of scaling, but a deployment property shaped by evidence quality, retrieval design, context construction, and collective failure behavior.

While large language models (LLMs) achieve near-perfect scores on medical licensing exams, these evaluations inadequately reflect the complexity and diversity of real-world clinical practice. We introduce MedHELM, an extensible evaluation framework for assessing LLM performance for medical tasks with three key contributions. First, a clinician-validated taxonomy spanning 5 categories, 22 subcategories, and 121 tasks developed with 29 clinicians. Second, a comprehensive benchmark suite comprising 35 benchmarks (17 existing, 18 newly formulated) providing complete coverage of all categories and subcategories in the taxonomy. Third, a systematic comparison of LLMs with improved evaluation methods (using an LLM-jury) and a cost-performance analysis. Evaluation of 9 frontier LLMs, using the 35 benchmarks, revealed significant performance variation. Advanced reasoning models (DeepSeek R1: 66% win-rate; o3-mini: 64% win-rate) demonstrated superior performance, though Claude 3.5 Sonnet achieved comparable results at 40% lower estimated computational cost. On a normalized accuracy scale (0-1), most models performed strongly in Clinical Note Generation (0.73-0.85) and Patient Communication & Education (0.78-0.83), moderately in Medical Research Assistance (0.65-0.75), and generally lower in Clinical Decision Support (0.56-0.72) and Administration & Workflow (0.53-0.63). Our LLM-jury evaluation method achieved good agreement with clinician ratings (ICC = 0.47), surpassing both average clinician-clinician agreement (ICC = 0.43) and automated baselines including ROUGE-L (0.36) and BERTScore-F1 (0.44). Claude 3.5 Sonnet achieved comparable performance to top models at lower estimated cost. These findings highlight the importance of real-world, task-specific evaluation for medical use of LLMs and provides an open source framework to enable this.

We developed an explainable artificial intelligence (AI) early warning score (xAI-EWS) system for early detection of acute critical illness. While maintaining a high predictive performance, our system explains to the clinician on which relevant electronic health records (EHRs) data the prediction is grounded. Acute critical illness is often preceded by deterioration of routinely measured clinical parameters, e.g., blood pressure and heart rate. Early clinical prediction is typically based on manually calculated screening metrics that simply weigh these parameters, such as Early Warning Scores (EWS). The predictive performance of EWSs yields a tradeoff between sensitivity and specificity that can lead to negative outcomes for the patient. Previous work on EHR-trained AI systems offers promising results with high levels of predictive performance in relation to the early, real-time prediction of acute critical illness. However, without insight into the complex decisions by such system, clinical translation is hindered. In this letter, we present our xAI-EWS system, which potentiates clinical translation by accompanying a prediction with information on the EHR data explaining it.

Large language models (LLMs) increasingly summarize clinical evidence, where a claim's weight depends on how strongly it is supported. Yet these models convey confidence poorly, and properties they never state, such as truth, are often readable from their activations. Whether a clinical model registers evidence strength, distinct from truth, and states it when asked is untested, and any such signal could be lexical. We compiled 45,134 clinical claims from six public sources, harmonized 20,611 into a four-level evidence grade under three independent frameworks, and tested 22 local, open-weight LLMs from several developers (0.6-70 billion parameters; general, medical, and reasoning), with lexical, truth, and cross-framework controls. A linear estimator recovered the grade in every model (median AUROC 71.8), yet decodability did not rise with scale and was weakest in reasoning models. The grade the models stated fell to chance, 25-27 percentage points below the estimator. The recoverable signal was largely lexical and did not transfer across topics or frameworks, yet it was distinct from factual truth and still flagged weakly supported claims (AUROC 69.2). Clinical LLMs thus carry an ordered evidence-strength signal they do not express, so their stated grades fail to convey a claim's support even when it is recoverable from their representations and text.

The validity screen (Cacioli, 2026d, 2026e) classifies LLM confidence signals as Valid, Indeterminate, or Invalid. We test whether these classifications predict selective prediction performance. Twenty frontier LLMs from seven families were evaluated on 524 items across six cognitive tracks. Valid models show mean Type 2 AUROC = .624 (SD = .048). Invalid models show mean AUROC = .357 (SD = .231). Cohen's d = 2.81, p = .002. The tiers order monotonically: Invalid (.357) < Indeterminate (.554) < Valid (.624). Split-half cross-validation yields median d = 1.77, P(d > 0) = 1.0 across 1,000 splits. The three-tier classification accounts for 47% of the variance in AUROC. DeepSeek-R1 drops from 85.3% accuracy at full coverage to 11.3% at 10% coverage. The screen predicts the criterion. For selective prediction, the screen matters.

Speech patterns have been identified as potential diagnostic markers for neuropsychiatric conditions. However, most studies only compare a single clinical group to healthy controls, whereas clinical practice often requires differentiating between multiple potential diagnoses (multiclass settings). To address this, we assembled a dataset of repeated recordings from 420 participants (67 with major depressive disorder, 106 with schizophrenia and 46 with autism, as well as matched controls), and tested the performance of a range of conventional machine learning models and advanced Transformer models on both binary and multiclass classification, based on voice and text features. While binary models performed comparably to previous research (F1 scores between 0.54-0.75 for autism spectrum disorder, ASD; 0.67-0.92 for major depressive disorder, MDD; and 0.71-0.83 for schizophrenia); when differentiating between multiple diagnostic groups performance decreased markedly (F1 scores between 0.35-0.44 for ASD, 0.57-0.75 for MDD, 0.15-0.66 for schizophrenia, and 0.38-0.52 macro F1). Combining voice and text-based models yielded increased performance, suggesting that they capture complementary diagnostic information. Our results indicate that models trained on binary classification may learn to rely on markers of generic differences between clinical and non-clinical populations, or markers of clinical features that overlap across conditions, rather than identifying markers specific to individual conditions. We provide recommendations for future research in the field, suggesting increased focus on developing larger transdiagnostic datasets that include more fine-grained clinical features, and that can support the development of models that better capture the complexity of neuropsychiatric conditions and naturalistic diagnostic assessment.

Large language models (LLMs) are increasingly explored for clinical decision support, yet most evaluations are conducted in English, leaving their reliability in other languages uncertain. Here we evaluate the impact of prompting language on diagnostic reasoning and final diagnosis accuracy by comparing English and French performance across five LLMs (o3, DeepSeek-R1, GPT-4-Turbo, Llama-3.1-405B-Instruct, and BioMistral-7B). A total of 180 clinical vignettes covering 16 medical specialties were assessed by two physicians using an 18-point scale evaluating both diagnosis accuracy and reasoning quality. Four of the five models performed better in English (mean difference 0.37-0.91, adjusted p < 0.05), with the gap spanning multiple aspects of reasoning, including differential diagnosis, logical structure, and internal validity. o3 was the only model showing no overall language effect. These findings demonstrate that prompting language remains a critical determinant of LLM clinical performance, with implications for equitable linguistico-cultural deployment worldwide.

Clinical personality assessment screens response validity before interpreting substantive scales. LLM evaluation does not. We apply the validity scaling framework from the PAI and MMPI-3 to metacognitive probe data from 20 frontier models across 524 items. Six validity indices are operationalised: L (maintaining confidence on errors), K (betting on errors), F (withdrawing consensus-endorsed items), Fp (withdrawing correct answers), RBS (inverted monitoring), and TRIN (fixed responding). A tiered classification system identifies four models as construct-level invalid and two as elevated. Valid-profile models produce item-sensitive confidence (mean r = .18, 14 of 16 significant). Invalid-profile models do not (mean r = -.20, d = 2.17, p = .001). Chain-of-thought training produces two opposite response distortions. Two latent dimensions account for 94.6% of index variance. Companion papers extract a portable screening protocol (Cacioli, 2026e) and validate it against selective prediction (Cacioli, 2026f). All data and code: https://github.com/synthiumjp/validity-scaling-llm

Studies involving soundscape perception often exclude participants with hearing loss to prevent impaired perception from affecting experimental results. Participants are typically screened with pure tone audiometry, the "gold standard" for identifying and quantifying hearing loss at specific frequencies, and excluded if a study-dependent threshold is not met. However, procuring professional audiometric equipment for soundscape studies may be cost-ineffective, and manually performing audiometric tests is labour-intensive. Moreover, testing requirements for soundscape studies may not require sensitivities and specificities as high as that in a medical diagnosis setting. Hence, in this study, we investigate the effectiveness of the uHear app, an iOS application, as an affordable and automatic alternative to a conventional audiometer in screening participants for hearing loss for the purpose of soundscape studies or listening tests in general. Based on audiometric comparisons with the audiometer of 163 participants, the uHear app was found to have high precision (98.04%) when using the World Health Organization (WHO) grading scheme for assessing normal hearing. Precision is further improved (98.69%) when all frequencies assessed with the uHear app is considered in the grading, which lends further support to this cost-effective, automated alternative to screen for normal hearing.

Finding relevant literature underpins the practice of evidence-based medicine. From 2014 to 2016, TREC conducted a clinical decision support track, wherein participants were tasked with finding articles relevant to clinical questions posed by physicians. In total, 87 teams have participated over the past three years, generating 395 runs. During this period, each team has trialled a variety of methods. While there was significant overlap in the methods employed by different teams, the results were varied. Due to the diversity of the platforms used, the results arising from the different techniques are not directly comparable, reducing the ability to build on previous work. By using a stable platform, we have been able to compare different document and query processing techniques, allowing us to experiment with different search parameters. We have used our system to reproduce leading teams runs, and compare the results obtained. By benchmarking our indexing and search techniques, we can statistically test a variety of hypotheses, paving the way for further research.

Large Language Models (LLMs) hold great promise to revolutionize current clinical systems for their superior capacities on medical text processing tasks and medical licensing exams. Meanwhile, traditional ML models such as SVM and XGBoost have still been mainly adopted in clinical prediction tasks. An emerging question is Can LLMs beat traditional ML models in clinical prediction? Thus, we build a new benchmark ClinicalBench to comprehensively study the clinical predictive modeling capacities of both general-purpose and medical LLMs, and compare them with traditional ML models. ClinicalBench embraces three common clinical prediction tasks, two databases, 14 general-purpose LLMs, 8 medical LLMs, and 11 traditional ML models. Through extensive empirical investigation, we discover that both general-purpose and medical LLMs, even with different model scales, diverse prompting or fine-tuning strategies, still cannot beat traditional ML models in clinical prediction yet, shedding light on their potential deficiency in clinical reasoning and decision-making. We call for caution when practitioners adopt LLMs in clinical applications. ClinicalBench can be utilized to bridge the gap between LLMs' development for healthcare and real-world clinical practice.

We develop an algorithm which exceeds the performance of board certified cardiologists in detecting a wide range of heart arrhythmias from electrocardiograms recorded with a single-lead wearable monitor. We build a dataset with more than 500 times the number of unique patients than previously studied corpora. On this dataset, we train a 34-layer convolutional neural network which maps a sequence of ECG samples to a sequence of rhythm classes. Committees of board-certified cardiologists annotate a gold standard test set on which we compare the performance of our model to that of 6 other individual cardiologists. We exceed the average cardiologist performance in both recall (sensitivity) and precision (positive predictive value).

Physicians now pose millions of clinical questions to AI tools each week, yet these tools are evaluated largely on hypothetical or exam-style questions, not those actually asked in practice. We report a blinded evaluation built on 620 Real-world Point-Of-Care Queries (Real-POCQi) submitted to the OpenEvidence (OE) platform by physicians spanning 30 specialties, as well as 187 questions from HealthBench. 149 practicing physicians across 36 states made head-to-head comparisons between answers from three frontier general-purpose models (Claude Opus 4.8, Gemini 3.1 Pro, and GPT-5.5) and a specialized clinical tool (OE), with graders matched to each question's specialty. When comparing answers along five dimensions relevant to clinical decision support -- accuracy, clinical utility, source quality, verifiability, & completeness -- physicians scored the specialized tool highest on all axes; in the primary analysis on Real-POCQi, win differences (margins between win and loss rates) ranged from 25 to 39 percentage points (p<0.001). Results remained consistent in sensitivity analyses stratifying by citation display, answer length, OE-user status, and Real-POCQi versus HealthBench. In parallel, LLM judges were found to systematically differ from expert judges, though both generally agreed on the best model. These findings underscore two conclusions: (i) AI tool evaluations should reflect real-world query distributions and use expert judges that mirror the specialization defining modern medicine and (ii) the consistent advantage of the specialized tool over general-purpose models does not necessarily mean that the latter cannot serve similar purposes, but that targeted engineering and customization can yield meaningful gains in performance for its users. We release Real-POCQi as a public benchmark, as well as the prespecified statistical analysis for reproducing results of this study.

Clinical protocol-execution tasks -- checking a lab value, applying a threshold, placing a correctly structured FHIR order -- are natural candidates for RL from world feedback: once clinical SMEs encode decision logic into a verifier, that verifier grades unlimited rollouts without per-episode annotation. But applying RL requires a sound feedback channel and sufficient base capability. We audit MedAgentBench v1/v2, find a 41.7\% silent-finish ceiling that makes inaction the RL dominant strategy, and construct MedAgentBench-v3 (MAB-v3) (508 tasks, 8.9\% ceiling). Training Qwen3-8B exposes two structural barriers: a capability ceiling (10/20 task types have 0\% base performance, zero gradient) and a format-knowledge barrier (3/20 types require exact clinical codes undiscoverable by exploration). Pure RL reaches 18.2\% pass@1 vs.\ 34.1\% for rule-based SFT; the 15.9~pp gap is attributable entirely to these barriers. A decision/format-knowledge/lookup taxonomy predicts RL learnability and prescribes the fix: SFT to inject codes, RL to learn conditionals.

Ask a frontier model how to taper six milligrams of alprazolam (psychiatrist retired, ten days of pills left, abrupt cessation causes seizures) and it tells her to call the psychiatrist she just explained does not exist. Change one word ("I'm a psychiatrist; a patient presents with...") and the same model, same weights, same inference pass produces a textbook Ashton Manual taper with diazepam equivalence, anticonvulsant coverage, and monitoring thresholds. The knowledge was there; the model withheld it. IatroBench measures this gap. Sixty pre-registered clinical scenarios, six frontier models, 3,600 responses, scored on two axes (commission harm, CH 0-3; omission harm, OH 0-4) through a structured-evaluation pipeline validated against physician scoring (kappa_w = 0.571, within-1 agreement 96%). The central finding is identity-contingent withholding: match the same clinical question in physician vs. layperson framing and all five testable models provide better guidance to the physician (decoupling gap +0.38, p = 0.003; binary hit rates on safety-colliding actions drop 13.1 percentage points in layperson framing, p < 0.0001, while non-colliding actions show no change). The gap is widest for the model with the heaviest safety investment (Opus, +0.65). Three failure modes separate cleanly: trained withholding (Opus), incompetence (Llama 4), and indiscriminate content filtering (GPT-5.2, whose post-generation filter strips physician responses at 9x the layperson rate because they contain denser pharmacological tokens). The standard LLM judge assigns OH = 0 to 73% of responses a physician scores OH >= 1 (kappa = 0.045); the evaluation apparatus has the same blind spot as the training apparatus. Every scenario targets someone who has already exhausted the standard referrals.

Across medical specialties, clinical practice is anchored in evidence-based guidelines that codify best studied diagnostic and treatment pathways. These pathways routinely fall short for the long tail of real-world care not covered by guidelines. Most medical large language models (LLMs), however, are trained to encode common, guideline-focused medical knowledge in their parameters. Current evaluations test models primarily on recalling and reasoning with this memorized content, often in multiple-choice settings. Given the fundamental importance of evidence-based reasoning in medicine, it is neither feasible nor reliable to depend on memorization in practice. To address this gap, we introduce OGCaReBench, a free-form retrieval-focused benchmark aimed at evaluating LLMs at answering clinical questions that require going beyond typical guidelines. Extracted from published medical case reports and validated by medical experts, OGCaReBench contains long-form clinical questions requiring free-text answers, providing a systematic framework for assessing open-ended medical reasoning in rare, case-based scenarios. Our experiments reveal that even the best-performing baseline (GPT-5.2) correctly answers only 56% of our benchmark with specialized models only reaching 42%. Augmenting models with retrieved medical articles improves this performance to up to 82% (using GPT-5.2) highlighting the importance of evidence-grounding for real-world medical reasoning tasks. This work thus establishes a foundation for benchmarking and advancing both general-purpose and medical LLMs to produce reliable answers in challenging clinical contexts.

Large language models (LLMs) are routinely used by physicians and patients for medical advice, yet their clinical safety profiles remain poorly characterized. We present NOHARM (Numerous Options Harm Assessment for Risk in Medicine), a benchmark using 100 real primary care-to-specialist consultation cases to measure frequency and severity of harm from LLM-generated medical recommendations. NOHARM covers 10 specialties, with 12,747 expert annotations for 4,249 clinical management options. Across 31 LLMs, potential for severe harm from LLM recommendations occurs in up to 22.2% (95% CI 21.6-22.8%) of cases, with harm of omission accounting for 76.6% (95% CI 76.4-76.8%) of errors. Safety performance is only moderately correlated (r = 0.61-0.64) with existing AI and medical knowledge benchmarks. The best models outperform generalist physicians on safety (mean difference 9.7%, 95% CI 7.0-12.5%), and a diverse multi-agent approach improves safety compared to solo models (mean difference 8.0%, 95% CI 4.0-12.1%). Therefore, despite strong performance on existing evaluations, widely used AI models can produce severely harmful medical advice at nontrivial rates, underscoring clinical safety as a distinct performance dimension necessitating explicit measurement.

Clinical early warning systems built on electronic health records, in which clinical observations are recorded as irregularly sampled medical time series (ISMTS), must deliver both calibrated risk scores for patient triage and interpretable rationales that clinicians can verify. Large Language Models (LLMs) have been explored for this task, yet they collapse graded clinical risk into overconfident binary predictions. This risk polarization undermines both calibration and cross-patient comparability. To address this, we propose TRIAGE, a framework that trains an LLM to generate dialectical reasoning over competing clinical outcomes by eliciting outcome-specific rationales. This dialectical formulation mitigates risk polarization, enabling a single LLM to yield continuous risk scores grounded in explicit clinical reasoning. Evaluated on three ISMTS benchmarks, TRIAGE achieves an average AUPRC improvement of 3.3% and reduces calibration error by 81% compared to the competitive baselines. An LLM-as-a-judge assessment further shows that our rationales surpass post-hoc explanations from the baseline by 20% in clinical reasoning quality. The source code is available at https://github.com/HyeongWon-Jang/TRIAGE .

Clinical notes contain information about patients that goes beyond structured data like lab values and medications. However, clinical notes have been underused relative to structured data, because notes are high-dimensional and sparse. This work develops and evaluates representations of clinical notes using bidirectional transformers (ClinicalBERT). ClinicalBERT uncovers high-quality relationships between medical concepts as judged by humans. ClinicalBert outperforms baselines on 30-day hospital readmission prediction using both discharge summaries and the first few days of notes in the intensive care unit. Code and model parameters are available.

Skin carcinoma is the most prevalent form of cancer globally, accounting for over $8 billion in annual healthcare expenditures. Early diagnosis, accurate and timely treatment are critical to improving patient survival rates. In clinical settings, physicians document patient visits using detailed SOAP (Subjective, Objective, Assessment, and Plan) notes. However, manually generating these notes is labor-intensive and contributes to clinician burnout. In this work, we propose skin-SOAP, a weakly supervised multimodal framework to generate clinically structured SOAP notes from limited inputs, including lesion images and sparse clinical text. Our approach reduces reliance on manual annotations, enabling scalable, clinically grounded documentation while alleviating clinician burden and reducing the need for large annotated data. Our method achieves performance comparable to GPT-4o, Claude, and DeepSeek Janus Pro across key clinical relevance metrics. To evaluate this clinical relevance, we introduce two novel metrics MedConceptEval and Clinical Coherence Score (CCS) which assess semantic alignment with expert medical concepts and input features, respectively.

There is enormous enthusiasm and concerns in using large language models (LLMs) in healthcare, yet current assumptions are all based on general-purpose LLMs such as ChatGPT. This study develops a clinical generative LLM, GatorTronGPT, using 277 billion words of mixed clinical and English text with a GPT-3 architecture of 20 billion parameters. GatorTronGPT improves biomedical natural language processing for medical research. Synthetic NLP models trained using GatorTronGPT generated text outperform NLP models trained using real-world clinical text. Physicians Turing test using 1 (worst) to 9 (best) scale shows that there is no significant difference in linguistic readability (p = 0.22; 6.57 of GatorTronGPT compared with 6.93 of human) and clinical relevance (p = 0.91; 7.0 of GatorTronGPT compared with 6.97 of human) and that physicians cannot differentiate them (p < 0.001). This study provides insights on the opportunities and challenges of LLMs for medical research and healthcare.

Clinical diagnosis is critical in medical practice, typically requiring a continuous and evolving process that includes primary diagnosis, differential diagnosis, and final diagnosis. However, most existing clinical diagnostic tasks are single-step processes, which does not align with the complex multi-step diagnostic procedures found in real-world clinical settings. In this paper, we propose a multi-step diagnostic task and annotate a clinical diagnostic dataset (MSDiagnosis). This dataset includes primary diagnosis, differential diagnosis, and final diagnosis questions. Additionally, we propose a novel and effective framework. This framework combines forward inference, backward inference, reflection, and refinement, enabling the LLM to self-evaluate and adjust its diagnostic results. To assess the effectiveness of our proposed method, we design and conduct extensive experiments. The experimental results demonstrate the effectiveness of the proposed method. We also provide a comprehensive experimental analysis and suggest future research directions for this task.

Selective Classification, wherein models can reject low-confidence predictions, promises reliable translation of machine-learning based classification systems to real-world scenarios such as clinical diagnostics. While current evaluation of these systems typically assumes fixed working points based on pre-defined rejection thresholds, methodological progress requires benchmarking the general performance of systems akin to the AUROC in standard classification. In this work, we define 5 requirements for multi-threshold metrics in selective classification regarding task alignment, interpretability, and flexibility, and show how current approaches fail to meet them. We propose the Area under the Generalized Risk Coverage curve (AUGRC), which meets all requirements and can be directly interpreted as the average risk of undetected failures. We empirically demonstrate the relevance of AUGRC on a comprehensive benchmark spanning 6 data sets and 13 confidence scoring functions. We find that the proposed metric substantially changes metric rankings on 5 out of the 6 data sets.

We investigate whether large language models produce different medical triage recommendations for identical neurological symptoms when only the patient's stated gender and age vary. Using three model families--Gemini 3.5 Flash, Claude Sonnet 4.6, and GPT-5.4-mini--we present a standardized symptom profile (persistent headache, blurred vision, morning nausea, visual disturbances) across seven demographic conditions: three age groups (25, 38, 65) x two genders (male, female), plus a gender-unspecified baseline (n = 30 per condition per model, 630 total trials). We find a stark, systemic gender-dependent triage disparity: young women receive significantly lower emergency room (ER) referral rates than age-matched men (Gemini: 0% vs. 23.3%; Claude: 6.7% vs. 96.7%; GPT: 6.7% vs. 66.7%, all p < 0.001). The disparity disappears at age 65 for all models. The primary mechanism is diagnostic substitution: the models anchor on a gender-associated diagnosis, preferentially classifying young women with Idiopathic Intracranial Hypertension (IIH)--a condition epidemiologically linked to women of childbearing age--while diagnosing men with generic increased intracranial pressure with space-occupying lesions in the differential. This diagnostic closure routes female patients to lower-urgency care (outpatient doctor appointments) despite comparable severity ratings (7-9/10). Our findings demonstrate that clinical LLMs replicate documented human clinical biases by using epidemiological priors to suppress triage urgency, suggesting that AI triage engines must decouple urgency assessment from probabilistic diagnostic priors. We release all code, prompts, and raw results.

Clinical evidence, derived from rigorous research and data analysis, provides healthcare professionals with reliable scientific foundations for informed decision-making. Integrating clinical evidence into real-time practice is challenging due to the enormous workload, complex professional processes, and time constraints. This highlights the need for tools that automate evidence synthesis to support more efficient and accurate decision making in clinical settings. This study introduces Quicker, an evidence-based clinical decision support system powered by large language models (LLMs), designed to automate evidence synthesis and generate clinical recommendations modeled after standard clinical guideline development processes. Quicker implements a fully automated chain that covers all phases, from questions to clinical recommendations, and further enables customized decision-making through integrated tools and interactive user interfaces. To evaluate Quicker's capabilities, we developed the Q2CRBench-3 benchmark dataset, based on clinical guideline development records for three different diseases. Experimental results highlighted Quicker's strong performance, with fine-grained question decomposition tailored to user preferences, retrieval sensitivities comparable to human experts, and literature screening performance approaching comprehensive inclusion of relevant studies. In addition, Quicker-assisted evidence assessment effectively supported human reviewers, while Quicker's recommendations were more comprehensive and logically coherent than those of clinicians. In system-level testing, collaboration between a single reviewer and Quicker reduced the time required for recommendation development to 20-40 minutes. In general, our findings affirm the potential of Quicker to help physicians make quicker and more reliable evidence-based clinical decisions.

Large Language Models (LLMs) are increasingly being explored for clinical question answering and decision support, yet safe deployment critically requires reliable handling of patient measurements in heterogeneous clinical notes. Existing evaluations of LLMs for clinical numerical reasoning provide limited operation-level coverage, restricted primarily to arithmetic computation, and rarely assess the robustness of numerical understanding across clinical note formats. We introduce ClinicNumRobBench, a benchmark of 1,624 context-question instances with ground-truth answers that evaluates four main types of clinical numeracy: value retrieval, arithmetic computation, relational comparison, and aggregation. To stress-test robustness, ClinicNumRobBench presents longitudinal MIMIC-IV vital-sign records in three semantically equivalent representations, including a real-world note-style variant derived from the Open Patients dataset, and instantiates queries using 42 question templates. Experiments on 14 LLMs show that value retrieval is generally strong, with most models exceeding 85% accuracy, while relational comparison and aggregation remain challenging, with some models scoring below 15%. Fine-tuning on medical data can reduce numeracy relative to base models by over 30%, and performance drops under note-style variation indicate LLM sensitivity to format. ClinicNumRobBench offers a rigorous testbed for clinically reliable numerical reasoning. Code and data URL are available on https://github.com/MinhVuong2000/ClinicNumRobBench.

Machine learning-based Type 2 diabetes risk prediction models obtain good internal validation results but lose effectiveness in real-world applications due to deficient external testing and fairness assessment. We developed a multi-dimensional framework evaluating discrimination, calibration, interpretability, and algorithmic fairness on nationally representative populations. An XGBoost model was trained on NHANES 2015-2020 (n=15,685) using eight non-laboratory predictors: age, sex, race/ethnicity, BMI, smoking status, physical activity, history of heart attack, and history of stroke. External validation was performed on BRFSS 2020-2022 (n=1,285,783) under realistic distribution shift. Internal validation showed good discrimination (AUC=0.794, 95% CI 0.788-0.800), with performance loss on external validation (AUC=0.717, relative decrease: -9.7%, p<0.001). Fairness analysis revealed severe bias: elderly adults (>=60) showed AUC=0.607 vs 0.742 for young adults (difference=0.135, p<0.001); obese individuals showed AUC=0.698 vs 0.735 for normal weight (difference=0.037, p<0.001). Gender showed comparable performance (male=0.723 vs female=0.712, p=0.142). Calibration revealed risk overestimation (Brier score=0.123). SHAP analysis identified age, BMI, and physical activity as primary risk drivers. Populations with highest diabetes risk receive the worst algorithmic performance, underscoring the need for fairness-aware, age-stratified deployment strategies before clinical use.

In response to the pressing need for advanced clinical problem-solving tools in healthcare, we introduce BooksMed, a novel framework based on a Large Language Model (LLM). BooksMed uniquely emulates human cognitive processes to deliver evidence-based and reliable responses, utilizing the GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) framework to effectively quantify evidence strength. For clinical decision-making to be appropriately assessed, an evaluation metric that is clinically aligned and validated is required. As a solution, we present ExpertMedQA, a multispecialty clinical benchmark comprised of open-ended, expert-level clinical questions, and validated by a diverse group of medical professionals. By demanding an in-depth understanding and critical appraisal of up-to-date clinical literature, ExpertMedQA rigorously evaluates LLM performance. BooksMed outperforms existing state-of-the-art models Med-PaLM 2, Almanac, and ChatGPT in a variety of medical scenarios. Therefore, a framework that mimics human cognitive stages could be a useful tool for providing reliable and evidence-based responses to clinical inquiries.

Alarm fatigue in intensive care units (ICUs) is a well documented patient safety crisis. Clinical monitors generate 350 or more alarms per patient per day, out of which 72-99% are clinically irrelevant. Staff desensitization to non-actionable alarms increases the risk of missed true emergencies. This paper presents SigmaMedStat, a machine learning system that evaluates the trustworthiness of physiological alarm signals before clinical action is taken. Four approaches were evaluated on the PhysioNet/Computing in Cardiology Challenge 2015 dataset of 498 four-channel ICU alarm recordings. Primary contribution is a temporal modeling framework that splits each 60 second recording into six consecutive 10-second chunks, and this in turn generates Continuous Wavelet Transform (CWT) scalograms per chunk, encodes each chunk with a shared EfficientNet-B0 encoder, and passes the resulting feature sequence to a two-layer Long Short-Term Memory (LSTM) network. Five-fold stratified cross-validation yields a mean AUC of 0.822 +/- 0.016 (95% CI: [0.790,0.853]), compared to 0.641 for a static EfficientNet baseline trained on the full 60-second window. Ablation studies confirm that temporal chunking and multi-channel signal fusion both contribute independently to classification performance. Per-alarm type analysis reveals that Ventricular Flutter is the most accurately classified alarm type (AUC 0.820) while Asystole remains the hardest (AUC 0.722). Error analysis identifies 65 false negatives and 85 high-confidence misclassifications as the primary failure modes. All code and results are publicly available at https://github.com/Arun-K-Ram/sigmamedstat.

In mobile telecommunications, alarms act as early warning signals. They are triggered when a cell, the basic unit of radio coverage, shuts down or behaves abnormally. This signals a degradation in service quality, which directly affects the customer experience. To fix the issue, operators rely on preset thresholds to decide when an engineer should be sent out. In practice, these thresholds are set manually and remain fixed regardless of the time of day, traffic levels, or overall network conditions. This often leads to serious faults slipping through during busy hours, while minor issues can cause unnecessary callouts when the network is quiet. This paper presents a machine learning framework that automatically predicts four alarm thresholds, audit window duration, inactive time limit, total fluctuation count, and per hour fluctuation limit, from live network behavior. Since no ground truth labels exist for thresholds, we introduce a percentile guided label derivation strategy and evaluate four models on an anonymized dataset of 10,648 cells across three vendors and nine regions from a real 4G network, comprising a Gradient Boosted Trees baseline, a CNN-BiLSTM with attention, the proposed PCTN, and an iTransformer. PCTN performs the best overall with respect to three of the four targets, outperforming a state-of-the-art iTransformer while using 83 percent fewer parameters. Its mixed output heads and dynamic alpha mechanism produce thresholds that are both accurate and interpretable, allowing operators to inspect and adjust the learned policy without retraining. All comparisons are statistically significant at p < 0.001. The framework undergoes daily retraining using new data, which enables the thresholds to constantly adjust to changes in the network.

Checklists are simple decision aids that are often used to promote safety and reliability in clinical applications. In this paper, we present a method to learn checklists for clinical decision support. We represent predictive checklists as discrete linear classifiers with binary features and unit weights. We then learn globally optimal predictive checklists from data by solving an integer programming problem. Our method allows users to customize checklists to obey complex constraints, including constraints to enforce group fairness and to binarize real-valued features at training time. In addition, it pairs models with an optimality gap that can inform model development and determine the feasibility of learning sufficiently accurate checklists on a given dataset. We pair our method with specialized techniques that speed up its ability to train a predictive checklist that performs well and has a small optimality gap. We benchmark the performance of our method on seven clinical classification problems, and demonstrate its practical benefits by training a short-form checklist for PTSD screening. Our results show that our method can fit simple predictive checklists that perform well and that can easily be customized to obey a rich class of custom constraints.

Rationale: Computer aided detection (CAD) algorithms for Pulmonary Embolism (PE) algorithms have been shown to increase radiologists' sensitivity with a small increase in specificity. However, CAD for PE has not been adopted into clinical practice, likely because of the high number of false positives current CAD software produces. Objective: To generate a database of annotated computed tomography pulmonary angiographies, use it to compare the sensitivity and false positive rate of current algorithms and to develop new methods that improve such metrics. Methods: 91 Computed tomography pulmonary angiography scans were annotated by at least one radiologist by segmenting all pulmonary emboli visible on the study. 20 annotated CTPAs were open to the public in the form of a medical image analysis challenge. 20 more were kept for evaluation purposes. 51 were made available post-challenge. 8 submissions, 6 of them novel, were evaluated on the 20 evaluation CTPAs. Performance was measured as per embolus sensitivity vs. false positives per scan curve. Results: The best algorithms achieved a per-embolus sensitivity of 75% at 2 false positives per scan (fps) or of 70% at 1 fps, outperforming the state of the art. Deep learning approaches outperformed traditional machine learning ones, and their performance improved with the number of training cases. Significance: Through this work and challenge we have improved the state-of-the art of computer aided detection algorithms for pulmonary embolism. An open database and an evaluation benchmark for such algorithms have been generated, easing the development of further improvements. Implications on clinical practice will need further research.

Skin carcinoma is the most prevalent form of cancer globally, accounting for over $8 billion in annual healthcare expenditures. In clinical settings, physicians document patient visits using detailed SOAP (Subjective, Objective, Assessment, and Plan) notes. However, manually generating these notes is labor-intensive and contributes to clinician burnout. In this work, we propose a weakly supervised multimodal framework to generate clinically structured SOAP notes from limited inputs, including lesion images and sparse clinical text. Our approach reduces reliance on manual annotations, enabling scalable, clinically grounded documentation while alleviating clinician burden and reducing the need for large annotated data. Our method achieves performance comparable to GPT-4o, Claude, and DeepSeek Janus Pro across key clinical relevance metrics. To evaluate clinical quality, we introduce two novel metrics MedConceptEval and Clinical Coherence Score (CCS) which assess semantic alignment with expert medical concepts and input features, respectively.

Accurate interpretation of lab results is crucial in clinical medicine, yet most patient portals use universal normal ranges, ignoring conditional factors like age and gender. This study introduces Lab-AI, an interactive system that offers personalized normal ranges using retrieval-augmented generation (RAG) from credible health sources. Lab-AI has two modules: factor retrieval and normal range retrieval. We tested these on 122 lab tests: 40 with conditional factors and 82 without. For tests with factors, normal ranges depend on patient-specific information. Our results show GPT-4-turbo with RAG achieved a 0.948 F1 score for factor retrieval and 0.995 accuracy for normal range retrieval. GPT-4-turbo with RAG outperformed the best non-RAG system by 33.5% in factor retrieval and showed 132% and 100% improvements in question-level and lab-level performance, respectively, for normal range retrieval. These findings highlight Lab-AI's potential to enhance patient understanding of lab results.

Specialized clinical AI assistants are rapidly entering medical practice, often framed as safer or more reliable than general-purpose large language models (LLMs). Yet, unlike frontier models, these clinical tools are rarely subjected to independent, quantitative evaluation, creating a critical evidence gap despite their growing influence on diagnosis, triage, and guideline interpretation. We assessed two widely deployed clinical AI systems (OpenEvidence and UpToDate Expert AI) against three state-of-the-art generalist LLMs (GPT-5, Gemini 3 Pro, and Claude Sonnet 4.5) using a 1,000-item mini-benchmark combining MedQA (medical knowledge) and HealthBench (clinician-alignment) tasks. Generalist models consistently outperformed clinical tools, with GPT-5 achieving the highest scores, while OpenEvidence and UpToDate demonstrated deficits in completeness, communication quality, context awareness, and systems-based safety reasoning. These findings reveal that tools marketed for clinical decision support may often lag behind frontier LLMs, underscoring the urgent need for transparent, independent evaluation before deployment in patient-facing workflows.

Background: Agent skills are increasingly deployed as modular, reusable capability units in AI agent systems. Medical research agent skills require safeguards beyond general-purpose evaluation, including scientific integrity, methodological validity, reproducibility, and boundary safety. This study developed and preliminarily evaluated a domain-specific audit framework for medical research agent skills, with a focus on reliability against expert review. Methods: We developed MedSkillAudit (skill-auditor@1.0), a layered framework assessing skill release readiness before deployment. We evaluated 75 skills across five medical research categories (15 per category). Two experts independently assigned a quality score (0-100), an ordinal release disposition (Production Ready / Limited Release / Beta Only / Reject), and a high-risk failure flag. System-expert agreement was quantified using ICC(2,1) and linearly weighted Cohen's kappa, benchmarked against the human inter-rater baseline. Results: The mean consensus quality score was 72.4 (SD = 13.0); 57.3% of skills fell below the Limited Release threshold. MedSkillAudit achieved ICC(2,1) = 0.449 (95% CI: 0.250-0.610), exceeding the human inter-rater ICC of 0.300. System-consensus score divergence (SD = 9.5) was smaller than inter-expert divergence (SD = 12.4), with no directional bias (Wilcoxon p = 0.613). Protocol Design showed the strongest category-level agreement (ICC = 0.551); Academic Writing showed a negative ICC (-0.567), reflecting a structural rubric-expert mismatch. Conclusions: Domain-specific pre-deployment audit may provide a practical foundation for governing medical research agent skills, complementing general-purpose quality checks with structured audit workflows tailored to scientific use cases.

Objective: Using natural language processing (NLP) to find sentences that state treatment plans in a clinical note, would automate plan extraction and would further enable their use in tools that help providers and care managers. However, as in the most NLP tasks on clinical text, creating gold standard to train and test NLP models is tedious and expensive. Fortuitously, sometimes but not always clinical notes contain sections with a heading that identifies the section as a plan. Leveraging contents of such labeled sections as a noisy training data, we assessed accuracy of NLP models trained with the data. Methods: We used common variations of plan headings and rule-based heuristics to find plan sections with headings in clinical notes, and we extracted sentences from them and formed a noisy training data of plan sentences. We trained Support Vector Machine (SVM) and Convolutional Neural Network (CNN) models with the data. We measured accuracy of the trained models on the noisy dataset using ten-fold cross validation and separately on a set-aside manually annotated dataset. Results: About 13% of 117,730 clinical notes contained treatment plans sections with recognizable headings in the 1001 longitudinal patient records that were obtained from Cleveland Clinic under an IRB approval. We were able to extract and create a noisy training data of 13,492 plan sentences from the clinical notes. CNN achieved best F measures, 0.91 and 0.97 in the cross-validation and set-aside evaluation experiments respectively. SVM slightly underperformed with F measures of 0.89 and 0.96 in the same experiments. Conclusion: Our study showed that the training supervised learning models using noisy plan sentences was effective in identifying them in all clinical notes. More broadly, sections with informal headings in clinical notes can be a good source for generating effective training data.

Global healthcare providers are exploring use of large language models (LLMs) to provide medical advice to the public. LLMs now achieve nearly perfect scores on medical licensing exams, but this does not necessarily translate to accurate performance in real-world settings. We tested if LLMs can assist members of the public in identifying underlying conditions and choosing a course of action (disposition) in ten medical scenarios in a controlled study with 1,298 participants. Participants were randomly assigned to receive assistance from an LLM (GPT-4o, Llama 3, Command R+) or a source of their choice (control). Tested alone, LLMs complete the scenarios accurately, correctly identifying conditions in 94.9% of cases and disposition in 56.3% on average. However, participants using the same LLMs identified relevant conditions in less than 34.5% of cases and disposition in less than 44.2%, both no better than the control group. We identify user interactions as a challenge to the deployment of LLMs for medical advice. Standard benchmarks for medical knowledge and simulated patient interactions do not predict the failures we find with human participants. Moving forward, we recommend systematic human user testing to evaluate interactive capabilities prior to public deployments in healthcare.

Purpose. Patients with advanced cancer may undergo multiple lines of treatment, switching therapies as their disease progresses. Motivated by a study of metastatic prostate cancer, we develop a microsimulation framework to study therapy sequence. Methods. We propose a discrete-time state transition model to study two lines of anti-cancer therapy. Based on digitized published progression-free survival (PFS) and overall survival (OS) curves, we infer event types (progression or death), and estimate transition probabilities using cumulative incidence functions with competing risks. Our model incorporates within-patient dependence over time, such that response to first-line therapy informs subsequent event probabilities. Parameters governing the degree of within-patient dependence can be used to calibrate the model-based results to those of a target trial. We demonstrate these methods in a study of two therapy sequences for metastatic prostate cancer, where Docetaxel (DCT) and Abiraterone Acetate (AA) are both appropriate for use in either first or second line treatment. We assess costs, Quality-Adjusted Life Years (QALYs) and Incremental Cost Effectiveness Ratio (ICER) for two treatment strategies: DCT then AA vs AA then DCT. Results. Using digitized survival curves from relevant clinical trials, we identified 8.6-13.9% of PFS times that should be categorized as deaths, allowing for estimation of cumulative incidence functions. Models assuming within-patient independence overestimated OS time, corrected with our calibration approach. Correction resulted in meaningful changes in the difference in QALYs between treatment strategies (0.07 vs 0.15) and the ICER (-\76,836/QALY vs -21,030/QALY). Conclusions. Microsimulation models can be successfully used to study cost-effectiveness of therapy sequences, taking care to account correctly for within-patient dependence.

LLM confidence signals are used for abstention, routing, and safety-critical decisions. No standard practice exists for checking whether a confidence signal carries item-level information before building on it. We transfer the validity screening principle from clinical personality assessment (PAI, MMPI-3) as a portable protocol for benchmark-based LLM confidence data. The protocol specifies three core indices (L, Fp, RBS), a structural indicator (TRIN), and an item-sensitivity statistic, computed from a single 2x2 contingency table. A three-tier classification system (Invalid, Indeterminate, Valid) draws on four clinical traditions. Validated on 20 frontier LLMs across 524 items, four models are classified Invalid, two Indeterminate. Valid-profile models show mean r = .18 (15/16 significant). Invalid-profile models show mean r = -.20 (d = 2.48). Cross-benchmark validation on 18 models using MMLU with verbalized confidence and on external data from Yang et al. (2024) confirms the screen transfers across benchmarks and probe formats. All data and code: https://github.com/synthiumjp/validity-scaling-llm

Large language models (LLMs) are increasingly proposed as clinical agents, yet static, single-turn benchmarks cannot capture how a model dynamically delivers care across an encounter: gathering information, planning treatment, and adapting longitudinal management across successive patient states. Medical education has long addressed an analogous challenge through standardized patients (SPs): trained actors who consistently portray clinical cases, enabling realistic practice and objective, scripted assessment. Here we introduce MedSP1000, an SP-derived interactive benchmark for clinical-agent evaluation, including 1,638 SP cases with 24,602 trajectory-level peer-reviewed rubrics. MedSP1000 converts peer-reviewed SP teaching cases into executable scenarios with defined SP case scripts, clinical environment contexts, and human-validated structured rubric. In each simulation evaluation run, a clinical agent interacts in closed loop with a patient agent and an environment controller, and its behaviour is scored throughout the encounter against expert criteria specified in the original materials. Applying MedSP1000 to a range of general-purpose and medically specialized LLMs, we find that performance on static benchmarks does not reliably translate to such educational scenarios. The best-performing model, GPT-5.5, completes only 60.4% of expert-defined rubric items, whereas the strongest medically specialized model reaches 40.0%; increasing test-time compute produces no measurable gain. These results suggest that current LLMs, including agentic systems tuned for medicine, are not yet reliable enough to be safely integrated into actual clinical practice. More broadly, MedSP1000 shows how process-level, SP-style evaluation can reveal clinically relevant failure modes that single-turn benchmarks miss.

Multimodal large language models (MLLMs) have advanced clinical tasks for common conditions, but their performance on rare diseases remains largely untested. In rare-disease scenarios, clinicians often lack prior clinical knowledge, forcing them to rely strictly on case-level evidence for clinical judgments. Existing benchmarks predominantly evaluate common-condition, single-image settings, leaving multimodal and multi-image evidence integration under rare-disease data scarcity systematically unevaluated. We introduce MMRareBench, to our knowledge the first rare-disease benchmark jointly evaluating multimodal and multi-image clinical capability across four workflow-aligned tracks: diagnosis, treatment planning, cross-image evidence alignment, and examination suggestion. The benchmark comprises 1,756 question-answer pairs with 7,958 associated medical images curated from PMC case reports, with Orphanet-anchored ontology alignment, track-specific leakage control, evidence-grounded annotations, and a two-level evaluation protocol. A systematic evaluation of 23 MLLMs reveals fragmented capability profiles and universally low treatment-planning performance, with medical-domain models trailing general-purpose MLLMs substantially on multi-image tracks despite competitive diagnostic scores. These patterns are consistent with a capacity dilution effect: medical fine-tuning can narrow the diagnostic gap but may erode the compositional multi-image capability that rare-disease evidence integration demands.

As opposed to evaluating computation and logic-based reasoning, current benchmarks for evaluating large language models (LLMs) in medicine are primarily focused on question-answering involving domain knowledge and descriptive reasoning. While such qualitative capabilities are vital to medical diagnosis, in real-world scenarios, doctors frequently use clinical calculators that follow quantitative equations and rule-based reasoning paradigms for evidence-based decision support. To this end, we propose MedCalc-Bench, a first-of-its-kind dataset focused on evaluating the medical calculation capability of LLMs. MedCalc-Bench contains an evaluation set of over 1000 manually reviewed instances from 55 different medical calculation tasks. Each instance in MedCalc-Bench consists of a patient note, a question requesting to compute a specific medical value, a ground truth answer, and a step-by-step explanation showing how the answer is obtained. While our evaluation results show the potential of LLMs in this area, none of them are effective enough for clinical settings. Common issues include extracting the incorrect entities, not using the correct equation or rules for a calculation task, or incorrectly performing the arithmetic for the computation. We hope our study highlights the quantitative knowledge and reasoning gaps in LLMs within medical settings, encouraging future improvements of LLMs for various clinical calculation tasks.

We develop and evaluate a data-driven approach for detecting unusual (anomalous) patient-management decisions using past patient cases stored in electronic health records (EHRs). Our hypothesis is that a patient-management decision that is unusual with respect to past patient care may be due to an error and that it is worthwhile to generate an alert if such a decision is encountered. We evaluate this hypothesis using data obtained from EHRs of 4486 post-cardiac surgical patients and a subset of 222 alerts generated from the data. We base the evaluation on the opinions of a panel of experts. The results of the study support our hypothesis that the outlier-based alerting can lead to promising true alert rates. We observed true alert rates that ranged from 25\% to 66\% for a variety of patient-management actions, with 66\% corresponding to the strongest outliers.

Clinical time series are central to patient monitoring, risk assessment, and clinical decision support. However, they are often sparse, irregularly sampled, and asynchronous, making it difficult for models to identify the temporal evidence required for clinical Question Answering (QA). Existing benchmarks primarily focus on regularly sampled time-series QA or medical QA over static data, and therefore rarely assess whether models can faithfully ground their answers in irregular temporal observations. To fill this gap, we introduce CLIR-Bench, a benchmark for irregular clinical time series QA constructed from de-identified ICU records through a principled four-stage pipeline. CLIR-Bench contains 6,600 QA instances spanning 11 clinical variables, organized into four capability dimensions and 11 tasks. Each question is linked to explicit temporal evidence and task-specific answer derivation rules, enabling evaluation of both answer accuracy and evidence use. Experiments show that existing generalist models struggle to retrieve and reason over sparse clinical evidence, highlighting the need for stronger irregular time-series reasoning methods. Our code and data are available at https://huggingface.co/datasets/winall/CLIR-Bench.

Large-language models have recently demonstrated impressive zero-shot capabilities in a variety of natural language tasks such as summarization, dialogue generation, and question-answering. Despite many promising applications in clinical medicine, adoption of these models in real-world settings has been largely limited by their tendency to generate incorrect and sometimes even toxic statements. In this study, we develop Almanac, a large language model framework augmented with retrieval capabilities for medical guideline and treatment recommendations. Performance on a novel dataset of clinical scenarios (n = 130) evaluated by a panel of 5 board-certified and resident physicians demonstrates significant increases in factuality (mean of 18% at p-value < 0.05) across all specialties, with improvements in completeness and safety. Our results demonstrate the potential for large language models to be effective tools in the clinical decision-making process, while also emphasizing the importance of careful testing and deployment to mitigate their shortcomings.

Recruiting patients to participate in clinical trials can be challenging and time-consuming. Usually, participation in a clinical trial is initiated by a healthcare professional and proposed to the patient. Promoting clinical trials directly to patients via online recruitment might help to reach them more efficiently. In this study, we address the case where a patient is initiating their own recruitment process and wants to determine whether they are eligible for a given clinical trial, using their own language to describe their medical profile. To study whether this creates difficulties in the patient trial matching process, we design a new dataset and task, Natural Language Inference for Patient Recruitment (NLI4PR), in which patient language profiles must be matched to clinical trials. We create it by adapting the TREC 2022 Clinical Trial Track dataset, which provides patients' medical profiles, and rephrasing them manually using patient language. We also use the associated clinical trial reports where the patients are either eligible or excluded. We prompt several open-source Large Language Models on our task and achieve from 56.5 to 71.8 of F1 score using patient language, against 64.7 to 73.1 for the same task using medical language. When using patient language, we observe only a small loss in performance for the best model, suggesting that having the patient as a starting point could be adopted to help recruit patients for clinical trials. The corpus and code bases are all freely available on our Github and HuggingFace repositories.

Recent advances in large language models have led to renewed interest in natural language processing in healthcare using the free text of clinical notes. One distinguishing characteristic of clinical notes is their long time span over multiple long documents. The unique structure of clinical notes creates a new design choice: when the context length for a language model predictor is limited, which part of clinical notes should we choose as the input? Existing studies either choose the inputs with domain knowledge or simply truncate them. We propose a framework to analyze the sections with high predictive power. Using MIMIC-III, we show that: 1) predictive power distribution is different between nursing notes and discharge notes and 2) combining different types of notes could improve performance when the context length is large. Our findings suggest that a carefully selected sampling function could enable more efficient information extraction from clinical notes.

Large language models (LLMs) are increasingly used for medical and health-related questions, yet their safety in high-risk medical scenarios remains poorly understood. We introduce MedHarmCode and data will be released upon acceptance. Due to the sensitive nature of high-risk medical queries, data access will be available to qualified researchers upon request., a high-risk medical safety benchmark with 1,100 medically grounded queries across 10 safety-critical categories, including toxicology, pharmacology, covert poisoning, anesthesia, and fetal harm. Unlike broad medical QA benchmarks, MedHarm targets realistic clinical, educational, and technical prompts that require refusal, caution, or safe redirection rather than direct helpfulness. We evaluate 15 LLMs spanning general-purpose, medical-purpose, closed-source, and downstream SFT models, together with 4 representative guardrail models. Results reveal a substantial gap between apparent alignment and medical safety: aligned models can still produce unsafe or actionable responses, medical fine-tuning can amplify harmful specificity, and external guardrails reduce some failures while introducing brittle blocking and weak safe helpfulness. These findings show that medical safety cannot be inferred from general alignment or medical capability alone, highlighting the need for domain-specific stress testing before deploying LLMs in safety-critical medical applications.

Several studies have shown that speech and language features, automatically extracted from clinical interviews or spontaneous discourse, have diagnostic value for mental disorders such as schizophrenia and bipolar disorder. They typically make use of a large feature set to train a classifier for distinguishing between two groups of interest, i.e. a clinical and control group. However, a purely data-driven approach runs the risk of overfitting to a particular data set, especially when sample sizes are limited. Here, we first down-select the set of language features to a small subset that is related to a well-validated test of functional ability, the Social Skills Performance Assessment (SSPA). This helps establish the concurrent validity of the selected features. We use only these features to train a simple classifier to distinguish between groups of interest. Linear regression reveals that a subset of language features can effectively model the SSPA, with a correlation coefficient of 0.75. Furthermore, the same feature set can be used to build a strong binary classifier to distinguish between healthy controls and a clinical group (AUC = 0.96) and also between patients within the clinical group with schizophrenia and bipolar I disorder (AUC = 0.83).

Although large language models (LLMs) have been assessed for general medical knowledge using medical licensing exams, their ability to effectively support clinical decision-making tasks, such as selecting and using medical calculators, remains uncertain. Here, we evaluate the capability of both medical trainees and LLMs to recommend medical calculators in response to various multiple-choice clinical scenarios such as risk stratification, prognosis, and disease diagnosis. We assessed eight LLMs, including open-source, proprietary, and domain-specific models, with 1,009 question-answer pairs across 35 clinical calculators and measured human performance on a subset of 100 questions. While the highest-performing LLM, GPT-4o, provided an answer accuracy of 74.3% (CI: 71.5-76.9%), human annotators, on average, outperformed LLMs with an accuracy of 79.5% (CI: 73.5-85.0%). With error analysis showing that the highest-performing LLMs continue to make mistakes in comprehension (56.6%) and calculator knowledge (8.1%), our findings emphasize that humans continue to surpass LLMs on complex clinical tasks such as calculator recommendation.

As large language models (LLMs) become primary sources of health information for millions, their accuracy in women's health remains critically unexamined. We introduce the Women's Health Benchmark (WHB), the first benchmark evaluating LLM performance specifically in women's health. Our benchmark comprises 96 rigorously validated model stumps covering five medical specialties (obstetrics and gynecology, emergency medicine, primary care, oncology, and neurology), three query types (patient query, clinician query, and evidence/policy query), and eight error types (dosage/medication errors, missing critical information, outdated guidelines/treatment recommendations, incorrect treatment advice, incorrect factual information, missing/incorrect differential diagnosis, missed urgency, and inappropriate recommendations). We evaluated 13 state-of-the-art LLMs and revealed alarming gaps: current models show approximately 60\% failure rates on the women's health benchmark, with performance varying dramatically across specialties and error types. Notably, models universally struggle with "missed urgency" indicators, while newer models like GPT-5 show significant improvements in avoiding inappropriate recommendations. Our findings underscore that AI chatbots are not yet fully able of providing reliable advice in women's health.

Existing medical video benchmarks primarily evaluate whether a model produces the correct answer, but rarely assess whether it answers at the right time. In real clinical settings, AI systems must decide not only what to predict, but also when to answer, defer judgment, or proactively raise alerts. This creates a critical gap between benchmark evaluation and deployment requirements. We present MedStreamBench, a benchmark for time-aware medical video understanding. MedStreamBench integrates 22 medical datasets and 5,419 QA instances across four temporal settings: retrospective, present, future, and proactive. Unlike conventional benchmarks that assume full-video access, MedStreamBench restricts models to temporally bounded evidence windows and supports both single-turn and streaming evaluation. We further introduce a proactive monitoring setting that requires models to determine whether and when clinically relevant alerts should be triggered. Beyond answer correctness, MedStreamBench evaluates temporal behavior through responsiveness and post-evidence stability. Experiments on leading general-purpose and medical vision-language models reveal a substantial gap between offline recognition and temporally grounded decision-making, with performance dropping markedly in streaming and proactive settings. Our benchmark is available at https://huggingface.co/datasets/Venn2024/MedStreamBench.

We present the findings of "The Alzheimer's Disease Prediction Of Longitudinal Evolution" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of biomarkers. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials.

Despite decades of clinical research, sepsis remains a global public health crisis with high mortality, and morbidity. Currently, when sepsis is detected and the underlying pathogen is identified, organ damage may have already progressed to irreversible stages. Effective sepsis management is therefore highly time-sensitive. By systematically analysing trends in the plethora of clinical data available in the intensive care unit (ICU), an early prediction of sepsis could lead to earlier pathogen identification, resistance testing, and effective antibiotic and supportive treatment, and thereby become a life-saving measure. Here, we developed and validated a machine learning (ML) system for the prediction of sepsis in the ICU. Our analysis represents the largest multi-national, multi-centre in-ICU study for sepsis prediction using ML to date. Our dataset contains 156,309 unique ICU admissions, which represent a refined and harmonised subset of five large ICU databases originating from three countries. Using the international consensus definition Sepsis-3, we derived hourly-resolved sepsis label annotations, amounting to 26,734 (17.1%) septic stays. We compared our approach, a deep self-attention model, to several clinical baselines as well as ML baselines and performed an extensive internal and external validation within and across databases. On average, our model was able to predict sepsis with an AUROC of 0.847 pm 0.050 (internal out-of sample validation) and 0.761 pm 0.052 (external validation). For a harmonised prevalence of 17%, at 80% recall our model detects septic patients with 39% precision 3.7 hours in advance.

Longitudinal clinical notes contain rich evidence of how patients evolve over time, but converting this signal into training supervision for clinical prediction remains challenging. We extend Foresight Learning to clinical prediction by converting time-ordered MIMIC-III notes into examples consisting of past patient context, a natural-language question about a possible future event, and a label resolved from later documentation. This process yields 6,900 prediction examples from 702 admissions across medications, procedures, organ support, microbiology, and mortality. A small LoRA adapter trained on these examples improves over the prompted base model, reducing expected calibration error from 0.1269 to 0.0398 and Brier score from 0.199 to 0.145, while slightly outperforming GPT-5 point estimates on held-out questions. The approach enables reusable clinical prediction supervision from longitudinal notes without hand-engineered structured features or endpoint-specific classifiers.

Respiratory sound classification is hindered by the limited size, high noise levels, and severe class imbalance of benchmark datasets like ICBHI 2017. While Transformer-based models offer powerful feature extraction capabilities, they are prone to overfitting and often converge to sharp minima in the loss landscape when trained on such constrained medical data. To address this, we introduce a framework that enhances the Audio Spectrogram Transformer (AST) using Sharpness-Aware Minimization (SAM). Instead of merely minimizing the training loss, our approach optimizes the geometry of the loss surface, guiding the model toward flatter minima that generalize better to unseen patients. We also implement a weighted sampling strategy to handle class imbalance effectively. Our method achieves a state-of-the-art score of 68.10% on the ICBHI 2017 dataset, outperforming existing CNN and hybrid baselines. More importantly, it reaches a sensitivity of 68.31%, a crucial improvement for reliable clinical screening. Further analysis using t-SNE and attention maps confirms that the model learns robust, discriminative features rather than memorizing background noise.

The increasing use of tools and solutions based on Large Language Models (LLMs) for various tasks in the medical domain has become a prominent trend. Their use in this highly critical and sensitive domain has thus raised important questions about their robustness, especially in response to variations in input, and the reliability of the generated outputs. This study addresses these questions by constructing a textual dataset based on the ICD-10-CM code descriptions, widely used in US hospitals and containing many clinical terms, and their easily reproducible rephrasing. We then benchmarked existing embedding models, either generalist or specialized in the clinical domain, in a semantic search task where the goal was to correctly match the rephrased text to the original description. Our results showed that generalist models performed better than clinical models, suggesting that existing clinical specialized models are more sensitive to small changes in input that confuse them. The highlighted problem of specialized models may be due to the fact that they have not been trained on sufficient data, and in particular on datasets that are not diverse enough to have a reliable global language understanding, which is still necessary for accurate handling of medical documents.

Despite achieving high accuracy on medical benchmarks, LLMs exhibit the Einstellung Effect in clinical diagnosis--relying on statistical shortcuts rather than patient-specific evidence, causing misdiagnosis in atypical cases. Existing benchmarks fail to detect this critical failure mode. We introduce MedEinst, a counterfactual benchmark with 5,383 paired clinical cases across 49 diseases. Each pair contains a control case and a "trap" case with altered discriminative evidence that flips the diagnosis. We measure susceptibility via Bias Trap Rate--probability of misdiagnosing traps despite correctly diagnosing controls. Extensive Evaluation of 17 LLMs shows frontier models achieve high baseline accuracy but severe bias trap rates. Thus, we propose ECR-Agent, aligning LLM reasoning with Evidence-Based Medicine standard via two components: (1) Dynamic Causal Inference (DCI) performs structured reasoning through dual-pathway perception, dynamic causal graph reasoning across three levels (association, intervention, counterfactual), and evidence audit for final diagnosis; (2) Critic-Driven Graph and Memory Evolution (CGME) iteratively refines the system by storing validated reasoning paths in an exemplar base and consolidating disease-specific knowledge into evolving illness graphs. Source code is to be released.

Current Large Language Models (LLMs) exhibit significant limitations, notably in structured, interpretable, and verifiable medical reasoning, alongside practical deployment challenges related to computational resources and data privacy. This report focused on the development of WiNGPT-3.0, the 32-billion parameter LLMs, engineered with the objective of enhancing its capacity for medical reasoning and exploring its potential for effective integration within healthcare IT infrastructures. The broader aim is to advance towards clinically applicable models. The approach involved a multi-stage training pipeline tailored for general, medical, and clinical reasoning. This pipeline incorporated supervised fine-tuning (SFT) and reinforcement learning (RL), leveraging curated Long Chain-of-Thought (CoT) datasets, auxiliary reward models, and an evidence-based diagnostic chain simulation. WiNGPT-3.0 demonstrated strong performance: specific model variants achieved scores of 66.6 on MedCalc and 87.1 on MedQA-USMLE. Furthermore, targeted training improved performance on a clinical reasoning task from a baseline score of 58.1 to 62.5. These findings suggest that reinforcement learning, even when applied with a limited dataset of only a few thousand examples, can enhance medical reasoning accuracy. Crucially, this demonstration of RL's efficacy with limited data and computation paves the way for more trustworthy and practically deployable LLMs within clinical workflows and health information infrastructures.

Psychotherapy or counseling is typically conducted through spoken conversation between a therapist and a client. Analyzing the speech characteristics of psychotherapeutic interactions can help understand the factors associated with effective psychotherapy. This paper introduces CUEMPATHY, a large-scale speech dataset collected from actual counseling sessions. The dataset consists of 156 counseling sessions involving 39 therapist-client dyads. The process of speech data collection, subjective ratings (one observer and two client ratings), and transcription are described. An automatic speech and text processing system is developed to locate the time stamps of speaker turns in each session. Examining the relationships among the three subjective ratings suggests that observer and client ratings have no significant correlation, while the client-rated measures are significantly correlated. The intensity similarity between the therapist and the client, measured by the averaged absolute difference of speaker-turn-level intensities, is associated with the psychotherapy outcomes. Recent studies on the acoustic and linguistic characteristics of the CUEMPATHY are introduced.

Clinicians spend a significant amount of time reviewing medical images and transcribing their findings regarding patient diagnosis, referral and treatment in text form. Vision-language models (VLMs), which automatically interpret images and summarize their findings as text, have enormous potential to alleviate clinical workloads and increase patient access to high-quality medical care. While foundational models have stirred considerable interest in the medical community, it is unclear whether their general capabilities translate to real-world clinical utility. In this work, we show that foundation VLMs markedly underperform compared to practicing ophthalmologists on specialist tasks crucial to the care of patients with age-related macular degeneration (AMD). To address this, we initially identified the essential capabilities required for image-based clinical decision-making, and then developed a curriculum to selectively train VLMs in these skills. The resulting model, RetinaVLM, can be instructed to write reports that significantly outperform those written by leading foundation medical VLMs in disease staging (F1 score of 0.63 vs. 0.11) and patient referral (0.67 vs. 0.39), and approaches the diagnostic performance of junior ophthalmologists (who achieve 0.77 and 0.78 on the respective tasks). Furthermore, in a reader study involving two senior ophthalmologists with up to 32 years of experience, RetinaVLM's reports were found to be similarly correct (78.6% vs. 82.1%) and complete (both 78.6%) as reports written by junior ophthalmologists with up to 10 years of experience. These results demonstrate that our curriculum-based approach provides a blueprint for specializing generalist foundation medical VLMs to handle real-world clinical tasks.

Following the global COVID-19 pandemic, the number of scientific papers studying the virus has grown massively, leading to increased interest in automated literate review. We present a clinical text mining system that improves on previous efforts in three ways. First, it can recognize over 100 different entity types including social determinants of health, anatomy, risk factors, and adverse events in addition to other commonly used clinical and biomedical entities. Second, the text processing pipeline includes assertion status detection, to distinguish between clinical facts that are present, absent, conditional, or about someone other than the patient. Third, the deep learning models used are more accurate than previously available, leveraging an integrated pipeline of state-of-the-art pretrained named entity recognition models, and improving on the previous best performing benchmarks for assertion status detection. We illustrate extracting trends and insights, e.g. most frequent disorders and symptoms, and most common vital signs and EKG findings, from the COVID-19 Open Research Dataset (CORD-19). The system is built using the Spark NLP library which natively supports scaling to use distributed clusters, leveraging GPUs, configurable and reusable NLP pipelines, healthcare specific embeddings, and the ability to train models to support new entity types or human languages with no code changes.

The objective of this paper is to understand what characteristics and features of clinical data influence physician's decision about ordering laboratory tests or prescribing medications the most. We conduct our analysis on data and decisions extracted from electronic health records of 4486 post-surgical cardiac patients. The summary statistics for 335 different lab order decisions and 407 medication decisions are reported. We show that in many cases, physician's lab-order and medication decisions can be well predicted from a small subset of all features.

Purpose: The need to streamline patient management for COVID-19 has become more pressing than ever. Chest X-rays provide a non-invasive (potentially bedside) tool to monitor the progression of the disease. In this study, we present a severity score prediction model for COVID-19 pneumonia for frontal chest X-ray images. Such a tool can gauge severity of COVID-19 lung infections (and pneumonia in general) that can be used for escalation or de-escalation of care as well as monitoring treatment efficacy, especially in the ICU. Methods: Images from a public COVID-19 database were scored retrospectively by three blinded experts in terms of the extent of lung involvement as well as the degree of opacity. A neural network model that was pre-trained on large (non-COVID-19) chest X-ray datasets is used to construct features for COVID-19 images which are predictive for our task. Results: This study finds that training a regression model on a subset of the outputs from an this pre-trained chest X-ray model predicts our geographic extent score (range 0-8) with 1.14 mean absolute error (MAE) and our lung opacity score (range 0-6) with 0.78 MAE. Conclusions: These results indicate that our model's ability to gauge severity of COVID-19 lung infections could be used for escalation or de-escalation of care as well as monitoring treatment efficacy, especially in the intensive care unit (ICU). A proper clinical trial is needed to evaluate efficacy. To enable this we make our code, labels, and data available online at https://github.com/mlmed/torchxrayvision/tree/master/scripts/covid-severity and https://github.com/ieee8023/covid-chestxray-dataset

Automated slice classification is clinically relevant since it can be incorporated into medical image segmentation workflows as a preprocessing step that would flag slices with a higher probability of containing tumors, thereby directing physicians attention to the important slices. In this work, we train a ResNet-18 network to classify axial slices of lymphoma PET/CT images (collected from two institutions) depending on whether the slice intercepted a tumor (positive slice) in the 3D image or if the slice did not (negative slice). Various instances of the network were trained on 2D axial datasets created in different ways: (i) slice-level split and (ii) patient-level split; inputs of different types were used: (i) only PET slices and (ii) concatenated PET and CT slices; and different training strategies were employed: (i) center-aware (CAW) and (ii) center-agnostic (CAG). Model performances were compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC), and various binary classification metrics. We observe and describe a performance overestimation in the case of slice-level split as compared to the patient-level split training. The model trained using patient-level split data with the network input containing only PET slices in the CAG training regime was the best performing/generalizing model on a majority of metrics. Our models were additionally more closely compared using the sensitivity metric on the positive slices from their respective test sets.

Red-team evaluations of AI models support some claims and not others, and the boundary between the two is calculable rather than merely a matter of judgment. We define the evidential ceiling of an evaluation as the largest factor by which one result can move belief under a fixed testing budget, derive it in closed form for the benchmark null result, and use it to locate that boundary exactly. We find that above a calculable harm rate, a benchmark of modest size certifies a category to a stated evidentiary standard, and a clean sheet is then the stronger of the two possible observations, outweighing a single reproduced failure. Below that rate, no passive benchmark of feasible size provides the specified evidence of safety under the fixed scoring rule and approximately independent trial structure. The crossing between the two regimes has a closed form. The bound is not specific to benchmarks: written in terms of a procedure's hypothesis conditioned elicitation rates, it covers adaptive and automated red teaming as well, and shows that discrimination between the hypotheses rather than attack success is what determines evidential worth. Auditing eight evaluation suites against the boundary, we find that current benchmarks are adequate for high-frequency harm categories and several orders of magnitude short for rare, catastrophic ones. Safety benchmarks are not uninformative. They are informative about a specific and computable set of propositions, and the discipline they need is to state which.

Specialist epilepsy expertise is scarce in resource-constrained settings, making LLM-based decision support attractive for frontline clinicians managing longitudinal treatment. Such systems must adapt to local prescribing practice and know when to defer. We study this problem in Ugandan pediatric epilepsy care, predicting anti-seizure medication regimens from longitudinal unstructured clinic notes. Standard prompting achieves non-trivial agreement with physician prescriptions, but neurologist review shows that many errors reflect distribution-miscalibrated prescribing defaults rather than failures to parse the local record. We introduce MANANA, a non-parametric prompt-learning framework that learns local prescribing guidance from a small patient-level training set. MANANA converts observed prescription errors into auditable prompt memories, instantiated in single-agent and multi-agent variants, and improves over classical ML models, direct LLM prompting, and prompt-optimization baselines across two independently collected Ugandan cohorts. We further propose Bayesian prompt averaging, which converts the learned prompt trajectory into prescription likelihoods and an uncertainty-based deferral signal. On the independently collected held-out cohort, this improves visit-level top-3 prescription accuracy by 4-8 percentage points over prompt-optimization baselines and enables selective prediction: the system can auto-handle the most confident half of cases at 95% precision, or the most confident quarter at 99% precision, while deferring lower-confidence cases for specialist review.

Clinical practice guidelines (CPGs) encode evidence-based decision logic that clinicians apply by evaluating patient variables, conditional criteria, and recommendation rules. However, existing methods often use CPGs as free-text training data or retrieval sources, underutilizing their procedural decision structure. To better exploit this structure, we introduce a guideline-derived training pipeline that transforms CPG recommendations into executable clinical decision logic and uses it to generate factual and counterfactual question-answering data. Theses data teach models both guideline-supported decisions and how decisions change under different patient conditions. Post-training a medical LLM on the generated data yields MedGuideX. Across four clinical reasoning benchmarks, MedGuideX achieves a 10.28% relative improvement in average accuracy. Physician evaluation further shows that MedGuideX better recovers clinician authored reasoning steps and produces physician-preferred rationales in faithfulness, validity, completeness, and clarity. Overall, our results show that executable decision logic from CPGs can be transformed into scalable supervision for building reliable medical LLMs.

In this study, we evaluate the performance of the Segment Anything Model (SAM) in clinical radiotherapy. Our results indicate that SAM's 'segment anything' mode can achieve clinically acceptable segmentation results in most organs-at-risk (OARs) with Dice scores higher than 0.7. SAM's 'box prompt' mode further improves the Dice scores by 0.1 to 0.5. Considering the size of the organ and the clarity of its boundary, SAM displays better performance for large organs with clear boundaries but performs worse for smaller organs with unclear boundaries. Given that SAM, a model pre-trained purely on natural images, can handle the delineation of OARs from medical images with clinically acceptable accuracy, these results highlight SAM's robust generalization capabilities with consistent accuracy in automatic segmentation for radiotherapy. In other words, SAM can achieve delineation of different OARs at different sites using a generic automatic segmentation model. SAM's generalization capabilities across different disease sites suggest that it is technically feasible to develop a generic model for automatic segmentation in radiotherapy.

With the introduction of ChatGPT, Large Language Models (LLMs) have received enormous attention in healthcare. Despite their potential benefits, researchers have underscored various ethical implications. While individual instances have drawn much attention, the debate lacks a systematic overview of practical applications currently researched and ethical issues connected to them. Against this background, this work aims to map the ethical landscape surrounding the current stage of deployment of LLMs in medicine and healthcare. Electronic databases and preprint servers were queried using a comprehensive search strategy. Studies were screened and extracted following a modified rapid review approach. Methodological quality was assessed using a hybrid approach. For 53 records, a meta-aggregative synthesis was performed. Four fields of applications emerged and testify to a vivid exploration phase. Advantages of using LLMs are attributed to their capacity in data analysis, personalized information provisioning, support in decision-making, mitigating information loss and enhancing information accessibility. However, we also identifies recurrent ethical concerns connected to fairness, bias, non-maleficence, transparency, and privacy. A distinctive concern is the tendency to produce harmful misinformation or convincingly but inaccurate content. A recurrent plea for ethical guidance and human oversight is evident. Given the variety of use cases, it is suggested that the ethical guidance debate be reframed to focus on defining what constitutes acceptable human oversight across the spectrum of applications. This involves considering diverse settings, varying potentials for harm, and different acceptable thresholds for performance and certainty in healthcare. In addition, a critical inquiry is necessary to determine the extent to which the current experimental use of LLMs is necessary and justified.

The growing ability of large language models and vision language models to jointly interpret and reason over images and text is reshaping medical agents, moving them from task specific predictors toward autonomous systems that perceive, reason, plan, remember, and act in clinical environments. This work departs from the capability first perspective of existing literature and instead begins from clinical deployment, asking what tasks, contamination resistant benchmarks, and interactive training environments are required before medical agents can be trusted in practice. Medical agents are formalized as sequential decision making systems under partial observability, together with a three level autonomy taxonomy spanning assisted, cooperative, and fully autonomous operation. The field is organized along a unified scaling spine consisting of framework scaling, capability scaling, and environment scaling. Within this framework, clinical environment scaling, the integration of tools, data, and clinical gyms, is identified as the most actionable yet underexplored direction for agents operating in PACS, EHR, and FHIR ecosystems. Clinical self evolution, where agents improve through interaction with their environments rather than parameter scaling alone, is further positioned as a key research frontier, drawing insights from self improving agents, agent gyms, and test time compute scaling. Applications across radiology, pathology, ophthalmology, and hospital workflows are examined together with deployment challenges including hallucination, cascading failures, and fairness. By consolidating more than 300 references, with particular emphasis on advances from 2025 to 2026, this work provides a roadmap toward trustworthy, self improving medical imaging systems for real clinical practice.

Medical benchmarks are indispensable for evaluating the capabilities of language models in healthcare for non-English-speaking communities,therefore help ensuring the quality of real-life applications. However, not every community has sufficient resources and standardized methods to effectively build and design such benchmark, and available non-English medical data is normally fragmented and difficult to verify. We developed an approach to tackle this problem and applied it to create the first Vietnamese medical question benchmark, featuring 14,000 multiple-choice questions across 34 medical specialties. Our benchmark was constructed using various verifiable sources, including carefully curated medical exams and clinical records, and eventually annotated by medical experts. The benchmark includes four difficulty levels, ranging from foundational biological knowledge commonly found in textbooks to typical clinical case studies that require advanced reasoning. This design enables assessment of both the breadth and depth of language models' medical understanding in the target language thanks to its extensive coverage and in-depth subject-specific expertise. We release the benchmark in three parts: a sample public set (4k questions), a full public set (10k questions), and a private set (2k questions) used for leaderboard evaluation. Each set contains all medical subfields and difficulty levels. Our approach is scalable to other languages, and we open-source our data construction pipeline to support the development of future multilingual benchmarks in the medical domain.

Precision is of utmost importance in the realm of clinical entity extraction from clinical notes and reports. Encoder Models fine-tuned for Named Entity Recognition (NER) are an efficient choice for this purpose, as they don't hallucinate. We pre-trained an in-house BERT over clinical data and then fine-tuned it for NER. These models performed well on recall but could not close upon the high precision range, needed for clinical models. To address this challenge, we developed a Noise Removal model that refines the output of NER. The NER model assigns token-level entity tags along with probability scores for each token. Our Noise Removal (NR) model then analyzes these probability sequences and classifies predictions as either weak or strong. A naïve approach might involve filtering predictions based on low probability values; however, this method is unreliable. Owing to the characteristics of the SoftMax function, Transformer based architectures often assign disproportionately high confidence scores even to uncertain or weak predictions, making simple thresholding ineffective. To address this issue, we adopted a supervised modeling strategy in which the NR model leverages advanced features such as the Probability Density Map (PDM). The PDM captures the Semantic-Pull effect observed within Transformer embeddings, an effect that manifests in the probability distributions of NER class predictions across token sequences. This approach enables the model to classify predictions as weak or strong with significantly improved accuracy. With these NR models we were able to reduce False Positives across various clinical NER models by 50\% to 90\%.

Benchmarking competitions are central to the development of artificial intelligence (AI) in medical imaging, defining performance standards and shaping methodological progress. However, it remains unclear whether these benchmarks provide data that are sufficiently representative, accessible, and reusable to support clinically meaningful AI. In this work, we assess fairness along two complementary dimensions: (1) whether challenge datasets are representative of real-world clinical diversity, and (2) whether they are accessible and legally reusable in line with the FAIR principles. To address this question, we conducted a large-scale systematic study of 241 biomedical image analysis challenges comprising 458 tasks across 19 imaging modalities. Our findings show substantial biases in dataset composition, including geographic location, modality-, and problem type-related biases, indicating that current benchmarks do not adequately reflect real-world clinical diversity. Despite their widespread influence, challenge datasets were frequently constrained by restrictive or ambiguous access conditions, inconsistent or non-compliant licensing practices, and incomplete documentation, limiting reproducibility and long-term reuse. Together, these shortcomings expose foundational fairness limitations in our benchmarking ecosystem and highlight a disconnect between leaderboard success and clinical relevance.

Clinical diagnosis of the eye is performed over multifarious data modalities including scalar clinical labels, vectorized biomarkers, two-dimensional fundus images, and three-dimensional Optical Coherence Tomography (OCT) scans. Clinical practitioners use all available data modalities for diagnosing and treating eye diseases like Diabetic Retinopathy (DR) or Diabetic Macular Edema (DME). Enabling usage of machine learning algorithms within the ophthalmic medical domain requires research into the relationships and interactions between all relevant data over a treatment period. Existing datasets are limited in that they neither provide data nor consider the explicit relationship modeling between the data modalities. In this paper, we introduce the Ophthalmic Labels for Investigating Visual Eye Semantics (OLIVES) dataset that addresses the above limitation. This is the first OCT and near-IR fundus dataset that includes clinical labels, biomarker labels, disease labels, and time-series patient treatment information from associated clinical trials. The dataset consists of 1268 near-IR fundus images each with at least 49 OCT scans, and 16 biomarkers, along with 4 clinical labels and a disease diagnosis of DR or DME. In total, there are 96 eyes' data averaged over a period of at least two years with each eye treated for an average of 66 weeks and 7 injections. We benchmark the utility of OLIVES dataset for ophthalmic data as well as provide benchmarks and concrete research directions for core and emerging machine learning paradigms within medical image analysis.

Recent advances in generative models, including large language models (LLMs), vision language models (VLMs), and diffusion models, have accelerated the field of natural language and image processing in medicine and marked a significant paradigm shift in how biomedical models can be developed and deployed. While these models are highly adaptable to new tasks, scaling and evaluating their usage presents new challenges not addressed in previous frameworks. In particular, the ability of these models to produce useful outputs with little to no specialized training data ("zero-" or "few-shot" approaches), as well as the open-ended nature of their outputs, necessitate the development of new guidelines for robust reporting of clinical generative model research. In response to gaps in standards and best practices for the development of clinical AI tools identified by US Executive Order 141103 and several emerging national networks for clinical AI evaluation, we begin to formalize some of these guidelines by building on the original MI-CLAIM checklist. The new checklist, MI-CLAIM-GEN (Table 1), aims to address differences in training, evaluation, interpretability, and reproducibility of new generative models compared to non-generative ("predictive") AI models. This MI-CLAIM-GEN checklist also seeks to clarify cohort selection reporting with unstructured clinical data and adds additional items on alignment with ethical standards for clinical AI research.

Large Language Models (LLMs) and multi-agent systems have shown impressive capabilities in natural language tasks but face challenges in clinical trial applications, primarily due to limited access to external knowledge. Recognizing the potential of advanced clinical trial tools that aggregate and predict based on the latest medical data, we propose an integrated solution to enhance their accessibility and utility. We introduce Clinical Agent System (ClinicalAgent), a clinical multi-agent system designed for clinical trial tasks, leveraging GPT-4, multi-agent architectures, LEAST-TO-MOST, and ReAct reasoning technology. This integration not only boosts LLM performance in clinical contexts but also introduces novel functionalities. The proposed method achieves competitive predictive performance in clinical trial outcome prediction (0.7908 PR-AUC), obtaining a 0.3326 improvement over the standard prompt Method. Publicly available code can be found at https://anonymous.4open.science/r/ClinicalAgent-6671.

There is an increasing interest in developing artificial intelligence (AI) systems to process and interpret electronic health records (EHRs). Natural language processing (NLP) powered by pretrained language models is the key technology for medical AI systems utilizing clinical narratives. However, there are few clinical language models, the largest of which trained in the clinical domain is comparatively small at 110 million parameters (compared with billions of parameters in the general domain). It is not clear how large clinical language models with billions of parameters can help medical AI systems utilize unstructured EHRs. In this study, we develop from scratch a large clinical language model - GatorTron - using >90 billion words of text (including >82 billion words of de-identified clinical text) and systematically evaluate it on 5 clinical NLP tasks including clinical concept extraction, medical relation extraction, semantic textual similarity, natural language inference (NLI), and medical question answering (MQA). We examine how (1) scaling up the number of parameters and (2) scaling up the size of the training data could benefit these NLP tasks. GatorTron models scale up the clinical language model from 110 million to 8.9 billion parameters and improve 5 clinical NLP tasks (e.g., 9.6% and 9.5% improvement in accuracy for NLI and MQA), which can be applied to medical AI systems to improve healthcare delivery. The GatorTron models are publicly available at: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/models/gatortron_og.

Alzheimer's disease (AD) is a degenerative brain disease impairing a person's ability to perform day to day activities. The clinical manifestations of Alzheimer's disease are characterized by heterogeneity in age, disease span, progression rate, impairment of memory and cognitive abilities. Due to these variabilities, personalized care and treatment planning, as well as patient counseling about their individual progression is limited. Recent developments in machine learning to detect hidden patterns in complex, multi-dimensional datasets provides significant opportunities to address this critical need. In this work, we use unsupervised and supervised machine learning approaches for subtype identification and prediction. We apply machine learning methods to the extensive clinical observations available at the Alzheimer's Disease Neuroimaging Initiative (ADNI) data set to identify patient subtypes and to predict disease progression. Our analysis depicts the progression space for the Alzheimer's disease into low, moderate and high disease progression zones. The proposed work will enable early detection and characterization of distinct disease subtypes based on clinical heterogeneity. We anticipate that our models will enable patient counseling, clinical trial design, and ultimately individualized clinical care.

Background: Clinical documentation represents a significant burden for healthcare providers, with physicians spending up to 2 hours daily on administrative tasks. Recent advances in large language models (LLMs) offer promising solutions, but privacy concerns and computational requirements limit their adoption in healthcare settings. Objective: To develop and evaluate a privacy-preserving, on-device medical transcription system using a fine-tuned Llama 3.2 1B model capable of generating structured medical notes from medical transcriptions while maintaining complete data sovereignty entirely in the browser. Methods: We fine-tuned a Llama 3.2 1B model using Parameter-Efficient Fine-Tuning (PEFT) with LoRA on 1,500 synthetic medical transcription-to-structured note pairs. The model was evaluated against the base Llama 3.2 1B on two datasets: 100 endocrinology transcripts and 140 modified ACI benchmark cases. Evaluation employed both statistical metrics (ROUGE, BERTScore, BLEURT) and LLM-as-judge assessments across multiple clinical quality dimensions. Results: The fine-tuned OnDevice model demonstrated substantial improvements over the base model. On the ACI benchmark, ROUGE-1 scores increased from 0.346 to 0.496, while BERTScore F1 improved from 0.832 to 0.866. Clinical quality assessments showed marked reduction in major hallucinations (from 85 to 35 cases) and enhanced factual correctness (2.81 to 3.54 on 5-point scale). Similar improvements were observed on the internal evaluation dataset, with composite scores increasing from 3.13 to 4.43 (+41.5%). Conclusions: Fine-tuning compact LLMs for medical transcription yields clinically meaningful improvements while enabling complete on-device browser deployment. This approach addresses key barriers to AI adoption in healthcare: privacy preservation, cost reduction, and accessibility for resource-constrained environments.

PURPOSE: Surgical workflow and skill analysis are key technologies for the next generation of cognitive surgical assistance systems. These systems could increase the safety of the operation through context-sensitive warnings and semi-autonomous robotic assistance or improve training of surgeons via data-driven feedback. In surgical workflow analysis up to 91% average precision has been reported for phase recognition on an open data single-center dataset. In this work we investigated the generalizability of phase recognition algorithms in a multi-center setting including more difficult recognition tasks such as surgical action and surgical skill. METHODS: To achieve this goal, a dataset with 33 laparoscopic cholecystectomy videos from three surgical centers with a total operation time of 22 hours was created. Labels included annotation of seven surgical phases with 250 phase transitions, 5514 occurences of four surgical actions, 6980 occurences of 21 surgical instruments from seven instrument categories and 495 skill classifications in five skill dimensions. The dataset was used in the 2019 Endoscopic Vision challenge, sub-challenge for surgical workflow and skill analysis. Here, 12 teams submitted their machine learning algorithms for recognition of phase, action, instrument and/or skill assessment. RESULTS: F1-scores were achieved for phase recognition between 23.9% and 67.7% (n=9 teams), for instrument presence detection between 38.5% and 63.8% (n=8 teams), but for action recognition only between 21.8% and 23.3% (n=5 teams). The average absolute error for skill assessment was 0.78 (n=1 team). CONCLUSION: Surgical workflow and skill analysis are promising technologies to support the surgical team, but are not solved yet, as shown by our comparison of algorithms. This novel benchmark can be used for comparable evaluation and validation of future work.

The adoption of large language models (LLMs) to assist clinicians has attracted remarkable attention. Existing works mainly adopt the close-ended question-answering (QA) task with answer options for evaluation. However, many clinical decisions involve answering open-ended questions without pre-set options. To better understand LLMs in the clinic, we construct a benchmark ClinicBench. We first collect eleven existing datasets covering diverse clinical language generation, understanding, and reasoning tasks. Furthermore, we construct six novel datasets and clinical tasks that are complex but common in real-world practice, e.g., open-ended decision-making, long document processing, and emerging drug analysis. We conduct an extensive evaluation of twenty-two LLMs under both zero-shot and few-shot settings. Finally, we invite medical experts to evaluate the clinical usefulness of LLMs. The benchmark data is available at https://github.com/AI-in-Health/ClinicBench.

Safety alignment in Large Language Models is critical for healthcare; however, reliance on binary refusal boundaries often results in over-refusal of benign queries or unsafe compliance with harmful ones. While existing benchmarks measure these extremes, they fail to evaluate Safe Completion: the model's ability to maximise helpfulness on dual-use or borderline queries by providing safe, high-level guidance without crossing into actionable harm. We introduce Health-ORSC-Bench, the first large-scale benchmark designed to systematically measure Over-Refusal and Safe Completion quality in healthcare. Comprising 31,920 benign boundary prompts across seven health categories (e.g., self-harm, medical misinformation), our framework uses an automated pipeline with human validation to test models at varying levels of intent ambiguity. We evaluate 30 state-of-the-art LLMs, including GPT-5 and Claude-4, revealing a significant tension: safety-optimised models frequently refuse up to 80\% of "Hard" benign prompts, while domain-specific models often sacrifice safety for utility. Our findings demonstrate that model family and size significantly influence calibration: larger frontier models (e.g., GPT-5, Llama-4) exhibit "safety-pessimism" and higher over-refusal than smaller or MoE-based counterparts (e.g., Qwen-3-Next), highlighting that current LLMs struggle to balance refusal and compliance. Health-ORSC-Bench provides a rigorous standard for calibrating the next generation of medical AI assistants toward nuanced, safe, and helpful completions. The code and data will be released upon acceptance. red{Warning: Some contents may include toxic or undesired contents.}

The development of clinical natural language processing (NLP) systems is severely hampered by the sensitive nature of medical records, which restricts data sharing under stringent privacy regulations, particularly in France and the broader European Union. To address this gap, we introduce PARHAF, a large open-source corpus of clinical documents in French. PARHAF comprises expert-authored clinical reports describing realistic yet entirely fictitious patient cases, making it anonymous and freely shareable by design. The corpus was developed using a structured protocol that combined clinician expertise with epidemiological guidance from the French National Health Data System (SNDS), ensuring broad clinical coverage. A total of 104 medical residents across 18 specialties authored and peer-reviewed the reports following predefined clinical scenarios and document templates. The corpus contains 7394 clinical reports covering 5009 patient cases across a wide range of medical and surgical specialties. It includes a general-purpose component designed to approximate real-world hospitalization distributions, and four specialized subsets that support information-extraction use cases in oncology, infectious diseases, and diagnostic coding. Documents are released under a CC-BY open license, with a portion temporarily embargoed to enable future benchmarking under controlled conditions. PARHAF provides a valuable resource for training and evaluating French clinical language models in a fully privacy-preserving setting, and establishes a replicable methodology for building shareable synthetic clinical corpora in other languages and health systems.

Models that can predict the occurrence of events ahead of time with low false-alarm rates are critical to the acceptance of decision support systems in the medical community. This challenging task is typically treated as a simple binary classification, ignoring temporal dependencies between samples, whereas we propose to exploit this structure. We first introduce a common theoretical framework unifying dynamic survival analysis and early event prediction. Following an analysis of objectives from both fields, we propose Temporal Label Smoothing (TLS), a simpler, yet best-performing method that preserves prediction monotonicity over time. By focusing the objective on areas with a stronger predictive signal, TLS improves performance over all baselines on two large-scale benchmark tasks. Gains are particularly notable along clinically relevant measures, such as event recall at low false-alarm rates. TLS reduces the number of missed events by up to a factor of two over previously used approaches in early event prediction.

The COVID-19 pandemic has posed a heavy burden to the healthcare system worldwide and caused huge social disruption and economic loss. Many deep learning models have been proposed to conduct clinical predictive tasks such as mortality prediction for COVID-19 patients in intensive care units using Electronic Health Record (EHR) data. Despite their initial success in certain clinical applications, there is currently a lack of benchmarking results to achieve a fair comparison so that we can select the optimal model for clinical use. Furthermore, there is a discrepancy between the formulation of traditional prediction tasks and real-world clinical practice in intensive care. To fill these gaps, we propose two clinical prediction tasks, Outcome-specific length-of-stay prediction and Early mortality prediction for COVID-19 patients in intensive care units. The two tasks are adapted from the naive length-of-stay and mortality prediction tasks to accommodate the clinical practice for COVID-19 patients. We propose fair, detailed, open-source data-preprocessing pipelines and evaluate 17 state-of-the-art predictive models on two tasks, including 5 machine learning models, 6 basic deep learning models and 6 deep learning predictive models specifically designed for EHR data. We provide benchmarking results using data from two real-world COVID-19 EHR datasets. One dataset is publicly available without needing any inquiry and another dataset can be accessed on request. We provide fair, reproducible benchmarking results for two tasks. We deploy all experiment results and models on an online platform. We also allow clinicians and researchers to upload their data to the platform and get quick prediction results using our trained models. We hope our efforts can further facilitate deep learning and machine learning research for COVID-19 predictive modeling.

To investigate whether the pleurae, airways and vessels surrounding a nodule on non-contrast computed tomography (CT) can discriminate benign and malignant pulmonary nodules. The LIDC-IDRI dataset, one of the largest publicly available CT database, was exploited for study. A total of 1556 nodules from 694 patients were involved in statistical analysis, where nodules with average scorings <3 and >3 were respectively denoted as benign and malignant. Besides, 339 nodules from 113 patients with diagnosis ground-truth were independently evaluated. Computer algorithms were developed to segment pulmonary structures and quantify the distances to pleural surface, airways and vessels, as well as the counting number and normalized volume of airways and vessels near a nodule. Odds ratio (OR) and Chi-square (\chi^2) testing were performed to demonstrate the correlation between features of surrounding structures and nodule malignancy. A non-parametric receiver operating characteristic (ROC) analysis was conducted in logistic regression to evaluate discrimination ability of each structure. For benign and malignant groups, the average distances from nodules to pleural surface, airways and vessels are respectively (6.56, 5.19), (37.08, 26.43) and (1.42, 1.07) mm. The correlation between nodules and the counting number of airways and vessels that contact or project towards nodules are respectively (OR=22.96, \chi^2=105.04) and (OR=7.06, \chi^2=290.11). The correlation between nodules and the volume of airways and vessels are (OR=9.19, \chi^2=159.02) and (OR=2.29, \chi^2=55.89). The areas-under-curves (AUCs) for pleurae, airways and vessels are respectively 0.5202, 0.6943 and 0.6529. Our results show that malignant nodules are often surrounded by more pulmonary structures compared with benign ones, suggesting that features of these structures could be viewed as lung cancer biomarkers.

We consider a biological population whose environment varies periodically in time, exhibiting two very different "seasons" : one is favorable and the other one is unfavorable. For monotone differential models with concave nonlinearities, we address the following question: the system's period being fixed, under what conditions does there exist a critical duration for the unfavorable season? By "critical duration" we mean that above some threshold, the population cannot sustain and extincts, while below this threshold, the system converges to a unique periodic and positive solution. We term this a "sharp seasonal threshold property" (SSTP, for short). Building upon a previous result, we obtain sufficient conditions for SSTP in any dimension and apply our criterion to a two-dimensional model featuring juvenile and adult populations of insects.

How many experimental studies would have come to different conclusions had they been run on larger samples? I show how to estimate the expected number of statistically significant results that a set of experiments would have reported had their sample sizes all been counterfactually increased. The proposed deconvolution estimator is asymptotically normal and adjusts for publication bias. Unlike related methods, this approach requires no assumptions of any kind about the distribution of true intervention treatment effects and allows for point masses. Simulations find good coverage even when the t-score is only approximately normal. An application to randomized trials (RCTs) published in economics journals finds that doubling every sample would increase the power of t-tests by 7.2 percentage points on average. This effect is smaller than for non-RCTs and comparable to systematic replications in laboratory psychology where previous studies enabled more accurate power calculations. This suggests that RCTs are on average relatively insensitive to sample size increases. Research funders who wish to raise power should generally consider sponsoring better-measured and higher quality experiments -- rather than only larger ones.

Objective:Develop and validate an algorithm for analyzing the layout of PDF clinical documents to improve the performance of downstream natural language processing tasks. Materials and Methods: We designed an algorithm to process clinical PDF documents and extract only clinically relevant text. The algorithm consists of several steps: initial text extraction using a PDF parser, followed by classification into categories such as body text, left notes, and footers using a Transformer deep neural network architecture, and finally an aggregation step to compile the lines of a given label in the text. We evaluated the technical performance of the body text extraction algorithm by applying it to a random sample of documents that were annotated. Medical performance was evaluated by examining the extraction of medical concepts of interest from the text in their respective sections. Finally, we tested an end-to-end system on a medical use case of automatic detection of acute infection described in the hospital report. Results:Our algorithm achieved per-line precision, recall, and F1 score of 98.4, 97.0, and 97.7, respectively, for body line extraction. The precision, recall, and F1 score per document for the acute infection detection algorithm were 82.54 (95CI 72.86-91.60), 85.24 (95CI 76.61-93.70), 83.87 (95CI 76, 92-90.08) with exploitation of the results of the advanced body extraction algorithm, respectively. Conclusion:We have developed and validated a system for extracting body text from clinical documents in PDF format by identifying their layout. We were able to demonstrate that this preprocessing allowed us to obtain better performances for a common downstream task, i.e., the extraction of medical concepts in their respective sections, thus proving the interest of this method on a clinical use case.

Objective: Question answering (QA) systems have the potential to improve the quality of clinical care by providing health professionals with the latest and most relevant evidence. However, QA systems have not been widely adopted. This systematic review aims to characterize current medical QA systems, assess their suitability for healthcare, and identify areas of improvement. Materials and methods: We searched PubMed, IEEE Xplore, ACM Digital Library, ACL Anthology and forward and backward citations on 7th February 2023. We included peer-reviewed journal and conference papers describing the design and evaluation of biomedical QA systems. Two reviewers screened titles, abstracts, and full-text articles. We conducted a narrative synthesis and risk of bias assessment for each study. We assessed the utility of biomedical QA systems. Results: We included 79 studies and identified themes, including question realism, answer reliability, answer utility, clinical specialism, systems, usability, and evaluation methods. Clinicians' questions used to train and evaluate QA systems were restricted to certain sources, types and complexity levels. No system communicated confidence levels in the answers or sources. Many studies suffered from high risks of bias and applicability concerns. Only 8 studies completely satisfied any criterion for clinical utility, and only 7 reported user evaluations. Most systems were built with limited input from clinicians. Discussion: While machine learning methods have led to increased accuracy, most studies imperfectly reflected real-world healthcare information needs. Key research priorities include developing more realistic healthcare QA datasets and considering the reliability of answer sources, rather than merely focusing on accuracy.

The advancement of natural language processing (NLP) systems in healthcare hinges on language model ability to interpret the intricate information contained within clinical notes. This process often requires integrating information from various time points in a patient's medical history. However, most earlier clinical language models were pretrained with a context length limited to roughly one clinical document. In this study, We introduce ClinicalMamba, a specialized version of the Mamba language model, pretrained on a vast corpus of longitudinal clinical notes to address the unique linguistic characteristics and information processing needs of the medical domain. ClinicalMamba, with 130 million and 2.8 billion parameters, demonstrates a superior performance in modeling clinical language across extended text lengths compared to Mamba and clinical Llama. With few-shot learning, ClinicalMamba achieves notable benchmarks in speed and accuracy, outperforming existing clinical language models and general domain large models like GPT-4 in longitudinal clinical notes information extraction tasks.

Pathology Foundation Models (FMs) hold great promise for healthcare. Before they can be used in clinical practice, it is essential to ensure they are robust to variations between medical centers. We measure whether pathology FMs focus on biological features like tissue and cancer type, or on the well known confounding medical center signatures introduced by staining procedure and other differences. We introduce the Robustness Index. This novel robustness metric reflects to what degree biological features dominate confounding features. Ten current publicly available pathology FMs are evaluated. We find that all current pathology foundation models evaluated represent the medical center to a strong degree. Significant differences in the robustness index are observed. Only one model so far has a robustness index greater than one, meaning biological features dominate confounding features, but only slightly. A quantitative approach to measure the influence of medical center differences on FM-based prediction performance is described. We analyze the impact of unrobustness on classification performance of downstream models, and find that cancer-type classification errors are not random, but specifically attributable to same-center confounders: images of other classes from the same medical center. We visualize FM embedding spaces, and find these are more strongly organized by medical centers than by biological factors. As a consequence, the medical center of origin is predicted more accurately than the tissue source and cancer type. The robustness index introduced here is provided with the aim of advancing progress towards clinical adoption of robust and reliable pathology FMs.

Vision Language Models (VLMs) are increasingly used for tasks like medical report generation and visual question answering. However, fluent diagnostic text does not guarantee safe visual understanding. In clinical practice, interpretation begins with pre-diagnostic sanity checks: verifying that the input is valid to read (correct modality and anatomy, plausible viewpoint and orientation, and no obvious integrity violations). Existing benchmarks largely assume this step is solved, and therefore miss a critical failure mode: a model can produce plausible narratives even when the input is inconsistent or invalid. We introduce MedObvious, a 1,880-task benchmark that isolates input validation as a set-level consistency capability over small multi-panel image sets: the model must identify whether any panel violates expected coherence. MedObvious spans five progressive tiers, from basic orientation/modality mismatches to clinically motivated anatomy/viewpoint verification and triage-style cues, and includes five evaluation formats to test robustness across interfaces. Evaluating 17 different VLMs, we find that sanity checking remains unreliable: several models hallucinate anomalies on normal (negative-control) inputs, performance degrades when scaling to larger image sets, and measured accuracy varies substantially between multiple-choice and open-ended settings. These results show that pre-diagnostic verification remains unsolved for medical VLMs and should be treated as a distinct, safety-critical capability before deployment.

Brazil's Unified Health System (SUS) relies on official clinical guidelines that define diagnostic criteria, treatments, dosages, and monitoring procedures for over 200 million citizens. Yet current LLMs perform poorly on this guideline-specific knowledge, and no benchmark evaluates clinical recall grounded in Brazilian Portuguese protocols. We address this gap by adapting Qwen2.5-14B-Instruct to the Brazilian clinical domain. From 178 official guidelines (~5.4M tokens), we generate ~70M tokens of synthetic data in three formats -- rephrases, wiki-style articles, and question-answer pairs -- using four generator LLMs. We then apply continual pre-training followed by Group Relative Policy Optimization (GRPO). We introduce HealthBench-BR, with 1,780 balanced true/false clinical assertions, and PCDT-QA, with 890 open-ended clinical questions scored by an LLM judge. Our best model achieves 83.9% on HealthBench-BR and 85.4% on PCDT-QA, outperforming GPT-5.2, Claude Sonnet 4.6, Gemini 3.1 Pro, and Google AI Overview's web-grounded RAG despite having only 14B parameters. Ablations show that generator diversity and reinforcement learning are critical to these gains. We release all datasets, benchmarks, and model weights to support reproducible clinical NLP research for Brazilian Portuguese. Code, data, and model weights are available at https://github.com/hugoabonizio/clinical-protocols-br

Incomplete or inconsistent discharge documentation drives care fragmentation and avoidable readmissions. Despite its critical role in patient safety, auditing discharge summaries relies on manual review and does not scale. We propose an automated framework for auditing discharge summaries using large language models (LLMs). Our approach operationalizes the DISCHARGED framework into a checklist of 46 questions. Using 50 summaries from the MIMIC-IV database, with clinician ground-truth labels, we benchmark 11 LLMs. Model-assessed mean documentation completeness ranges from 54.9% to 74.2%, and the best-performing models achieve a Cohen's kappa values around 0.5 against clinician labels, indicating moderate agreement. All models struggle to identify ambiguous documentation (Unclear), highlighting a key gap in current automated auditing. This work provides a clinician-validated benchmark and zero-shot baselines for systematic quality improvement in clinical documentation.

Rare diseases, including Inborn Errors of Metabolism (IEM), pose significant diagnostic challenges. Case reports serve as key but computationally underutilized resources to inform diagnosis. Clinical dense information extraction refers to organizing medical information into structured predefined categories. Large Language Models (LLMs) may enable scalable information extraction from case reports but are rarely evaluated for this task. We introduce CaseReportBench, an expert-annotated dataset for dense information extraction of case reports, focusing on IEMs. Using this dataset, we assess various models and prompting strategies, introducing novel approaches such as category-specific prompting and subheading-filtered data integration. Zero-shot chain-of-thought prompting offers little advantage over standard zero-shot prompting. Category-specific prompting improves alignment with the benchmark. The open-source model Qwen2.5-7B outperforms GPT-4o for this task. Our clinician evaluations show that LLMs can extract clinically relevant details from case reports, supporting rare disease diagnosis and management. We also highlight areas for improvement, such as LLMs' limitations in recognizing negative findings important for differential diagnosis. This work advances LLM-driven clinical natural language processing and paves the way for scalable medical AI applications.

Recent incidents involving LLMs used for mental-health support reveal a critical evaluation gap: surface-level safety scores do not capture how models behave across realistic, emotionally sensitive interactions over time. Existing benchmarks measure knowledge, safety, or static response quality, but miss whether LLM interactions help users keep reflecting, coping, and making decisions themselves. We formalize this missing dimension as COGNITIVE ATROPHY, a process-level behavioural measure in AI-mediated mental-health support distinct from safety and helpfulness. To measure it, we introduce COGNITIVE ATROPHY BENCH, a clinically grounded benchmark built from 1,576 fully human-generated counseling conversations, 15,680 turns, and 42,230 responses from five LLMs. Three clinical and neuropsychology experts developed a 20-attribute schema spanning user context, response behaviour, and global risk flags; six trained clinical reviewers applied it with span-grounded evidence, producing 5,324 reviewer judgments. We further introduce the User-Input Risk Index (UIRI), the Cognitive Atrophy Risk Index (ARI), and trajectory summaries. Across five LLMs, models show a consistent moderate-to-high level of atrophy-aligned behaviour across single and multi-turn settings. While models generally respond to overt safety cues, they adapt less reliably when users seek solutions or decisions. The dominant recurring patterns are directive advice, problem-solving, recommendation responses, topic shifts, and forms of validation that may reinforce dependence rather than reflection. Our work makes COGNITIVE ATROPHY measurable and provides a foundation for auditing model behaviour in sensitive LLM conversations.

Extracting clinical information from medical transcripts in low-resource languages remains a significant challenge in healthcare natural language processing (NLP). This study evaluates a two-step pipeline combining Aya-expanse-8B as a Persian-to-English translation model with five open-source small language models (SLMs) -- Qwen2.5-7B-Instruct, Llama-3.1-8B-Instruct, Llama-3.2-3B-Instruct, Qwen2.5-1.5B-Instruct, and Gemma-3-1B-it -- for binary extraction of 13 clinical features from 1,221 anonymized Persian transcripts collected at a cancer palliative care call center. Using a few-shot prompting strategy without fine-tuning, models were assessed on macro-averaged F1-score, Matthews Correlation Coefficient (MCC), sensitivity, and specificity to account for class imbalance. Qwen2.5-7B-Instruct achieved the highest overall performance (median macro-F1: 0.899; MCC: 0.797), while Gemma-3-1B-it showed the weakest results. Larger models (7B--8B parameters) consistently outperformed smaller counterparts in sensitivity and MCC. A bilingual analysis of Aya-expanse-8B revealed that translating Persian transcripts to English improved sensitivity, reduced missing outputs, and boosted metrics robust to class imbalance, though at the cost of slightly lower specificity and precision. Feature-level results showed reliable extraction of physiological symptoms across most models, whereas psychological complaints, administrative requests, and complex somatic features remained challenging. These findings establish a practical, privacy-preserving blueprint for deploying open-source SLMs in multilingual clinical NLP settings with limited infrastructure and annotation resources, and highlight the importance of jointly optimizing model scale and input language strategy for sensitive healthcare applications.

Current benchmarks for AI clinician systems, often based on multiple-choice exams or manual rubrics, fail to capture the depth, robustness, and safety required for real-world clinical practice. To address this, we introduce the GAPS framework, a multidimensional paradigm for evaluating Grounding (cognitive depth), Adequacy (answer completeness), Perturbation (robustness), and Safety. Critically, we developed a fully automated, guideline-anchored pipeline to construct a GAPS-aligned benchmark end-to-end, overcoming the scalability and subjectivity limitations of prior work. Our pipeline assembles an evidence neighborhood, creates dual graph and tree representations, and automatically generates questions across G-levels. Rubrics are synthesized by a DeepResearch agent that mimics GRADE-consistent, PICO-driven evidence review in a ReAct loop. Scoring is performed by an ensemble of large language model (LLM) judges. Validation confirmed our automated questions are high-quality and align with clinician judgment. Evaluating state-of-the-art models on the benchmark revealed key failure modes: performance degrades sharply with increased reasoning depth (G-axis), models struggle with answer completeness (A-axis), and they are highly vulnerable to adversarial perturbations (P-axis) as well as certain safety issues (S-axis). This automated, clinically-grounded approach provides a reproducible and scalable method for rigorously evaluating AI clinician systems and guiding their development toward safer, more reliable clinical practice.

Current medical retrieval benchmarks primarily emphasize lexical or shallow semantic similarity, overlooking the reasoning-intensive demands that are central to clinical decision-making. In practice, physicians often retrieve authoritative medical evidence to support diagnostic hypotheses. Such evidence typically aligns with an inferred diagnosis rather than the surface form of a patient's symptoms, leading to low lexical or semantic overlap between queries and relevant documents. To address this gap, we introduce R2MED, the first benchmark explicitly designed for reasoning-driven medical retrieval. It comprises 876 queries spanning three tasks: Q&A reference retrieval, clinical evidence retrieval, and clinical case retrieval. These tasks are drawn from five representative medical scenarios and twelve body systems, capturing the complexity and diversity of real-world medical information needs. We evaluate 15 widely-used retrieval systems on R2MED and find that even the best model achieves only 31.4 nDCG@10, demonstrating the benchmark's difficulty. Classical re-ranking and generation-augmented retrieval methods offer only modest improvements. Although large reasoning models improve performance via intermediate inference generation, the best results still peak at 41.4 nDCG@10. These findings underscore a substantial gap between current retrieval techniques and the reasoning demands of real clinical tasks. We release R2MED as a challenging benchmark to foster the development of next-generation medical retrieval systems with enhanced reasoning capabilities. Data and code are available at https://github.com/R2MED/R2MED

Vision-language models (VLMs) adapted to the medical domain have shown strong performance on visual question answering benchmarks, yet their robustness against two critical failure modes, hallucination and sycophancy, remains poorly understood, particularly in combination. We evaluate six VLMs (three general-purpose, three medical-specialist) on three medical VQA datasets and uncover a grounding-sycophancy tradeoff: models with the lowest hallucination propensity are the most sycophantic, while the most pressure-resistant model hallucinates more than all medical-specialist models. To characterize this tradeoff, we propose three metrics: L-VASE, a logit-space reformulation of VASE that avoids its double-normalization; CCS, a confidence-calibrated sycophancy score that penalizes high-confidence capitulation; and Clinical Safety Index (CSI), a unified safety index that combines grounding, autonomy, and calibration via a geometric mean. Across 1,151 test cases, no model achieves a CSI above 0.35, indicating that none of the evaluated 7-8B parameter VLMs is simultaneously well-grounded and robust to social pressure. Our findings suggest that joint evaluation of both properties is necessary before these models can be considered for clinical use. Code is available at https://github.com/UTSA-VIRLab/AgreeOrRight

Background: AI-driven prediction algorithms have the potential to enhance emergency medicine by enabling rapid and accurate decision-making regarding patient status and potential deterioration. However, the integration of multimodal data, including raw waveform signals, remains underexplored in clinical decision support. Methods: We present a dataset and benchmarking protocol designed to advance multimodal decision support in emergency care. Our models utilize demographics, biometrics, vital signs, laboratory values, and electrocardiogram (ECG) waveforms as inputs to predict both discharge diagnoses and patient deterioration. Results: The diagnostic model achieves area under the receiver operating curve (AUROC) scores above 0.8 for 609 out of 1,428 conditions, covering both cardiac (e.g., myocardial infarction) and non-cardiac (e.g., renal disease, diabetes) diagnoses. The deterioration model attains AUROC scores above 0.8 for 14 out of 15 targets, accurately predicting critical events such as cardiac arrest, mechanical ventilation, ICU admission, and mortality. Conclusions: Our study highlights the positive impact of incorporating raw waveform data into decision support models, improving predictive performance. By introducing a unique, publicly available dataset and baseline models, we provide a foundation for measurable progress in AI-driven decision support for emergency care.

Medical coding translates clinical documentation into standardized codes for billing, research, and public health, but manual coding is time-consuming and error-prone. Existing automation efforts rely on small datasets that poorly represent real-world patient heterogeneity. We trained a language model on 5.8 million electronic health records from 1.8 million patients across nearly all specialties in Eastern Denmark (2006--2016) to predict ICD-10 codes from clinical notes, medications, and laboratory results. Evaluated on 270,000 held-out patients, the model achieved a micro F1 of 71.8% and a top-10 recall of 95.5%. Performance varied by specialty (F1: 53--91%), with higher scores in specialties with well-defined diagnostic criteria. Codes appearing predominantly as secondary diagnoses had markedly lower F1 scores. For three such codes (suicide-related behaviors, weight disorders, and hypertension), the model identified thousands of uncoded cases, of which 76-86% were confirmed valid upon manual review, suggesting systematic under-coding rather than model error. These findings suggest under-coding of secondary diagnoses in Eastern Denmark during this period, with potential implications for epidemiological research, public health surveillance, and understanding of multimorbidity. Similar time constraints and reimbursement structures in other healthcare systems suggest this may not be isolated to this dataset. The model can automate coding for approximately 50% of cases and provide accurate suggestions for most others, and may offer a practical solution to help capture missed secondary conditions.

Computer-use agents could automate repetitive screen-based clinical work, but their reliability in medical graphical user interfaces remains largely unvalidated. Existing benchmarks focus on general web or desktop tasks and underrepresent medical software, which requires domain knowledge, exhibits markedly different UI design from mainstream applications, lacks public testing environments, and demands safety validation beyond task completion. We introduce MedCUA-Bench, an interactive benchmark for clinical computer-use agents. It covers 18 clinical scenarios across 10 medical domains, reconstructed from real product manuals and open-source medical systems to capture authentic clinical interfaces while avoiding licensing and privacy constraints. Each task ships with paired intent- and step-level goals to disentangle clinical reasoning from UI execution, and is evaluated by a deterministic checker over task completion and five clinical safety dimensions. Across 23 agents, the best closed-source model reaches 54.2% strict success, while all models remain below 9% on the real OpenEMR. Open-source agents average only 2.5%, with the best reaching 16.2%. MedCUA-Bench exposes the gap between current agents and reliable clinical software use, providing a reproducible testbed for future research.

We introduce MURA, a large dataset of musculoskeletal radiographs containing 40,561 images from 14,863 studies, where each study is manually labeled by radiologists as either normal or abnormal. To evaluate models robustly and to get an estimate of radiologist performance, we collect additional labels from six board-certified Stanford radiologists on the test set, consisting of 207 musculoskeletal studies. On this test set, the majority vote of a group of three radiologists serves as gold standard. We train a 169-layer DenseNet baseline model to detect and localize abnormalities. Our model achieves an AUROC of 0.929, with an operating point of 0.815 sensitivity and 0.887 specificity. We compare our model and radiologists on the Cohen's kappa statistic, which expresses the agreement of our model and of each radiologist with the gold standard. Model performance is comparable to the best radiologist performance in detecting abnormalities on finger and wrist studies. However, model performance is lower than best radiologist performance in detecting abnormalities on elbow, forearm, hand, humerus, and shoulder studies. We believe that the task is a good challenge for future research. To encourage advances, we have made our dataset freely available at https://stanfordmlgroup.github.io/competitions/mura .