Dataset Viewer
Auto-converted to Parquet Duplicate
The dataset viewer is not available for this split.
External: Generic Unexpected error: Unexpected (temporary) at read, context: { url: https://huggingface.co/datasets/Mobiusi/Medical-Decision-Making-Capacity-Assessment-Constraint-Verification-Benchmark/resolve/refs%2Fconvert%2Fparquet/default/train/0000.parquet, called: reqwest::fetch, service: hf, path: default/train/0000.parquet, range: 2879-3992, read: 0 } => read data from http response, source: error decoding response body
Error code:   UnexpectedError

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Medical Decision-Making Capacity Assessment Constraint Verification Benchmark

This benchmark presents structured JSON cases containing specific medical decisions, patient interview evidence, clinical context, and claimed decision-making capacity assessments to be verified. Cases focus on the elements required for each decision, enabling evaluation of whether conclusions are supported by the evidence and whether material information gaps or conflicts are identified. It is designed for model evaluation, constraint reasoning research, and clinical AI validation in consultation-liaison psychiatry.

Technical Specifications

Field Type Description
evidence_gaps array Material information missing from the record that prevents or limits validation of the assessment.
overall_result string The overall judgment on whether the claimed assessment is supported by the available decision-specific evidence.
clinical_context object Relevant clinical and communication information for interpreting the interview evidence.
decision_context object The medical decision being considered and the stakes specific to that decision.
constraint_checks array Explicit checks of whether the proposed assessment follows the constraints of a decision-specific capacity evaluation.
overall_rationale string A concise synthesis of the required elements, constraint checks, evidence gaps, and conflicts supporting the overall result.
required_elements array Decision-specific capacity elements and an evidence-based assessment of whether each element is addressed.
claimed_assessment object The capacity determination and supporting rationale to evaluate against the specific decision and interview evidence.
evidence_conflicts array Material contradictions among the interview evidence, clinical context, and claimed assessment.
interview_evidence array Patient statements and observations from the interview that may support or challenge capacity elements.

Compliance Statement

Authorization TypeCC-BY-NC-SA 4.0 (Attribution–NonCommercial–ShareAlike)
Commercial UseRequires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and AnonymizationNo PII, no real company names, simulated scenarios follow industry standards
Compliance SystemCompliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Source & Contact

contact@mobiusi.com

Downloads last month
54