Upload README.md with huggingface_hub

389313c verified 20 days ago

5.95 kB

language:
  - en
license: mit
library_name: timm
pipeline_tag: image-classification
tags:
  - medical-imaging
  - knee-mri
  - acl-tear-detection
  - deep-learning
  - convnext
  - self-attention
  - masked-slice-modeling
  - radiology
  - orthopedics
datasets:
  - stanford-mrnet
metrics:
  - roc_auc
  - accuracy
  - f1
model-index:
  - name: ACL-LKNet
    results:
      - task:
          type: image-classification
          name: Knee MRI ACL Tear Detection
        dataset:
          type: stanford-mrnet
          name: Stanford MRNet Locked Test Cohort (N=120)
        metrics:
          - type: roc_auc
            value: 0.9639
            name: AUROC
          - type: precision_recall_auc
            value: 0.9293
            name: AUPRC
          - type: accuracy
            value: 0.8167
            name: Accuracy
          - type: specificity
            value: 0.9394
            name: Specificity
          - type: sensitivity
            value: 0.6667
            name: Sensitivity
          - type: f1
            value: 0.766
            name: F1 Score

ACL-LKNet: Self-Supervised Large-Kernel Network for ACL Tear Detection in Knee MRI

ACL-LKNet is an anatomically grounded deep learning architecture specifically engineered for automated Anterior Cruciate Ligament (ACL) tear detection from tri-planar (Sagittal, Coronal, and Axial) volumetric knee MRI examinations.

Developed as part of a doctoral investigation in computational musculoskeletal radiology, ACL-LKNet combines:

Large-Kernel 2D Backbone (ConvNeXt-Tiny): Large $7 \times 7$ depthwise convolutions capturing the complete oblique trajectory of intra-articular ligaments.
Masked Slice Modeling (MSM): Volumetric self-supervised pretext reconstruction across anisotropic slice stacks.
Parametric Slice Attention: Dynamic slice sequence pooling that outputs explicit, interpretable slice attention weights $\alpha_{p,s}$.
Tri-Planar Cross-Attention Fusion: 2-head self-attention operating over learned plane positional embeddings ($e_{\text{sag}}, e_{\text{cor}}, e_{\text{axi}}$).
Strict Anatomical Invariants: No horizontal/vertical flipping during training to preserve internal knee joint chirality and oblique ACL orientation.

Benchmark Performance on Stanford MRNet

Evaluated on the locked, official Stanford MRNet benchmark test set ($N=120$ examinations, 54 tears, 66 controls) with empirical 95% bootstrap confidence intervals ($N=1{,}000$ iterations):

Diagnostic Metric	ACL-LKNet (5-Fold Ensemble)	95% Bootstrap CI	Stanford MRNet Baseline (Bien et al., 2018)	Absolute $\Delta$ Gain
AUROC	0.9639	[0.9277, 0.9919]	0.9370	+0.0269
AUPRC	0.9293	[0.8492, 0.9889]	--	--
Accuracy	81.67%	[75.00%, 88.33%]	82.50%	$-0.0083$
Specificity	93.94%	[87.69%, 98.59%]	96.80%	$-0.0286$
Sensitivity	66.67%	[53.22%, 79.25%]	75.90%	$-0.0923$
F1-Score	0.7660	[0.6585, 0.8519]	--	--
Brier Score	0.1184	[0.0891, 0.1520]	--	Well-Calibrated

Statistical Significance (RQ1): Paired DeLong test comparing ConvNeXt-Tiny against ResNet-18 yields $z = 3.864, p = 0.000104$ ($p < 0.001$), establishing the statistical superiority of large receptive fields for elongated ligament structures.

Quickstart: Python Inference via Hugging Face Hub

import torch
from huggingface_hub import hf_hub_download

# 1. Download model weights from Hugging Face Hub
checkpoint_path = hf_hub_download(
    repo_id="shareefch1413/ACL-LKNet",
    filename="finetune_best.pt"
)

# 2. Instantiate model architecture
from src.config import Config
from src.models.acl_lknet import create_model_from_config

config = Config(backbone_name="convnext_tiny")
model = create_model_from_config(config)

state_dict = torch.load(checkpoint_path, map_location="cpu")
model.load_state_dict(state_dict["ema_state_dict"] if "ema_state_dict" in state_dict else state_dict["model_state_dict"])
model.eval()

# 3. Predict on tri-planar MRI volume (Sagittal, Coronal, Axial)
# Each volume tensor is shape: (1, 24, 3, 224, 224)
dummy_exam = {
    "sagittal": torch.randn(1, 24, 3, 224, 224),
    "coronal":  torch.randn(1, 24, 3, 224, 224),
    "axial":    torch.randn(1, 24, 3, 224, 224)
}

with torch.no_grad():
    output = model(dummy_exam)
    tear_probability = torch.sigmoid(output["logits"]).item()

print(f"Predicted ACL Tear Probability: {tear_probability * 100:.2f}%")

Clinical Interpretability: Grad-CAM++ and Slice Attention

Parametric Slice Attention Profiles: Learns autonomous focus on central intercondylar notch slices (11--15/24) where the ACL is anatomically situated without requiring slice-level bounding box supervision.
High-Resolution Grad-CAM++: Hooks into ConvNeXt-Tiny Stage 2 ($14 \times 14$ feature map) to generate intra-articular gradient heatmaps localized to the femoral footprint and midsubstance tear site.
Decision Curve Analysis (DCA): Demonstrates superior clinical net benefit over "treat all" and "treat none" policies across all relevant surgical intervention thresholds ($p_t \in [0.10, 0.75]$).

Citation

@article{acl_lknet2026,
  title={ACL-LKNet: Anatomically Constrained Large-Kernel Network with Multi-Plane Self-Attention for Volumetric ACL Tear Detection in Knee MRI},
  author={PhD Candidate in Biomedical Engineering and Computational Medicine},
  journal={IEEE Transactions on Medical Imaging (Preprint / PhD Dissertation Protocol)},
  year={2026}
}