Key result
A deep learning-based tool accurately quantified coronary artery stenosis on whole-slide images with a mean absolute error of 1.01 percentage points, outperforming visual estimates by pathologists.
Why the study?
Accurate quantification of coronary stenosis in forensic autopsies is crucial for determining causality but is hindered by subjective visual assessments and inter-observer variability.
Does an AI-driven tool using whole-slide images improve the accuracy and reproducibility of coronary stenosis measurement in forensic autopsies compared to visual estimates?
Observational (n=98)
Single-blind
Yes
Does an AI-driven tool using whole-slide images improve the accuracy and reproducibility of coronary stenosis measurement in forensic autopsies compared to visual estimates?
A deep learning-based tool using whole-slide images provides rapid, reproducible, and accurate coronary stenosis measurements in forensic autopsies, outperforming visual estimates by pathologists.
Answers are grounded in this paper and cite their sources.
Supports objective CAD causality assessment in forensic autopsies; extends AI validation to whole-slide imaging in legal medicine.
Cardiovascular disease is a leading cause of mortality, with coronary artery disease accounting for over 60% of cases in adults. Accurate quantification of coronary stenosis in forensic autopsies is crucial for determining causality between pathological findings and death, but is hindered by subjective visual assessments and inter-observer variability. This study aimed to develop an AI-driven tool using whole-slide images for objective stenosis measurement in forensic investigations. From 98 anonymized H&E-stained autopsy slides (234 coronary sections), 103 high-quality regions of interest were selected and split into training (n = 82), validation (n = 14), and test (n = 7) datasets. Annotations delineated lumen, internal elastic lamina, and external elastic lamina using QuPath. A SegFormer-B0 transformer model was trained with data augmentation, weighted cross-entropy loss, and AdamW optimization. Post-processing enforced anatomical structural hierarchy and generated hybrid confidence maps. On validation dataset, agreement with ground truth was excellent (MAE 3.22% points; RMSE 4.00; MAPE 6.07%; r = 0.986; ICC = 0.986), with slight underestimation. Bland–Altman bias was − 1.36 pp (95% LoA − 9.01 to 6.28 pp), indicating performance across severities. On the test set, accuracy improved (MAE 1.01 pp; RMSE 1.50; MAPE 2.11%; r = 0.998; ICC = 0.995) and outperformed three pathologists’ visual estimates (MAE 16.11, 8.18, 4.79 pp). Bland–Altman bias for the model was − 0.82 pp with tight limits of agreement (− 3.48 to 1.83 pp). Total inference time was 199.08 s for seven cases (28.44 s/image). WSI-based transformer pipeline enables rapid, auditable, and reproducible coronary stenosis measurement, reducing inter-observer variability and supporting standardized interpretation in forensic investigations.
No takes yet. Share an insight, caveat, or question.
Pigaiani et al. (2026) conducted an observational in Coronary artery stenosis in forensic autopsies (n=98). Deep learning-based tool (SegFormer-B0) vs. Visual estimation by forensic pathologists was evaluated on Correlation of predicted coronary stenosis percentage with ground truth on the test set (r 0.9975, 95% CI 0.9823-0.9996, p=<0.0001). A deep learning-based tool accurately quantified coronary artery stenosis on whole-slide images with a mean absolute error of 1.01 percentage points, outperforming visual estimates by pathologists.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: