Title : Hierarchical Multi-Instance Learning for Automated Screening and Subtype Classification of Congenital Heart Disease From Echocardiography Videos
Abstract:
Background: Transthoracic echocardiography is the first-line test for congenital heart disease (CHD), but accurate screening and lesion subtyping require expertise and synthesis across multiple heterogeneous cine clips. We developed a video-based deep learning approach for examination-level CHD screening and subtype classification of common left-to-right shunt lesions.
Methods: We retrospectively assembled 2,373 echocardiography examinations (800 normal; 493 VSD, 751 ASD, and 329 PDA) from The First Affiliated Hospital of Shandong First Medical University. Each examination comprised multiple cine clips acquired across standard views. We proposed Echocardiography Hierarchical Multiple-Instance Learning network (EchoHMIL), which encodes clips with a spatiotemporal backbone and aggregates variable numbers of clips using attention-based multi-instance learning to form an examination-level representation. A hierarchical dual-head design was optimized with a gated multi-task objective to perform: (i) normal-versus-defect screening, and (ii) VSD/ASD/PDA subtype classification conditional on defect status. Data were split at the patient level (70%/10%/20%) and evaluated on an independent test set using AUC, sensitivity, specificity, accuracy, and macro-F1, with bootstrap 95% confidence intervals.
Results: On the test set (n = 475), EchoHMIL achieved an AUC of 0.957 for normal-versus-defect screening. At a sensitivity-prioritized operating point, sensitivity was 92.2% and specificity was 82.4%. For VSD/ASD/PDA subtype classification among defect cases, EchoHMIL achieved an overall accuracy of 88.8% with a macro-F1 of 0.885. Attention weights and gradient-based saliency maps highlighted clinically plausible regions associated with septal and ductal anatomy.
Conclusions: EchoHMIL enables automated, examination-level CHD screening and shunt-lesion subtyping from routine echocardiography videos. This clinically aligned hierarchical MIL
framework may support triage and improve diagnostic consistency, warranting external and prospective validation.
Keywords: congenital heart disease; deep learning; hierarchical classification; multiple-instance learning
