ProtoECGNet: Case-Based Interpretable Deep Learning for Multi-Label ECG Classification with Contrastive Learning

2025-04-11

AI Summary

Existing ECG automatic classification models lack clinical interpretability, particularly for multi-label diagnostic scenarios. Method: We propose a prototype-based learning framework tailored to multi-label ECG diagnosis. It introduces the first contrastive prototype loss for multi-label settings—jointly optimizing inter-class separation, co-occurring diagnosis clustering, and prototype diversity. We extend structured multi-branch prototype networks to the full PTB-XL dataset (71 labels), integrating 1D/2D CNNs with global-local prototype embeddings. Results: Our model achieves state-of-the-art black-box performance across all 71 diagnostic labels. Clinical expert evaluation confirms that learned prototypes are highly representative of real ECG cases and intuitively interpretable. This work represents the first high-accuracy, multi-label, prototype-based ECG diagnosis system grounded in authentic waveform segments, delivering transparent, case-driven decision support for rhythm, morphological, and diffuse abnormalities.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Multi-modal Vision
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Deep learning-based electrocardiogram (ECG) classification has shown impressive performance but clinical adoption has been slowed by the lack of transparent and faithful explanations. Post hoc methods such as saliency maps may fail to reflect a model's true decision process. Prototype-based reasoning offers a more transparent alternative by grounding decisions in similarity to learned representations of real ECG segments, enabling faithful, case-based explanations. We introduce ProtoECGNet, a prototype-based deep learning model for interpretable, multi-label ECG classification. ProtoECGNet employs a structured, multi-branch architecture that reflects clinical interpretation workflows: it integrates a 1D CNN with global prototypes for rhythm classification, a 2D CNN with time-localized prototypes for morphology-based reasoning, and a 2D CNN with global prototypes for diffuse abnormalities. Each branch is trained with a prototype loss designed for multi-label learning, combining clustering, separation, diversity, and a novel contrastive loss that encourages appropriate separation between prototypes of unrelated classes while allowing clustering for frequently co-occurring diagnoses. We evaluate ProtoECGNet on all 71 diagnostic labels from the PTB-XL dataset, demonstrating competitive performance relative to state-of-the-art black-box models while providing structured, case-based explanations. To assess prototype quality, we conduct a structured clinician review of the final model's projected prototypes, finding that they are rated as representative and clear. ProtoECGNet shows that prototype learning can be effectively scaled to complex, multi-label time-series classification, offering a practical path toward transparent and trustworthy deep learning models for clinical decision support.

Problem

• Lack of transparent explanations in deep learning ECG classification
• Need for faithful case-based interpretability in multi-label ECG diagnosis
• Challenges in scaling prototype learning for complex time-series classification

Innovation

• 1D and 2D CNNs with global and localized prototypes
• Multi-branch architecture reflecting clinical workflows
• Contrastive loss for multi-label prototype separation