From Prototypes to Sparse ECG Explanations: SHAP-Driven Counterfactuals for Multivariate Time-Series Multi-class Classification

TL;DR

This paper introduces a SHAP-driven framework for generating sparse, physiologically-aligned counterfactual explanations for multivariate ECG classification, improving interpretability and stability in clinical AI models.

Contribution

It presents a novel prototype-driven, SHAP-based method for creating sparse, physiologically-coherent counterfactual explanations tailored to ECG models, with real-time capabilities.

Findings

Counterfactuals modify 78% of signals while maintaining 81.3% validity

Achieves 43% improvement in temporal stability of explanations

Class-specific validity ranges from 98.9% for MI to 13.2% for HYP

Abstract

In eXplainable Artificial Intelligence (XAI), instance-based explanations for time series have gained increasing attention due to their potential for actionable and interpretable insights in domains such as healthcare. Addressing the challenges of explainability of state-of-the-art models, we propose a prototype-driven framework for generating sparse counterfactual explanations tailored to 12-lead ECG classification models. Our method employs SHAP-based thresholds to identify critical signal segments and convert them into interval rules, uses Dynamic Time Warping (DTW) and medoid clustering to extract representative prototypes, and aligns these prototypes to query R-peaks for coherence with the sample being explained. The framework generates counterfactuals that modify only 78% of the original signal while maintaining 81.3% validity across all classes and achieving 43% improvement in temporal stability. We evaluate three variants of our approach, Original, Sparse, and Aligned Sparse, with class-specific performance ranging from 98.9% validity for myocardial infarction (MI) to challenges with hypertrophy (HYP) detection (13.2%). This approach supports near realtime generation (< 1 second) of clinically valid counterfactuals and provides a foundation for interactive explanation platforms. Our findings establish design principles for physiologically-aware counterfactual explanations in AI-based diagnosis systems and outline pathways toward user-controlled explanation interfaces for clinical deployment.