ECG-Reasoning-Benchmark: A Benchmark for Evaluating Clinical Reasoning Capabilities in ECG Interpretation

TL;DR

This paper introduces ECG-Reasoning-Benchmark, a comprehensive evaluation framework revealing that current multimodal models lack genuine step-by-step reasoning in ECG interpretation, often relying on superficial cues instead of true visual understanding.

Contribution

The paper presents a new benchmark with over 6,400 samples to systematically assess reasoning in ECG diagnosis, exposing significant gaps in current models' logical deduction capabilities.

Findings

Models have high medical knowledge but poor reasoning chain completion (6%).

Models fail to ground ECG findings to visual evidence.

Current training paradigms do not foster true visual reasoning in ECG interpretation.

Abstract

While Multimodal Large Language Models (MLLMs) show promising performance in automated electrocardiogram interpretation, it remains unclear whether they genuinely perform actual step-by-step reasoning or just rely on superficial visual cues. To investigate this, we introduce \textbf{ECG-Reasoning-Benchmark}, a novel multi-turn evaluation framework comprising over 6,400 samples to systematically assess step-by-step reasoning across 17 core ECG diagnoses. Our comprehensive evaluation of state-of-the-art models reveals a critical failure in executing multi-step logical deduction. Although models possess the medical knowledge to retrieve clinical criteria for a diagnosis, they exhibit near-zero success rates (6% Completion) in maintaining a complete reasoning chain, primarily failing to ground the corresponding ECG findings to the actual visual evidence in the ECG signal. These results demonstrate that current MLLMs bypass actual visual interpretation, exposing a critical flaw in existing training paradigms and underscoring the necessity for robust, reasoning-centric medical AI. The code and data are available at https://github.com/Jwoo5/ecg-reasoning-benchmark.