DeepArrhythmia: Segment-Contextualized ECG Arrhythmia Classification via Selective Evidence Acquisition

TL;DR

DeepArrhythmia is a multimodal ECG arrhythmia classification framework that leverages segment context, beat localization, and selective evidence to improve beat-level predictions.

Contribution

It introduces a novel agentic, multimodal approach that combines raw signals and waveform images with selective evidence acquisition for ECG classification.

Findings

Effective multi-beat rhythm context utilization

Improved beat-level classification accuracy

Segment-level confidence routing enhances decision quality

Abstract

Beat-level Electrocardiography (ECG) arrhythmia detection aims to assign an arrhythmia class to each beat in a recording, yet many existing systems treat beats as isolated local instances. This is limiting because beat labels often depend on multi-beat rhythm context, including timing, compensatory pauses, and beat-to-beat morphological consistency. We present DeepArrhythmia, a tool-grounded multimodal framework for segment-contextualized beat-level ECG arrhythmia classification. Given a multi-beat ECG segment, DeepArrhythmia combines the raw ECG signal and a rendered waveform image, localizes R peaks to identify beat instances, and produces structured beat-level predictions. The framework decouples physiological measurement from evidence integration using specialized tools for beat localization, numerical rhythm--morphology extraction, and morphology-focused textual analysis. DeepArrhythmia uses segment-level confidence to route between minimal and rich evidence states, since richer physiological evidence is not uniformly useful. This agentic design integrates rhythm context, explicit physiological grounding, and selective evidence acquisition for decision making.