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

2026-05-19 · Source: cs.LG updates on arXiv.org · Field: Health & Wellbeing — Medical Specialties & Subspecialties, Medical Devices & Health Technology, Health & Medical Research · Depth: Expert, quick

Summary

DeepArrhythmia is a novel multimodal framework designed for segment-contextualized beat-level Electrocardiography (ECG) arrhythmia classification. It addresses the limitation of existing systems that treat individual beats in isolation, by incorporating multi-beat rhythm context, including timing, compensatory pauses, and beat-to-beat morphological consistency. The framework processes multi-beat ECG segments, combining raw ECG signals with rendered waveform images. It localizes R peaks to identify beat instances and generates structured beat-level predictions. DeepArrhythmia decouples physiological measurement from evidence integration, employing specialized tools for beat localization, numerical rhythm-morphology extraction, and morphology-focused textual analysis. Its agentic design utilizes segment-level confidence to route between minimal and rich evidence states, selectively acquiring physiological evidence for improved decision-making.

Key takeaway

For AI Scientists developing medical diagnostic tools, DeepArrhythmia demonstrates the critical need to move beyond isolated data points and incorporate contextual information. Your models for ECG arrhythmia detection should explicitly account for multi-beat rhythm and morphology, rather than treating beats as independent events. Consider implementing selective evidence acquisition to optimize processing and improve diagnostic accuracy, as richer evidence is not always beneficial.

Key insights

DeepArrhythmia improves ECG arrhythmia classification by integrating multi-beat context and selective evidence acquisition.

Principles

• Beat labels depend on multi-beat rhythm context.
• Richer physiological evidence is not uniformly useful.

Method

DeepArrhythmia combines raw ECG and waveform images, localizes R peaks, and uses specialized tools for rhythm-morphology extraction and textual analysis, routing evidence based on segment-level confidence.

In practice

• Integrate multi-beat context for ECG analysis.
• Use multimodal data (signal + image) for diagnosis.

Topics

• DeepArrhythmia
• ECG Arrhythmia Classification
• Beat-level Detection
• Multimodal Framework
• R Peak Localization

Best for: AI Scientist, Machine Learning Engineer, Research Scientist

Related on AIssential

• Echo2ECG: Enhancing ECG Representations with Cardiac Morphology from Multi-View Echos — Aligning global ECGs with multi-view Echos resolves representational mismatch, yielding powerful, lightweight cardiac morphology features.
• A hidden predictor of sudden cardiac death uncovered by deep learning — Deep learning identifies novel ECG features to predict sudden cardiac death risk more accurately than current methods.
• PULSE: Privileged Knowledge Transfer from Rich to Deployable Sensors for Embodied Multi-Sensory Learning — PULSE transfers privileged EDA knowledge to low-cost sensors, improving stress detection without requiring EDA at inference.
• EVL-ECG: Efficient ECG Interpretation With Multi-Aspect Heterogeneous Knowledge Distillation — EVL-ECG enables efficient, accurate ECG interpretation by distilling knowledge across diverse model architectures.
• SL-S4Wave: Self-Supervised Learning of Physiological Waveforms with Structured State Space Models — SL-S4Wave uses tailored S4 models and contrastive learning for robust, label-efficient physiological waveform analysis.
• Evidential learning driven Breast Tumor Segmentation with Stage-divided Vision-Language Interaction — TextBCS enhances breast tumor segmentation in MRI by combining text prompts with evidential learning to address low contrast and blurred boundaries.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by cs.LG updates on arXiv.org.