Ecg scroll streams long heart monitor data for real time event detection
ECG-Scroll: A Long-Horizon, Streaming Benchmark and Agent Environment for Interpretation of Ambulatory Electrocardiograms
Machine LearningArtificial IntelligenceComputation and Language
Summary
Long heart monitor recordings are hard to analyze because important events happen briefly and unpredictably over many hours. The authors treat this analysis like a continuous decision-making process where a program watches the data as it arrives and must detect problems quickly without future information. They created ECG-Scroll, a new test setup where programs get to measure, remember, and decide what to check in these long signals in real time. They also provide a large set of real recordings to train and evaluate such programs, showing how well simple rule-based and language model-based methods perform.
What this means in practice
- •For cardiology device developers: Build real-time heart monitoring software that detects brief arrhythmia episodes as they occur in long, streaming ECG data.$Commercial implications: Enables ambulatory ECG device makers to develop responsive diagnostic tools leveraging continuous data interpretation for timely medical alerts.
- •For healthcare data engineers: Develop pipelines to process and analyze long-term two-lead ECG streams using tools that integrate memory and planning for event detection.
Authors
Haitao Li, Chenglin Li, Zhengyao Ding, Ziyu Li, Yiheng Mao, Zhengxing Huang
Abstract
Multimodal large language models (MLLMs) can now interpret a standard ten-second, twelve-lead electrocardiogram (ECG) with clinically grounded, reward-verified reasoning. Real cardiac monitoring is different. Ambulatory (Holter) and telemetry recordings span hours to days and are read as they stream in, and their clinically decisive findings are paroxysmal, brief episodes buried in an otherwise unremarkable trace. Such a recording cannot be held in one context at diagnostic resolution, and its future has not yet happened, so a reader must work online, deciding what to measure now, committing evidence to memory as it passes, and reporting events as they occur. We recast long-duration ECG interpretation as a long-horizon, online (streaming, causal) sequential decision process and introduce ECG-Scroll. As a benchmark, long ambulatory recordings are streamed to an agent chunk by chunk, and it must localize, quantify, and promptly flag paroxysmal events without access to future signal; because the underlying signal is retained, every answer is checkable against objective ground truth, giving rule-based rather than judge-based rewards, and the streaming formulation adds a metric batch evaluation cannot express, the detection latency between an event's onset and the moment the agent records it. As an agent environment, it is a fixed, gym-style interaction layer that exercises three competencies single-glance ECG models never touch: Memory, Tool use through signal-grounded measurement rather than reading pixels, and Planning of what to measure now and when to commit. We release 390 whole-recording instances spanning 2,536 hours of two-lead ambulatory ECG and evaluate a signal-threshold rule agent alongside off-the-shelf LLM agents online, characterizing how they use memory, tools, and planning and where the benchmark's head-room lies.