VitalAgent: A Tool-Augmented Agent for Reactive and Proactive Physiological Monitoring over Wearable Health Data

2026-05-28

AI Summary

Existing mobile health systems struggle to perform temporal reasoning and proactive monitoring on long-term ECG/PPG signals collected from wearable devices, often limited to static summaries or task-specific predictions. This work proposes VitalAgent—the first tool-augmented agent framework capable of both reactive question answering and proactive alerting—integrating longitudinal physiological memory, dynamic signal computation interfaces, and a collaborative mechanism with large language models. Concurrently, we introduce VitalBench, the first longitudinal benchmark dataset pairing continuous physiological signals with question-answer pairs. Experimental results demonstrate that VitalAgent outperforms prompt-based and ReAct baselines by over 30% on reactive question-answering tasks and achieves, for the first time, effective proactive monitoring of long-term physiological signals.

Abstract

Wearable devices enable continuous monitoring of physiological signals such as ECG and PPG, but existing mHealth systems are largely limited to task-specific prediction pipelines or reactive question answering over static summaries. They lack the ability to support temporal reasoning, persistent physiological context, and proactive monitoring over long-term signal streams. We propose VitalAgent, a tool-augmented agentic framework for ECG/PPG-based mHealth that supports both reactive question answering and proactive monitoring. VitalAgent is built on a longitudinal physiological memory and a tool-augmented reasoning interface that enables dynamic computation over raw signals. We further introduce VitalBench, a longitudinal physiological monitoring benchmark dataset comprising 1,862 QA pairs for reactive question answering and 90.2 hours of continuous ECG/PPG recordings for proactive monitoring, covering cardiac, physical activity, and stress-related tasks. Experiments demonstrate that VitalAgent achieves over 30% improvement over prompt-based and ReAct baselines in reactive evaluation and supports proactive alert monitoring over long-term physiological signals, highlighting the importance of dynamic tool use and long-term physiological monitoring.

Problem

• physiological monitoring
• wearable health data
• temporal reasoning
• proactive monitoring
• mHealth

Innovation

• tool-augmented agent
• longitudinal physiological monitoring
• dynamic signal reasoning
• proactive health alerting
• wearable ECG/PPG