Cardiologent: Multi-Agent System for Arrhythmia Decision Support
Researchers have developed Cardiologent, a multi-agent clinical decision support system for patient-level arrhythmia assessment. The system addresses the limitation of existing large language models paired with ECG data, which only analyze single recordings without assembling a patient-level finding. Cardiologent spans from arrhythmia detection to clinical decision-making, using agents for each signal, including a single ECG lead and photoplethysmography. The system is designed to interpret what an arrhythmia means for a specific patient and recommend management, moving beyond mere rhythm naming. This work was published on arXiv under ID 2607.25340.
Key facts
- Cardiologent is a multi-agent system for arrhythmia decision support.
- It addresses patient-level assessment beyond single ECG recordings.
- The system uses agents for ECG lead and photoplethysmography signals.
- It spans from detection to clinical decision-making.
- Published on arXiv with ID 2607.25340.
- Existing LLM-ECG systems stop short of patient-level findings.
- The same arrhythmia can lead to opposite decisions based on patient context.
- Cardiologent targets the full decision task from detection to management.
Entities
Institutions
- arXiv