Risk Stratification on Retrospective multimodal cardiovascular clinical decision-making dataset (test)
0.866AUROCHetMedAgent
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| HetMedAgentSystem Type=Multi-Agent Systems, oversight=w/o Clinician2026.05 | 0.866 | 84.4 | |
| MedAgentsSystem Type=Multi-Agent Systems2026.05 | 0.823 | 78.9 | |
| AgentClinicSystem Type=Multi-Agent Systems2026.05 | 0.817 | 78.1 | |
| MetaGPTSystem Type=Multi-Agent Systems2026.05 | 0.811 | 77.3 | |
| AutoGenSystem Type=Multi-Agent Systems2026.05 | 0.804 | 76.5 | |
| MeditronSystem Type=Medical LLMs2026.05 | 0.801 | 76.8 | |
| Llama-3-MeditronSystem Type=Medical LLMs2026.05 | 0.795 | 75.6 | |
| PMC-LLaMASystem Type=Medical LLMs2026.05 | 0.782 | 74.5 | |
| BioMistralSystem Type=Medical LLMs2026.05 | 0.776 | 73.1 | |
| DoctorGLMSystem Type=Medical LLMs2026.05 | 0.768 | 72.4 | |
| GatorTronSystem Type=Medical LLMs2026.05 | 0.753 | 70.8 | |
| HetMedAgentSystem Type=Multi-Agent Systems, oversight=w/o Clinician2026.05 | 0.727 | 71.9 | |
| MedAgentsSystem Type=Multi-Agent Systems2026.05 | 0.692 | 65.3 | |
| AgentClinicSystem Type=Multi-Agent Systems2026.05 | 0.681 | 64.1 | |
| MeditronSystem Type=Medical LLMs2026.05 | 0.673 | 63.4 | |
| MetaGPTSystem Type=Multi-Agent Systems2026.05 | 0.673 | 63.2 | |
| AutoGenSystem Type=Multi-Agent Systems2026.05 | 0.665 | 62.4 | |
| Llama-3-MeditronSystem Type=Medical LLMs2026.05 | 0.658 | 61.9 | |
| PMC-LLaMASystem Type=Medical LLMs2026.05 | 0.641 | 59.8 | |
| BioMistralSystem Type=Medical LLMs2026.05 | 0.628 | 58.2 | |
| DoctorGLMSystem Type=Medical LLMs2026.05 | 0.614 | 57.1 | |
| GatorTronSystem Type=Medical LLMs2026.05 | 0.597 | 55.3 |