Overall Clinical Outcome Prediction on Real Clinical Data (n=35 patients) 1.0 (test)
75.7AccuracyTheraAgent
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TheraAgentBase LLM=GPT-4o, Method Category=Medical AI Agents2026.03 | 75.7 | 82.1 | |
| GPT-4o + RAGBase LLM=GPT-4o, Method Category=Single LLM2026.03 | 65.7 | 68.3 | |
| MDAgentsBase LLM=GPT-4o, Method Category=Medical AI Agents2026.03 | 62.9 | 68.6 | |
| MedAgent-ProBase LLM=GPT-4o, Method Category=Medical AI Agents2026.03 | 57.1 | 62.8 | |
| MedAgentsBase LLM=GPT-4o, Method Category=Medical AI Agents2026.03 | 54.3 | 67.3 | |
| GPT-4o + CoTBase LLM=GPT-4o, Method Category=Single LLM2026.03 | 52.9 | 54 | |
| GPT-4oBase LLM=GPT-4o, Method Category=Single LLM2026.03 | 48.6 | 37.3 | |
| GPT-4o + Self-ConsistencyBase LLM=GPT-4o, Method Category=Single LLM2026.03 | 47.1 | 39.6 |