Binary Disease Diagnosis on MIMIC-CXR
81Macro AccuracyCSRL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CSRL2026.05 | 81 | 81 | |
| CheXagent-8B2026.05 | 80.8 | 80.2 | |
| Best Tool per Pathology2026.05 | 80.4 | 81.2 | |
| Logistic Regression2026.05 | 79.1 | 80.5 | |
| InternVL3.5-38B-Instruct2026.05 | 79 | 71 | |
| Majority Voting (tie → pos)2026.05 | 77.9 | 77.9 | |
| Majority Voting (tie → neg)2026.05 | 77.4 | 76.2 | |
| Gemini-3.1-Pro-Preview2026.05 | 76.9 | 76 | |
| Hulu-Med-Flash-Preview-27B2026.05 | 76.7 | 76.7 | |
| CheXzero2026.05 | 74.6 | 75.6 | |
| MedKLIP2026.05 | 73.7 | 76.2 | |
| MedGemma-27B-IT2026.05 | 72.1 | 73.3 | |
| Qwen2.5-VL-72B-Instruct2026.05 | 71.3 | 70.3 | |
| Lingshu-7B2026.05 | 71.1 | 75.1 | |
| Claude-Sonnet-4.62026.05 | 70.9 | 70.8 | |
| DenseNet-CheX2026.05 | 70.4 | 71.1 | |
| DenseNet-MIMIC2026.05 | 70.1 | 68.6 | |
| GPT-4o2026.05 | 69.9 | 67.2 | |
| HuatuoGPT-Vision-7B2026.05 | 69.8 | 66.7 | |
| DenseNet-NIH2026.05 | 66.5 | 68.8 | |
| Qwen2.5-VL-7B-Instruct2026.05 | 53.9 | 37.1 |