Pediatric appendicitis diagnosis on Internal Regensburg cohort
85.1AccuracyClaMPAPP
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| ClaMPAPP2026.06 | 85.1 | 97.7 | 75.9 | 75 | 84.8 | 84 | 88 | 28 | 2 | |
| OpenAI o3-pro2026.06 | 75.2 | 91.9 | 62.9 | 64.8 | 76 | 79 | 73 | 43 | 7 | |
| OpenAI GPT-5.52026.06 | 74.8 | 80.2 | 70.7 | 66.9 | 73 | 69 | 82 | 34 | 17 | |
| MedGemma-4b-it2026.06 | 73.8 | 53.5 | 88.8 | 78 | 63.4 | 46 | 103 | 13 | 40 | |
| Anthropic Claude Opus 4.72026.06 | 69.3 | 58.1 | 77.6 | 65.8 | 61.7 | 50 | 90 | 26 | 36 | |
| Llama-3.1-8b-instant2026.06 | 66.3 | 75.6 | 59.5 | 58 | 65.7 | 65 | 69 | 47 | 21 | |
| OASST-Pythia-12b2026.06 | 43.6 | 98.8 | 2.6 | 42.9 | 59.9 | 85 | 3 | 113 | 1 |