Radiology Report Human Evaluation on Open-I (test)
4.97Clarity Mean ScoreHuman
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| HumanReport Type=Human, n=362025.12 | 4.97 | 0.972 | 4.78 | 0.889 | 4.94 | 0.972 | 0.889 | |
| GPT-4 (YOLO-LLM pipeline)Report Type=GPT-4, n=582025.12 | 4.88 | 0.879 | 3.36 | 0.655 | 2.81 | 0.379 | 0.707 |