Table 4.
Evaluation of images closest to or furthest from the recognition of the 10 surgeon evaluators.
| Evaluator | Closest to your recognition | Furthest from your recognition | ||||
|---|---|---|---|---|---|---|
| Manual annotation | AI prediction | Manual annotation | AI prediction | |||
| Internal | External | Internal | External | |||
| Surgeon 1 | 9 | 5 | 6 | 2 | 3 | 9 |
| Surgeon 2 | 3 | 6 | 9 | 0 | 3 | 7 |
| Surgeon 3 | 5 | 2 | 14 | 6 | 7 | 3 |
| Surgeon 4 | 3 | 9 | 10 | 9 | 4 | 9 |
| Surgeon 5 | 5 | 11 | 8 | 5 | 4 | 16 |
| Surgeon 6 | 8 | 10 | 6 | 5 | 2 | 11 |
| Surgeon 7 | 6 | 2 | 12 | 3 | 12 | 8 |
| Surgeon 8 | 7 | 14 | 1 | 3 | 1 | 16 |
| Surgeon 9 | 2 | 3 | 19 | 6 | 10 | 3 |
| Surgeon 10 | 10 | 12 | 3 | 0 | 5 | 20 |
| Total | 1/10 | 4/10 | 5/10 | 0/10 | 3/10 | 7/10 |
Underlined numbers indicate the most votes among the 25 images.