Table 4.
Prediction of the CAS by the ML-assisted system using the test set constructed from the entire dataset.
| Predicted CAS | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Reference CAS | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | |
| 0 | 214 | 156 | 101 | 16 | 4 | 3 | 0 | 0 | |
| 1 | 253 | 733 | 589 | 144 | 41 | 2 | 0 | 0 | |
| 2 | 74 | 349 | 713 | 324 | 86 | 20 | 1 | 0 | |
| 3 | 21 | 84 | 219 | 308 | 227 | 54 | 1 | 0 | |
| 4 | 7 | 49 | 63 | 178 | 161 | 86 | 19 | 2 | |
| 5 | 0 | 1 | 12 | 29 | 42 | 44 | 25 | 11 | |
| 6 | 0 | 0 | 0 | 0 | 2 | 1 | 17 | 0 | |
| 7 | 0 | 0 | 0 | 0 | 0 | 1 | 3 | 7 | |
All of 5497 test results from 30 repeated experiments were presented. The predicted CAS was generated using the ML-assisted system, and the reference CAS was evaluated by three ophthalmologists using the patients’ facial images.
CAS clinical activity score, ML machine learning.