Table 6.
Number of Positive Scans, Eyes, and Subjects versus the Total Cases, as Defined by the Consensus of the Two Experts on Dataset A (Training) and B (Testing)
| NM | MH* | ME | AMD | |
|---|---|---|---|---|
| Training statistics | ||||
| Scan | 80/316 | 49/287 | 190/287 | 59/295 |
| Eye | 66/187 | 27/176 | 109/180 | 27/178 |
| Subject | 65/133 | 26/128 | 84/130 | 21/133 |
| Testing statistics | ||||
| Scan | 22/124 | 21/153 | 59/118 | 81/111 |
| Eye | 13/54 | 8/66 | 29/54 | 31/50 |
| Subject | 10/36 | 6/43 | 23/34 | 20/33 |
The consensus includes 96.9%, 95.4%, 88.0%, and 90.5% of 326 scans from dataset A, and 94.7%, 100%, 90.0%, and 84.7% of 131 scans in dataset B. In each cell, the number of positive cases versus the total cases is shown.
For the MH category, unfortunately, dataset B did not contain MH cases, which coincides with real clinical situations, since MH has a very low prevalence rate. Therefore, for MH diagnosis performance testing only, the training and testing datasets were organized in a way that 80% of MH cases originating from dataset A were randomly sampled and included in the training set, and the rest (six subjects) were included in the testing set.