TABLE 1.
Summary of the study subjects and data distribution.
| Total | yes GON and yes MF | no GON and yes MF | yes GON and no MF | no GON and no MF | p-value | |
| Training dataset | ||||||
| No. of SDOCT scans | 3,919 | 1,679 | 890 | 629 | 721 | \ |
| No. of patients | 951 | 390 | 216 | 166 | 179 | \ |
| Gender (male/female) | 402/549 | 183/207 | 83/133 | 72/94 | 64/115 | 0.047 |
| Age, years (mean ± SD) | 65.6 ± 12.6 | 67.3 ± 12.9 | 64.7 ± 12.3 | 66.6 ± 11.7 | 62.4 ± 12.2 | < 0.001 |
| No. of eyes | 1,196 | 495 | 280 | 192 | 229 | \ |
| Eye (right/left) | 579/617 | 242/253 | 142/138 | 90/102 | 105/124 | 0.70 |
| Tuning dataset | ||||||
| No. of SDOCT scans | 460 | 195 | 163 | 32 | 70 | \ |
| No. of patients | 132 | 56 | 43 | 13 | 20 | \ |
| Gender (male/female) | 52/80 | 22/34 | 16/27 | 8/5 | 6/14 | 0.32 |
| Age, years (mean ± SD) | 64.7 ± 12.7 | 66.3 ± 13.2 | 64.0 ± 13.0 | 63.8 ± 13.0 | 66.6 ± 10.3 | 0.66 |
| No. of eyes | 180 | 83 | 51 | 17 | 29 | \ |
| Eye (right/left) | 89/91 | 39/44 | 27/24 | 8/9 | 15/14 | 0.91 |
| Internal validation dataset | ||||||
| No. of SDOCT scans | 454 | 205 | 114 | 36 | 99 | \ |
| No. of patients | 143 | 63 | 38 | 19 | 23 | \ |
| Gender (male/female) | 71/72 | 32/31 | 17/21 | 9/10 | 13/10 | 0.83 |
| Age, years (mean ± SD) | 67.0 ± 12.2 | 68.9 ± 12.6 | 63.3 ± 13.4 | 70.6 ± 11.1 | 65.1 ± 7.7 | 0.06 |
| No. of eyes | 196 | 96 | 44 | 22 | 34 | \ |
| Eye (right/left) | 103/93 | 51/45 | 25/19 | 9/13 | 18/16 | 0.68 |
| External testing dataset 1 | ||||||
| No. of SDOCT scans | 545 | 215 | 125 | 79 | 126 | \ |
| No. of patients | 307 | 114 | 79 | 47 | 67 | \ |
| Gender (male/female) | 133/174 | 55/59 | 35/44 | 19/28 | 24/43 | 0.42 |
| Age, years (mean ± SD) | 68.6 ± 11.6 | 70.4 ± 11.8 | 67.3 ± 11.6 | 72.6 ± 10.6 | 64.2 ± 10.4 | < 0.001 |
| No. of eyes | 417 | 156 | 104 | 63 | 94 | \ |
| Eye (right/left) | 211/206 | 80/76 | 55/49 | 28/35 | 48/46 | 0.75 |
| External testing dataset 2 | ||||||
| No. of SDOCT scans | 267 | 74 | 61 | 69 | 63 | \ |
| No. of patients | 193 | 52 | 48 | 51 | 42 | \ |
| Gender (male/female) | 97/96 | 30/22 | 23/25 | 23/28 | 21/21 | 0.62 |
| Age, years (mean ± SD) | 60.5 ± 12.4 | 66.6 ± 7.1 | 55.6 ± 16.8 | 62.0 ± 9.6 | 56.5 ± 11.3 | < 0.001 |
| No. of eyes | 261 | 70 | 61 | 69 | 61 | \ |
| Eye (right/left) | 137/124 | 39/31 | 33/28 | 31/38 | 34/27 | 0.53 |
| External testing dataset 3 | ||||||
| No. of SDOCT scans | 515 | 296 | 101 | 66 | 52 | \ |
| No. of patients | 294 | 156 | 66 | 41 | 31 | \ |
| Gender (male/female) | 168/126 | 95/61 | 37/29 | 22/19 | 14/17 | 0.40 |
| Age, years (mean ± SD) | 62.8 ± 12.5 | 64.2 ± 10.9 | 60.4 ± 14.0 | 65.9 ± 10.8 | 57.2 ± 16.4 | 0.004 |
| No. of eyes | 428 | 243 | 86 | 54 | 45 | \ |
| Eye (right/left) | 213/215 | 121/122 | 44/42 | 26/28 | 22/23 | 0.99 |
| External testing dataset 4 | ||||||
| No. of SDOCT scans | 933 | 477 | 57 | 241 | 158 | \ |
| No. of patients | 249 | 108 | 19 | 58 | 64 | \ |
| Gender (male/female) | 109/136 | 53/54 | 11/8 | 22/35 | 23/39 | 0.40 |
| Age, years (mean ± SD) | 66.0 ± 14.9 | 70.2 ± 13.6 | 64.5 ± 11.8 | 67.0 ± 16.6 | 58.5 ± 13.3 | < 0.001 |
| No. of eyes | 399 | 187 | 25 | 92 | 95 | \ |
| Eye (right/left) | 205/194 | 96/91 | 13/12 | 48/44 | 48/47 | 0.99 |
| External testing dataset 5 | ||||||
| No. of SDOCT scans | 1058 | 287 | 171 | 222 | 378 | \ |
| No. of patients | 272 | 65 | 56 | 53 | 98 | \ |
| Gender (male/female) | 155/117 | 43/22 | 32/24 | 35/18 | 45/53 | 0.02 |
| Age, years (mean ± SD) | 68.6 ± 8.5 | 67.5 ± 8.9 | 70.6 ± 8.4 | 69.2 ± 8.3 | 67.9 ± 8.4 | 0.17 |
| No. of eyes | 532 | 126 | 88 | 119 | 199 | \ |
| Eye (right/left) | 263/269 | 61/65 | 56/32 | 52/67 | 94/105 | 0.80 |
GON, glaucomatous optic neuropathy; MF, myopic features.
Training, Tuning, and Internal validation: CUHK Eye Centre and Hong Kong Eye Hospital (HKEH), Hong Kong SAR.
External testing dataset 1: Prince of Wales Hospital (PWH), Hong Kong SAR.
External testing dataset 2: Tuen Mun Eye Centre (TMEC), Hong Kong SAR.
External testing dataset 3: Alice Ho Miu Ling Nethersole Hospital (AHNH), Hong Kong SAR.
External testing dataset 4: Byers Eye Institute, Stanford University (Stanford), the United States.
External testing dataset 5: Singapore Eye Research Institute (SERI), Singapore.
One-way ANOVA and χ2 test were used for numerical and categorical data, respectively for comparison between “yes GON and yes MF,” “yes GON and no MF,” “no GON and yes MF,” and “no GON and no MF” groups. p-values in bold were AUROC values with significant difference.