From: A comparison of Covid-19 early detection between convolutional neural networks and radiologists
Training/validation | Test | ||||||
---|---|---|---|---|---|---|---|
Overall | Total | Covid19-pos | Covid19-neg | Total | Covid19-pos | Covid19-neg | |
Num patients | 6339 | 5555 | 3647 | 1908 | 784 | 296 | 488 |
Age (mean ± std) | 63 ± 18 | 63 ± 18 | 62 ± 17 | 63 ± 18 | 62 ± 18 | 62 ± 17 | 62 ± 19 |
Age (quantiles) | 64 (38–86) | 64 (38–86) | 64 (40–86) | 64 (37–87) | 64 (38–86) | 64 (38–83) | 64 (38–87) |
Female sex | 3188 (50%) | 2761(50%) | 1754(48%) | 1007(53%) | 427 (54%) | 151 (51%) | 276 (56%) |
AP, PA Chest radiographs per patient | 1 (1–4) | 1 (1–4) | 2 (1–5) | 1 (1–1) | 1 (1–1) | 1 (1–1) | 1 (1–1) |
Num. images | 12,599 | 9855 | 2744 | 946 | 347 | 599 | |
Num. images acquired with a mobile device | 3699 (29%) | 2970 (30%) | 729 (27%) | 230 (24%) | 78 (22%) | 152 (25%) |