Table 1 Baseline characteristics of the development dataset, internal and external test datasets
From: Deep learning approach for screening neonatal cerebral lesions on ultrasound in China
Parameter | Development dataset | Internal test set | External test dataset | ||||
|---|---|---|---|---|---|---|---|
Train | Test | P value | SCMCH | CSMCH | GZMCH | ||
Count | 1206 (79.4%) | 312 (20.6%) | – | 199 | 109 (30.6%) | 117 (32.8%) | 130 (36.5%) |
Sex | |||||||
Male | 726 (60.2%) | 194 (62.2%) | 0.567 | 140 (70.3%) | 61 (56.0%) | 67 (57.3%) | 62 (47.7%) |
Female | 480 (39.8%) | 118 (37.8%) | 59 (29.7%) | 48 (44.0%) | 50 (42.7%) | 68 (52.3%) | |
GA (week) | 34.7 (25.1– 40.7) | 36.4 (26.9– 41.4) | 0.979 | 36.1 (27.2– 41.0) | 35.3 (24.9– 40.3) | 38.7 (27.3–40.6) | 36.7 (26.3– 41.0) |
BW (gram) | 2200.0 (686.5– 4017.0) | 2420.0 (891.0– 4024.0) | 0.031 | 2505.0 (710.0– 4021.8) | 2280.0 (750.0– 3474.4) | 3000.0 (1050.0– 4050.0) | 2470.0 (743.5–3879.0) |
Age (day) | 12.0 (1.0– 148.4) | 13.0 (1.0– 177.2) | 0.744 | 11.0 (1.0– 47.2) | 2.0 (1.0– 54.2) | 6.0 (1.0– 30.4) | 2.5 (1.0– 22.3) |
Delivery | |||||||
CS | 665 (55.1%) | 174 (55.8%) | 0.937 | 122 (61.3%) | 79 (72.5%) | 45 (38.5%) | 66 (50.8%) |
VD | 541 (44.9%) | 138 (44.2%) | 77 (38.7%) | 30 (27.5%) | 72 (61.5%) | 64 (49.2%) | |
Apgar score | |||||||
1 min | 9.0 (1.00– 10.0) | 9.0 (1.0– 10.0) | 0.764 | 9.0 (3.0– 10.0) | 10.0 (3.6– 10.0) | 10.0 (2.0– 10.0) | 9.0 (2.1– 10.0) |
5 min | 10.0 (4.0– 10.0) | 10.0 (2.6– 10.0) | 0.631 | 10.0 (5.0– 10.0) | 10.0 (8.0– 10.0) | 10.0 (7.0– 10.0) | 10.0 (5.0– 10.0) |
10 min | 10.0 (5.0– 10.0) | 10.0 (3.0– 10.0) | 0.716 | 10.0 (6.0– 10.0) | 10.0 (8.4– 10.0) | 5.0 (5.00– 7.00) | 10.0 (5.3– 10.0) |
