Table 1 Data characteristics including patient demographics, image properties, and data sources. Values with a plus/minus sign represent the means ± standard deviation. Note that the values of age and sex in the external testing set are not provided because they were de-identified before data collection. \(\chi ^{2}\) test was used for P-value of categorical variables such as label, sex, and manufacturer. Analysis of variance (marked as \(\dagger \)) was used for age.
From: Deep learning based dual stage model for accurate nasogastric tube positioning in chest radiographs
Variable | Training & validation dataset | Internal testing dataset | External testing dataset | P-value |
|---|---|---|---|---|
Total (n) | 1799 | 450 | 378 | |
Label (n, %) | 0.12 | |||
Complete | 1566 (87.1) | 391 (86.9) | 314 (83.1) | |
Incomplete | 233 (12.9) | 59 (13.1) | 64 (16.9) | |
Sex (n, %) | 0.77 | |||
Male | 974 (54.1) | 275 (61.1) | – | |
Female | 600 (33.4) | 175 (38.9) | – | |
Unknown | 225 (12.5) | – | – | |
Age (years, mean ± SD) | \(77.50 \pm 11.20\) | \(77.51 \pm 10.96\) | - | \(0.98^{\dagger }\) |
Manufacturer (n, %) | 0.0001 | |||
Canon | 608 (33.8) | 176 (39.1) | – | |
DongKang | 582 (32.3) | 166 (36.9) | – | |
DK Medical Systems | 156 (8.7) | 44 (9.8) | – | |
Samsung | 66 (3.7) | 21 (4.7) | 172 (45.5) | |
SIEMENS | – | – | 158 (41.8) | |
Other | 387 (21.5) | 43 (9.6) | 48 (12.7) |
