Table 1 Cohort Characteristics after various Sampling Routines
From: Deep learning predicts hip fracture using confounding patient and healthcare variables
Cohort | cs-train | cs-test | cc-Random | cc-Dem | cc-Pt | cc-PtHp |
|---|---|---|---|---|---|---|
Sampling | Cross-sectional | Cross-sectional | Case–control | Case–control | Case–control | Case–control |
Matching | NA | NA | NA | Age + gender | PT | PT + HP |
Partition | Train | Test | Test | Test | Test | Test |
No. of radiographs | 17,587 | 5970 | 416 | 405 | 416 | 411 |
No. of patients | 6768 | 2256 | 275 | 252 | 217 | 186 |
No. of scanners | 11 | 11 | 10 | 9 | 8 | 6 |
No. of scanner manufacturers | 4 | 4 | 4 | 4 | 4 | 4 |
Age, mean (SD), years | 61 (22) | 61 (22) | 67 (24) | 75 (20) | 75 (21) | 74 (19) |
Female frequency, no. (%) | 11,647 (66) | 3873 (65) | 260 (62) | 249 (61) | 263 (63) | 253 (62) |
Fracture frequency, no. (%) | 572 (3) | 207 (3) | 207 (50) | 207 (51) | 207 (50) | 207 (50) |
BMI, mean (SD) | 28 (7) | 28 (7) | 25 (5) | 25 (5) | 24 (5) | 24 (4) |
Fall frequency, no. (%) | 3214 (18) | 1139 (19) | 133 (32) | 160 (40) | 174 (42) | 165 (40) |
Pain frequency, no. (%) | 9010 (51) | 2960 (50) | 164 (39) | 137 (34) | 117 (28) | 104 (25) |
