Skip to main content
. 2022 Mar 18;9:854677. doi: 10.3389/fmed.2022.854677

Table 1.

Patient and image characteristics in the training and validation datasets.

Patients Normal training datasets
(n = 474)
Normal validation datasets
(n = 82)
UC training datasets
(n = 440)
UC validation datasets
(n = 72)
CD training datasets
(n = 444)
CD validation datasets
(n = 64)
Males, n
Age, y, mean (SD) (%)
238 (42.81%)
42.6 (13.5)
202 (39.45%)
41.8 (14.1)
192 (37.80%)
43.4 (14.7)
BBPS (mean, median, range) 2.85, 3, (1-3) 2.61, 3, (1-3) 2.50, 3, (1-3) 1.92, 2, (1-3) 2.4, 2, (1-3) 2.0, 2, (1-3)
Disease duration, y, mean (SD) 2.58 (1.24) 2.38 (1.11) 4.63 (2.35) 4.11 (1.17)
Disease manifestation
Diarrhea, n (%) 346 (78.64%) 61 (86.11%) 353 (79.50%) 43 (67.19%)
Abdominal pain, n (%) 322 (73.18%) 56 (77.78%) 262 (66.00%) 49 (76.56%)
CDAI, mean (SD) 277.56 (104.00) 284.06 (88.01)
SES-CD, mean (SD) 6.52 (2.52) 5.75 (2.66)
Mayo endoscopy score
1, n (%) 94 (21.36%) 10 (13.89%)
2, n (%) 48 (10.91%) 11 (15.28%)
3, n (%) 298 (67.73%) 51 (70.83%)
Number of images per patient, mean (SD) 23 (2.4) 21 (1.7) 21 (1.2)

UC, ulcerative colitis; CD, Crohn's disease, BBPS, Boston bowel preparation score; y, year; SD, standard deviation.