Table 1 Clinical characteristics of patients in deep-learning set.

From: Predicting multiple linear stapler firings in double stapling technique with an MRI-based deep-learning model

Number of linear stapler firings

 ≥ 3

 ≤ 2

p value

n = 58 (17.7%)

n = 270 (82.3%)

Baseline characteristics

 Age [y]

  

0.218

  > 70

16 (27.6)

54 (20.0)

 

  ≤ 70

42 (72.4)

216 (80.0)

 

 Gender, n (%)

  

0.084

  Male

46 (79.3)

181 (67.0)

 

  Female

12 (20.7)

89 (33.0)

 

 BMI [kg/m2]

  

0.745

  > 25

17 (29.3)

72 (26.7)

 

  ≤ 25

41 (70.7)

198 (73.3)

 

 Diabetes mellitus, n (%)

  

0.873

  Yes

9 (15.5)

32 (11.9)

 

  No

49 (84.5)

238 (88.1)

 

 nCRT, n (%)

  

0.709

  Yes

14 (24.1)

71 (26.3)

 

  No

44 (75.9)

199 (73.7)

 

 Operation time [min]

  

0.246

  > 150

32 (55.2)

124 (45.9)

 

  ≤ 150

26 (44.8)

146 (54.1)

 

 Anastomotic leakage, n (%)

  

0.021

  Yes

14 (24.1)

34 (12.6)

 

  No

44 (75.9)

236 (87.4)

 

Biochemical data

 Hemoglobin [g/L]

  

0.893

  < 120

12 (20.7)

58 (21.5)

 

  ≥ 120

46 (79.3)

212 (78.5)

 

 Albumin [g/L]

  

0.207

  < 35

8 (13.8)

22 (8.1)

 

  ≥ 35

50 (86.2)

248 (91.9)

 

 CEA [ng/mL]

  

0.007

  > 5

26 (44.8)

71 (26.3)

 

  ≤ 5

32 (55.2)

199 (73.7)

 

Tumor characteristics

 Distance from anus [cm]

  

0.138

  < 5

4 (6.9)

39 (14.4)

 

  ≥ 5

54 (93.1)

231 (85.6)

 

 Tumor size [cm]

  

0.004

  ≥ 5

14 (24.1)

26 (9.6)

 

  < 5

44 (75.8)

244 (90.4)

 

 T stage, n (%)

  

0.865

  T ≤ 2

14 (24.1)

68 (25.2)

 

  T3-4

44 (75.9)

229 (84.8)

 

 N stage, n (%)

  

0.663

  N0

30(51.7)

152 (56.3)

 

  N+

28(48.3)

118 (43.7)

 

 CRM (evaluated by MRI), n (%)

  

0.014

  Positive

20 (34.5)

52 (19.3)

 

  Negative

38 (65.5)

218 (80.7)

 
  1. MRI: magnetic resonance imaging; CRM: circumferential resection margin; N: node; T: tumor; CEA: carcinoma embryonic antigen; nCRT: neoadjuvant chemoradiotherapy; BMI: body mass index.