Skip to main content

Table 1 The demographics of clinical factors for patients in the training and validation sets

From: Predicting microvascular invasion in hepatocellular carcinoma: a deep learning model validated across hospitals

  Training Set (N = 216) Validation Set (N = 93) p-value
  MVI+ (N = 68) MVI- (N = 148) MVI+ (N = 28) MVI- (N = 65) 0.41
Median of age in years (Q.25-Q.75) 58 (53–68) 60 (50–68.3) 65 (58.3–72.5) 59 (54–67) 0.32
Gender 0.12
 Male 55 (80.9%) 103 (69.6%) 21 (75%) 53 (81.5%)  
 Female 13 (19.1%) 45 (30.4%) 7 (25%) 12 (18.5%)  
 Median of MTD in mm (Q.25-Q.75) 4.5 (3–7.4) 2.2 (1.5–3.4) 5.9 (3.9–6.9) 2.5 (2–4.2) 0.12
 Median of AFP in ng/ml (Q.25-Q.75) 33.5 (6.5–1294.9) 11.4 (3.7–209.6) 9.2 (3.8–328.1) 7 (3.13–71.85) 0.87
Child-Pugh score 0.12
 A 64 (94.1%) 139 (93.9%) 25 (89.3%) 59 (90.6%)  
 B 4 (5.8%) 7 (4.7%) 2 (7.1%) 4 (6.2%)  
 C 0 (0%) 2 (1.4%) 1 (3.6%) 2 (3.1%)  
HBsAg 0.22
 Positive 37 (54.4%) 78 (52.7%) 11 (39.3%) 43 (66.2%)  
 Negative 31 (45.6%) 70 (47.3%) 17 (60.7%) 22 (33.8%)  
HCsAg 0.27
 Positive 23 (33.8%) 52 (35.1%) 11 (39.3%) 18 (27.7%)  
 Negative 45 (66.2%) 96 (64.9%) 17 (60.7%) 47 (72.3%)  
  1. Data are presented as n (%) unless otherwise noted. P-value were derived from statistical comparison between the training and validation sets using Mann–Whitney–Wilcoxon text,
  2. p-value > 0.05 indicates no significant difference exists between these two data sets. MVI+ microvascular invasion positive; MVI- microvascular invasion negative; MTD maximum tumor diameter; AFP alpha-fetoprotein; Q.25 25 % quantile; Q.75 75 % quantile; HBsAg Hepatitis B surface antigen; HCsAg Hepatitis C surface antigen