Skip to main content

Table 1 Clinicopathological characteristics of breast cancer and benign patients

From: Independent validation of machine learning in diagnosing breast Cancer on magnetic resonance imaging within a single institution

Clinicopathological characteristics of breast cancer and benign patients
  Training data Testing data
Malignant Benign Malignant Benign
Total 1073 382 421 114
Age, years (mean, range) 47.6 (19–77) 42.2 (16–76) 49.3 (25–75) 41.9 (19–65)
Lesion type
 Mass 716 (66.7%) 230 (60.2%) 293 (69.6%) 70 (61.4%)
 Non-mass 357 (33.3%) 152 (39.8%) 128 (30.4%) 44 (38.6%)
MRI-BI-RADS category
 0 0 (0%) 2 (0.5%) 0 (0%) 0 (0%)
 1 0 (0%) 1 (0.3%) 0 (0%) 2 (1.8%)
 2 0 (0%) 4 (1.0%) 0 (0%) 0 (0%)
 3 4 (0.3%) 202 (52.9%) 0 (0%) 50 (43.8%)
 4 351 (33.1%) 170 (44.5%) 113 (26.8%) 60 (52.6%)
 5 529 (49.8%) 3 (0.8%) 221 (52.5%) 2 (1.8%)
 6 178 (16.8%) 0 (0%) 87 (20.7%) 0 (0%)
Pre or Post Biopsy MRI
 Pre 868 (81.7%) 362 (94.8%) 330 (78.4%) 112 (98.2%)
 Post 194 (18.3%) 20 (5.2%) 91 (21.6%) 2 (1.8%)
Histology
 IDC 914 (85.2%)   366 (86.9%)  
 ILC 22 (2.1%)   4 (1.0%)  
 DCIS 76 (7.1%)   18 (4.3%)  
 Other malignant lesions 61 (5.6%)   33 (7.8%)  
 Fibroadenoma   165 (43.2%)   46 (40.4%)
 Papilloma   66 (17.3%)   28 (24.6%)
 Inflammation   19 (5.0%)   10 (8.8%)
 Other benign lesions   132 (34.5%)   30 (26.3%)
Grade of IDC
 I 56 (6.2%)   13 (3.7%)  
 II 683 (75.1%)   275 (77.2%)  
 III 171 (18.7%)   68 (19.1%)  
Lymph node status (n = 1468)
 Negative 734 (70.3%)   295 (70.7%)  
 Positive 310 (29.7%)   122 (29.3%)  
Estrogen receptor
  < 1% 193 (18.1%)   77 (18.3%)  
  ≥ 1% 876 (81.9%)   344 (81.7%)  
Progesterone receptor
  < 1% 222 (20.8%)   104 (24.7%)  
  ≥ 1% 846 (79.2%)   317 (75.3%)  
Her2
 0 or 1+ 632 (59.2%)   243 (57.7%)  
 2+ or 3+ 436 (40.8%)   178 (42.3%)  
Ki-67
  < 14% 180 (16.9%)   60 (14.3%)  
  ≥ 14% 887 (83.1%)   361 (85.7%)