Skip to main content

Table 1 Characteristics of patients in the training and testing sets. P values correspond to measures computed using Wilcoxon rank-sum tests performed to compare the training and testing cohorts

From: A machine learning model that classifies breast cancer pathologic complete response on MRI post-neoadjuvant chemotherapy

Characteristics

Cohort

 

Training

Testing

Pvalue

n

222

56

 

Age, mean ± SD, years

51.8 (11.8)

51.3 (11.8)

0.90

Pathologic response (%)

  

0.50

 pCR

61 (27.5)

13 (23.2)

 

 no-pCR

161 (72.5)

43 (76.8)

 

Histology, no (%)

  

0.90

 Invasive ductal

203 (91.4)

51 (91.1)

 

 Invasive lobular

8 (3.6)

3 (5.4)

 

 Mix

5 (2.3)

0 (0)

 

 Invasive NOS

6 (2.7)

2 (3.5)

 

Molecular subtype, no (%)

  

0.40

 HR+HER2−

76 (34.2)

22 (39.3)

 

 HR+HER2+

52 (23.4)

9 (16.1)

 

 HR−HER2+

36 (16.2)

8 (14.3)

 

 Triple negative

58 (26.2)

17 (30.3)

 
  1. Abbreviations: pCR pathologic complete response, (+) positive, (−) negative; NOS not otherwise specified
  2. *P value < 0.05