Skip to main content

Table 1 Characteristics of the cases collected for the study

From: Multi-modal artificial intelligence for the combination of automated 3D breast ultrasound and mammograms in a population of women with predominantly dense breasts

 

Full cohort (n = 430)

Observer study Cohort (n = 152)

Age (y)

48 (30–70)

50 (33–70)

DM compressed breast thickness (mm)

47 (19–78)

48 (20–76)

BI-RADS density category

A: 10

B: 106

C: 217

D: 97

A: 6

B: 31

C: 69

D: 46

Ground truth

Malignant: 42

Benign: 114

Normal: 274

Malignant: 42

Benign: 30

Normal: 80

Histology of malignant lesions (42 lesions in 42 patients)

Invasive ductal carcinoma: 32

Ductal carcinoma in situ: 3

Mucinous carcinoma: 2

Intraductal papillary carcinoma: 2

Metaplastic carcinoma: 1

Neuroendocrine carcinoma: 1

Apocrine carcinoma: 1

Invasive ductal carcinoma: 32

Ductal carcinoma in situ: 3

Mucinous carcinoma: 2

Intraductal papillary carcinoma: 2

Metaplastic carcinoma: 1

Neuroendocrine carcinoma: 1

Apocrine carcinoma: 1

Histology of benign lesions (Full Cohort: 152 lesions in 114 patients, 38 patients had two benign lesions; Observer Study Cohort: 35 lesions in 30 patients, 5 patients had two benign lesions)

Fibroadenoma: 79

Cyst: 1

Intraductal papillary: 11

Radial scar: 1

Atypical Ductal Hyperplasia: 6

Lipoma: 1

Phyllodes tumor: 2

Hamartoma: 1

Sclerosing adenosis: 2

Benign hyperplasia: 48

Fibroadenoma: 14

Intraductal papillary: 3

Atypical Ductal Hyperplasia: 2

Lipoma: 1

Phyllodes tumor: 2

Hamartoma: 1

Benign hyperplasia: 12

  1. When available, median and range (within parentheses) are given
  2. BI-RADS Breast imaging reporting and data system, DM Digital mammography