Skip to main content

Table 2 Summary of characteristics of the in-house dataset

From: A YOLO-based AI system for classifying calcifications on spot magnification mammograms

Characteristics

No. of cases

No. of spot magnification images*

1269

Age at examination (y)†

29–81 (54.4 ± 8.42)

Mean time interval between image acquisition and biopsy (d)‡

48.71 (46.3, 51.1)

BI-RADS assessment

n (%)

 4A (low suspicion for malignancy)

435 (68.4)

 4B (moderate suspicion for malignancy)

178 (27.99)

 4C (high suspicion for malignancy)

23 (3.61)

Histologic type of malignancy

n (%)

 Invasive ductal carcinoma

61 (29.9)

 Invasive lobular carcinoma

3 (1.5)

 Ductal carcinoma in situ

140 (68.6)

  1. Except where indicated, data are numbers of cases, with percentages in parenthesis
  2. BI-RADS, Breast Imaging Reporting and Data System
  3. *Three of the cases had only one image for each
  4. †Data in parenthesis are means ± standard deviations
  5. ‡Data in parenthesis are 95% confidence intervals