Table 1.
Clinical characteristics of the dataset. Patient age is summarized on a patient basis, and lesion information (malignancy status and subtypes) is summarized on a lesion basis. The full set is a mixture of cases imaged using either two or three sequences, and the DWI subset contains cases imaged using three sequences.
| Full set () | DWI subset () | |
|---|---|---|
| Benign/malignant prevalence | Benign: 195 (22.9) | Benign: 66 (17.0) |
| Malignant: 657 (77.1) | Malignant: 323 (83.0) | |
| Age (years): |
||
| Unknown: 96 |
Unknown: 12 |
|
| Benign lesion characteristics | ||
| Lesion subtypes |
Fibroadenoma: 60 (30.8) | Fibroadenoma: 18 (27.3) |
| Columnar change: 15 (7.7) | Columnar change: 5 (7.6) | |
| Papilloma: 13 (6.7) | Papilloma: 6 (9.1) | |
| Parenchyma tissue: 11 (5.6) | Parenchyma tissue: 8 (12.1) | |
| Fibrotic tissue: 10 (5.1) | Fibrotic tissue: 5 (7.6) | |
| Hyperplasia: 8 (4.1) | Hyperplasia: 5 (7.6) | |
| Cystic change: 6 (3.1) | Cystic change: 3 (4.5) | |
| Fat necrosis: 4 (2.1) | Fat necrosis: 3 (4.5) | |
| Other: 26 (13.3) | Other: 12 (18.2) | |
| Unknown: 42 (21.5) |
Unknown: 1 (1.5) |
|
| Malignant lesion characteristics | ||
| Lesion subtypes | IDC: 133 (20.2) | IDC: 71 (22.0) |
| DCIS: 118 (18.0) | DCIS: 20 (6.2) | |
| IDC + DCIS: 316 (48.1) | IDC + DCIS: 197 (61.0) | |
| ILC: 27 (4.1) | ILC: 15 (4.6) | |
| ILC + LCIS: 24 (3.7) | ILC + LCIS: 5 (1.5) | |
| Other: 28 (4.3) | Other: 15 (4.6) | |
| Unknown: 11 (1.7) | ||
| Estrogen receptor status | Positive: 408 (62.1) | Positive: 235 (72.8) |
| Negative: 127 (19.3) | Negative: 83 (25.7) | |
| Unknown: 122 (18.6) | Unknown: 5 (1.5) | |
| Progesterone receptor status | Positive: 350 (53.3) | Positive: 209 (64.7) |
| Negative: 183 (27.9) | Negative: 108 (33.4) | |
| Unknown: 124 (18.9) | Unknown: 6 (1.9) | |
| HER-2 status | Positive: 87 (13.2) | Positive: 54 (16.7) |
| Negative: 401 (61.0) | Negative: 240 (74.3) | |
| Equivocal: 5 (0.8) | Equivocal: 2 (0.6) | |
| Unknown: 164 (25.0) | Unknown: 27 (8.4) | |
Note: Numbers in parentheses are percentages. For some subjects, only the decade of age was available (e.g., 60 s) as part of the patient information deidentification process. In these situations, the middle of the decade was used for the calculation of the mean subject age. DCE, dynamic contrast-enhanced sequence; T2w, T2-weighted sequence; DWI, diffusion-weighted imaging sequence; std, standard deviation; IDC, invasive ductal carcinoma; DCIS, ductal carcinoma in situ; ILC, invasive lobular carcinoma; and HER-2, human epidermal growth factor receptor 2.