Skip to main content
. 2022 Jan 11;1:796078. doi: 10.3389/fradi.2021.796078

Table 1.

Composition of datasets DHMItrain, DHMItest, DINB, and DVTB, per task and per class as well as the total amount of samples.

breastbin ACR BIRADS view Total
0 1 1 2 3 4 5 6 A B C D CC MLO Images Patients
D HMItrain 1,016 1,000 397 619 11 186 284 508 372 925 602 117 1,210 806 2,016 997
D HMItest 255 249 96 159 2 45 75 123 91 221 159 33 296 208 504 409
D INB 310 100 67 220 23 43 49 8 136 147 99 28 204 206 410 115
D VTB NA NA 3,152 1,994 94 102 0 22 667 2,489 2,082 198 2,791 2,669 5,460 606