BreakHis Breast Cancer Dataset

BreakHis stands for Breast Cancer Histopathological Image Classification, and the dataset contains 7,909 breast tissue pathology images from 82 patients. The dataset contains 2,480 benign samples and 5,429 malignant samples (700 X 460 pixels, 3-channel RGB, 8-bit depth per channel, PNG format).

The images were obtained by using 4 magnification factors (40X, 100X, 200X, 400X), corresponding to the objectives (4X, 10X, 20X, 40X).