The project provides pretrained deep neural network models for breast cancer screening from mammography images. It includes two model variants: an image-only model that takes mammography images as input, and an image-and-heatmaps model that additionally incorporates patch-level prediction heatmaps. The models process four standard mammography views (L-CC, R-CC, L-MLO, R-MLO) and output probability predictions for benign and malignant findings per breast. Implemented in PyTorch.