This is the EfficientNet-B3 based multiple-view model from the paper: https://arxiv.org/abs/2503.19945.
This model achieved the best (to my knowledge and by the time of the paper) AUC for CBIS-DDSM dataset, using original test division.
This model achieves an AUC of 0.8658. Considering only the mammograms that has CC and MLO views (N=2.694 total, for training and testing).
Please find code and inference instructions here: https://github.com/dpetrini/multiple-view
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support