Annotation-free deep-learning framework for microcalcifications detection on mammograms

International audience ; Breast cancer detection at an early stage significantly increases the chances of recovery for patients. Mammography(MG) is one of the most popular non-invasive and high-resolution imaging allowing radiologists to depictearly signs of the disease. Microcalcifications (MCs) often occupy less than 1mm in size and can represent ahigh risk of suspicion depending on the spatial distribution, morphology, and their evolution over time. Theirdetection is challenging both the clinicians and computer-aided detection tools. In this work, we propose anovel annotation-free framework designed specifically for the MCs detection and trained in a self-supervisedmanner thanks to the generation of synthetic MCs. Inspired by the UNet3+ architecture, we reduced its numberof parameters to make it applicable in practice and added multi-scale features to enrich fine-grained detailswith more global context information. Both multi-channel segmentation and multi-class classification tasks areimplemented in a multi-scale output approach to catch MC of various sizes. We perform a comparison withseveral state-of-the-art methods, including different flavors of ResNet-22, ConvNeXt, and UNet3+. An analysisof classification and segmentation performances has been done, using the Gradient-weighted Class ActivationMapping method to make classifiers visually explainable. In this study, we used two public datasets, INBreastand Breast MicroCalcifications Dataset for validation and test purposes. We achieved an AUC score of 0.93 inthe characterization of malignant MCs while having a semantic segmentation precision of 0.70. To the best ofour knowledge, we are the first study claiming segmentation performances on the BMCD dataset.