Cross-Domain Transfer Learning Architecture for Microcalcification Cluster Detection Using the MEXBreast Multiresolution Mammography Dataset
Ricardo Salvador Luna Lozoya, Humberto de Jesús Ochoa Domínguez, Humberto Sossa, Vianey Guadalupe Cruz Sánchez, Osslan Osíris Vergara Villegas, Karina Núñez Barragán · Mathematics · 2025
Microcalcification clusters (MCCs) are key indicators of breast cancer, with studies showing that approximately 50% of mammograms with MCCs confirm a cancer diagnosis. Early detection is critical, as it ensures a five-year survival rate of up to 99%. However, MCC detection remains challenging due to their features, such as small size, texture, shape, and impalpability. Convolutional neural networks (CNNs) offer a solution for MCC detection. Nevertheless, CNNs are typically trained on single-resolution images, limiting their generalizability across different image resolutions. We propose a CNN trained on digital mammograms with three common resolutions: 50, 70, and 100 μm. The architecture processes individual 1 cm2 patches extracted from the mammograms as input samples and includes a MobileNetV2 backbone, followed by a flattening layer, a dense layer, and a sigmoid activation function. This architecture was trained to detect MCCs using patches extracted from the INbreast database, which has a resolution of 70 μm, and achieved an accuracy of 99.84%. We applied transfer learning (TL) and trained on 50, 70, and 100 μm resolution patches from the MEXBreast database, achieving accuracies of 98.32%, 99.27%, and 89.17%, respectively. For comparison purposes, models trained from scratch, without leveraging knowledge from the pretrained model, achieved 96.07%, 99.20%, and 83.59% accuracy for 50, 70, and 100 μm, respectively. Results demonstrate that TL improves MCC detection across resolutions by reusing pretrained knowledge.