Breast cancer is the cancer with the highest incidence in women worldwide, and seriously threatens the lives and health of women. Mammography, which is commonly used for screening, is considered to be the most effective means of diagnosing breast cancer. Currently, computer-assisted breast mass systems based on mammography can help doctors improve film reading efficiency, but improving the accuracy of assisted diagnostic systems and reducing the false positive rate are still challenging tasks. In the image classification field, convolutional neural networks have obvious advantages over other classification algorithms. Aiming at the very small percentage of breast lesion area in breast X-ray images, in this paper, the classical VGG16 network model is improved by simplifying the network structure, optimizing the convolution form and introducing an attention mechanism. The improved model achieves 99.8 and 98.05% accuracy on the Mammographic Image Analysis Society (MIAS) and The Digital Database for Screening Mammography (DDSM), respectively, which is obviously superior to some methods of recent studies.

Breast cancer has a very high incidence rate worldwide, and effective screening and early diagnosis are particularly important. In this paper, two improved You Only Look Once version 8 (YOLOv8) models, the YOLOv8-GHOST and YOLOv8-P2 models, are proposed to address the difficulty of distinguishing lesions from normal tissues in mammography images. The YOLOv8-GHOST model incorporates GHOSTConv and C3GHOST modules into the original YOLOv8 model to capture richer feature information while using only 57% of the number of parameters required by the original model. The YOLOv8-P2 algorithm significantly reduces the number of necessary parameters by streamlining the number of channels in the feature map. This paper proposes the YOLOv8-GHOST-P2 model by combining the above two improvements. Experiments conducted on the MIAS and DDSM datasets show that the new models achieved significantly improved computational efficiency while maintaining high detection accuracy. Compared with the traditional YOLOv8 method, the three new models improved and achieved F1 scores of 98.38%, 98.8%, and 98.57%, while the number of parameters reduced by 42.9%, 46.64%, and 2.8%. These improvements provide a more efficient and accurate tool for clinical breast cancer screening and lay the foundation for subsequent studies. Future work will explore the potential applications of the developed models to other medical image analysis tasks.