Advanced Deep Learning Based Breast Cancer Detection and Classification Using Enhanced YOLOv11 and YOLOv12 Architectures A Comparative Study on Accuracy, Speed, and Real-Time Performance for Medical Imaging Applications

Abstract

Medical imaging enables early discovery of problems, therefore greatly enhancing patient results. This study investigates the use of modern deep learning
methods particularly YOLOv11 and YOLOv12 for the identification and categorization of breast cancer. In order to determine how well the proposed models
work when it comes to their speed, accuracy, and immediate diagnosis capability, a comparison study is conducted. Both models are trained and evaluated using
mammography and histopathology image datasets with annotations. The results provide insight into the best model structure suitable for implementation in
clinical settings. Technology that can detect objects in real time aids in developing intelligent devices for clinical use.