<![CDATA[Background: In recent years, image detection has gained substantial importance within the medical field, particularly in diagnosing and interpreting diseases such as breast cancer. Breast cancer stands as a formidable threat, constituting approximately 30% of newly diagnosed cancer cases in the United States and contributing to 12.5% of all new cancer cases. Early detection is crucial for preventing the progression of this severe ailment. Method: This research endeavours to leverage deep learning methodologies to ascertain the presence or absence of breast cancer in patients, utilizing the mammograph breast cancer RSNA dataset. The Contrast Limited Adaptive Histogram Equalization (CLAHE) technique enhances the dataset's images for more precise results. The focus of this study is specifically on the biopsy status related to breast cancer. The deep learning algorithms implemented encompass ResNet50, VGG16, and EfficientNetB0. Notably, the dataset is confined to biopsy status, streamlining the investigation to this critical aspect. Result: The experimental results reveal that the ResNet50 model achieved the highest accuracy at 61%, coupled with an F1-Score of 0.47%. These findings underscore the potential of deep learning techniques, particularly ResNet50, in aiding the early detection of breast cancer. Conclusion: Incorporating image enhancement techniques like CLAHE adds an extra layer of refinement to the dataset, contributing to the overall accuracy and reliability of the diagnostic process. As medical image analysis continues to evolve, such studies pave the way for advancements in early disease detection and intervention strategies.]]>
Copyrights © 2026
