Abstract: Breast cancer is the type of cancer that is originates in cells of breast. Breast cancer can occur in both men and women, but it is far more common in women. Breast cancer detection is critical for early diagnosis and treatment.  Mammography is the most known and effective process to detect early signs of breast cancer. Convolutional Neural Networks (CNNs) have emerged as powerful tools for mammogram image analysis due to their ability to automatically learn and extract relevant features from complex data. This study explores the application of CNNs for detecting and classifying the stages of breast cancer from mammographic images. By employing a deep learning framework, we trained a CNN Pre-train model like EfficientNet B4, Inception V4 model on a labeled dataset of mammograms, where the images were preprocessed to enhance feature extraction. The model's performance was evaluated using metrics such as accuracy, sensitivity, specificity, and the area under the ROC curve (AUC). Results demonstrated that the CNN achieved high accuracy in distinguishing between different stages of breast cancer, highlighting its potential as an effective diagnostic aid. Further improvements and validations with larger datasets are necessary to enhance the model's robustness and generalizability. This approach promises to support radiologists in making more accurate and timely diagnoses, ultimately improving patient outcomes.

Keywords: Breast cancer, Convolutional Neural Networks (CNNs), mammogram images, CNN Pre-train model like EfficientNet B4, Inception V4 model

| DOI: 10.17148/IARJSET.2024.11574