Breast cancer is a common and fatal disease that affects many women around the world every year. Detecting breast cancer at an early stage is crucial for effective treatment and improved survival rates. Researchers have explored various methods, including neural networks and machine learning techniques, to help detect diseases. However, due to limited data availability, leveraging pre-trained models trained on diverse image datasets has become a common practice. This paper presents a new method for breast cancer detection using a deep learning model based on the ResNet50 architecture, combined with heat mapping and gradient-weighted class activation mapping (Grad-Cam). The proposed method involves applying ResNet50 optimization based on Grad-Cam results to the augmented data. The results obtained from the proposed model were compared with three other previously known models: VGG16, VGG19, and ResNet50. The proposed model achieved an accuracy of 0.88, superior to the rest of the models. Grad-Cam demonstrated near-perfect feature extraction in a breast cancer classification task. This research contributes to the advancement of breast cancer detection by combining the power of deep learning, the ResNet50 architecture, and visualization techniques such as heat maps and Grad-Cam.