A Comprehensive CNN Approach for Reliable Breast Cancer Diagnosis

Breast cancer is a common type of cancer in women worldwide, and early detection is critical for effective treatment. Conventional breast cancer diagnostic methods, such as mammography and biopsy, rely heavily on human interpretation, which can be time-consuming and error-prone. Recently, deep learning methods, specifically Convolutional Neural Networks (CNNs), have shown promise in improving the precision and effectiveness of breast cancer diagnosis. This study describes an innovative Convolutional Neural Network (CNN) method for detecting breast cancer using histopathology images. A total of 8,598 images were collected and divided into two datasets: training (6,860 images) and testing (1,738 images). The CNN structure was explicitly designed for analyzing breast cancer images with Conv2D, MaxPooling2D, Dropout, Flatten, and Dense layers. The model underwent 25 training epochs and was evaluated using loss, accuracy, precision, recall, and F1-Score metrics. A confusion matrix was constructed to assess the model's classification accuracy, and performance metrics were calculated. The CNN model achieved an overall accuracy of 85% when diagnosing breast cancer, demonstrating its effectiveness in automated breast cancer diagnosis.