Classification of Histopathological Images of Breast Cancer Using Convolutional Neural Networks

Convolutional neural networks (CNNs) play a crucial role in early diagnosis detection, aiding healthcare professionals in decision-making. This study utilizes different CNN architectures (AlexNet, ResNet-50, and EfficientNet) to classify breast cancer histopathological images as benign or malignant, using the BreakHis dataset. The models were trained and evaluated with various magnifications and epochs, measuring the performance of each model based on metrics such as accuracy, recall, and specificity in image classification. The results showed that EfficientNet achieved an average of 98.15%, ResNet-50 reached 98.18%, and AlexNet obtained 95.47%.