Advanced Breast Cancer Detection and Classification using Hybrid Deep Learning Model

Breast cancer is still one of the most common and deadly diseases in women. Early detection is key to better survival rates. This study proposes a hybrid deep learning model combining Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Vision Transformers (ViTs) for better breast cancer detection and classification. The model is trained on several image modalities like mammograms, histopathology slides, and MRI scans and uses image pre-processing methods like Contrast Limited Adaptive Histogram Equalization (CLAHE), Gaussian Blur, and Unsharp Masking. We also use U-Net for tumor segmentation and YOLOv8 for tumor detection. The proposed model is tested on the CBIS-DDSM dataset and attains high accuracy, precision, recall, and F1-score. Results show the utility of the model in detecting malignant and benign cases with enhanced explainability through Grad-CAM heatmaps.