Author : Md Arafat Mostafiz

A common kind of cancer is breast cancer. Raising the survival rate of breast cancer patients is mainly dependent on breast cancer recurrence prognosis. The accuracy of cancer detection and diagnosis has increased with the progress of technology and ML approaches. Machine learning (ML) provides a number of statistical and probabilistic approaches. This study introduces a deep learning-based approach to automatically classify breast cancer images from the BreakHis dataset. Feature extraction was performed using a Convolutional Neural Network (CNN) to automatically detect significant tissue structures. The MobileNetV2 architecture was employed for its efficiency in handling large-scale data while maintaining high classification accuracy. The model achieved an impressive accuracy of 98.18%, with precision of 98.38%, sensitivity of 97.37%, and an F1score of 97.85%. When compared to other architectures, MobileNetV2 outperformed Xception, ResNet101, and EfficientNet, which demonstrated lower accuracy and sensitivity. These results highlight the potential of MobileNetV2 for reliable, fast, and cost-effective breast cancer detection, offering a promising tool for clinical applications.

Keywords: Healthcare, Breast Cancer, Cancer Classification, Early Detection, Medical Imaging, Machine Learning (ML), Breast Cancer Histopathological Image Dataset.