Enhanced Convolutional Neural Networks for Breast Cancer Detection
Ashok Reddy Kandula, Pujitha Ganta, Grace Apoorva Gali, Sandeep Kasagani, Devi Venkata Karthik Tikkisetti · 2024
For women all around the world, breast cancer is a serious concern. Catching it early is like having a superhero move - it can make all the difference, turning a potentially life-threatening situation into one with a shot at successful treatment. So, finding it sooner rather than later is a real lifesaver. To evaluate the performance of the proposed model, the researchers trained and tested the model on the Breast ultrasound Images dataset. These images, meticulously prepared through resizing, normalization, and augmentation, have been widely utilized in prior studies focusing on breast cancer prediction. The study focuses on refining accuracy through the utilization of Convolutional Neural Networks (CNNs) in combination with Transfer Learning for initial improvement. Following this, the approach shifts to leveraging the potential of the Adam optimizer with the CNN and Transfer Learning architecture to further boost accuracy. The implications of this study transcend the conventional confines of traditional medical image analysis, providing a glimpse into a prospective era where intelligent algorithms harmoniously collaborate with medical professionals. This collaboration serves to elevate the benchmarks of diagnostic accuracy. With the augmentation of early detection potential, patient outcomes stand to witness marked enhancement, thus underscoring the transformative prowess embedded within the capabilities of deep learning in the domain of breast cancer diagnosis.