Exploring the Potential of Deep Neural Networks with Gamma and Discrete Wavelet Transform in Breast Cancer Detection.

Breast cancer is a prevalent and life-threatening disease affecting millions of women worldwide. Early and accurate identification of breast cancer plays a crucial role in improving patient outcomes and survival rates. In recent years, deep learning techniques and image-processing algorithms have emerged as powerful breast cancer identification and diagnosis tools. The proposed approach involves a multi-step process, starting with preprocessing the mammographic images using gamma correction and wavelet analysis algorithms to enhance the features related to cancerous tissues. A CNN architecture is employed to extract meaningful features from the preprocessed images. AlexNet, VGG16, Googlenet, Resnet18, and Resnet50 were trained on the Breast Ultrasound Images Dataset to classify malignant, benign, and normal cases. AlexNet has achieved higher accuracy than the rest of the networks, up to $93.58\% $ , in detecting breast cancer while reducing false positives and negatives. Combining image processing and deep learning technologies enables more accurate and reliable identification of abnormalities, enabling healthcare professionals to make better-informed decisions and deliver timely patient interventions.