Convolutional Neural Network for Breast Cancer Prediction using MRI Images

This research study initiates a multifaceted analysis of breast cancer, a pervasive malignancy originating in the mammary gland cells. Globally, breast cancer ranks as the second most commonly diagnosed cancer, surpassed only by skin cancer. While breast cancer can affect individuals of any gender, it disproportionately impacts women. In light of this significant health concern, this research study examines the notable advancements in breast cancer detection, encompassing a wide array of techniques designed to enhance early diagnosis. The success of this effort relies on old-fashioned techniques like mammograms and check-ups, which are still very important tools. Additionally, this study introduces emerging technologies, specifically AI-powered analysis of mammograms. The technologies, such as Convolutional Neural Networks (CNN) utilizing Residual Networks (ResNet), Visual Geometry Group Networks (VGG16), and Siamese Neural Network architectures (SNN) and Convolutional Recurrent Neural Network (CRNN) have shown promising potential to substantially enhance the diagnostic accuracy. Moreover, this study explores the application of various machine learning algorithms to collect datasets, aiming to predict the early development of breast cancer. This comprehensive approach underscores the promising future of breast cancer detection and emphasizes the importance of a multidimensional strategy in combating this global health challenge. This study utilizes the pre-existing research on Magnetic Resonance Imaging (MRI) scans of breast datasets, seeking to improve the accuracy and early detection by ultimately contributing to establish a more effective breast cancer management.