Development of Deep Learning Based Self Examination of Breast Cancer

The Breast cancer is one of the leading causes of death among women worldwide. Early detection significantly increases the chances of successful treatment. However, in many regions, access to diagnostic tools and medical professionals is limited. This paper presents a deep learning- based system designed to aid in the self-examination of breast cancer using smartphone-acquired photographic images. The acquired images are encrypted using the RSA algorithm to ensure privacy. The proposed system utilizes convolutional neural networks (CNNs), specifically employing a Decision Tree Classifier, to predict the likelihood of cancer based on user-provided images. The model supports multilingual capabilities to cater to non-technical users. Experimental results show promising accuracy, suggesting that deep learning can support accessible and affordable early detection methods.