We applied a machine learning and deep neural network algorithm that learns patterns in existing skin cancer data, feed it with unseen skin cancer scan data, and conclude how well it classifies the scans based on the evulation metrics decided for this study.
Cancer is a disease that marks an irregular growth of cells with an inclination to spread transversely across the body. We recognize this specific disease by the visible presence of a tumor, atypical bleeding, and a prolonged cough. There is a vast pool of cancer-causing agents. Besides exposure to certain chemicals, lifestyle behaviors such as smoking and excessive alcohol consumption may be risks. The most prevalent approaches for diagnosing cancer in a patient involve MRI and ultrasound scans. In its developmental stages, cancer can be treated by chemotherapy, surgery, and radiation therapy, among others. Provided the above, it is crucial to conduct periodic medical checks to diagnose cancer.
The figure depicts the dominant forms of skin cancer that our deep convolutional neural network will differentiate.