Whale Optimized Deep Learning Technique for Accurate Skin Cancer Identification

Early skin cancer diagnosis is crucial because it can stop some types of the disease, including melanoma and focal cell carcinoma. Nevertheless, there are a number of factors that negatively affect the detection accuracy. Due to its tendency to spread to other cells, cancer has become a dangerous disease in recent years, and it is often not detected at an early stage. Typically, a biopsy method which is a painful procedure is used to find malignancy. Due to the advancement of technological advances, image processing techniques are being used to identify it. Deep Learning (DL) based computer aided diagnosis (CAD) models are helpful for correctly classifying skin cancer. Here, both typical and atypical nevi, as well as melanoma, are identified and classified using an image processing approach. There are numerous different categories of cancer, including skin, stomach, bone, and blood cancers. Among these cancer forms, skin cancer can be a terrible condition that is identified early on and then treated. As a result, this article proposed a threshold technique for classifying skin cancer. For the purpose of identifying the cancer region, the threshold-based segmentation is employed. GLCM (Grey Level Co-occurrence Matrix) is the feature removal method employed here. These characteristics are employed in categorization. For classification, CNN (Convolutional Neural Network) classifier is employed. The CNN is optimized with the WOA (Whale optimization Algorithm) to increase classification accuracy. The Accuracy of the WOA-CNN is 92.6%.