Detection and Classification of Skin Cancer Using Back Propagation Ann

Cancer of the skin is the most common form of the disease and is to blame for the deaths of millions of people each year. The early detection of potentially hazardous skin cancer cases and the administration of suitable treatments are essential components in assuring a relatively low overall death rate while maintaining a very high percentage of those that survive. A significant portion of the relevant studies concentrate on algorithms that are relied on machine learning, however these algorithms are unable to deliver the highest possible level of accuracy and specificity. At the preprocessing step, enhancing procedures including sharpening filters and smoothing filters are employed to reduce noise from the image. After that, Ostu segmentation was used for accurate diagnosis of the cancerous area of the skin. A Back-Propagation Artificial Neural Network (BP-ANN) was constructed with relation to the categorization of skin cancer using the Spatially Grey Level Dependency matrix (SGLD) characteristics, in order to capture the maximum efficiency of the system in the proposal. As a result, the findings of the proposal may be efficiently used to the categorization of both benign and malignant forms of skin cancer. Comparing the suggested technique to the state of the art methodologies, it has been shown via the modelling exercise that the proposed algorithm yields superior results qualitatively and quantitatively.