Melanoma is the most dangerous and deadly type of skin cancer. Melanoma can be cured if detected early, but the form of melanoma that resembles a mole makes it difficult to distinguish. Early treatment performed by a dermatologist on melanoma through a biopsy process. However, a lack of biopsy is a long preparation and laboratory results that take a little longer. This fear will make cancer cells spread more widely. With this problem, this final project will design an Android-based mobile application that can detect melanoma early. Applications designed using expert systems are forward chaining methods and image processing methods K-Nearest Neighbor (KNN). From the results of image processing testing that has been done this application has an accuracy of 72% by using training data totaling 70 images. While the expert system with the forward chaining method is suitable to be used in this application because the decisions taken are in accordance with the knowledge representation that has been entered into the system.
Copyrights © 2018