Decision Support System for Skin Cancer Diagnosis

Abstract Computer aided medical diagnosis is mostly based on very advanced analysis of huge amounts of measured data. Application of computational intelligence methods helps medical doctors to process data faster, and give better, more reliable diagnoses. In dermatology computer analysis is getting more important every year. Differentiation of malignant melanoma images requires very fast image processing and feature extraction/classification algorithms. Applying artificial intelligence algorithms to explore and search large database of dermatoscopic images allows doctors to semantically filter out images with specified characteristics. This paper describes a decision-support system which is based on semantic analysis of melanoma images and further classification of characteristic objects commonly found in pigmented skin lesions. For classification Artificial Neural Networks and Support Vector Machines are used and their performance is evaluated and compared