Skin Cancer Disease Detection and Classification using Probabilistic Neural Networks and Multimodal Features

Abstract

Skin cancer disease detection and analysis heavily rely on human visual inspection, limited by
microscopic behavior. Computer-based image recognition systems have been instrumental in
achieving accurate classification and identification of skin cancer diseases. This research incorporates
K-means clustering for real-time skin lesion image detection, followed by feature extraction through
Gray Level Co-occurrence Matrix (GLCM)-based texture features, Discrete Wavelet Transform
(DWT)-based low-level features, and Statistical Color features. While classification is typically
conducted using SVM-based methods, these are less accurate with respect to texture features. To
improve feature-based matching, an advanced artificial intelligence approach based on Probabilistic
Neural Networks (PNN) is introduced for classification. The proposed methodology is implemented
in the MATLAB environment and demonstrates significantly better accuracy compared to
conventional approaches.