Automatic skin disease detection using modified level set and dragonfly based neural network

Abstract

Dermatology is an essential fields that are to be analyzed, monitored, and treated to avoid skin disorders. The skin disease is caused because many factors influence humans such as age, sex, and lifestyle. Also, less amount of exposure to sunlight, bacteria, hot weather leads to skin diseases. Hence it is important to detect the skin disease at the earlier stage to avoid the fading condition of skin. In this paper we have developed an efficient automatic detection of skin disease using a two-stage adaptive process. At first stage modified level set approach is used for segmentation of skin images, later using color, shape and texture features are extracted and in the final stage dragonfly optimization-based Neural network is used for classification of types of skin diseases such as normal or abnormal. The proposed dragonfly based NN is evaluated using existing methods such as SVM, ANN for different evaluation metrics such as accuracy, sensitivity, and specificity to show the system efficiency.