Prediction of the Success of Wart Treatment Methods

Rukiye uzun, Yalçın İşler, Mualla Toksan


Wart is a dermatoses originated by Human Papilloma Virus (HPV), with which people can be infected by direct or indirect contact. Almost all age groups, especially children and young adults suffer from warts. Recently, new treatment methods including cryotherapy and immunotherapy have been developed as alternatives to the conventional methods. Although treatment decision process is very important, there is no validated decision strategy yet except for only a few studies. In this study, an expert system is proposed to predict whether the selected wart treatment method will be successful or not. The publicly available datasets are applied to the Multi-Layer Perceptron (MLP) and the Extreme Learning Machine (ELM) classification algorithms. We compute the classifier performances by the 10-fold cross-validation method. As a result, the proposed MLP-based approach results in 78.95% of sensitivity, 98.60% of specificity, and 94.45% of accuracy to predict the success of a wart treatment method.


Human papilloma virus (HPV), Wart treatment, Immunotherapy, Cryotherapy, Prediction, Multi-layer perceptron, Extreme learning machine

Full Text:



© 2011 Karaelmas Fen ve Mühendislik Dergisi