Paper Title
Prediction of Antiviral Treatment Response of Hepatitis B Egyptian Patients using Machine Learning

Abstract
Abstract - Problem: Knowing the patient’s response to treatment is a very important issue. Therefore, it is necessary to predict the treatment response to know the effect of drugs. Recently, many machine learning techniques were applied for treatment response prediction. Methods: The aim of this study is to efficiently predict the antiviral treatment response of Hepatitis B Egyptian patients. This was achieved by applying multiple machine learning techniques, which are Decision Tree, Random Forest, k-Nearest Neighbor, Gradient Boosting. The input features include clinical laboratory features, quantitative level for Hepatitis B virus (DNAHBV) plus fibrosis stage and the antiviral drug type (Entecavir, Tenofovir and Lamivudine). Also, two over-sampling techniques were applied to overcome the data imbalance issues. They are Synthetic Minority Over-sampling Technique (SMOTE) and Adaptive Synthetic Sampling (ADASYN). Attribute selection method was also applied to reduce the dimensionality of features. Results: The highest accuracies achieved by Decision Tree, Random Forest, k-Nearest Neighbor, Gradient Boosting were 85.7%, 85.7%, 92.9% and 85.7% respectively. Conclusion: The best classification model for antiviral treatment response on HBV Egyptian patients was kNN classifier model. It achieved an accuracy of 92.9%, recall of 1.0 for response class and 0.75 for non-response class, and precision of 0.91 for response class and 1.0 for non-response class. Keywords - Machine-Learning; Treatment Response Prediction; Hepatitis B Virus Egyptian Patients; Decision Tree; Random Forest; K nearest Neighbor; Gradient Boosting.