An Automatic Diagnosis System Based on Machine Learning Models for Predicting Hepatitis C from Blood Samples

Hepatitis C is an irresistible disease that has affected about 70 million people around the world. It is also the reason of 400 thousand deaths annually. The proper diagnosis of diabetic patients and the detection of hepatitis C in earlier stages help in saving lives. Electronic health records (EHRs) are very useful for understanding the hepatitis C and its prognosis in a better way. It usually contains the statistical data of clinical practice that is generated by the computer base techniques. These techniques unveiled the new trends and patterns that are unnoticeable by the doctors. In this chapter, EHR of 615 patients are examined using proposed automatic diagnosis system based on different Machine Learning (ML) models to detect the hepatitis C. The basic aim of this study is to choose the best ML model for detecting the blood donor, suspected blood donor, hepatitis, fibrosis, and cirrhosis from real-time laboratory data of blood samples. The comparative study between ML models such as Support Vector Machine (SVM), Multivariate Adaptive Regression Splines (MARS), Bayesian Generalized Linear Model (BGLM), Random Forest (RF), and Decision Tree (DT) is carried out using accuracy performance matrix. A brief study on the abovementioned ML models is accomplished according to prediction accuracy of disease diagnosis. All the ML models are validated and implemented using R programming language.