Hepatitis C Prediction Using Logistic Regression

Chong, Stanley Shi Wen (2025) Hepatitis C Prediction Using Logistic Regression. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RMM_Stanley Chong Shi Wen_Full Text.pdf Restricted to Registered users only Download (1MB)

Abstract

In this project, the prediction of Hepatitis C infection has been investigated using Logistic Regression. A dataset comprising demographic and clinical variables from blood donors and Hepatitis C patients was analyzed. The report consists of five chapters. Chapter 1 introduces the project, highlighting the significance of Hepatitis C as a global health concern and the need for early detection. The objective of this project is to develop a logistic regression model to predict Hepatitis C infection with an accuracy of at least 85%, using key clinical and demographic indicators. Chapter 2 reviews recent literature on predictive modeling for Hepatitis C, comparing machine learning techniques such as logistic regression, Random Forest, and XGBoost. The chapter identifies gaps in existing research and underscores the suitability of logistic regression for binary classification tasks in medical diagnostics. Chapter 3 details the methodology, including data preprocessing, feature selection, and logistic regression modeling. Key statistical tests, such as the Hosmer-Lemeshow test and Wald statistic, are employed to validate the model. Evaluation metrics, including accuracy, sensitivity, and specificity, are used to assess performance. Chapter 4 presents the results of Exploratory Data Analysis (EDA) and logistic regression modeling. The final model, refined through iterative feature selection, achieved an accuracy of 94.8% and identified significant predictors such as AST, BIL, and GGT. The chapter also confirms that the model meets regression assumptions, including multicollinearity checks. Chapter 5 summarizes the findings, emphasizing the model’s potential as a preliminary screening tool for Hepatitis C. Recommendations for future research include expanding the dataset, exploring advanced machine learning techniques, and validating the model in clinical settings. This study aims to support healthcare providers in early diagnosis and intervention, ultimately improving patient outcomes. Keywords: Hepatitis C, logistic regression, prediction