Chronic Liver Disease Classification Using Machine Learning Techniques

Abstract

The chronic liver disease provides significant challenges to the global healthcare system, as it consumes the majority of healthcare budget, particularly in low-income nations. Early classifying and prevention based on disease severity is among the most popular significant issues facing the health sector, especially in developing countries such as Ethiopia. Machine learning has an important role in analyzing a huge amount of medical information and addressing complex problems for early disease classification. The goal of this research was to use machine-learning algorithms to classify chronic liver disease as not chronic liver disease or chronic liver disease, depending on whether the disease was present or not. In this study, the experimental design have been used with an experimental approach to determine the most appropriate performance model. The data for this study purpose was collected from Nekemte Referal Hospital. The three algorithms employed in the study were support Vector Machine, Logistic Regression, and Nave Bayes. Evaluation of the models was done using 10-fold cross-validation and classification performance in order to compare the models. The performance of the given classification techniques was evaluated using accuracy, precision, recall, F1-score, sensitivity, and specificity. Logistic Regression (LR), Support Vector Machine (SVM), and Naive Bayes have accuracy rates of 84%, 92%, and 83%, respectively. The study findings show that, Support Vector Machine record better performance compared to Logistic Regression and Na??ve Bayes models based on accuracy and Fl-score.