Machine Learning Framework for Liver Cirrhosis Stage Prediction Using Clinical and Biochemical Features

Abstract

Chronic liver disease is characterized by liver cirrhosis, a progressive condition associated with high morbidity and mortality rates globally. Accurate classification of disease stage is essential for timely clinical intervention, comprehensive disease assessment, and the development of individualized treatment plans. This study proposes an interpretable machine learning model to predict the stage of liver cirrhosis using routinely collected clinical and biochemical parameters. A comprehensive exploratory data analysis was conducted to identify patterns and relationships among variables and to examine feature distributions across different stages of disease progression. After data preprocessing, including handling missing values, feature encoding, and normalization, multiple machine learning algorithms were evaluated, including Logistic Regression, Random Forest, Support Vector Machine, K-Nearest Neighbors, Gradient Boosting and Extreme Gradient Boosting (XGBoost). Model performance was assessed using accuracy, precision, recall, F1-score, and confusion matrix analysis. Experimental results demonstrated that ensemble-based models outperformed traditional machine learning methods, with XGBoost achieving the highest performance in multi-class stage prediction. Feature importance analysis identified clinical blood parameters such as bilirubin, albumin, platelet count, prothrombin time, and cholesterol as significant predictors of disease severity. The proposed framework is reliable, interpretable, and suitable for automated liver cirrhosis stage classification, offering valuable support to clinicians in early diagnosis and risk stratification. Results show the promising contributions of machine learning methods to decision support systems in healthcare, as well as their applicability to enhance the management of chronic liver diseases via leveraging data and predictive modelling.