Machine Learning-Based Analysis and Prediction of Liver Cirrhosis

Abstract

Liver cirrhosis poses a significant threat as a highly infectious blood-borne illness, often remaining asymptomatic in its initial stages, thereby complicating early diagnosis and treatment. As the disease advances to its later stages, diagnostic and therapeutic interventions become increasingly daunting. This work endeavors to provide a robust solution by introducing an Artificial Intelligence (AI) system driven by state-of-the-art Machine Learning (ML) algorithms, aiming to aid healthcare professionals in the early detection of liver cirrhosis. Multiple ML algorithms are under development with the primary objective of predicting the likelihood of liver cirrhosis infection. Through this research, seven distinct models have been crafted, leveraging diverse parameters and employing different ML algorithms such as Logistic Regression (LR), Linear Discriminant Analysis (LDA), k-Nearest Neighbors (KNN), Random Forest (RF), Multi-layer Perceptron (MLP), AdaBoost, and Bernoulli Naive Bayes (BernoulliNB). Among these, the RF algorithm emerged as the frontrunner, exhibiting an impressive accuracy rate of approximately 98 percent. Utilizing an open-access liver cirrhosis dataset, this methodology surpasses earlier research endeavors, showcasing a substantial improvement in predictive accuracy. Rigorous comparisons between models underscore their robustness, affirming their reliability and establishing a clear pathway forward for future investigations in this domain.

Topics & Concepts