Automated Prediction of Liver Disease using Machine Learning (ML) Algorithms
Aviral Srivastava, V Vineeth Kumar, T R Mahesh, V Vivek
Abstract
Information systems and strategic tools have lately been offered as new strategies in medical research to improve the disease detection process. Enzyme activation, bile synthesis, lipid metabolism, and vitamin, glycogen, and mineral storage are all crucial functions of the liver. Because liver problems are difficult to diagnose in the early stages due to a lack of particular symptoms, they are commonly neglected. Hyperbilirubinemia is common symptom in many liver diseases, and it's tough to tell the difference early on. However, this is not guaranteed, and understanding enzyme levels is necessary to detect and confirm the presence of liver illness. A variety of machine learning (ML) algorithms are being used to predict liver diseases. We recommend employing Logistics Regression (LR), Naive Bayes Model (NB), K-Nearest Neighbor (Knn) in this project. The data was separated into two categories: patients with liver disease and sicknesses the most accurate machine learning method was used to predict the final result. The different algorithms are compared against various performance metrics and the best one is identified.