Abstract:

It's crucial to predict liver disease in its early stages. In contrast to urban areas and major cities, where liver disease is more severe and currently more prevalent, rural areas have a low prevalence of the condition. Millions of people die each year as a result of liver disease. Early liver illness does not reveal any concerns. Patients' chances of survival can be improved by early detection of liver disease complications. In this project, we used a dataset of Indian patients with liver disease that included information on age, gender, total and direct bilirubin levels, alkaline phosphatase, alanine and aspartate aminotransferases, total proteins, albumin and globulin ratios, results, and 416 patients with liver disease and 167 patients without liver disease. The project seeks to forecast liver illness based on the user's blood test report results. The project's primary area of focus is machine learning, which covers data science and artificial intelligence. We estimate the risk of liver illness using machine learning methods. Random Forest and Logistic Regression are two machine learning methods used in this project. The research then utilizes the model's training data to predict whether or not a person has liver disease.

Description:

Random Forest Algorithm, Logistic Regression Algorithm, Indian liver patient datasets