Cirrhosis Disease Prediction Using Machine Learning

Cirrhosis is a serious liver condition that can be fatal. It frequently advances silently until symptoms appear in the late stages. For prompt intervention and better patient outcomes, early detection and precise prediction of cirrhosis are essential. Liver disease is a serious worldwide health issue, and effective treatment depends on early detection. In this project, we use a variety of machine learning techniques to predict liver illness using clinical variables from the Indian Cirrhosis Patient Records dataset available from Kaggle. These techniques include Regression, Logistic Regression, Random Forest, KNN, and Support Vector Machines. We start by thoroughly preparing the data for model dependability, managing missing values, outliers, and normalizing the data. Then, feature selection is carried out to determine the factors most closely related to liver illness. The predictive models are then developed, improved upon, and evaluated using pertinent metrics and cross-validation procedures. The objective of the study is to improve patient care and results in the field of medical diagnostics by developing accurate predictive models for the early diagnosis of liver disease, as well as gaining insights into the associations between clinical characteristics and liver disease.