Hepatitis C Disease Prediction Using Machine Learning Approach

The Hepatitis C virus (HCV), which affects millions of people globally is a serious global health issue. Effective management and diagnosis of HCV infection depend on early detection and precise prediction of its effects. The historical patient data that comprises demographic data (e.g., age, gender), clinical data (e.g., symptoms, medical history), and laboratory results (e.g., viral load, liver function tests) is used to train logistic regression models. Logistic regression methods calculate the chance or probability of HCV infection, disease progression, or treatment response by analyzing these factors. Logistic regression models provide targeted treatments and public health planning by prioritizing prevention efforts and resource allocation by considering factors like age, gender, lifestyle, and location. In order to accurately anticipate and categories diseases, machine learning (ML) techniques are frequently applied, serving as a useful tool for doctors and specialists. In this study, we suggest that to improve accuracy, a different model should be trained for each scenario. We made use of the accessible real-world data. The 615 patients in the collected dataset have 13 different traits. For model evaluation, the logistic regression approach has been used. The accuracy of the logistic regression model was initially 88.66%; however, it was increased to 90.24% after the hyper parameter values were changed.