ANALYSIS OF MEASURES FOR ACCURATE DIAGNOSIS OF VIRAL HEPATITIS AT AN EARLY STAGE AND THE NEED FOR THEIR IMPROVEMENT

Abstract

Due to its high rates of morbidity and mortality, hepatitis remains a serious worldwide health concern.  Early prediction of hepatitis outcomes is still a crucial area for development, despite advancements in diagnosis and therapy.  By using a variety of cutting-edge machine learning (ML) algorithms to predict hepatitis, this study aims to close this gap and support international initiatives to improve public health outcomes. The hepatitis dataset from the UCI repository, which comprises 155 people and 20 characteristics pertaining to clinical information, test results, and demographics, was used in the study. Despite advancements in antiviral medication and effective vaccinations, viral hepatitis continues to have a significant global burden.  Hepatitis A, B, C, D, and E are the five types of hepatitis viruses.  Along with HIV infection, malaria, and tuberculosis, hepatitis B and hepatitis C virus infections rank among the top four infectious diseases in the world in terms of mortality.  About 47% of such deaths can be attributed to the hepatitis B virus, 48% to the hepatitis C virus, and the remaining portion to the hepatitis A and hepatitis E viruses.