Machine learning based automated liver fibrosis stage diagnosis with prediction

The healthcare sector generates large amounts of data. Hepatitis $\mathbf{C}$ virus-related chronic liver disorders frequently show considerable signs of fibrosis. The significant increase in the number of HCV-infected patients, the expensive price, and the problems with the method of evaluating liver fibrosis are all contributing to it becoming a global issue. This study intends to identify the characteristics that substantially aid in identifying the phases of liver fibrosis and to produce rules to support medical professionals in treating patients using a clinically noninvasive method. It provides a lot of potential for automated illness prediction using health-related data analysis. Liver illness is one of the main reasons for the nation’s high death rates and is regarded as a life-threatening condition everywhere. People might experience the effects of liver illness early in life. Liver diseases account for around $2.4 \%$ of yearly deaths in India. Due to the early-stage weak signs of liver disease, detection is especially challenging. The warning indicators always appear exclusively in late. Due to the increased issues that liver-related disease creates for individuals today, it is crucial to understand its origins and begin the diagnosis process. This means that a more accurate and reliable automated programme is required for the early diagnosis of liver disease. Finding patterns in large amounts of data or large data sets in order to make decisions is known as machine learning. This allows machines to learn (either supervised, unsupervised, semi-supervised, or reinforced). The Liver Patient data set used in this paper comes from the UCIRepository at the University of California, Irvine. In order to predict the disease, for this, certain machine learning models are developed. Performance analysis uses accuracy, specificity, and precision as parameters.