Comparative Analysis of Liver Disease Detection using Diverse Machine Learning Techniques

Liver disease is one of the threatening diseases that will cause severe damage to human life if it is not detected in its early stages. Liver disease depends on the habits of people, such as alcohol addiction, environmental pollution, medical drugs, and fast food. The detection of liver disease is one of the major challenges in the healthcare system due to its late symptoms, such as blood vomiting, stomach pain, jaundice, etc. Healthcare professionals use different methodologies to make decisions based on the patient’s medical reports to find liver diseases. Early diagnosis and prognosis of liver disease are crucial factors for rapid treatment and a reduction in serious health consequences. In this modern technological world, machine learning (ML) based systems can be very useful to medical professionals to diagnose liver diseases in their initial stages and help them recover rapidly. This research work presents a ML-based models to analyze and predict liver disease in its early stages using patients' blood report data sets. The comparative analysis implementation was carried out with more than 30,000 data sets using various ML techniques such as Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Artificial Neural Networks (ANN). The experiment results show that the Random Forest technique achieved better performance in terms of good accuracy and precision when compared to other techniques.