An Intelligent Web-Based Liver Disease Prediction System Using Machine Learning Techniques

Abstract

Liver disease has emerged as one of the major health concerns worldwide due to lifestyle changes, alcohol consumption, and viral infections. Early detection of liver disorders is crucial for effective treatment and prevention of severe complications. Traditional diagnostic methods often rely on clinical expertise and laboratory tests, which may be time-consuming and prone to human error. To address these challenges, this project presents an intelligent web-based system for liver disease prediction using machine learning techniques. The proposed system integrates a Django-based web application with machine learning models to provide a user-friendly platform for disease prediction. The system allows users to register, log in, and input medical parameters such as age, bilirubin levels, enzyme levels, protein content, and other relevant features. These inputs are processed and analyzed using classification algorithms to determine whether the user is likely to have liver disease. The dataset used in this system is the Indian Liver Patient Dataset (ILPD), which contains both liver disease and non-liver disease cases. Data preprocessing techniques such as handling missing values, normalization using StandardScaler, and feature encoding are applied to enhance model performance. Multiple machine learning algorithms including Decision Tree, Logistic Regression, and Support Vector Machine (SVM) are implemented and evaluated. The Decision Tree classifier is used for real-time prediction due to its simplicity and interpretability. Logistic Regression and SVM models are used for performance comparison, providing accuracy, precision, sensitivity, and specificity metrics. The system includes an admin module for managing users and viewing classification results, ensuring efficient system monitoring. The proposed solution offers several advantages such as automated diagnosis support, reduced dependency on manual analysis, and improved prediction accuracy. It also provides scalability and accessibility through a web interface. The system can assist healthcare professionals and individuals in early detection and decision-making. In conclusion, the integration of machine learning with web technologies provides an efficient and reliable solution for liver disease prediction. Future enhancements may include deep learning models, real-time hospital integration, and mobile application support to further improve usability and accuracy.