Predictive Modeling of Liver Disease using Machine Learning on Indian Patient Data
Abstract
Liver diseases are a growing concern in developing countries like India, where diagnosis is often delayed due to limited access to specialized healthcare. This study utilizes the Indian Liver Patient dataset to build predictive models for early detection of liver disease. We employ logistic regression, decision trees, and random forest classifiers to predict whether a patient has a liver condition. Our results show that the random forest classifier achieves the highest accuracy of 79.4%, highlighting the potential of machine learning in clinical decision support systems.
Keywords
Download Options
Introduction
Liver disease represents a significant health burden globally and particularly in India, where conditions such as cirrhosis and hepatitis are common. Early diagnosis is crucial, yet challenging due to vague symptoms and poor health infrastructure in rural areas. Machine learning can offer a non-invasive, cost-effective solution for liver disease detection using routine blood tests and demographic data.
In this paper, we analyze the Indian Liver Patient dataset to build predictive models for liver disease diagnosis, aiming to assist healthcare providers in making faster and more accurate decisions.
Conclusion
The study demonstrates that machine learning, particularly Random Forest, can effectively predict liver disease based on routine clinical data. With an accuracy of nearly 80%, such models can serve as useful screening tools in rural or resourcelimited settings. Future work could explore ensemble stacking or deep learning models for further improvement.