Utilizing Diverse Machine Learning Models for Liver Disease Patient Prediction

• Published in: International Conference on Computing Communication Control and Automation (Online) pp. 1 - 6
• Main Authors: Shah, Khushi, Barage, Abhay, Maluskar, Aditya, Nagare, Gajanan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 23.08.2024
• Subjects: Accuracy; bagging; ensemble learning; extra tree classifier; feature scaling; Forestry; gradient boosting; hepatitis; Imputation; Liver diseases; machine learning; multivariate imputation; Predictive models; random forest; Random forests; Real-time systems; Solids; stacking; Testing; Viruses (medical); XGBoost
• ISSN: 2771-1358
• DOI: 10.1109/ICCUBEA61740.2024.10774747

The liver-damaging virus known as hepatitis C is still a major global health concern. The need for better early detection strategies is highlighted by the serious consequences that can arise from delayed diagnosis and treatment. This study investigates the possibility of using real-time data analysis and machine learning (ML) models to forecast cases of hepatitis C. The recommended method makes use of a number of data preprocessing strategies, such as resolving data imbalance, scaling features, and picking pertinent features to improve accuracy via appropriate imputations. Multivariate imputation fills in the missing values. Skewed columns are subjected to standardisation, min-max scaling, maximum absolute scaling, and robust scaling techniques in addition to the log1p transformation. The model's output was compared both internally and to other models that had been employed in earlier studies. The suggested model outperformed alternative techniques with the highest testing accuracies of 91.82% and 86.06%, respectively, using extra tree classifier and random forest, indicating its potential as a practical solution for liver disease detection.