A Hybrid Machine Learning Framework for Early Diagnosis of Chronic Liver Disease

View at publisher Summarise with VIRAT
Authors: T. V. N. L. Aswini, Kamesh Sonti, Purnima K. Sharma, A V N Kumara Swamy, T. J. V. Subrahmanyeswara Rao
Cited by: 0
References: 8

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Chronic liver disease (CLD) is a severe global health concern that is frequently recognized only in the late stages due to limitations in standard diagnostic approaches. Reducing mortality and improving patient outcomes require early, accurate, and non-invasive CLD prediction. For the early diagnosis of CLD, this study offers a novel hybrid machine learning (ML) framework that combines Extreme Gradient Boosting and K-Nearest Neighbors. A dataset with demographic, lifestyle, and biochemical information for 1,700 people was examined. With an accuracy of 98%, precision of 0.96, recall of 0.95, and an AUC of 0.99, the suggested hybrid model performed better than any other. These results demonstrate how hybrid ensemble models can improve prediction accuracy, provide a dependable non-invasive method for early CLD screening, and open the door for sophisticated healthcare decision-support systems. Notably, the suggested paradigm shows potential for early-age screening as well, considering the rising prevalence of pediatric and adolescent CLD associated with obesity and lifestyle modifications.

Fields of Study & Subjects

Artificial Intelligence in Healthcare Liver Disease Diagnosis and Treatment Machine Learning and ELM

Publication details

Journal / source
International Conference on Signal Processing and Electronics Design (ICSPED)
Pages
241–245
Publication date
23 January 2026
Language
EN