Early Prediction and Detection of Chronic Kidney Disease Using Multi-Input Model

View at publisher Summarise with VIRAT
Authors: S. Sangeetha, S. Deepajothi
Cited by: 0

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 kidney disease (CKD) is a major global health problem which requires reliable early detection and prediction strategies for better patient outcome and better treatment choices. In this paper, we present a novel framework for predicting CKD utilizing a multi input deep learning architecture combining clinical and non-clinical factors including demographic data, lifestyle, medication history, quality of life score, and medical checkup frequency. Here we had a separate layer for processing the numerical and categorical features. Embedding layers are used for categorical features and Attention layers are used to gain the feature importance to make the model more interpretable. Random Forest was used as a feature selection technique to determine the most appropriate predictors. We empirically evaluate the multi-input deep learning network framework and achieve an accuracy of 97.16%, which shows that it is capable of learning complex interactions across diverse data. Additionally, an ensemble model employing XG-Boost in combination with the multi-input deep learning network was built in order to improve the accuracy up to 98.03%. This comprehensive framework achieves this by unifying heterogeneous data and introducing advanced deep learning architecture and ensemble techniques for early detection and early prediction of CKD. The results demonstrate the power of proposed multi input deep learning and ensemble framework to establish a new standard by integrating a wide range of health indicators (clinical and non-clinical factors) for predicting the CKD at early, enable timely interventions and aid better patient care in clinical domains.

Fields of Study & Subjects

Artificial Intelligence in Healthcare Machine Learning in Healthcare Chronic Kidney Disease and Diabetes

Publication details

Journal / source
CRC Press eBooks
Pages
38–45
Publication date
31 March 2026
Language
EN
Document type
Book chapter