"Enhancing Chronic Kidney Disease Diagnostics through a Novel Hybrid Multi-Model Fusion Approach in Machine Learning"
Year: 2024
Type: article
Abstract: In recent years, intelligent classification techniques, especially those based on machine learning (ML), have gained significant traction, transforming various sectors, notably the medical industry. ML has emerged as a critical tool in healthcare, enhancing the diagnosis and prediction of illnesses. This paper focuses on applying ML algorithms for diagnosing chronic kidney disease (CKD), a significant challenge in medical diagnostics. We explored the efficacy of Artificial Neural Networks (ANNs), Support Vector Machines (SVM), Logistic Regression (LR), and introduced a novel Hybrid Multi-Model Fusion (HMMF) method, utilizing a dataset of 153 cases with 11 CKD patient attributes. The objective was to assess and compare these classifiers’ predictive capabilities for CKD diagnosis through several metrics. The study revealed that ANNs outperformed the LR model, with ANNs achieving 84.44% accuracy, 84.21% sensitivity, 84.61% specificity, and an AUC-ROC of 84.41%, highlighting the robustness of ANNs in complex medical prediction tasks.
Author: Pinar Karadayi Ataş
Institution: Istanbul Arel University
Language: English
Subfield: Health Information Management
Field: Health Professions
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W4392371963