Advanced CKD detection through optimized metaheuristic modeling in healthcare informatics

Area of Science:

• Medical Informatics
• Machine Learning
• Computational Biology

Background:

• Accurate medical data categorization is crucial for illness prediction in healthcare informatics.
• Machine learning and deep learning models show promise but face challenges like high dimensionality and computational complexity.
• Existing methods for Chronic Kidney Disease (CKD) diagnosis require improvement in efficiency and accuracy.

Purpose of the Study:

• To present a novel classification model using metaheuristic methods for efficient Chronic Kidney Disease (CKD) diagnosis.
• To address the limitations of existing techniques in medical data categorization and illness detection.
• To improve the accuracy and efficiency of CKD prediction through optimized feature selection and classification.

Main Methods:

• Extensive pre-processing of medical data, including missing value imputation, transformation, and normalization.
• Application of the Binary Grey Wolf Optimization method for reliable subset feature selection.
• Utilizing the Extreme Learning Machine (ELM) with optimized hidden nodes for CKD classification.

Main Results:

• The proposed model demonstrated high accuracy in Chronic Kidney Disease (CKD) diagnosis.
• The metaheuristic feature selection significantly improved prediction accuracy compared to existing models.
• The optimized Extreme Learning Machine (ELM) effectively classified CKD presence.

Conclusions:

• The novel metaheuristic-based classification model offers a more accurate and efficient approach to Chronic Kidney Disease (CKD) diagnosis.
• Optimized feature selection and data processing are key to enhancing machine learning performance in medical informatics.
• This study provides a valuable contribution to improving illness detection and healthcare informatics systems.