Intelligent Diagnostic Prediction and Classification System for Chronic Kidney Disease

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Computational Biology

Background:

• Healthcare systems are integrating advanced technologies like machine learning (ML) and artificial intelligence (AI) for enhanced services.
• Accurate prediction and classification of chronic kidney disease (CKD) are crucial for timely intervention and patient management.
• Existing methods for CKD prediction may suffer from feature redundancy and suboptimal classification accuracy.

Purpose of the Study:

• To propose an intelligent prediction and classification system for chronic kidney disease (CKD).
• To develop a novel framework, Density based Feature Selection (DFS) with Ant Colony based Optimization (D-ACO), for improved CKD diagnosis.
• To evaluate the efficacy of the D-ACO algorithm in enhancing classification accuracy while reducing feature dimensionality.

Main Methods:

• The D-ACO framework comprises three phases: preprocessing, feature selection (FS) using DFS, and classification using ACO.
• Density based Feature Selection (DFS) is employed to eliminate irrelevant or redundant features prior to classifier construction.
• The Ant Colony Optimization (ACO) algorithm is utilized for building the classification model.

Main Results:

• The D-ACO algorithm was tested on a benchmark CKD dataset.
• Performance evaluation demonstrated significant improvements in classification accuracy compared to existing methods.
• The proposed system achieved higher accuracy using a reduced set of features, indicating efficient feature selection.

Conclusions:

• The D-ACO algorithm presents a robust and intelligent approach for CKD prediction and classification.
• The integration of DFS and ACO effectively addresses feature redundancy and enhances diagnostic performance.
• This intelligent system offers a promising advancement in healthcare services for chronic kidney disease management.