Disease Diagnosis Based on Improved Gray Wolf Optimization (IGWO) and Ensemble Classification.

Area of Science:

• Computational biology
• Medical informatics
• Machine learning

Background:

• Accurate disease diagnosis is crucial for effective treatment.
• Traditional feature selection methods may not be optimal for complex medical datasets.
• Ensemble classification offers potential for improved predictive performance.

Purpose of the Study:

• To introduce an improved Gray Wolf Optimization (IGWO) algorithm for enhanced disease diagnosis.
• To develop a two-phase strategy involving feature selection and ensemble classification.
• To evaluate the efficacy of IGWO in selecting optimal features for disease diagnosis.

Main Methods:

• A novel Improved Gray Wolf Optimization (IGWO) algorithm was developed.
• The IGWO algorithm was applied to the Feature Selection Phase (FSP) to identify key diagnostic features.
• An Ensemble Classification Phase (ECP) utilized voting from five classifiers: Naïve Bayes, SVM, DNN, Decision Tree, and KNN.

Main Results:

• The proposed IGWO method demonstrated superior performance in feature selection compared to existing techniques.
• The integrated IGWO and ensemble classification strategy significantly improved disease diagnosis accuracy.
• Key performance metrics including precision, recall, and overall accuracy were enhanced.

Conclusions:

• The IGWO algorithm is an effective optimization technique for feature selection in disease diagnosis.
• Ensemble classification, powered by IGWO-selected features, enhances diagnostic strategy performance.
• This approach offers a promising tool for improving the accuracy and reliability of disease diagnosis systems.