Quantum-Inspired gravitationally guided particle swarm optimization for feature selection and classification.

Area of Science:

• Computational Intelligence
• Medical Informatics
• Optimization Algorithms

Background:

• Population-based metaheuristic algorithms balance exploration and exploitation for complex problems.
• Existing methods like Genetic Algorithms, Particle Swarm Optimization, and Gravitational Search Algorithm face limitations such as premature convergence and parameter sensitivity.
• Accurate diagnosis of Non-Communicable Diseases (NCDs) is crucial for effective patient treatment.

Purpose of the Study:

• To introduce Quantum-Inspired Gravitationally Guided Particle Swarm Optimization (QIGPSO) for complex optimization challenges.
• To enhance the diagnosis of Non-Communicable Diseases (NCDs) using advanced metaheuristic optimization in medical data analysis.
• To leverage the strengths of Quantum Particle Swarm Optimization (QPSO) and Gravitational Search Algorithm (GSA) for improved search processes.

Main Methods:

• Developed QIGPSO by integrating QPSO and GSA to harness global and local search capabilities.
• Implemented an absolute Gaussian random variable and modified position update equations to refine the search mechanism.
• Utilized a wrapper-based method with Support Vector Machine (SVM) for feature selection and classification of NCD datasets.

Main Results:

• QIGPSO demonstrated effectiveness in identifying critical features within medical datasets for NCD diagnosis.
• The algorithm achieved high accuracy rates and reduced misclassification numbers across various NCD datasets.
• QIGPSO exhibited faster convergence compared to conventional optimization methods, improving the exploration-exploitation balance.

Conclusions:

• QIGPSO offers a robust and efficient approach for feature selection and classification in NCD medical data analysis.
• The enhanced optimization technique provides valuable data insights, supporting clinicians in making informed treatment decisions.
• QIGPSO effectively addresses limitations of traditional algorithms, paving the way for improved diagnostic tools in healthcare.