PSO-based support vector machine with cuckoo search technique for clinical disease diagnoses

Area of Science:

• Computational intelligence
• Machine learning for healthcare
• Biomedical data analysis

Background:

• Accurate disease diagnosis is crucial for effective treatment.
• Traditional diagnostic methods can be time-consuming and may lack precision.
• Machine learning offers potential for automated and improved diagnostic accuracy.

Purpose of the Study:

• To develop a novel hybrid machine learning model for enhanced disease diagnosis.
• To improve the classification accuracy and efficiency of diagnostic algorithms.
• To evaluate the performance of the proposed model against existing methods.

Main Methods:

• Hybridization of support vector machine (SVM) with cuckoo search (CS) and particle swarm optimization (PSO).
• A two-stage optimization process: CS for initial SVM kernel parameter tuning, followed by PSO for further SVM parameter optimization.
• Comparative analysis against Particle Swarm Optimization-Support Vector Machine (PSO-SVM) and Genetic Algorithm-Support Vector Machine (GA-SVM) models.

Main Results:

• The proposed CS-PSO-SVM model demonstrated superior classification accuracy.
• The hybrid model achieved a higher F-measure compared to PSO-SVM and GA-SVM.
• Experimental results confirm the model's effectiveness in disease diagnosis.

Conclusions:

• The CS-PSO-SVM model is a highly efficient and accurate method for disease diagnosis.
• This novel approach offers significant advantages over previously reported machine learning algorithms.
• The findings suggest a promising direction for advancing machine learning applications in medical diagnostics.