Prediction of Heart Attack Risk Using GA-ANFIS Expert System Prototype

Area of Science:

• Artificial Intelligence
• Medical Informatics
• Computational Intelligence

Background:

• Accurate heart disease diagnosis is crucial for effective patient management.
• Existing diagnostic methods can be improved with advanced computational techniques.
• Expert systems offer potential for enhanced medical decision-support.

Purpose of the Study:

• To develop and validate a novel GA-ANFIS expert system for classifying heart disease severity.
• To optimize fuzzy models using both ANFIS and GA for improved diagnostic performance.
• To compare the efficacy of the GA-ANFIS system against the traditional ANFIS model.

Main Methods:

• Utilized thirteen heart disease attributes as input features for classification.
• Employed Adaptive Neuro-Fuzzy Inference Systems (ANFIS) for initial fuzzy model optimization.
• Integrated Genetic Algorithm (GA) for second-level fuzzy model optimization within the GA-ANFIS framework.
• Performed system modeling and validation in the MATLAB environment.

Main Results:

• The GA-ANFIS system demonstrated superior efficiency in heart disease diagnosis compared to the ANFIS model.
• The proposed GA-ANFIS model, incorporating a predicted value technique, enhanced diagnostic accuracy.
• Comparative analysis confirmed the improved performance of the GA-ANFIS approach.

Conclusions:

• The GA-ANFIS expert system is a more efficient and accurate tool for heart disease diagnosis.
• This novel approach offers significant advantages over traditional ANFIS methods.
• The developed system shows promise for real-world clinical application in heart disease assessment.