Machine Learning for Ischemic Heart Disease Diagnosis Aided by Evolutionary Computing

Area of Science:

• Cardiology
• Artificial Intelligence
• Medical Diagnostics

Background:

• Heart disease is a leading cause of mortality globally, particularly in developing nations.
• Accurate and timely diagnosis is often hindered by limited resources and healthcare professionals.
• Technological advancements are crucial for improving diagnostic accuracy in patient care.

Purpose of the Study:

• To develop a hybrid diagnostic tool for detecting ischemic heart disease.
• To leverage computational intelligence techniques for analyzing patient data and electrocardiogram (ECG) signals.
• To enhance diagnostic capabilities in areas with limited medical resources.

Main Methods:

• Developed a hybrid diagnostic system integrating multiple machine learning techniques.
• Utilized a database of clinical histories from approximately 1020 patients.
• Analyzed 92 electrocardiogram (ECG) signal images using an Artificial Neural Network.

Main Results:

• The hybrid tool achieved up to 97.01% accuracy in diagnosing ischemic heart disease.
• The system effectively analyzed both clinical data and ECG images.
• The tool demonstrated high diagnostic performance in preliminary testing.

Conclusions:

• The developed hybrid tool shows significant potential for accurate ischemic heart disease diagnosis.
• This technology can aid medical professionals, especially in resource-limited settings.
• Positive feedback from medical experts validates the tool's effectiveness and clinical utility.