Cardiac state diagnosis using adaptive neuro-fuzzy technique

Area of Science:

• Cardiology
• Biomedical Engineering
• Artificial Intelligence in Medicine

Background:

• Heart rate variability analysis is a key noninvasive method for assessing autonomic nervous system (ANS) function.
• Cardiac signals can reveal current diseases or predict future health issues.
• Manual analysis of large heart rate datasets for abnormalities is labor-intensive and time-consuming.

Purpose of the Study:

• To develop and evaluate an automated system for classifying heart abnormalities.
• To assess the effectiveness of an adaptive neuro-fuzzy network in analyzing cardiac states.
• To improve the efficiency and accuracy of diagnosing heart conditions using heart rate data.

Main Methods:

• An adaptive neuro-fuzzy network was employed for the classification task.
• The system was trained and tested on data representing 10 distinct cardiac states.
• Performance was evaluated based on classification accuracy.

Main Results:

• The adaptive neuro-fuzzy network demonstrated high efficacy in classifying cardiac states.
• The system achieved an accuracy level exceeding 94% in identifying heart abnormalities.
• The proposed method proved effective in handling voluminous heart rate data.

Conclusions:

• Automated analysis of heart rate using adaptive neuro-fuzzy networks offers a powerful tool for detecting cardiac abnormalities.
• This approach significantly reduces the time and effort required for manual data analysis.
• The high accuracy achieved suggests potential for clinical application in early disease detection and patient monitoring.