Related Experiment Videos
Radial basis function neural network approach for the diagnosis of coronary artery disease based on the standard
1Institute of Precision and Biomedical Engineering, University of Technology, Warsaw. lewenk@mech.pw.edu.pl
Insights
This study evaluated a radial basis function neural network for diagnosing coronary artery disease using ECG exercise tests. The optimized network achieved over 97% accuracy, identifying both patient conditions and specific stenosed vessels.
Area of Science:
- Cardiology
- Artificial Intelligence
- Medical Diagnostics
Background:
- Coronary artery disease (CAD) diagnosis relies on various clinical data.
- Electrocardiogram (ECG) exercise tests are a common diagnostic tool.
- Computer-aided diagnosis (CAD) systems can enhance diagnostic accuracy.
Purpose of the Study:
- To evaluate a radial basis function neural network (RBFNN) for computer-aided diagnosis of CAD.
- To assess the RBFNN's performance using data from traditional ECG exercise tests.
- To optimize the RBFNN for improved diagnostic capabilities.
Main Methods:
- Utilized 776 patient records (297 healthy, 479 CAD) confirmed by coronary arteriography.
- Input data included ECG ST segment levels/slopes, heart rate, blood pressure, and patient symptoms.
- Optimized RBFNN by selecting radial function type, training methods, and regularization parameters.
Main Results:
- The best-performing RBFNN achieved over 97% accuracy on a 400-element test set.
- The network accurately diagnosed overall patient condition (healthy/unhealthy).
- The system identified specific stenosed coronary vessels with high accuracy.
Conclusions:
- Radial basis function neural networks are effective tools for computer-aided CAD diagnosis.
- Optimized RBFNNs can significantly improve diagnostic accuracy beyond simple classification.
- This approach offers potential for detailed identification of individual vessel stenosis.
Abstract:
The purpose of the paper is the evaluation of a radial basis function neural network as a tool for computer aided coronary artery disease diagnosis based on the results of the traditional ECG exercise test. The research was performed using 776 data records from an exercise test (297 records from healthy patients and 479 from ill patients) confirmed by coronary arteriography results. Each record described the state of the patient, provided input data for the neural network, included the level and slope of an ST segment of a 12-lead ECG signal made at rest and after effort, heart rate, blood pressure, load during the test, and occurrence of coronary pain, coronary arteriography, correct output pattern for the neural network, and verified the existence (or not) of more than 50% stenosis of the particular coronary vessels. Radial basis function neural networks for coronary artery disease diagnosis were optimised by choosing the type of radial function, the method of training (setting the number of centres and their dimensions), and regularisation. The best network correctly recognised over 97% of cases from a 400-element test set, diagnosing not only the patients' condition (simple 'healthy/unhealthy' diagnosis), but also pointing out individual unhealthy/stenosed vessels.