Authors:
Mohamed A. Mneimneh
;
Micheal T. Johnson
and
Richard J. Povinelli
Affiliation:
Electrical and Computer Engineering, Marquette University, United States
Keyword(s):
Inverse problem, Ischemia, Decision Tree.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Business Analytics
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Decision Support Systems
;
Enterprise Information Systems
;
Health Information Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
Due to the increasing prices of medical care, and especially due to cardiovascular injury; scientists are looking for inexpensive and less invasive ways to diagnose myocardial ischemia. Many studies have shown that the variations of the ST-segment in the ECG signal are an indicator for ischemia. For this purpose, this work proposes an approach based on a heart cell group model and principle component analysis, using a decision tree classifier to differentiate between the ischemic and healthy beats. The cardiac based model is based on a physiological model of the electrical cycle of depolarization and repolarization of the atria and ventricles. The model parameters are estimated by minimizing the squared error between the generated signal and the recorded ECG. The approach is applied to beats from the Long-Term ST database, which consists of 86 subjects and more than 20,000 beats in which 80% of the beats are ischemic and 20% are healthy. A 10-fold cross validation test is performed o
ver the dataset. The accuracy of this approach is 91.62%, with sensitivity of 95.09% and specificity of 75.66%.
(More)