Date of Award

Fall 2001

Document Type


Degree Name

Doctor of Philosophy (PhD)


Biomedical Engineering

First Advisor

Stan A. Napper


In the present dissertation, an automated neural network-based ECG diagnosing system was designed to detect the presence of myocardial infarction based on the hypothesis that an artificial neural network-based ECG interpretation system may improve the clinical myocardial infarction. 137 patients were included. Among them 122 had myocardial infarction, but the remaining 15 were normal. The sensitivity and the specificity of present system were 92.2% and 50.7% respectively. The sensitivity was consistent with relevant research. The relatively low specificity results from the rippling of the low pass filtering. We can conclude that neural network-based system is a promising aid for the myocardial infarction diagnosis.