Detection of Myocardial Infarction using High-Frequency QRS Components

This paper presents a method for detecting Myocardial Infarction. one hundred subjects were selected from PTB database in which reason for admission of 50 people is myocardial infarction and 50 people is healthy control. In order to extract highfrequency components, signal averaging was applied to the signal with the aid of cross-correlation function. Averaged signal is then passed through a bandpass filter (150-250 Hz). Seven features were extracted from 7 different leads. A linear classifier was used in order to distinguish between patients and healthy people .The accuracy of the proposed method is 91% with 90% sensitivity and 92% specificity. Simplicity and speed of proposed method makes it useful in practical diagnosis.

Keywords:

High-frequency QRS , Cross Correlation , Statistical feature , Linear classifier , Averaging