Towards real-time detection of myocardial infarction by digital analysis of electrocardiograms

Myocardial infarction (MI) is one of the most common sudden-onset heart diseases. Early diagnosis and management of heart ischemia result in good prognosis. Early changes in the heart muscle activity after ischemia reflect in ST segment elevation on electrocardiogram (ECG) recordings. With the development of signal processing techniques and the portable devices, there is a need to develop a real-time algorithm that accurately detects MI non-invasively. In this paper, we propose a computer algorithm that employs digital analysis scheme towards the real-time detection of MI. The proposed algorithm extract features based on clinical diagnosis conditions allowing the continuous analysis of ST segment and simultaneous detection of abnormal heart activity resulting from MI. Using an online ECG library of patient data, the signals were filtered for high frequency noise, baseline drift then features of interest (Q, R, S waves and J points) were extracted. These were used to measure the ST segment elevation and depression as an important indicator of MI defined in clinical guideline for MI diagnosis. The developed algorithm was capable of detecting MI with 85% sensitivity and 100% specificity in a test set of 40 ECG recordings.