Early Diagnosis and Automated Analysis of Myocardial Infarction (STEMI) by Detection of ST Segment Elevation Using Wavelet Transform and Feature Extraction

摘要

Myocardial Infarction(MI) has remained the predominant and most serious of all the cardiovascular diseases over the years. Myocardial Infarction also termed as heart attack, happens when the flow of blood decreases or ceases to the heart or a part of it, causing trauma to the muscles of the heart. Early diagnosis of heart attack, getting immediate medical help with remote monitoring and end user notification is aimed in this paper. Broad pathological Q- wave, inversion of T- wave, elevated ST-segment in electrocardiogram (ECG) are the predominant features that must be taken into consideration while foretelling MI. Morphological changes namely ST-segment diversion and T-wave changes are inspected for any potential threat of MI. This paper suggests a novel approach in identifying the variation in ST segment by Wavelet transform, gradient decent methods, and some custom algorithms. Wavelet transform of the live ECG signals obtained from these leads disintegrates recorded wave into sub-bands of different order. Wavelet transform is an efficient method in capturing time and frequency simultaneously. Baseline wandering reduction, denoising and filtering of the live ECG and detection of PQRST positions on the ECG is executed primarily. Identification of various segments namely ST, PT, PQ etc. and implementing various algorithms for any anomalies in ST-segment which deviates beyond a threshold with reference to the normal sinus rhythm is noticed. Having fewer leads and targeting lightweight mobile healthcare application, real-time analyses for remote recording, diagnosing and notifying is aimed in this paper.