A Novel Non Invasive Myocardial Infarction Vigilant System by using Raspberry Pi

In recent years, the number of patients suffering from cardiac diseases has increased at an alarming rate. Cardio related ailments are the prime cause of large death losses in most of the world. Accurate classification of Electrocardiogram (ECG) signals can play a pivotal role in the diagnosis of heart disease. Human life can be saved if the disease is diagnosed at an early stage. The most common reason for the delay in treatment is patient unawareness. This calls for the use of highly sophisticated techniques providing quick and accurate results. This article recommends a non-invasive technique based on electrocardiogram signal analysis for detecting anomalies. Daubechies wavelet decomposition using 10 decomposition levels is used for denoising the ECG signal by dropping the noise coefficients. The proposed algorithm is used to detect ST-segment deviation from the isoelectric line which is the major cause of myocardial infarction. A Raspberry Pi running the algorithm sends AT commands to the GSM module to call for emergency help and send the GPS location via an SMS, in case a heart attack is detected. The purpose of this technique is to bridge the delay between the onset of heart attack symptoms and the arrival of medical aid. The accuracy of the system was found to be 95.23%.