Ischemic and non-ischemic heartbeat classifier for portable automatic detection systems
Keywords:ASIC, Cardiac Ischemic, Fuzzy Logic, COVID-19
This article presents a new heartbeat classifier oriented to an automatic cardiac ischemia detection system aimed at the medical control of the population with cardiac disorders (a higher risk group of severe COVID-19). The algorithm focuses on distinguishing ischemic heartbeats from non-ischemic based on methods and techniques proposed by medical experts in cardiac ischemia diagnoses. The proposed algorithm was tested with 80 electrocardiogram recordings obtained from the Long-term ST database of Physionet with a 100,112 heartbeat population. The heartbeat population contains different types of non-ischemic cases. The algorithm exhibits a sensitivity of 98.7 percent and specificity of 97.23 percent. The digital synthesis, in a 180nm CMOS process with a dynamic power consumption of 13.15 micro Watt demanding an area of 490x400 micrometers, attests to the algorithm benefits.