ADAPTIVE FILTERING FOR DETECTING MYOCARDIAL INFARCTION USING NONINVASIVE CONDUCTING POLYMER COMPOSITE SENSORS

摘要

Continuous electrocardiographic (ECG) monitoring usingrnconducting polymer composite sensors (CPS) presents arnnon-invasive way to detect cardiac irregularities such asrnmyocardial infarction (MI). Electromyography (EMG),rnwhich measures muscle activity in the human body, has arnfrequency range that overlaps that of the ECG wave. As arnresult, both EMG and ECG data are present when CPSsrncollect ECG signals. When measuring ECG waves of anrnindividual during motion, we account for EMG byrnremoving the motion artifact from the ECG signal. Withrnthe use of a normalized least mean square (NLMS)rnalgorithm and known signal characteristics, we show thatrnEMG noise can be successfully filtered from an ECGrnsignal that is collected using our CPSs in the standard 12rnlead ECG placement. Our software produces arndiagnostic-friendly ECG signal and then determines thernpatient’s heart rate. When applied to the arrhythmiarndatabase from the Massachusetts Institute of Technologyrnand Beth Israel Hospital (MIT-BIH), our heartbeatrndetection logic has an accuracy of 99.6% with only 199rnfalse beats and 240 missed beats out of 109,494 totalrnheartbeats taken from 48 individual recordings.