ECG Biometric with Abnormal Cardiac Conditions in Remote Monitoring System

摘要: This paper presents a person identification mechanism using electrocardiogram (ECG) signals with abnormal cardiac conditions in network environments. A total of 164 subjects were used this three different databases containing various irregular heart states from MIT-BIH arrhythmia database (MITDB), supraventricular (SVDB), and Charles Sturt diabetes complication screening initiative (DiSciRi) database. We proposed simple yet effective biometric sample extraction technique for ECG samples to improve the process. These points then applied four classifiers verify robustness identification. Varying numbers enrollment recognition QRS complexes validate stability method. Our experimentation results show that outperforms existing methods lacking ability efficiently extract features matching. is evident by obtaining high accuracy 96.7% MITDB, 96.4% SVDB, 99.3% DiSciRi. Moreover, sensitivity, specificity, positive predictive value, Youden Index’s values further verifies reliability also suggests possibility improving classification performance recordings low sampling frequency increased number samples.