Key result
Novel ECG biometric recognition system achieves ~99% authentication accuracy across multiple databases.
Why the study?
An efficient method for ECG analysis and biometric recognition using lead I ECG with area under curve features was needed to improve authentication performance.
A novel biometric recognition system using area under curve analysis of lead I ECG demonstrates high authentication accuracy across multiple standard databases.
Answers are grounded in this paper and cite their sources.
May support ECG biometrics for authentication; leaves open prospective clinical validation before adoption.
In this paper, we introduce a human recognition system that utilizes lead I electrocardiogram (ECG). It proposes an efficient method for ECG analysis that corrects the signal and extract all major features of its waveform. FIR equiripple high pass filter is used for denoising ECG signal. R peak is detected using Haar wavelet transform. A novel class of features called as area under curve are computed from dominant fiducials of ECG waveform along with other known class of features such as interval features, amplitude features and angle features. The feasibility of an electrocardiogram as a new biometric is tested on selected features that reports the authentication performance 99.49% on QT database, 98.96% on PTB database and 98.48%on MIT-BIH arrhythmia database. The results obtained from the proposed approach surpasses the other conventional methods of biometric applications.
No takes yet. Share an insight, caveat, or question.
Pal et al. (2019) studied Biometric identification (n=352). Area under curve analysis of ECG with Kernel PCA was evaluated on Authentication performance (classification accuracy). The proposed ECG biometric recognition system using area under curve features achieved authentication accuracies of 99.49% on the QT database, 98.96% on the PTB database, and 98.48% on the MIT-BIH database.
Synapse has enriched 10 related papers. Consider them for comparative context: