Title: Involutional neural networks for ECG spectrogram classification and person identification

Authors: Hatem Zehir; Toufik Hafs; Sara Daas

Addresses: Faculty of Technology, LERICA Laboratory, Badji Mokhtar-Annaba University, P.O. Box 12, Annaba, 23000, Algeria ' Faculty of Technology, LERICA Laboratory, Badji Mokhtar-Annaba University, P.O. Box 12, Annaba, 23000, Algeria ' Faculty of Technology, LERICA Laboratory, Badji Mokhtar-Annaba University, P.O. Box 12, Annaba, 23000, Algeria

Abstract: Reliable personal identification is crucial. This study explores an ECG-based system using spectrograms and involutional neural networks (INNs) for accurate person identification. The system first preprocesses ECG signals with a Butterworth filter and segments them using the Pan-Tompkins++ algorithm. Each segment, representing a heartbeat, is then converted into a spectrogram using the short-time Fourier transform (STFT). Finally, an INN classifies the spectrograms for identification. Tested on the MIT-BIH Arrhythmia and Physikalisch-Technische Bundesanstalt (PTB) databases, the system achieved accuracies of 97.93% and 97.63%, respectively, surpassing conventional methods like convolutional neural networks (CNNs). This improvement is attributed to INNs' ability to better capture both local and temporal patterns in the spectrogram data.

Keywords: biometrics; deep learning; electrocardiogram; INNs; involutional neural networks; Person Identification.

DOI: 10.1504/IJSISE.2024.140031

International Journal of Signal and Imaging Systems Engineering, 2024 Vol.13 No.1, pp.41 - 53

Received: 03 Mar 2024
Accepted: 07 May 2024
Published online: 15 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article