Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks

The electrocardiogram (ECG) is relatively easy to acquire and has been used for reliable biometric authentication. Despite growing interest in ECG authentication, there are still two main problems that need to be tackled, i.e., the accuracy and processing speed. Therefore, this paper proposed a fast...

Full description

Bibliographic Details
Main Authors: Ihsanto, Eko, Ramli, Kalamullah, Sudiana, Dodi, Gunawan, Teddy Surya
Format: Article
Language:English
English
Published: MDPI 2020
Subjects:
Online Access:http://irep.iium.edu.my/80641/
http://irep.iium.edu.my/80641/1/80641_Fast%20and%20Accurate%20Algorithm%20for%20ECG.pdf
http://irep.iium.edu.my/80641/7/80641_Fast%20and%20accurate%20algorithm%20for%20ECG_Scopus.pdf
_version_ 1848788996094689280
author Ihsanto, Eko
Ramli, Kalamullah
Sudiana, Dodi
Gunawan, Teddy Surya
author_facet Ihsanto, Eko
Ramli, Kalamullah
Sudiana, Dodi
Gunawan, Teddy Surya
author_sort Ihsanto, Eko
building IIUM Repository
collection Online Access
description The electrocardiogram (ECG) is relatively easy to acquire and has been used for reliable biometric authentication. Despite growing interest in ECG authentication, there are still two main problems that need to be tackled, i.e., the accuracy and processing speed. Therefore, this paper proposed a fast and accurate ECG authentication utilizing only two stages, i.e., ECG beat detection and classification. By minimizing time-consuming ECG signal pre-processing and feature extraction, our proposed two-stage algorithm can authenticate the ECG signal around 660 μs. Hamilton’s method was used for ECG beat detection, while the Residual Depthwise Separable Convolutional Neural Network (RDSCNN) algorithm was used for classification. It was found that between six and eight ECG beats were required for authentication of different databases. Results showed that our proposed algorithm achieved 100% accuracy when evaluated with 48 patients in the MIT-BIH database and 90 people in the ECG ID database. These results showed that our proposed algorithm outperformed other state-of-the-art methods.
first_indexed 2025-11-14T17:49:41Z
format Article
id iium-80641
institution International Islamic University Malaysia
institution_category Local University
language English
English
last_indexed 2025-11-14T17:49:41Z
publishDate 2020
publisher MDPI
recordtype eprints
repository_type Digital Repository
spelling iium-806412020-07-11T11:34:30Z http://irep.iium.edu.my/80641/ Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks Ihsanto, Eko Ramli, Kalamullah Sudiana, Dodi Gunawan, Teddy Surya TK7885 Computer engineering The electrocardiogram (ECG) is relatively easy to acquire and has been used for reliable biometric authentication. Despite growing interest in ECG authentication, there are still two main problems that need to be tackled, i.e., the accuracy and processing speed. Therefore, this paper proposed a fast and accurate ECG authentication utilizing only two stages, i.e., ECG beat detection and classification. By minimizing time-consuming ECG signal pre-processing and feature extraction, our proposed two-stage algorithm can authenticate the ECG signal around 660 μs. Hamilton’s method was used for ECG beat detection, while the Residual Depthwise Separable Convolutional Neural Network (RDSCNN) algorithm was used for classification. It was found that between six and eight ECG beats were required for authentication of different databases. Results showed that our proposed algorithm achieved 100% accuracy when evaluated with 48 patients in the MIT-BIH database and 90 people in the ECG ID database. These results showed that our proposed algorithm outperformed other state-of-the-art methods. MDPI 2020-05-09 Article PeerReviewed application/pdf en http://irep.iium.edu.my/80641/1/80641_Fast%20and%20Accurate%20Algorithm%20for%20ECG.pdf application/pdf en http://irep.iium.edu.my/80641/7/80641_Fast%20and%20accurate%20algorithm%20for%20ECG_Scopus.pdf Ihsanto, Eko and Ramli, Kalamullah and Sudiana, Dodi and Gunawan, Teddy Surya (2020) Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks. Applied Sciences, 10 (9). pp. 1-15. ISSN 2076-3417 https://www.mdpi.com/2076-3417/10/9/3304/htm http://dx.doi.org/10.3390/app10093304
spellingShingle TK7885 Computer engineering
Ihsanto, Eko
Ramli, Kalamullah
Sudiana, Dodi
Gunawan, Teddy Surya
Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks
title Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks
title_full Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks
title_fullStr Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks
title_full_unstemmed Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks
title_short Fast and accurate algorithm for ECG authentication using residual depthwise separable convolutional neural networks
title_sort fast and accurate algorithm for ecg authentication using residual depthwise separable convolutional neural networks
topic TK7885 Computer engineering
url http://irep.iium.edu.my/80641/
http://irep.iium.edu.my/80641/
http://irep.iium.edu.my/80641/
http://irep.iium.edu.my/80641/1/80641_Fast%20and%20Accurate%20Algorithm%20for%20ECG.pdf
http://irep.iium.edu.my/80641/7/80641_Fast%20and%20accurate%20algorithm%20for%20ECG_Scopus.pdf