EKG Sinyalleriyle Biyometrik Kimlik Doğrulama: Evrişimsel Otokoderlere Dayalı Güvenli Bir Kimlik Doğrulama Modeli

Öz

Elektrokardiyografi (EKG) sinyalleri, kalbin bireysel elektriksel özelliklerini yakalayarak biyometrik tanımlama için benzersiz bir fırsat sağlar. Bu çalışma, konvolüsyonel oto kodlayıcılar (CAE) kullanarak EKG tabanlı kimlik tanımayı araştırmaktadır. Önerilen yöntem, EKG sinyallerinden özellikleri verimli bir şekilde çıkararak kimlik tespiti için kompakt ve anlamlı bir temsil oluşturmaktadır. Geleneksel yöntemlerin aksine CAE, özellik genişletme ve dönüm noktası tespiti ile aralıkları ayırarak mevcut literatürdeki sınırlamaları ele alır. Çalışmada, çeşitli ve temsili eğitim verileri sağlayan MIT-BIH Aritmi EKG veri seti kullanılmıştır. Temel özellikleri öğrenerek ve boyutluluğu azaltarak model, hassas sınıflandırma için girdi verilerini sıkıştırır ve yeniden yapılandırır. Kişisel tıbbi verilerin hassasiyeti göz önünde bulundurularak, şifreleme ve sıkıştırma dahil olmak üzere sağlam veri koruma stratejileri uygulanmaktadır. Deneysel sonuçlar EKG tabanlı tanımlamada 98.46% gibi yüksek bir doğruluk oranı göstererek yaklaşımın etkili bir biyometrik kimlik doğrulama yöntemi olduğunu doğrulamaktadır. Bulgular, ayırt edici bir biyometrik tanımlayıcı olarak kardiyak elektriksel aktivitenin potansiyelini vurgulamaktadır. Önerilen model, teknolojik ilerleme ve veri gizliliği arasında dengeli bir yaklaşım sunarak makine öğrenimi tekniklerini ve sıkı güvenlik önlemlerini entegre ederek biyometrik tanımaya katkıda bulunmaktadır. Bu araştırma, EKG sinyallerini kullanarak güvenli ve güvenilir kişisel tanımlamanın yolunu açmaktadır.

Anahtar Kelimeler

• Biyometrik kimlik doğrulama
• Evrişimsel otokodlayıcılar (CAE)
• özellik çıkarma
• sinyal sıkıştırma
• gizlilik ve veri güvenliği.

Biometric Authentication with ECG Signals: A Secure Identification Model Based on Convolutional Autoencoders

Öz

Electrocardiography (ECG) signals provide a unique opportunity for biometric identification by capturing individual electrical properties of the heart. This study explores ECG-based identity recognition using convolutional autoencoders (CAE). The proposed method efficiently extracts features from ECG signals, constructing a compact and meaningful representation for identification. Unlike traditional methods, CAE separates intervals with feature expansion and landmark detection, addressing limitations in existing literature. The study employs the MIT-BIH Arrhythmia ECG dataset, ensuring diverse and representative training data. By learning key features and reducing dimensionality, the model compresses and reconstructs input data for precise classification. Recognizing the sensitivity of personal medical data, robust data protection strategies, including encryption and compression, are implemented. Experimental results show a high accuracy of 98.46% in ECG-based identification, validating the approach as an effective biometric authentication method. The findings highlight the potential of cardiac electrical activity as a distinctive biometric identifier. The proposed model contributes to biometric recognition by integrating machine learning techniques and stringent security measures, offering a balanced approach between technological advancement and data privacy. This research paves the way for secure and reliable personal identification using ECG signals.

Anahtar Kelimeler

• Biometric authentication
• convolutional autoencoders (CAE)
• feature extraction
• data dimensionality reduction
• privacy and data security
• identification accuracy

Kaynakça

1. Martis RJ, Acharya UR, Adeli H. Current methods in electrocardiogram characterization. Comput Biol Med 2014; 54: 95-109.
2. Ince T, Kiranyaz S, Gabbouj M. A generic and robust system for automated patient-specific classification of ECG signals. IEEE Trans Biomed Eng 2009; 56(5): 1415-1426.
3. Biel L, Pettersson O, Philipson L, Wide P. ECG analysis: A new approach in human identification. IEEE Trans Instrum Meas 2001; 50(3): 808-812.
4. Jyotishi D, Dandapat S. An LSTM-based model for person identification using ECG signal. IEEE Sens Lett 2020; 4(8): 1-4.
5. Sancho J, Alesanco Á, García J. Biometric authentication using the PPG: A long-term feasibility study. Sensors 2018; 18(5): 1525.
6. Khushk KP, Iqbal AA. An overview of leading biometrics technologies used for human identity. Proc Student Conf Eng Sci Technol (SCONEST), Karachi, Pakistan, 2005; 1-4.
7. Ribeiro Pinto J, Cardoso JS, Lourenco A. Evolution, current challenges, and future possibilities in ECG biometrics. IEEE Access 2018; 6: 34746-34776.
8. Sadhukhan D, Mitra M. R-peak detection algorithm for ECG using double difference and RR interval processing. Procedia Technol 2012; 4: 873-877.

9. Bassiouni MM, El-Dahshan ESA, Khalefa W, Salem AM. Intelligent hybrid approaches for human ECG signals identification. Signal Image Video Process 2018; 12(5): 941-949.
10. Zhang Q, Zhou D, Zeng X. HeartID: A multiresolution convolutional neural network for ECG-based biometric human identification in smart health applications. IEEE Access 2017; 5: 11805-11816.
11. Prakash AJ, Patro KK, Samantray S, Pławiak P, Hammad M. A deep learning technique for biometric authentication using ECG beat template matching. Information 2023; 14(2): 65.
12. Patro KK, Prakash AJ, Rao MJ, Kumar PR. An efficient optimized feature selection with machine learning approach for ECG biometric recognition. IETE J Res 2022; 68(4): 2743-2754.
13. Hejazi M, Al-Haddad SAR, Hashim SJ, Aziz AFA, Singh YP. Non-fiducial based ECG biometric authentication using one-class support vector machine. Proc Int Conf Signal Process Algorithms Archit Arrang Appl (SPA), Poznan, Poland, 2017; 190-194.
14. Li N, He F, Ma W, Wang R, Jiang L, Zhang X. The identification of ECG signals using WT-UKF and IPSO-SVM. Sensors 2022; 22(5): 1962.
15. Li N, He F, Ma W, Wang R, Jiang L, Zhang X. The identification of ECG signals using wavelet transform and WOA-PNN. Sensors 2022; 22(12): 4343.
16. Fatimah B, Singh P, Singhal A, Pachori RB. Biometric identification from ECG signals using Fourier decomposition and machine learning. IEEE Trans Instrum Meas 2022; 71: 1-10.
17. Chu Y, Shen H, Huang K. ECG authentication method based on parallel multi-scale one-dimensional residual network with center and margin loss. IEEE Access 2019; 7: 51598-51607.
18. Al-Jibreen A, Al-Ahmadi S, Islam S, Artoli AM. Person identification with arrhythmic ECG signals using deep convolution neural network. Sci Rep 2024; 14: 55066.
19. Maleki Lonbar S, Beigi A, Bagheri N, Peris-Lopez P, Camara C. Deep learning-based biometric authentication system using a high temporal/frequency resolution transform. Front Digit Health 2024; 6: 1463713.
20. Wang G, Shanker S, Nag A, Lian Y, John D. ECG biometric authentication using self-supervised learning for IoT edge sensors. IEEE J Biomed Health Inform 2024; 28(1): 1-10.
21. Belo D, Bento N, Silva H, Fred A, Gamboa H. ECG biometrics using deep learning and relative score threshold classification. Sensors 2020; 20(15): 4078.
22. Israel SA, Irvine JM, Cheng A, Wiederhold MD, Wiederhold BK. ECG to identify individuals. Pattern Recognit 2005; 38(1): 133-142.
23. Mostefai L, Mohamed B, Merzoug B, Lotfi M, Mouloud D. Enhanced local patterns using deep learning techniques for ECG-based identity recognition system. Preprint, 2021.
24. Donida Labati R, Muñoz E, Piuri V, Sassi R, Scotti F. Deep-ECG: Convolutional neural networks for ECG biometric recognition. Pattern Recognit Lett 2019; 126: 78-85.
25. Hua X, Han J, Zhao C, et al. A novel method for ECG signal classification via one-dimensional convolutional neural network. Multimed Syst 2022; 28: 1387-1399.
26. Liu CL, Hsaio WH, Tu YC. Time series classification with multivariate convolutional neural network. IEEE Trans Ind Electron 2019; 66(6): 4788-4797.
27. Kiranyaz S, Avci O, Abdeljaber O, Ince T, Gabbouj M, Inman DJ. 1D convolutional neural networks and applications: A survey. Mech Syst Signal Process 2021; 151: 107398.
28. Saleem MA, Senan N, Wahid F, Aamir M, Samad A, Khan M. Comparative analysis of recent architecture of convolutional neural network. Math Probl Eng 2022; 2022: 7313612.
29. Yıldırım Ö, Pławiak P, Tan RS, Acharya UR. Arrhythmia detection using deep convolutional neural network with long duration ECG signals. Comput Biol Med 2018; 102: 411-420.
30. Eigen D, Rolfe J, Fergus R, LeCun Y. Understanding deep architectures using a recursive convolutional network. arXiv preprint, 2013. Available: http://arxiv.org/abs/1312.1847.
31. Abdi M, Nahavandi S. Multi-residual networks: Improving the speed and accuracy of residual networks. arXiv preprint, 2016. Available: http://arxiv.org/abs/1609.05672.
32. Huang G, Liu Z, Pleiss G, Van Der Maaten L, Weinberger KQ. Convolutional networks with dense connectivity. IEEE Trans Pattern Anal Mach Intell 2022; 44(12): 8704-8716.
33. Chen J, He T, Zhuo W, Ma L, Ha S, Chan SHG. TVConv: Efficient translation variant convolution for layout-aware visual processing. arXiv preprint, 2022. Available: http://arxiv.org/abs/2203.10489.
34. Molchanov P, Tyree S, Karras T, Aila T, Kautz J. Pruning convolutional neural networks for resource efficient inference. arXiv preprint, 2016. Available: http://arxiv.org/abs/1611.06440.
35. Howard AG, Zhu M, Chen B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint, 2017. Available: http://arxiv.org/abs/1704.04861.
36. Murat F, Yildirim O, Talo M, Baloglu UB, Demir Y, Acharya UR. Application of deep learning techniques for heartbeats detection using ECG signals—analysis and review. Comput Biol Med 2020; 120: 103726.
37. Ortigosa N, Giménez VM. Raw data extraction from electrocardiograms with portable document format. Comput Methods Programs Biomed 2014; 113(1): 284-289.
38. Almalchy MT, Ciobanu V, Popescu N. Noise removal from ECG signal based on filtering techniques. Proc Int Conf Control Syst Comput Sci (CSCS), Bucharest, Romania, 2019; 176-181.
39. Cao M, Zhao T, Li Y. ECG heartbeat classification using deep transfer learning with convolutional neural network and STFT technique. J Phys Conf Ser 2023; 2547(1): 012031.
40. Dalianis H. Evaluation metrics and evaluation. In: Clinical Text Mining. Springer Int Publ, 2018; 45-53.
41. Wang F, et al. A novel ECG signal compression method using spindle convolutional auto-encoder. Comput Methods Programs Biomed 2019; 175: 139-150.
42. Jyotishi D, Dandapat S. An LSTM-based model for person identification using ECG signal. IEEE Sens Lett 2020; 4(8): 1-4.
43. Abdeldayem SS, Bourlai T. A novel approach for ECG-based human identification using spectral correlation and deep learning. IEEE Trans Biom Behav Identity Sci 2020; 2(1): 1-14.
44. Singh YN, Singh SK. Identifying individuals using eigenbeat features of electrocardiogram. J Eng 2013; 2013: 539284.