A deformable CNN architecture for predicting clinical acceptability of ECG signal

• Autorzy: Allam Jaya Prakash , Samantray Saunak , Sahoo Suraj Prakash , Ari Samit

Identyfikatory

DOI

10.1016/j.bbe.2023.01.006

• Języki publikacji: EN

Abstrakty

EN

The degraded quality of the electrocardiogram (ECG) signals is the main source of false alarms in critical care units. Therefore, a preliminary analysis of the ECG signal is required to decide its clinical acceptability. In conventional techniques, different handcrafted features are extracted from the ECG signal based on signal quality indices (SQIs) to predict clinical acceptability. A one-dimensional deformable convolutional neural network (1DDCNN) is proposed in this work to extract features automatically, without manual interference, to detect the clinical acceptability of ECG signals efficiently. In order to create DCNN, the deformable convolution and pooling layers are merged into the regular convolutional neural network (CNN) architecture. In DCNN, the equidistant sampling locations of a regular CNN are replaced with adaptive sampling locations, which improves the network’s ability to learn based on the input. Deformable convolution layers concentrate more on significant segments of the ECG signals rather than giving equal attention to all segments. The proposed method is able to detect acceptable and unacceptable ECG signals with an accuracy of 99.50%, recall of 99.78%, specificity of 99.60%, precision of 99.47%, and F-score of 0.999. Experimental results show that the proposed method performs better than earlier state-of-the-art techniques.

Słowa kluczowe

EN

convolutional neural network; CNN; deep learning; deformable convolution; electrocardiogram; ECG; false alarm; signal quality analysis; SQA

PL

sieć neuronowa konwolucyjna; CNN; uczenie głębokie; elektrokardiogram; EKG; fałszywy alarm; analiza jakości sygnału; SQA

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2023
• Tom: Vol. 43, no. 1
• Strony: 335--351
• Opis fizyczny: Bibliogr. 65 poz., rys., tab., wykr.

Twórcy

autor

Allam Jaya Prakash

• Department of Electronics and Communication, National Institute of Technology Rourkela, Rourkela, Odisha 769008, India

autor

Samantray Saunak

• Department of ETC, IIIT Bhubaneswar, Bhubaneswar, Odisha, India

autor

Sahoo Suraj Prakash

• School of Electronics Engineering (SENSE), VIT Vellore Vellore, Tamil Nadu, India

autor

Ari Samit

• Department of Electronics and Communication, National Institute of Technology Rourkela, Rourkela, Odisha, India

Bibliografia

• [1] Soliman EZ, Elsalam MA, Li Y. The relationship between high resting heart rate and ventricular arrhythmogenesis in patients referred to ambulatory 24 h electrocardiographic recording. Europace 2010;12(2):261-5.
• [2] Zhao Z, Zhang Y. SQI quality evaluation mechanism of singlelead ECG signal based on simple heuristic fusion and fuzzy comprehensive evaluation. Front Physiol 2018;06:9.
• [3] Singh P, Pradhan G. A New ECG denoising framework using generative adversarial network. IEEE/ACM Trans Comput Biol Bioinformat 2021;18(2):759-64.
• [4] Zhao Z, Zhang Y. SQI quality evaluation mechanism of singlelead ECG signal based on simple heuristic fusion and fuzzy comprehensive evaluation. Front Physiol 2018;9:727.
• [5] Yea Bai. Is the sequence of superalarm triggers more predictive than sequence of the currently utilized patient monitor alarms? IEEE Trans Bio-medical Eng 2017;64 (5):1023-32.
• [6] Alam S, Gupta R, Sharma KD. On-board signal quality assessment guided compression of photoplethysmogram for personal health monitoring. IEEE Trans Instrument Measur 2021;70:1-9.
• [7] zhu W, Chen X, Wang Y, Wang L. Arrhythmia recognition and classification using ECG morphology and segment feature analysis. IEEE/ACM Trans Comput Biol Bioinformat 2019;16 (1):131-8.
• [8] Qaisar SM, Khan SI, Dallet D, Tadeusiewicz R, Pławiak P. Signal-piloted processing metaheuristic optimization and wavelet decomposition based elucidation of arrhythmia for mobile healthcare. Biocybernet Biomedi Eng 2022;42 (2):681-94.
• [9] Zivanovic M, González-Izal M. Simultaneous powerline interference and baseline wander removal from ECG and EMG signals by sinusoidal modeling. Medical Eng Phys 2013;35(10):1431-41. Available from https:// www.sciencedirect.com/science/article/pii/S1350453313000775.
• [10] Łaôski JM, Henzel N. ECG baseline wander and powerline interference reduction using nonlinear filter bank. Signal Process 2005;85(4):781-793. Available from: http:// www.sciencedirect.com/science/article/pii/S0165168404003251.
• [11] Satija U, Ramkumar B, Manikandan MS. A review of signal processing techniques for electrocardiogram signal quality assessment. IEEE Rev Biomedi Eng 2018;11:36-52.
• [12] Oster J, Behar J, Sayadi O, Nemati S, Johnson AE, Clifford GD. Semisupervised ECG ventricular beat classification with novelty detection based on switching Kalman filters. IEEE Trans Biomed Eng 2015;62(9):2125-34.
• [13] Satija U, Ramkumar B, Manikandan MS. A unified sparse signal decomposition and reconstruction framework for elimination of muscle artifacts from ECG signal. In: 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE; 2016. p. 779-83.
• [14] Prabhakararao E, Manikandan MS. On the use of variational mode decomposition for removal of baseline wander in ECG signals. In: 2016 Twenty Second National Conference on Communication (NCC). IEEE; 2016. p. 1-6.
• [15] Satija U, Ramkumar B, Manikandan MS. A robust sparse signal decomposition framework for baseline wander removal from ECG signal. 2016 IEEE Region 10 Conference (TENCON). IEEE; 2016. p. 2470-3.
• [16] Orphanidou C, Bonnici T, Charlton P, Clifton D, Vallance D, Tarassenko L. Signal-quality indices for the electrocardiogram and photoplethysmogram: derivation and applications to wireless monitoring. IEEE J Biomed Health Informat 2015;19(3):832-8.
• [17] Quesnel PX, Chan ADC, Yang H. Real-time biosignal quality analysis of ambulatory ECG for detection of myocardial ischemia. In: 2013 IEEE International Symposium on Medical Measurements and Applications (MeMeA). p. 1-5.
• [18] Hayn D, Jammerbund B, Schreier G. QRS detection based ECG quality assessment. Physiol Measur 2012;33(9):1449.
• [19] Naseri H, Homaeinezhad M. Electrocardiogram signal quality assessment using an artificially reconstructed target lead. Comput Methods Biomech Biomed Eng 2015;18(10):1126-41.
• [20] Behar J, Oster J, Li Q, Clifford GD. ECG signal quality during arrhythmia and its application to false alarm reduction. IEEE Trans Biomed Eng 2013;60(6):1660-6.
• [21] Clifford G, Behar J, Li Q, Rezek I. Signal quality indices and data fusion for determining clinical acceptability of electrocardiograms. Physiol Measur 2012;33(9):1419.
• [22] Li Q, Mark RG, Clifford GD. Robust heart rate estimation from multiple asynchronous noisy sources using signal quality indices and a Kalman filter. Physiol Measurem 2007 dec;29 (1):15-32. Available from: https://doi.org/10.1088%2F0967- 3334%2F29%2F1%2F002.
• [23] Orphanidou C, Drobnjak I. Quality assessment of ambulatory ECG using wavelet entropy of the HRV signal. IEEE J Biomed Health Informat 2017;21(5):1216-23.
• [24] Smital L, Vítek M, Kozumplík J, Provazník I. Adaptive wavelet wiener filtering of ECG signals. IEEE Trans Biomed Eng 2013;60(2):437-45.
• [25] Bartolo A, Clymer BD, Burgess RC, Turnbull JP, Golish JA, Perry MC. An arrhythmia detector and heart rate estimator for overnight polysomnography studies. IEEE Trans Biomed Eng 2001;48(5):513-21.
• [26] Eea Morgado. Quality estimation of the electrocardiogram using cross-correlation among leads. Biomed Eng Online 2015;14. June.
• [27] Smital L, Haider CR, Vitek M, Leinveber P, Jurak P, Nemcova A, et al. Real-time quality assessment of long-term ECG signals recorded by wearables in free-living conditions. IEEE Trans Biomed Eng 2020. 1-1.
• [28] Zaunseder S, Huhle R, Malberg H. CinC challenge-Assessing the usability of ECG by ensemble decision trees. 2011 Computing in Cardiology. IEEE; 2011. p. 277-80.
• [29] Li Q, Clifford GD. Signal quality and data fusion for false alarm reduction in the intensive care unit. J Electrocardiol 2012;45(6):596-603.
• [30] Xu H, Yan W, Lan K, Ma C, Wu D, Wu A, et al. Assessing electrocardiogram and respiratory signal quality of a wearable device (SensEcho): semisupervised machine learning-based validation study. JMIR Mhealth Uhealth 2021;9(8):e25415. Aug, Available from: https://mhealth.jmir. org/2021/8/e25415.
• [31] Kužílek J, Huptych M, Chudáček V, Spilka J, Lhotská L. Data driven approach to ECG signal quality assessment using multistep SVM classification. 2011 Computing in Cardiology. IEEE; 2011. p. 453-5.
• [32] Tobón DP, Falk TH. Online ECG quality assessment for context-aware wireless body area networks. In: 2015 IEEE 28th Canadian Conference on Electrical and Computer Engineering (CCECE). IEEE; 2015. p. 587-92.
• [33] Behar J, Oster J, Li Q, Clifford GD. ECG signal quality during arrhythmia and its application to false alarm reduction. IEEE Trans Biomed Eng 2013;60(6):1660-6.
• [34] Nanni L, Ghidoni S, Brahnam S. Handcrafted vs. non-handcrafted features for computer vision classification. Pattern Recognit 2017;71:158-72.
• [35] Wang CC, Kang Y. Feature extraction techniques of nonstationary signals for fault diagnosis in machinery systems. 2012.
• [36] Biswal B, Dash PK, Panigrahi BK. Non-stationary power signal processing for pattern recognition using HS-transform. Appl Soft Comput 2009;9(1):107-17.
• [37] Rathod V, Tiwari A, Kakde O. Feature extraction and classification for electro-encephalography based BCI: A review. Helix 2020;10(04):345-9.
• [38] Acharya UR, Fujita H, Oh SL, Hagiwara Y, Tan JH, Adam M. Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals. Informat Sci 2017;415:190-8.
• [39] Zhou X, Zhu X, Nakamura K, Mahito N. ECG quality assessment using 1D-convolutional neural network. In: 2018 14th IEEE International Conference on Signal Processing (ICSP). IEEE; 2018. p. 780-784.
• [40] Seeuws N, De Vos M, Bertrand A. Electrocardiogram quality assessment using unsupervised deep learning. IEEE Trans Biomed Eng 2022;69(2):882-93.
• [41] Acharya UR, Fujita H, Oh SL, Hagiwara Y, Tan JH, Adam M, et al. Deep convolutional neural network for the automated diagnosis of congestive heart failure using ECG signals. Appl Intell 2019;49(1):16-27.
• [42] Zhou X, Zhu X, Nakamura K, Noro M. Electrocardiogram quality assessment with a generalized deep learning model assisted by conditional generative adversarial networks. Life 2021;11(10):1013.
• [43] Zhang Q, Fu L, Gu L. A cascaded convolutional neural network for assessing signal quality of dynamic ECG. Comput Mathe Methods Med. 2019;2019.
• [44] Jin Y, Li Z, Qin C, Liu J, Liu Y, Zhao L, et al. A novel attentional deep neural network-based assessment method for ECG quality. Biomed Signal Process Control 2023;79:104064.
• [45] Liu G, Han X, Tian L, Zhou W, Liu H. ECG quality assessment based on hand-crafted statistics and deep-learned Stransform spectrogram features. Comput Methods Programs Biomed 2021;208:106269. Available from https:// www.sciencedirect.com/science/article/pii/S0169260721003436.
• [46] Zhang J, Wang L, Zhang W, Yao J. A signal quality assessment method for electrocardiography acquired by mobile device. In: 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). p. 1-3.
• [47] Tan H, Lai J, Liu Y, Song Y, Wang J, Chen M, et al. Neural architecture search for real-time quality assessment of wearable multi-lead ECG on mobile devices. Biomed Signal Process Control 2022;74:103495.
• [48] Cimr D, Fujita H, Tomaskova H, Cimler R, Selamat A. Automatic seizure detection by convolutional neural networks with computational complexity analysis. Comput Methods Programs Biomed 2023;229:107277.
• [49] Fujita H, Cimr D. Computer aided detection for fibrillations and flutters using deep convolutional neural network. Informat Sci 2019;486:231-9.
• [50] Kwon O, Jeong J, Kim HB, Kwon IH, Park SY, Kim JE, et al. Electrocardiogram sampling frequency range acceptable for heart rate variability analysis. Healthcare Informat Res 2018;24(3):198-206.
• [51] Apandi ZFM, Ikeura R, Hayakawa S. Arrhythmia Detection Using MIT-BIH Dataset: A Review. In: 2018 International Conference on Computational Approach in Smart Systems Design and Applications (ICASSDA). p. 1-5.
• [52] Al G. PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. PhysioBank 2000;101(23):e215-20.
• [53] Moody GB, The Mark RG. MIT-BIH arrhythmia database on CD-ROM and software for use with it. [1990] Proceedings Computers in Cardiology. IEEE; 1990. p. 185-8.
• [54] Mohagheghian F, Han D, Peitzsch A, Nishita N, Ding E, Dickson EL, et al. Optimized signal quality assessment for photoplethysmogram signals using feature selection. IEEE Trans Biomed Eng 2022;69(9):2982-93.
• [55] Keskes N, Fakhfakh S, Kanoun O, Derbel N. Representativeness consideration in the selection of classification algorithms for the ECG signal quality assessment. Biomed Signal Process Control 2022;76:103686.
• [56] Ong MEH, Padmanabhan P, Chan YH, Lin Z, Overton J, Ward KR, et al. An observational, prospective study exploring the use of heart rate variability as a predictor of clinical outcomes in pre-hospital ambulance patients. Resuscitation 2008;78(3):289-97.
• [57] Mohebbi M, Ghassemian H. Prediction of paroxysmal atrial fibrillation based on non-linear analysis and spectrum and bispectrum features of the heart rate variability signal. Comput Methods Programs Biomed 2012;105(1):40-9.
• [58] Acharya UR, Fujita H, Lih OS, Hagiwara Y, Tan JH, Adam M. Automated detection of arrhythmias using different intervals of tachycardia ECG segments with convolutional neural network. Informat Sci 2017;405:81-90.
• [59] Allam JP, Samantray S, Ari S. SpEC: A system for patient specific ECG beat classification using deep residual network. Biocybernet Biomed Eng 2020;40(4):1446-57.
• [60] Dai J, Qi H, Xiong Y, Li Y, Zhang G, Hu H, et al. Deformable convolutional networks. In: Proceedings of the IEEE international conference on computer vision. p. 764-73.
• [61] Qin L, Xie Y, Liu X, Yuan X, Wang H. An End-to-End 12- Leading Electrocardiogram Diagnosis System Based on Deformable Convolutional Neural Network With Good Antinoise Ability. IEEE Trans Instrument Measur 2021;70:1-13.
• [62] Bian J, Yang S, Xiong H, Wang L, Fu Y, Sun Z, et al. CRLEDD: regularized causalities learning for early detection of diseases using electronic health record (EHR) data. IEEE Trans Emerg Top Comput Intell 2021;5(4):541-53.
• [63] Goyal M, Reeves ND, Davison AK, Rajbhandari S, Spragg J, Yap MH. DFUNet: convolutional neural networks for diabetic foot ulcer classification. IEEE Trans Emerg Top Comput Intell 2020;4(5):728-39.
• [64] Fujita H, Cimr D. Decision support system for arrhythmia prediction using convolutional neural network structure without preprocessing. Appl Intell 2019;49(9):3383-91.
• [65] Meyes R, Lu M, de Puiseau CW, Meisen T. Ablation studies in artificial neural networks. arXiv preprint arXiv:190108644. 2019.

Uwagi

PL

Opracowanie rekordu ze środków MEiN, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2022-2023)