Spatio-spectral independent component analysis for fetal ECG extraction from two-channel maternal abdominal signals

• Autorzy: Kotas Marian P. , AlShrouf Anwar M.

Identyfikatory

DOI

10.1016/j.bbe.2024.02.002

• Języki publikacji: EN

Abstrakty

EN

Independent component analysis (ICA) is widely used to separate maternal and fetal electrocardiograms. However, it has become less effective due to the efforts to reduce the number of recording electrodes. To address this issue, we propose an extension of ICA that can extract the fetal electrocardiogram from only two maternal abdominal electric signals. We solve this problem by increasing the dimension of the observed signals using the method of delays, followed by spatio-spectral filtering to separate the source signals. By iteratively applying this approach, we can extract signals that are not separable using the original observations alone. These signals are then clustered to create signal subspaces corresponding to different sources, allowing for a rough reconstruction of signal components produced by these sources. This initial decomposition can then be refined by using the reconstructed components as new observations, extending the original ones, and applying ICA to this extended signal representation. Applied to two-channel maternal abdominal signals, the proposed method was able to extract 3 source signals (two maternal and one fetal), resulting in the achievement of the goal of over-complete blind source separation (BSS). Furthermore, the method enabled the successful detection of fetal QRS (fQRS) complexes in experiments on two datasets of real-world maternal abdominal signals. For the ADFECGDB dataset, the method reached the sensitivity, positive predictivity, and F1 score of 100%, 99.97%, and 99.98%, respectively, outperforming all reference methods. For the PREGNANCY dataset, the corresponding values were 98.95%, 98.92%, and 98.93%, second only to one reference method.

Słowa kluczowe

EN

blind source separation; over complete mixture; spatio spectral filtering; independent component analysis; fetal ECG

PL

ślepa separacja sygnałów; analiza składowych niezależnych; EKG płodu

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2024
• Tom: Vol. 44, no. 1
• Strony: 247--263
• Opis fizyczny: Bibliogr. 52 poz., rys., tab., wykr.

Twórcy

autor

Kotas Marian P.

• Department of Cybernetics Nanotechnology and Data Processing of the Silesian University of Technology, Gliwice, Poland

autor

AlShrouf Anwar M.

• Department of Biomedical Technology, College of Applied Medical Science of the Prince Sattam Bin Abdulaziz University, Alkharj, Saudi Arabia

Bibliografia

• [1] Jezewski J, Wrobel J, Matonia A, Horoba A, Martinek R, Kupka T, et al. Is abdominal fetal electrocardiography an alternative to Doppler ultrasound for FHR variability evaluation? Front Physiol 2017;8:305.
• [2] Ranjan R, Chandra B, Kumar Bhandari A. Ocular artifact elimination from electroencephalography signals: A systematic review. Biocybern Biomed Eng 2021;41(3):960-96.
• [3] Khajuria A, Joshi D. EEG based cortical investigation for the limit of stability analysis in transfemoral amputees: A comparison with able-bodied individuals. Biocybern Biomed Eng 2022;42(1):362-74.
• [4] Lakshmi GP, Kumar PA. Independent component analysis application for fault detection in process industries: Literature review and an application case study for fault detection in multiphase flow systems. Measurement 2023;209:112504.
• [5] Allam JA, Samantray S, Sahoo SP, Ari S. A deformable CNN architecture for predicting clinical acceptability of ECG signal. Biocybern Biomed Eng 2023;43(1):335-51.
• [6] Hou Y, Liu R, ShuM, Chen C. An ECG denoising method based on adversarial denoising convolutional neural network. Biomed Signal Process Control 2023;84:104964.
• [7] Fan T, Qiu S, Wang Z, Zhao H, Jiang J, Wang Y, et al. A new deep convolutional neural network incorporating attentional mechanisms for ECG emotion recognition. Comput Biol Med 2023;59:106938.
• [8] Wang T, Qin Y. A novel multi-scale convolutional network with attention-based bidirectional gated recurrent unit for atrial fibrillation discrimination. Biocybern Biomed Eng 2021;41(2):445-55.
• [9] Rahul J, Sharma DS. Automatic cardiac arrhythmia classification based on hybrid 1-D CNN and Bi-LSTM model. Biocybern Biomed Eng 2022;42(1):312-24.
• [10] Zeng W, Shan L, Su B, Du S. Epileptic seizure detection with deep EEG features by convolutional neural network and shallow classifiers. Front Neurosci 2023;17:1145526.
• [11] Kusuma S, Jothi KR. ECG signals-based automated diagnosis of congestive heart failure using deep CNN and LSTM architecture. Biocybern Biomed Eng 2022;42(1):247-57.
• [12] Patro KK, Allam JP, Hammad M, Tadeusiewicz R, Pławiak P. SCovNet: A skip connection-based feature union deep learning technique with statistical approach analysis for the detection of COVID-19. Biocybern Biomed Eng 2023;43(1):352-68.
• [13] Cardoso JF, Souloumiac A. Blind beamforming for non-Gaussian signals. IEE Proc F Radar Signal Process 1993;140(6):362-70.
• [14] Jaros R, Barnova K, Vilimkova Kahankova R, Pelisek J, Litschmannova M, Martinek R. Independent component analysis algorithms for non-invasive fetal electrocardiography. PLoS One 2023;18(6):e0286858.
• [15] Theis F, Lang E, Puntonet C. A geometric algorithm for overcomplete linear ICA. Neurocomputing 2002;56:381-98.
• [16] Févotte C, Godsill SJ, Wolfe PJ. Bayesian approach for blind separation of underdetermined mixtures of sparse sources. Lecture Notes in Comput Sci 2004;3195:398-405.
• [17] Lewicki M, Sejnowski T. Learning overcomplete representations. Neural Comput 2000;12:337-65.
• [18] Kotas M. Combined application of independent component analysis and projective filtering to fetal ECG extraction. Biocybern Biomed Eng 2008;28(1):75-93.
• [19] Tamber KK, Hayes DJL, Carey SJ, Wijekoon JHB, Heazell AEP. A systematic scoping review to identify the design and assess the performance of devices for antenatal continuous fetal monitoring. PLoS One 2020;15(12):e0242983.
• [20] Kahankova R, Mikolasova M, Martinek R. Optimization of adaptive filter control parameters for non-invasive fetal electrocardiogram extraction. PLoS One 2022;17(4):e0266807.
• [21] Jaros R, Martinek R, Kahankova R, Koziorek J. Novel hybrid extraction systems for fetal heart rate variability monitoring based on non-invasive fetal electrocardiogram. IEEE Access 2019;1.
• [22] Barnova K, Martinek R, Jaros R, Kahankova R, Behbehani K, Snasel V. System for adaptive extraction of non-invasive fetal electrocardiogram. Appl Soft Comput 2021;113:107940.
• [23] Giraldo-Guzmán J, Kotas M, Castells F, Contreras-Ortiz SH, Urina-Triana M. Estimation of PQ distance dispersion for atrial fibrillation detection. Comput Methods Programs Biomed 2021;208:106167.
• [24] Kotas M, Piela M, Contreras-Ortiz SH. Modified spatio-temporal matched filtering for brain responses classification. IEEE Trans Hum-Mach Syst 2022;52:1-10.
• [25] Rao K, Prasanna S, Yegnanarayana B. Determination of instants of significant excitation in speech using Hilbert envelope and group delay function. Signal Process Lett IEEE 2007;14:762-5.
• [26] Antink CH, Mai Y, Aalto R, Brüser C, Leonhardt S, Oksala N, Vehkaoja A. Ballisto-cardiography can estimate beat-to-beat heart rate accurately at night in patients after vascular intervention. IEEE J Biomed Health Inf 2020;24(8):2230-7.
• [27] Yager RR. On ordered weighted averaging aggregation operators in multicriteria decisionmaking. IEEE Trans Syst Man Cybern 1988;18(1):183-90.
• [28] Leski JM, Kotas M. On robust fuzzy c-regression models. Fuzzy Sets and Systems 2015;279:112-29.
• [29] Leski JM, Kotas M. Hierarchical clustering with planar segments as prototypes. Pattern Recognit Lett 2015;54:1-10.
• [30] Cardoso JF. Multidimensional independent component analysis. In: Proc. of the 1998 IEEE international conference on acoustics, speech and signal processing. 1998.
• [31] De Lathauwer L, De Moor B, Vandewalle J. Fetal electrocardiogram extraction by blind source subspace separation. IEEE Trans Biomed Eng 2000;47(5):567-72.
• [32] Kotas M, Jezewski J, Matonia A, Kupka T. Towards noise immune detection of fetal QRS complexes. Comput Methods Programs Biomed 2010;97(3):241-56.
• [33] Jezewski J, Matonia A, Kupka T, Roj D, Czabanski R. Determination of the fetal heart rate from abdominal signals: evaluation of beat-to-beat accuracy in relation to the direct fetal electrocardiogram. Biomedizinische Techn (Biomed Eng) 2012;57(5):383-94.
• [34] Goldberger AL, Amaral LA, Glass L, Hausdorff JM, Ivanov PC, Mark RG, et al. PhysioNet: components of a new research resource for complex physiologic signals. Circulation 2000;101(23). 215-220.
• [35] Matonia A, Jezewski J, Kupka T, Jezewski M, Horoba K, Wrobel J, et al. Fetal electrocardiograms, direct and abdominal with reference heartbeat annotations. Sci Data 2020;7(1):200.
• [36] Gurve D, Krishnan S. Separation of fetal-ECG from single-channel abdominal ECG using activation scaled non-negative matrix factorization. IEEE J Biomed Health Inf 2020;24(3):669-80.
• [37] Sameni Reza, Clifford Gari D, Jutten Christian, Mohammad B. Multichannel ECG and noise modeling: Application to maternal and fetal ECG signals. 2007.
• [38] Martin Bland J, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;327(8476).
• [39] Lampros T, Kalafatakis K, Giannakeas N, Tsipouras MG, Glavas E, Tzallas AT. An optimized hybrid methodology for non-invasive fetal electrocardiogram signal extraction and monitoring. Array 2023;19:100302.
• [40] Hao J, Yang Y, Zhou Z, Wu S. Fetal electrocardiogram signal extraction based on fast independent component analysis and singular value decomposition. Sensors (Basel) 2022;22(10):3705.
• [41] Qiao L, Hu S, Xiao B, Bi X, Li W, Gao X. A dual self-calibrating framework for noninvasive fetal ECG R-peak detection. IEEE Internet Things J 2023;10(18):16579-93.
• [42] Zhang Y, Yu S. Single-lead noninvasive fetal ECG extraction by means of combining clustering and principal components analysis. Med Biol Eng Comput 2020;58(2):419-32.
• [43] Shokouhm A, Tavassolian N. Fetal electrocardiogram extraction using dual-path source separation of single-channel non-invasive abdominal recordings. IEEE Trans Biomed Eng 2023;70(1):283-95.
• [44] Lee KJ, Lee B. End-to-end deep learning architecture for separating maternal and fetal ECGs using W-net. IEEE Access 2022;10:39782-8.
• [45] Raj A, Brablik J, Kahankova R, Jaros R, Barnova K, Snasel V, Mirjalili S, Martinek R. Nature inspired method for noninvasive fetal ECG extraction. Sci Rep 2022;12(1):20159.
• [46] Mohebbian MR, Vedaei SS, Wahid KA, Dinh A, Marateb HR, Tavakolian K. Fetal ECG extraction from maternal ECG using attention-based CycleGAN. IEEE J Biomed Health Inf 2022;26(2):515-26.
• [47] Basak P, Sakib AN, Chowdhury ME, Al-Emadi N, Yalcin HC, Pedersen S, et al. A novel deep learning technique for morphology preserved fetal ECG extraction from mother ECG using 1D-CycleGAN. Expert Syst Appl 2024;235:121196.
• [48] Wang X, He Z, Lin Z, Han Y, Liu T, Lu J, Xie S. PA2net: Period-aware attention network for robust fetal ECG detection. IEEE Trans Instrum Meas 2022;71:1-12.
• [49] Rahman AA, Mahmud SB, Chowdhury MB, Yalcin HC, Khandakar AB, Mutlu OC, et al. Fetal ECG extraction from maternal ECG using deeply supervised LinkNet++ model. Eng Appl Artif Intell 2023;123(C):106414.
• [50] Wang X, Han Y, Deng Y. CSGSA-net: Canonical-structured graph sparse attention network for fetal ECG estimation. Biomed Signal Process Control 2023;82:104556.
• [51] Li R, Frasch MG, Wu HT. Efficient fetal-maternal ECG signal separation from two channel maternal abdominal ECG via diffusion-based channel selection. Front Physiol 2017;8:277.
• [52] Kotas MP, Dyrek A, Piela M, Leski JM. Clustering based multiple state-space projections. Signal Process 2021;178:107762.

Uwagi

PL

Opracowanie rekordu ze środków MNiSW, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2024).