Design of Deep Convolutional Neural Network Architectures for Denoising Electrocardiographic Signals

Deep Learning (DL) models have been used extensively in image processing and other domains with great success but only very recently have been used in processing electrocardiogram (ECG) signals. Effective and powerful methods for denoising real ECG signals are important for wearable sensors and devices. This paper presents the design of a DL model, the Convolutional Neural Networks (CNNs), together with the more conventional filtering methods (low pass filtering, high pass filtering, Notch filtering) and the standard wavelet-based technique for denoising ECG signals. This combination shows to produce better results than using only the CNN models. The methods are trained, tested and evaluated on real ECG datasets taken from the MIT PhysioNet database. The results show the CNN model is a performant model that can be used for denoising ECG signals, and also that before applying the CNN model, standard filtering of the original ECG signal has a beneficial effect for the final results. All CNN models used an NVIDIA TITAN V Graphical Processing Unit (GPU) with 12 GB RAM, which reduces drastically the computational times.

[1]  N. S. Lingayat,et al.  Detection of QRS complexes in electrocardiogram using support vector machine , 2008, Journal of medical engineering & technology.

[2]  Gayadhar Pradhan,et al.  A New ECG Denoising Framework Using Generative Adversarial Network , 2020, IEEE/ACM Transactions on Computational Biology and Bioinformatics.

[3]  Nicholas S. Peters,et al.  ECG Diagnosis in Clinical Practice , 2008 .

[4]  Jimeng Sun,et al.  Opportunities and Challenges in Deep Learning Methods on Electrocardiogram Data: A Systematic Review , 2020, ArXiv.

[5]  G.B. Moody,et al.  The impact of the MIT-BIH Arrhythmia Database , 2001, IEEE Engineering in Medicine and Biology Magazine.

[6]  Khaled Daqrouq,et al.  ECG Signal Denoising By Wavelet Transform Thresholding , 2008 .

[7]  Rik Vullings,et al.  End-to-end trained encoder-decoder convolutional neural network for fetal electrocardiogram signal denoising. , 2020 .

[8]  Ki H. Chon,et al.  Automatic Motion and Noise Artifact Detection in Holter ECG Data Using Empirical Mode Decomposition and Statistical Approaches , 2012, IEEE Transactions on Biomedical Engineering.

[9]  Míriam Garcia Bermúdez,et al.  External 30 Days Holter Usefulness in Unexplained Syncope, Palpitations and Cardioembolic Suspected Cryptogenic Stroke Study. , 2017, European cardiology.

[10]  Rodrigo Castañeda-Miranda,et al.  DSP-based arrhythmia classification using wavelet transform and probabilistic neural network , 2017, Biomed. Signal Process. Control..

[11]  Hugo Gamboa,et al.  Noise detection on ECG based on agglomerative clustering of morphological features , 2017, Comput. Biol. Medicine.

[12]  Peter Holland,et al.  Removing ECG noise from surface EMG signals using adaptive filtering , 2009, Neuroscience Letters.

[13]  Ying Chen,et al.  A two-stage ECG signal denoising method based on deep convolutional network , 2020, bioRxiv.

[14]  Yu Tsao,et al.  Noise Reduction in ECG Signals Using Fully Convolutional Denoising Autoencoders , 2019, IEEE Access.

[15]  Corneliu T. C. Arsene,et al.  Complex Deep Learning Models for Denoising of Human Heart ECG signals , 2019, ArXiv.

[16]  Monisha Chakraborty,et al.  Determination of Signal to Noise Ratio of Electrocardiograms Filtered by Band Pass and Savitzky-Golay Filters , 2012 .

[17]  Feng Wan,et al.  Muscle and electrode motion artifacts reduction in ECG using adaptive Fourier decomposition , 2014, 2014 IEEE International Conference on Systems, Man, and Cybernetics (SMC).

[18]  Carlos Lastre-Domínguez,et al.  ECG Signal Denoising and Features Extraction Using Unbiased FIR Smoothing , 2019, BioMed research international.

[19]  Christian Jutten,et al.  A Nonlinear Bayesian Filtering Framework for ECG Denoising , 2007, IEEE Transactions on Biomedical Engineering.

[20]  Hujun Yin,et al.  Deep Learning Models for Denoising ECG Signals , 2019, 2019 27th European Signal Processing Conference (EUSIPCO).

[21]  Alexander J. Casson,et al.  An Ultra Low Power Personalizable Wrist Worn ECG Monitor Integrated With IoT Infrastructure , 2018, IEEE Access.

[22]  Karol Antczak,et al.  Deep Recurrent Neural Networks for ECG Signal Denoising , 2018, ArXiv.

[23]  Mohanasankar Sivaprakasam,et al.  Deep Network for Capacitive ECG Denoising , 2019, 2019 IEEE International Symposium on Medical Measurements and Applications (MeMeA).

[24]  Ming Liu,et al.  ECG signal enhancement based on improved denoising auto-encoder , 2016, Eng. Appl. Artif. Intell..

[25]  N.V. Thakor,et al.  Applications of adaptive filtering to ECG analysis: noise cancellation and arrhythmia detection , 1991, IEEE Transactions on Biomedical Engineering.

[26]  Xin Gao,et al.  Diagnosing Abnormal Electrocardiogram (ECG) via Deep Learning , 2019, Practical Applications of Electrocardiogram.

[27]  Kee Koo Kwon,et al.  Vital Block and Vital Sign Server for ECG and Vital Sign Monitoring in a Portable u-Vital System , 2020, Sensors.

[28]  Andrew Y. Ng,et al.  Cardiologist-Level Arrhythmia Detection with Convolutional Neural Networks , 2017, ArXiv.

[29]  D Panigrahy,et al.  Extended Kalman smoother with differential evolution technique for denoising of ECG signal , 2016, Australasian Physical & Engineering Sciences in Medicine.

[30]  Annisa Darmawahyuni,et al.  Deep Learning-Based Stacked Denoising and Autoencoder for ECG Heartbeat Classification , 2020, Electronics.

[31]  Rui Rodrigues,et al.  A Neural Network Approach to ECG Denoising , 2012, ArXiv.

[32]  Akash Ravi,et al.  Enhancing Biomedical Sensing in Ambulances using Neural Networks and Internet of Things , 2020, 2020 Fourth International Conference on Computing Methodologies and Communication (ICCMC).

[33]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

[34]  Willis J. Tompkins,et al.  A Real-Time QRS Detection Algorithm , 1985, IEEE Transactions on Biomedical Engineering.

[35]  Trygve Eftestøl,et al.  High Frequency Noise Detection and Handling in ECG Signals , 2018, 2018 26th European Signal Processing Conference (EUSIPCO).