ECG Fiducial Point Localization Using a Deep Learning Model

ECG signals are essential in diagnosing cardiovascular diseases (CVD). Automatic localization of ECG fiducial points helps in the end-point detection and tracking of CVD. Nowadays, collecting ECG signals is more accessible because of the availability of wearable devices. We develop an algorithm to estimate the peaks of P and T waves and the onset and offset of the QRS complex. We evaluate it using ECG signals collected using a wearable device named HEMOTAG. The algorithm combines a rule-based method for heartbeat detection and a deep convolutional neural network (CNN) for fiducial points localization. Three datasets were used to train and evaluate the proposed algorithm. The first and second datasets are QT and Lobachevsky University Electrocardiography Database (LUDB), which are used in ten-fold cross-validation. The third dataset was collected using HEMOTAG, which is used as a held-out set. A percentage of error (PoE) less than 1.75% was achieved based on the cross-validation, and PoE less than 2.42% is achieved based on the held-out set.

[1]  M. Di Rienzo,et al.  Editorial: Cardiac Vibration Signals: Old Techniques, New Tricks, and Applications , 2022, Frontiers in Physiology.

[2]  Arash Andalib,et al.  Non-invasive Assessment Of Elevated Left Ventricular End Diastolic Pressure Using A Small Portable Device: Design For Tele-Monitoring , 2022, Journal of Cardiac Failure.

[3]  Arash Andalib,et al.  A Novel Non-Invasive Device For Screening And Optimized Management To Improve Heart Failure Outcomes In Patients With Diabetes Mellitus , 2022, Journal of Cardiac Failure.

[4]  Wei Li,et al.  Wearable sensors and devices for real-time cardiovascular disease monitoring , 2021, Cell Reports Physical Science.

[5]  Murtadha D. Hssayeni,et al.  Ensemble deep model for continuous estimation of Unified Parkinson’s Disease Rating Scale III , 2021, Biomedical engineering online.

[6]  O. Camara,et al.  Delineation of the electrocardiogram with a mixed-quality-annotations dataset using convolutional neural networks , 2021, Scientific Reports.

[7]  Sadasivan Puthusserypady,et al.  DENS-ECG: A Deep Learning Approach for ECG Signal Delineation , 2020, Expert Syst. Appl..

[8]  Mario Konijnenburg,et al.  CorNET: Deep Learning Framework for PPG-Based Heart Rate Estimation and Biometric Identification in Ambulant Environment , 2019, IEEE Transactions on Biomedical Circuits and Systems.

[9]  I. Yusipov,et al.  LUDB: A New Open-Access Validation Tool for Electrocardiogram Delineation Algorithms , 2018, IEEE Access.

[10]  Mikhail V. Ivanchenko,et al.  Finding Morphology Points of Electrocardiographic-Signal Waves Using Wavelet Analysis , 2018, Radiophysics and Quantum Electronics.

[11]  Sudha Ramasamy,et al.  Wearable sensors for ECG measurement: a review , 2018, Sensor Review.

[12]  Feng Liu,et al.  Deep Learning and Its Applications in Biomedicine , 2018, Genom. Proteom. Bioinform..

[13]  Daniel Cremers,et al.  Regularization for Deep Learning: A Taxonomy , 2017, ArXiv.

[14]  Byunghan Lee,et al.  Deep learning in bioinformatics , 2016, Briefings Bioinform..

[15]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[16]  N. S. Lingayat,et al.  Detection of P and T-waves in Electrocardiogram , 2008 .

[17]  Pablo Laguna,et al.  A database for evaluation of algorithms for measurement of QT and other waveform intervals in the ECG , 1997, Computers in Cardiology 1997.

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

[19]  Murtadha D. Hssayeni,et al.  Multi-Modal Physiological Data Fusion for Affect Estimation Using Deep Learning , 2021, IEEE Access.

[20]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[21]  Xiao Hu,et al.  Detection of Onset and Offset of QRS Complex Based a Modified Triangle Morphology , 2014 .

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