1
|
Jimenez-Perez G, Acosta J, Alcaine A, Camara O. Generalising electrocardiogram detection and delineation: training convolutional neural networks with synthetic data augmentation. Front Cardiovasc Med 2024; 11:1341786. [PMID: 39100388 PMCID: PMC11294154 DOI: 10.3389/fcvm.2024.1341786] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/20/2023] [Accepted: 06/14/2024] [Indexed: 08/06/2024] Open
Abstract
Introduction Extracting beat-by-beat information from electrocardiograms (ECGs) is crucial for various downstream diagnostic tasks that rely on ECG-based measurements. However, these measurements can be expensive and time-consuming to produce, especially for long-term recordings. Traditional ECG detection and delineation methods, relying on classical signal processing algorithms such as those based on wavelet transforms, produce high-quality delineations but struggle to generalise to diverse ECG patterns. Machine learning (ML) techniques based on deep learning algorithms have emerged as promising alternatives, capable of achieving similar performance without handcrafted features or thresholds. However, supervised ML techniques require large annotated datasets for training, and existing datasets for ECG detection/delineation are limited in size and the range of pathological conditions they represent. Methods This article addresses this challenge by introducing two key innovations. First, we develop a synthetic data generation scheme that probabilistically constructs unseen ECG traces from "pools" of fundamental segments extracted from existing databases. A set of rules guides the arrangement of these segments into coherent synthetic traces, while expert domain knowledge ensures the realism of the generated traces, increasing the input variability for training the model. Second, we propose two novel segmentation-based loss functions that encourage the accurate prediction of the number of independent ECG structures and promote tighter segmentation boundaries by focusing on a reduced number of samples. Results The proposed approach achieves remarkable performance, with a F 1 -score of 99.38% and delineation errors of 2.19 ± 17.73 ms and 4.45 ± 18.32 ms for ECG segment onsets and offsets across the P, QRS, and T waves. These results, aggregated from three diverse freely available databases (QT, LU, and Zhejiang), surpass current state-of-the-art detection and delineation approaches. Discussion Notably, the model demonstrated exceptional performance despite variations in lead configurations, sampling frequencies, and represented pathophysiology mechanisms, underscoring its robust generalisation capabilities. Real-world examples, featuring clinical data with various pathologies, illustrate the potential of our approach to streamline ECG analysis across different medical settings, fostered by releasing the codes as open source.
Collapse
Affiliation(s)
- Guillermo Jimenez-Perez
- Department of Information and Communication Technologies, PhySense Research Group, BCN-MedTech, Barcelona, Spain
- Arrhythmia Unit, Department of Cardiology, Virgen Del Rocío University Hospital, Seville, Spain
- Institut d’Investigacions Biomèdiques August Pi i Sunyer, Barcelona, Spain
| | - Juan Acosta
- Arrhythmia Unit, Department of Cardiology, Virgen Del Rocío University Hospital, Seville, Spain
| | - Alejandro Alcaine
- Computing for Medical and Biological Applications (CoMBA) Group, Facultad de Ciencias de la Salud, Universidad San Jorge, Zaragoza, Spain
| | - Oscar Camara
- Department of Information and Communication Technologies, PhySense Research Group, BCN-MedTech, Barcelona, Spain
| |
Collapse
|
2
|
Krasteva V, Stoyanov T, Schmid R, Jekova I. Delineation of 12-Lead ECG Representative Beats Using Convolutional Encoder-Decoders with Residual and Recurrent Connections. SENSORS (BASEL, SWITZERLAND) 2024; 24:4645. [PMID: 39066042 PMCID: PMC11280871 DOI: 10.3390/s24144645] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/19/2024] [Revised: 07/11/2024] [Accepted: 07/16/2024] [Indexed: 07/28/2024]
Abstract
The aim of this study is to address the challenge of 12-lead ECG delineation by different encoder-decoder architectures of deep neural networks (DNNs). This study compares four concepts for encoder-decoders based on a fully convolutional architecture (CED-Net) and its modifications with a recurrent layer (CED-LSTM-Net), residual connections between symmetrical encoder and decoder feature maps (CED-U-Net), and sequential residual blocks (CED-Res-Net). All DNNs transform 12-lead representative beats to three diagnostic ECG intervals (P-wave, QRS-complex, QT-interval) used for the global delineation of the representative beat (P-onset, P-offset, QRS-onset, QRS-offset, T-offset). All DNNs were trained and optimized using the large PhysioNet ECG database (PTB-XL) under identical conditions, applying an advanced approach for machine-based supervised learning with a reference algorithm for ECG delineation (ETM, Schiller AG, Baar, Switzerland). The test results indicate that all DNN architectures are equally capable of reproducing the reference delineation algorithm's measurements in the diagnostic PTB database with an average P-wave detection accuracy (96.6%) and time and duration errors: mean values (-2.6 to 2.4 ms) and standard deviations (2.9 to 11.4 ms). The validation according to the standard-based evaluation practices of diagnostic electrocardiographs with the CSE database outlines a CED-Net model, which measures P-duration (2.6 ± 11.0 ms), PQ-interval (0.9 ± 5.8 ms), QRS-duration (-2.4 ± 5.4 ms), and QT-interval (-0.7 ± 10.3 ms), which meet all standard tolerances. Noise tests with high-frequency, low-frequency, and power-line frequency noise (50/60 Hz) confirm that CED-Net, CED-Res-Net, and CED-LSTM-Net are robust to all types of noise, mostly presenting a mean duration error < 2.5 ms when compared to measurements without noise. Reduced noise immunity is observed for the U-net architecture. Comparative analysis with other published studies scores this research within the lower range of time errors, highlighting its competitive performance.
Collapse
Affiliation(s)
- Vessela Krasteva
- Institute of Biophysics and Biomedical Engineering, Bulgarian Academy of Sciences, Acad. G. Bonchev Str. Bl. 105, 1113 Sofia, Bulgaria; (V.K.); (T.S.)
| | - Todor Stoyanov
- Institute of Biophysics and Biomedical Engineering, Bulgarian Academy of Sciences, Acad. G. Bonchev Str. Bl. 105, 1113 Sofia, Bulgaria; (V.K.); (T.S.)
| | - Ramun Schmid
- Signal Processing, Schiller AG, Altgasse 68, CH-6341 Baar, Switzerland;
| | - Irena Jekova
- Institute of Biophysics and Biomedical Engineering, Bulgarian Academy of Sciences, Acad. G. Bonchev Str. Bl. 105, 1113 Sofia, Bulgaria; (V.K.); (T.S.)
| |
Collapse
|
3
|
Grégoire JM, Gilon C, Vaneberg N, Bersini H, Carlier S. Machine learning-based atrial fibrillation detection and onset prediction using QT-dynamicity. Physiol Meas 2024; 45:075001. [PMID: 38848724 DOI: 10.1088/1361-6579/ad55a1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/26/2023] [Accepted: 06/07/2024] [Indexed: 06/09/2024]
Abstract
Objective. This study examines the value of ventricular repolarization using QT dynamicity for two different types of atrial fibrillation (AF) prediction.Approach. We studied the importance of QT-dynamicity (1) in the detection and (2) the onset prediction (i.e. forecasting) of paroxysmal AF episodes using gradient-boosted decision trees (GBDT), an interpretable machine learning technique. We labeled 176 paroxysmal AF onsets from 88 patients in our unselected Holter recordings database containing paroxysmal AF episodes. Raw ECG signals were delineated using a wavelet-based signal processing technique. A total of 44 ECG features related to interval and wave durations and amplitude were selected and the GBDT model was trained with a Bayesian hyperparameters selection for various windows. The dataset was split into two parts at the patient level, meaning that the recordings from each patient were only present in either the train or test set, but not both. We used 80% on the database for the training and the remaining 20% for the test of the trained model. The model was evaluated using 5-fold cross-validation.Main results.The mean age of the patients was 75.9 ± 11.9 (range 50-99), the number of episodes per patient was 2.3 ± 2.2 (range 1-11), and CHA2DS2-VASc score was 2.9 ± 1.7 (range 1-9). For the detection of AF, we obtained an area under the receiver operating curve (AUROC) of 0.99 (CI 95% 0.98-0.99) and an accuracy of 95% using a 30 s window. Features related to RR intervals were the most influential, followed by those on QT intervals. For the AF onset forecast, we obtained an AUROC of 0.739 (0.712-0.766) and an accuracy of 74% using a 120s window. R wave amplitude and QT dynamicity as assessed by Spearman's correlation of the QT-RR slope were the best predictors.Significance. The QT dynamicity can be used to accurately predict the onset of AF episodes. Ventricular repolarization, as assessed by QT dynamicity, adds information that allows for better short time prediction of AF onset, compared to relying only on RR intervals and heart rate variability. Communication between the ventricles and atria is mediated by the autonomic nervous system (ANS). The variations in intraventricular conduction and ventricular repolarization changes resulting from the influence of the ANS play a role in the initiation of AF.
Collapse
Affiliation(s)
- Jean-Marie Grégoire
- IRIDIA, Université Libre de Bruxelles, Av. Adolphe Buyl 87, 1050 Bruxelles, Belgium
- Cardiology Department, Université de Mons, Place du Parc 20, 7000 Mons, Belgium
| | - Cédric Gilon
- IRIDIA, Université Libre de Bruxelles, Av. Adolphe Buyl 87, 1050 Bruxelles, Belgium
| | - Nathan Vaneberg
- IRIDIA, Université Libre de Bruxelles, Av. Adolphe Buyl 87, 1050 Bruxelles, Belgium
| | - Hugues Bersini
- IRIDIA, Université Libre de Bruxelles, Av. Adolphe Buyl 87, 1050 Bruxelles, Belgium
| | - Stéphane Carlier
- Cardiology Department, Université de Mons, Place du Parc 20, 7000 Mons, Belgium
| |
Collapse
|
4
|
Joung C, Kim M, Paik T, Kong SH, Oh SY, Jeon WK, Jeon JH, Hong JS, Kim WJ, Kook W, Cha MJ, van Koert O. Deep learning based ECG segmentation for delineation of diverse arrhythmias. PLoS One 2024; 19:e0303178. [PMID: 38870233 PMCID: PMC11175442 DOI: 10.1371/journal.pone.0303178] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/15/2023] [Accepted: 04/20/2024] [Indexed: 06/15/2024] Open
Abstract
Accurate delineation of key waveforms in an ECG is a critical step in extracting relevant features to support the diagnosis and treatment of heart conditions. Although deep learning based methods using segmentation models to locate P, QRS, and T waves have shown promising results, their ability to handle arrhythmias has not been studied in any detail. In this paper we investigate the effect of arrhythmias on delineation quality and develop strategies to improve performance in such cases. We introduce a U-Net-like segmentation model for ECG delineation with a particular focus on diverse arrhythmias. This is followed by a post-processing algorithm which removes noise and automatically determines the boundaries of P, QRS, and T waves. Our model has been trained on a diverse dataset and evaluated against the LUDB and QTDB datasets to show strong performance, with F1-scores exceeding 99% for QRS and T waves, and over 97% for P waves in the LUDB dataset. Furthermore, we assess various models across a wide array of arrhythmias and observe that models with a strong performance on standard benchmarks may still perform poorly on arrhythmias that are underrepresented in these benchmarks, such as tachycardias. We propose solutions to address this discrepancy.
Collapse
Affiliation(s)
- Chankyu Joung
- Department of Mathematical Sciences and Research Institute of Mathematics, Seoul National University, Gwanak-gu, Seoul, South Korea
| | - Mijin Kim
- Division of Cardiology, Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea
| | - Taejin Paik
- Department of Mathematical Sciences and Research Institute of Mathematics, Seoul National University, Gwanak-gu, Seoul, South Korea
| | - Seong-Ho Kong
- AI Institute, Seoul National University, Seoul, South Korea
- Department of Surgery, Seoul National University Hospital and Seoul National University College of Medicine, Seoul, South Korea
| | - Seung-Young Oh
- Department of Surgery, Seoul National University Hospital and Seoul National University College of Medicine, Seoul, South Korea
- Department of Critical Care Medicine, Seoul National University Hospital, Seoul, South Korea
| | - Won Kyeong Jeon
- Department of Internal Medicine, Seoul National University Hospital, Seoul, South Korea
| | | | | | | | - Woong Kook
- Department of Mathematical Sciences and Research Institute of Mathematics, Seoul National University, Gwanak-gu, Seoul, South Korea
- AI Institute, Seoul National University, Seoul, South Korea
| | - Myung-Jin Cha
- Division of Cardiology, Department of Internal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea
| | - Otto van Koert
- Department of Mathematical Sciences and Research Institute of Mathematics, Seoul National University, Gwanak-gu, Seoul, South Korea
| |
Collapse
|
5
|
Jimenez-Perez G, Acosta J, Bocanegra-Pérez ÁJ, Arana-Rueda E, Frutos-López M, Sánchez-Brotons JA, Llamas-López H, Di Massa Pezzutti R, González de la Portilla Concha C, Camara O, Pedrote A. Delineation of intracavitary electrograms for the automatic quantification of decrement-evoked potentials in the coronary sinus with deep-learning techniques. Front Physiol 2024; 15:1331852. [PMID: 38818521 PMCID: PMC11138951 DOI: 10.3389/fphys.2024.1331852] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/01/2023] [Accepted: 04/10/2024] [Indexed: 06/01/2024] Open
Abstract
Cardiac arrhythmias cause depolarization waves to conduct unevenly on the myocardial surface, potentially delaying local components with respect to a previous beat when stimulated at faster frequencies. Despite the diagnostic value of localizing the distinct local electrocardiogram (EGM) components for identifying regions with decrement-evoked potentials (DEEPs), current software solutions do not perform automatic signal quantification. Electrophysiologists must manually measure distances on the EGM signals to assess the existence of DEEPs during pacing or extra-stimuli protocols. In this work, we present a deep learning (DL)-based algorithm to identify decrement in atrial components (measured in the coronary sinus) with respect to their ventricular counterparts from EGM signals, for disambiguating between accessory pathways (APs) and atrioventricular re-entrant tachycardias (AVRTs). Several U-Net and W-Net neural networks with different configurations were trained on a private dataset of signals from the coronary sinus (312 EGM recordings from 77 patients who underwent AP or AVRT ablation). A second, separate dataset was annotated for clinical validation, with clinical labels associated to EGM fragments in which decremental conduction was elucidated. To alleviate data scarcity, a synthetic data augmentation method was developed for generating EGM recordings. Moreover, two novel loss functions were developed to minimize false negatives and delineation errors. Finally, the addition of self-attention mechanisms and their effect on model performance was explored. The best performing model was a W-Net model with 6 levels, optimized solely with the Dice loss. The model obtained precisions of 91.28%, 77.78% and of 100.0%, and recalls of 94.86%, 95.25% and 100.0% for localizing local field, far field activations, and extra-stimuli, respectively. The clinical validation model demonstrated good overall agreement with respect to the evaluation of decremental properties. When compared to the criteria of electrophysiologists, the automatic exclusion step reached a sensitivity of 87.06% and a specificity of 97.03%. Out of the non-excluded signals, a sensitivity of 96.77% and a specificity of 95.24% was obtained for classifying them into decremental and non-decremental potentials. Current results show great promise while being, to the best of our knowledge, the first tool in the literature allowing the delineation of all local components present in an EGM recording. This is of capital importance at advancing processing for cardiac electrophysiological procedures and reducing intervention times, as many diagnosis procedures are performed by comparing segments or late potentials in subsequent cardiac cycles.
Collapse
Affiliation(s)
- Guillermo Jimenez-Perez
- PhySense Research Group, BCN MedTech, Universitat Pompeu Fabra, Barcelona, Spain
- Arrhythmia Unit, Department of Cardiology at Virgen Del Rocío University Hospital, Sevilla, Spain
| | - Juan Acosta
- Arrhythmia Unit, Department of Cardiology at Virgen Del Rocío University Hospital, Sevilla, Spain
| | | | - Eduardo Arana-Rueda
- Arrhythmia Unit, Department of Cardiology at Virgen Del Rocío University Hospital, Sevilla, Spain
| | - Manuel Frutos-López
- Arrhythmia Unit, Department of Cardiology at Virgen Del Rocío University Hospital, Sevilla, Spain
| | - Juan A. Sánchez-Brotons
- Arrhythmia Unit, Department of Cardiology at Virgen Del Rocío University Hospital, Sevilla, Spain
| | - Helena Llamas-López
- Arrhythmia Unit, Department of Cardiology at Virgen Del Rocío University Hospital, Sevilla, Spain
| | | | | | - Oscar Camara
- PhySense Research Group, BCN MedTech, Universitat Pompeu Fabra, Barcelona, Spain
| | - Alonso Pedrote
- Arrhythmia Unit, Department of Cardiology at Virgen Del Rocío University Hospital, Sevilla, Spain
| |
Collapse
|
6
|
Diaw MD, Papelier S, Durand-Salmon A, Felblinger J, Oster J. A Human-Centered AI Framework for Efficient Labelling of ECGs From Drug Safety Trials. IEEE Trans Biomed Eng 2024; 71:1697-1704. [PMID: 38157467 DOI: 10.1109/tbme.2023.3348329] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/03/2024]
Abstract
Drug safety trials require substantial ECG labelling like, in thorough QT studies, measurements of the QT interval, whose prolongation is a biomarker of proarrhythmic risk. The traditional method of manually measuring the QT interval is time-consuming and error-prone. Studies have demonstrated the potential of deep learning (DL)-based methods to automate this task but expert validation of these computerized measurements remains of paramount importance, particularly for abnormal ECG recordings. In this paper, we propose a highly automated framework that combines such a DL-based QT estimator with human expertise. The framework consists of 3 key components: (1) automated QT measurement with uncertainty quantification (2) expert review of a few DL-based measurements, mostly those with high model uncertainty and (3) recalibration of the unreviewed measurements based on the expert-validated data. We assess its effectiveness on 3 drug safety trials and show that it can significantly reduce effort required for ECG labelling-in our experiments only 10% of the data were reviewed per trial-while maintaining high levels of QT accuracy. Our study thus demonstrates the possibility of productive human-machine collaboration in ECG analysis without any compromise on the reliability of subsequent clinical interpretations.
Collapse
|
7
|
Bocanegra-Pérez ÁJ, Piella G, Sebastian R, Jimenez-Perez G, Falasconi G, Saglietto A, Soto-Iglesias D, Berruezo A, Penela D, Camara O. Automatic and interpretable prediction of the site of origin in outflow tract ventricular arrhythmias: machine learning integrating electrocardiograms and clinical data. Front Cardiovasc Med 2024; 11:1353096. [PMID: 38572307 PMCID: PMC10987867 DOI: 10.3389/fcvm.2024.1353096] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/09/2023] [Accepted: 03/07/2024] [Indexed: 04/05/2024] Open
Abstract
The treatment of outflow tract ventricular arrhythmias (OTVA) through radiofrequency ablation requires the precise identification of the site of origin (SOO). Pinpointing the SOO enhances the likelihood of a successful procedure, reducing intervention times and recurrence rates. Current clinical methods to identify the SOO are based on qualitative analysis of pre-operative electrocardiograms (ECG), heavily relying on physician's expertise. Although computational models and machine learning (ML) approaches have been proposed to assist OTVA procedures, they either consume substantial time, lack interpretability or do not use clinical information. Here, we propose an alternative strategy for automatically predicting the ventricular origin of OTVA patients using ML. Our objective was to classify ventricular (left/right) origin in the outflow tracts (LVOT and RVOT, respectively), integrating ECG and clinical data from each patient. Extending beyond differentiating ventricle origin, we explored specific SOO characterization. Utilizing four databases, we also trained supervised learning models on the QRS complexes of the ECGs, clinical data, and their combinations. The best model achieved an accuracy of 89%, highlighting the significance of precordial leads V1-V4, especially in the R/S transition and initiation of the QRS complex in V2. Unsupervised analysis revealed that some origins tended to group closer than others, e.g., right coronary cusp (RCC) with a less sparse group than the aortic cusp origins, suggesting identifiable patterns for specific SOOs.
Collapse
Affiliation(s)
- Álvaro J. Bocanegra-Pérez
- Physense, BCN Medtech, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, Spain
| | - Gemma Piella
- Physense, BCN Medtech, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, Spain
| | - Rafael Sebastian
- Computational Multiscale Simulation Lab (CoMMLab), Department of Computer Science, Universitat de Valencia, Valencia, Spain
| | - Guillermo Jimenez-Perez
- Physense, BCN Medtech, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, Spain
| | - Giulio Falasconi
- Cardiology Department, Heart Institute, Teknon Medical Center, Barcelona, Spain
| | - Andrea Saglietto
- Division of Cardiology, Department of Medical Sciences, University of Turin, Turin, Italy
| | - David Soto-Iglesias
- Cardiology Department, Heart Institute, Teknon Medical Center, Barcelona, Spain
| | - Antonio Berruezo
- Cardiology Department, Heart Institute, Teknon Medical Center, Barcelona, Spain
| | - Diego Penela
- Department of Arrhythmology, Humanitas Research Hospital, Milan, Italy
| | - Oscar Camara
- Physense, BCN Medtech, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, Spain
| |
Collapse
|
8
|
Li X, Cai W, Xu B, Jiang Y, Qi M, Wang M. SEResUTer: a deep learning approach for accurate ECG signal delineation and atrial fibrillation detection. Physiol Meas 2023; 44:125005. [PMID: 37827168 DOI: 10.1088/1361-6579/ad02da] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/09/2023] [Accepted: 10/12/2023] [Indexed: 10/14/2023]
Abstract
Objective.Accurate detection of electrocardiogram (ECG) waveforms is crucial for computer-aided diagnosis of cardiac abnormalities. This study introduces SEResUTer, an enhanced deep learning model designed for ECG delineation and atrial fibrillation (AF) detection.Approach. Built upon a U-Net architecture, SEResUTer incorporates ResNet modules and Transformer encoders to replace convolution blocks, resulting in improved optimization and encoding capabilities. A novel masking strategy is proposed to handle incomplete expert annotations. The model is trained on the QT database (QTDB) and evaluated on the Lobachevsky University Electrocardiography Database (LUDB) to assess its generalization performance. Additionally, the model's scope is extended to AF detection using the the China Physiological Signal Challenge 2021 (CPSC2021) and the China Physiological Signal Challenge 2018 (CPSC2018) datasets.Main results.The proposed model surpasses existing traditional and deep learning approaches in ECG waveform delineation on the QTDB. It achieves remarkable average F1 scores of 99.14%, 98.48%, and 98.46% for P wave, QRS wave, and T wave delineation, respectively. Moreover, the model demonstrates exceptional generalization ability on the LUDB, achieving average SE, positive prediction rate, and F1 scores of 99.05%, 94.59%, and 94.62%, respectively. By analyzing RR interval differences and the existence of P waves, our method achieves AF identification with 99.20% accuracy on the CPSC2021 test set and demonstrates strong generalization on CPSC2018 dataset.Significance.The proposed approach enables highly accurate ECG waveform delineation and AF detection, facilitating automated analysis of large-scale ECG recordings and improving the diagnosis of cardiac abnormalities.
Collapse
Affiliation(s)
- Xinyue Li
- School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, 200093, People's Republic of China
| | - Wenjie Cai
- School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, 200093, People's Republic of China
| | - Bolin Xu
- School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, 200093, People's Republic of China
| | - Yupeng Jiang
- School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, 200093, People's Republic of China
| | - Mengdi Qi
- School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai, 200093, People's Republic of China
| | - Mingjie Wang
- Shanghai Key Laboratory of Bioactive Small Molecules, School of Basic Medical Science, Fudan University, Shanghai, 200032, People's Republic of China
| |
Collapse
|
9
|
Nurmaini S, Darmawahyuni A, Rachmatullah MN, Firdaus F, Sapitri AI, Tutuko B, Tondas AE, Putra MHP, Islami A. Robust electrocardiogram delineation model for automatic morphological abnormality interpretation. Sci Rep 2023; 13:13736. [PMID: 37612382 PMCID: PMC10447439 DOI: 10.1038/s41598-023-40965-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/24/2023] [Accepted: 08/19/2023] [Indexed: 08/25/2023] Open
Abstract
Knowledge of electrocardiogram (ECG) wave signals is one of the essential steps in diagnosing heart abnormalities. Considerable performance with respect to obtaining the critical point of a signal waveform (P-QRS-T) through ECG delineation has been achieved in many studies. However, several deficiencies remain regarding previous methods, including the effects of noise interference on the performance degradation of delineation and the role of medical knowledge in reaching a delineation decision. To address these challenges, this paper proposes a robust delineation model based on a convolutional recurrent network with grid search optimization, aiming to classify the precise P-QRS-T waves. In order to make a delineation decision, the results from the ECG waveform classification model are utilized to interpret morphological abnormalities, based on medical knowledge. We generated 36 models, and the model with the best results achieved 99.97% accuracy, 99.92% sensitivity, and 99.93% precision for ECG waveform classification (P-wave, QRS-complex, T-wave, and isoelectric line class). To ensure the model robustness, we evaluated delineation model performance on seven different types of ECG datasets, namely the Lobachevsky University Electrocardiography Database (LUDB), QT Database (QTDB), the PhysioNet/Computing in Cardiology Challenge 2017, China Physiological Signal Challenge 2018, ECG Arrhythmia of Chapman University, MIT-BIH Arrhythmia Database and General Mohammad Hossein Hospital (Indonesia) databases. To detect the patterns of ECG morphological abnormalities through proposed delineation model, we focus on investigating arrhythmias. This process is based on two inputs examination: the P-wave and the regular/irregular rhythm of the RR interval. As the results, the proposed method has considerable capability to interpret the delineation result in cases with artifact noise, baseline drift and abnormal morphologies for delivering robust ECG delineation.
Collapse
Affiliation(s)
- Siti Nurmaini
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, 30139, Indonesia.
| | - Annisa Darmawahyuni
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, 30139, Indonesia.
| | - Muhammad Naufal Rachmatullah
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, 30139, Indonesia
| | - Firdaus Firdaus
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, 30139, Indonesia
| | - Ade Iriani Sapitri
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, 30139, Indonesia
| | - Bambang Tutuko
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, 30139, Indonesia.
| | - Alexander Edo Tondas
- Department of Cardiology and Vascular Medicine, Dr. Mohammad Hoesin Hospital, Palembang, 30126, Indonesia
| | | | - Anggun Islami
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, 30139, Indonesia
| |
Collapse
|
10
|
Rahman MM, Rivolta MW, Badilini F, Sassi R. A Systematic Survey of Data Augmentation of ECG Signals for AI Applications. SENSORS (BASEL, SWITZERLAND) 2023; 23:s23115237. [PMID: 37299964 DOI: 10.3390/s23115237] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 04/27/2023] [Revised: 05/23/2023] [Accepted: 05/27/2023] [Indexed: 06/12/2023]
Abstract
AI techniques have recently been put under the spotlight for analyzing electrocardiograms (ECGs). However, the performance of AI-based models relies on the accumulation of large-scale labeled datasets, which is challenging. To increase the performance of AI-based models, data augmentation (DA) strategies have been developed recently. The study presented a comprehensive systematic literature review of DA for ECG signals. We conducted a systematic search and categorized the selected documents by AI application, number of leads involved, DA method, classifier, performance improvements after DA, and datasets employed. With such information, this study provided a better understanding of the potential of ECG augmentation in enhancing the performance of AI-based ECG applications. This study adhered to the rigorous PRISMA guidelines for systematic reviews. To ensure comprehensive coverage, publications between 2013 and 2023 were searched across multiple databases, including IEEE Explore, PubMed, and Web of Science. The records were meticulously reviewed to determine their relevance to the study's objective, and those that met the inclusion criteria were selected for further analysis. Consequently, 119 papers were deemed relevant for further review. Overall, this study shed light on the potential of DA to advance the field of ECG diagnosis and monitoring.
Collapse
Affiliation(s)
- Md Moklesur Rahman
- Dipartimento di Informatica, Università degli Studi di Milano, 20133 Milan, Italy
| | | | - Fabio Badilini
- School of Nursing, University of California, San Francisco, CA 94143, USA
- AMPS-LLC, New York, NY 10025, USA
| | - Roberto Sassi
- Dipartimento di Informatica, Università degli Studi di Milano, 20133 Milan, Italy
| |
Collapse
|
11
|
Diaw MD, Papelier S, Durand-Salmon A, Felblinger J, Oster J. AI-Assisted QT Measurements for Highly Automated Drug Safety Studies. IEEE Trans Biomed Eng 2023; 70:1504-1515. [PMID: 36355743 DOI: 10.1109/tbme.2022.3221339] [Citation(s) in RCA: 6] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/12/2022]
Abstract
Rate-corrected QT interval (QTc) prolongation has been suggested as a biomarker for the risk of drug-induced torsades de pointes, and is therefore monitored during clinical trials for the assessment of drug safety. Manual QT measurements by expert ECG analysts are expensive, laborious and prone to errors. Wavelet-based delineators and other automatic methods do not generalize well to different T wave morphologies and may require laborious tuning. Our study investigates the robustness of convolutional neural networks (CNNs) for QT measurement. We trained 3 CNN-based deep learning models on a private ECG database with human expert-annotated QT intervals. Among these models, we propose a U-Net model, which is widely used for segmentation tasks, to build a novel clinically useful QT estimator that includes QT delineation for better interpretability. We tested the 3 models on four external databases, amongst which a clinical trial investigating four drugs. Our results show that the deep learning models are in stronger agreement with the experts than the state-of-the-art wavelet-based algorithm. Indeed, the deep learning models yielded up to 71% of accurate QT measurements (absolute difference between manual and automatic QT below 15 ms) whereas the wavelet-based algorithm only allowed 52% of QT accuracy. For the 2 studies of drugs with small to no QT prolonging effect, a mean absolute difference of 6 ms (std = 5 ms) was obtained between the manual and deep learning methods. For the other 2 drugs with more significant effect on the volunteers, a mean difference of up to 17 ms (std = 17 ms) was obtained. The proposed models are therefore promising for automated QT measurements during clinical trials. They can analyze various ECG morphologies from a diversity of individuals although some QT-prolonged ECGs can be challenging. The U-Net model is particularly interesting for our application as it facilitates expert review of automatic QT intervals, which is still required by regulatory bodies, by providing QRS onset and T offset positions that are consistent with the estimated QT intervals.
Collapse
|
12
|
Inter-patient ECG characteristic wave detection based on convolutional neural network combined with transformer. Biomed Signal Process Control 2023. [DOI: 10.1016/j.bspc.2022.104436] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
|
13
|
Mehri M, Calmon G, Odille F, Oster J. A Deep Learning Architecture Using 3D Vectorcardiogram to Detect R-Peaks in ECG with Enhanced Precision. SENSORS (BASEL, SWITZERLAND) 2023; 23:2288. [PMID: 36850889 PMCID: PMC9963088 DOI: 10.3390/s23042288] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 01/09/2023] [Revised: 02/14/2023] [Accepted: 02/15/2023] [Indexed: 06/18/2023]
Abstract
Providing reliable detection of QRS complexes is key in automated analyses of electrocardiograms (ECG). Accurate and timely R-peak detections provide a basis for ECG-based diagnoses and to synchronize radiologic, electrophysiologic, or other medical devices. Compared with classical algorithms, deep learning (DL) architectures have demonstrated superior accuracy and high generalization capacity. Furthermore, they can be embedded on edge devices for real-time inference. 3D vectorcardiograms (VCG) provide a unifying framework for detecting R-peaks regardless of the acquisition strategy or number of ECG leads. In this article, a DL architecture was demonstrated to provide enhanced precision when trained and applied on 3D VCG, with no pre-processing nor post-processing steps. Experiments were conducted on four different public databases. Using the proposed approach, high F1-scores of 99.80% and 99.64% were achieved in leave-one-out cross-validation and cross-database validation protocols, respectively. False detections, measured by a precision of 99.88% or more, were significantly reduced compared with recent state-of-the-art methods tested on the same databases, without penalty in the number of missed peaks, measured by a recall of 99.39% or more. This approach can provide new applications for devices where precision, or positive predictive value, is essential, for instance cardiac magnetic resonance imaging.
Collapse
Affiliation(s)
- Maroua Mehri
- Epsidy, 54000 Nancy, France
- Ecole Nationale d’Ingénieurs de Sousse, LATIS-Laboratory of Advanced Technology and Intelligent Systems, Université de Sousse, Sousse 4023, Tunisia
| | | | - Freddy Odille
- Epsidy, 54000 Nancy, France
- IADI-Imagerie Adaptative Diagnostique et Interventionnelle, Inserm U1254, Université de Lorraine, 54000 Nancy, France
- CIC-IT 1433, Inserm, CHRU de Nancy, Université de Lorraine, 54000 Nancy, France
| | - Julien Oster
- IADI-Imagerie Adaptative Diagnostique et Interventionnelle, Inserm U1254, Université de Lorraine, 54000 Nancy, France
- CIC-IT 1433, Inserm, CHRU de Nancy, Université de Lorraine, 54000 Nancy, France
| |
Collapse
|
14
|
Chen Z, Wang M, Zhang M, Huang W, Gu H, Xu J. Post-processing refined ECG delineation based on 1D-UNet. Biomed Signal Process Control 2023. [DOI: 10.1016/j.bspc.2022.104106] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
|
15
|
Tutuko B, Darmawahyuni A, Nurmaini S, Tondas AE, Naufal Rachmatullah M, Teguh SBP, Firdaus F, Sapitri AI, Passarella R. DAE-ConvBiLSTM: End-to-end learning single-lead electrocardiogram signal for heart abnormalities detection. PLoS One 2022; 17:e0277932. [PMID: 36584187 PMCID: PMC9803308 DOI: 10.1371/journal.pone.0277932] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/01/2022] [Accepted: 11/08/2022] [Indexed: 01/01/2023] Open
Abstract
BACKGROUND The electrocardiogram (ECG) is a widely used diagnostic that observes the heart activities of patients to ascertain a heart abnormality diagnosis. The artifacts or noises are primarily associated with the problem of ECG signal processing. Conventional denoising techniques have been proposed in previous literature; however, some lacks, such as the determination of suitable wavelet basis function and threshold, can be a time-consuming process. This paper presents end-to-end learning using a denoising auto-encoder (DAE) for denoising algorithms and convolutional-bidirectional long short-term memory (ConvBiLSTM) for ECG delineation to classify ECG waveforms in terms of the PQRST-wave and isoelectric lines. The denoising reconstruction using unsupervised learning based on the encoder-decoder process can be proposed to improve the drawbacks. First, The ECG signals are reduced to a low-dimensional vector in the encoder. Second, the decoder reconstructed the signals. The last, the reconstructed signals of ECG can be processed to ConvBiLSTM. The proposed architecture of DAE-ConvBiLSTM is the end-to-end diagnosis of heart abnormality detection. RESULTS As a result, the performance of DAE-ConvBiLSTM has obtained an average of above 98.59% accuracy, sensitivity, specificity, precision, and F1 score from the existing studies. The DAE-ConvBiLSTM has also experimented with detecting T-wave (due to ventricular repolarisation) morphology abnormalities. CONCLUSION The development architecture for detecting heart abnormalities using an unsupervised learning DAE and supervised learning ConvBiLSTM can be proposed for an end-to-end learning algorithm. In the future, the precise accuracy of the ECG main waveform will affect heart abnormalities detection in clinical practice.
Collapse
Affiliation(s)
- Bambang Tutuko
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, Indonesia
| | - Annisa Darmawahyuni
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, Indonesia
- * E-mail: , .id (SN); , .id (AD)
| | - Siti Nurmaini
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, Indonesia
- * E-mail: , .id (SN); , .id (AD)
| | - Alexander Edo Tondas
- Department of Cardiology & Vascular Medicine, Dr. Mohammad Hoesin Hospital, Palembang, Indonesia
| | | | | | - Firdaus Firdaus
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, Indonesia
| | - Ade Iriani Sapitri
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, Indonesia
| | - Rossi Passarella
- Intelligent System Research Group, Faculty of Computer Science, Universitas Sriwijaya, Palembang, Indonesia
| |
Collapse
|
16
|
Wu W, Huang Y, Wu X. A New Deep Learning Method with Self-Supervised Learning for Delineation of the Electrocardiogram. ENTROPY (BASEL, SWITZERLAND) 2022; 24:1828. [PMID: 36554233 PMCID: PMC9778283 DOI: 10.3390/e24121828] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 10/23/2022] [Revised: 12/11/2022] [Accepted: 12/12/2022] [Indexed: 06/17/2023]
Abstract
Heartbeat characteristic points are the main features of an electrocardiogram (ECG), which can provide important information for ECG-based cardiac diagnosis. In this manuscript, we propose a self-supervised deep learning framework with modified Densenet to detect ECG characteristic points, including the onset, peak and termination points of P-wave, QRS complex wave and T-wave. We extracted high-level features of ECG heartbeats from the QT Database (QTDB) and two other larger datasets, MIT-BIH Arrhythmia Database (MITDB) and MIT-BIH Normal Sinus Rhythm Database (NSRDB) with no human-annotated labels as pre-training. By applying different transformations to ECG signals, the task of discriminating signals before and after transformation was defined as the pretext task. Subsequently, the convolutional layer was frozen and the weights of the self-supervised network were transferred to the downstream task of characteristic point localizations on heart beats in the QT dataset. Finally, the mean ± standard deviation of the detection errors of our proposed self-supervised learning method in QTDB for detecting the onset, peak, and termination points of P-waves, the onset and termination points of QRS waves, and the peak and termination points of T-waves were -0.24 ± 10.04, -0.48 ± 11.69, -0.28 ± 10.19, -3.72 ± 8.18, -4.12 ± 13.54, -0.68 ± 20.42, and 1.34 ± 21.04. The results show that the deep learning network based on the self-supervised framework constructed in this manuscript can accurately detect the feature points of a heartbeat, laying the foundation for automatic extraction of key information related to ECG-based diagnosis.
Collapse
Affiliation(s)
- Wenwen Wu
- Center for Biomedical Engineering, School of Information Science and Technology, Fudan University, Shanghai 200433, China
| | - Yanqi Huang
- Center for Biomedical Engineering, School of Information Science and Technology, Fudan University, Shanghai 200433, China
| | - Xiaomei Wu
- Center for Biomedical Engineering, School of Information Science and Technology, Fudan University, Shanghai 200433, China
- Academy for Engineering and Technology, Fudan University, Shanghai 200433, China
- Key Laboratory of Medical Imaging Computing and Computer Assisted Intervention (MICCAI) of Shanghai, Shanghai 200433, China
- Yiwu Research Institute of Fudan University, Yiwu 322000, China
- Shanghai Engineering Research Center of Assistive Devices, Shanghai 200433, China
| |
Collapse
|
17
|
Ma G, Chen Y, Zhu W, Zheng L, Tang H, Yu Y, Wang L. Evaluating and Visualizing the Contribution of ECG Characteristic Waveforms for PPG-Based Blood Pressure Estimation. MICROMACHINES 2022; 13:1438. [PMID: 36144060 PMCID: PMC9502729 DOI: 10.3390/mi13091438] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 08/14/2022] [Revised: 08/27/2022] [Accepted: 08/27/2022] [Indexed: 06/16/2023]
Abstract
Non-invasive continuous blood pressure monitoring is of great significance for the preventing, diagnosing, and treating of cardiovascular diseases (CVDs). Studies have demonstrated that photoplethysmogram (PPG) and electrocardiogram (ECG) signals can effectively and continuously predict blood pressure (BP). However, most of the BP estimation models focus on the waveform features of the PPG signal, while the peak value of R-wave in ECG is only used as a time reference, and few references investigated the ECG waveforms. This paper aims to evaluate the influence of three characteristic waveforms in ECG on the improvement of BP estimation. PPG is the primary signal, and five input combinations are formed by adding ECG, P wave, QRS complex, T wave, and none. We employ five common convolutional neural networks (CNN) to validate the consistency of the contribution. Meanwhile, with the visualization of Gradient-weighted class activation mapping (Grad-CAM), we generate the heat maps and further visualize the distribution of CNN's attention to each waveform of PPG and ECG. The heat maps show that networks pay more attention to the QRS complex and T wave. In the comparison results, the QRS complex and T wave have more contribution to minimizing errors than P wave. By separately adding P wave, QRS complex, and T wave, the average MAE of these networks reaches 7.87 mmHg, 6.57 mmHg, and 6.21 mmHg for systolic blood pressure (SBP), and 4.27 mmHg, 3.65 mmHg, and 3.73 mmHg, respectively, for diastolic blood pressure (DBP). The results of the experiment show that QRS complex and T wave deserves more attention and feature extraction like PPG waveform features in the continuous BP estimation.
Collapse
Affiliation(s)
- Gang Ma
- School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei 230026, China
- Suzhou Institute of Biomedical Engineering and Technology, China Academy of Sciences, Suzhou 215163, China
| | - Yuhang Chen
- School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei 230026, China
- Suzhou Institute of Biomedical Engineering and Technology, China Academy of Sciences, Suzhou 215163, China
| | - Wenliang Zhu
- School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei 230026, China
- Suzhou Institute of Biomedical Engineering and Technology, China Academy of Sciences, Suzhou 215163, China
| | - Lesong Zheng
- School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei 230026, China
| | - Hui Tang
- School of Electronics and Information Technology, Soochow University, Suzhou 215006, China
| | - Yong Yu
- Suzhou Institute of Biomedical Engineering and Technology, China Academy of Sciences, Suzhou 215163, China
| | - Lirong Wang
- Suzhou Institute of Biomedical Engineering and Technology, China Academy of Sciences, Suzhou 215163, China
- School of Electronics and Information Technology, Soochow University, Suzhou 215006, China
| |
Collapse
|
18
|
Liu J, Li Z, Jin Y, Liu Y, Liu C, Zhao L, Chen X. A review of arrhythmia detection based on electrocardiogram with artificial intelligence. Expert Rev Med Devices 2022; 19:549-560. [DOI: 10.1080/17434440.2022.2115887] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/04/2022]
Affiliation(s)
- Jinlei Liu
- School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
| | - Zhiyuan Li
- School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
| | - Yanrui Jin
- School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
| | - Yunqing Liu
- School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
| | - Chengliang Liu
- School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
- MoE Key Lab of Artificial Intelligence, AI Institute, Shanghai Jiao Tong University, China
| | - Liqun Zhao
- Department of Cardiology, Shanghai First People’s Hospital Affiliated to Shanghai Jiao Tong University, 100 Haining Road, Shanghai 200080, China
| | - Xiaojun Chen
- School of Mechanical Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China
| |
Collapse
|
19
|
Doste R, Lozano M, Jimenez-Perez G, Mont L, Berruezo A, Penela D, Camara O, Sebastian R. Training machine learning models with synthetic data improves the prediction of ventricular origin in outflow tract ventricular arrhythmias. Front Physiol 2022; 13:909372. [PMID: 36035489 PMCID: PMC9412034 DOI: 10.3389/fphys.2022.909372] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/31/2022] [Accepted: 07/07/2022] [Indexed: 11/13/2022] Open
Abstract
In order to determine the site of origin (SOO) in outflow tract ventricular arrhythmias (OTVAs) before an ablation procedure, several algorithms based on manual identification of electrocardiogram (ECG) features, have been developed. However, the reported accuracy decreases when tested with different datasets. Machine learning algorithms can automatize the process and improve generalization, but their performance is hampered by the lack of large enough OTVA databases. We propose the use of detailed electrophysiological simulations of OTVAs to train a machine learning classification model to predict the ventricular origin of the SOO of ectopic beats. We generated a synthetic database of 12-lead ECGs (2,496 signals) by running multiple simulations from the most typical OTVA SOO in 16 patient-specific geometries. Two types of input data were considered in the classification, raw and feature ECG signals. From the simulated raw 12-lead ECG, we analyzed the contribution of each lead in the predictions, keeping the best ones for the training process. For feature-based analysis, we used entropy-based methods to rank the obtained features. A cross-validation process was included to evaluate the machine learning model. Following, two clinical OTVA databases from different hospitals, including ECGs from 365 patients, were used as test-sets to assess the generalization of the proposed approach. The results show that V2 was the best lead for classification. Prediction of the SOO in OTVA, using both raw signals or features for classification, presented high accuracy values (>0.96). Generalization of the network trained on simulated data was good for both patient datasets (accuracy of 0.86 and 0.84, respectively) and presented better values than using exclusively real ECGs for classification (accuracy of 0.84 and 0.76 for each dataset). The use of simulated ECG data for training machine learning-based classification algorithms is critical to obtain good SOO predictions in OTVA compared to real data alone. The fast implementation and generalization of the proposed methodology may contribute towards its application to a clinical routine.
Collapse
Affiliation(s)
- Ruben Doste
- Department of Computer Science, University of Oxford, Oxford, United Kingdom
- *Correspondence: Ruben Doste,
| | - Miguel Lozano
- Computational Multiscale Simulation Lab (CoMMLab), Department of Computer Science, Universitat de Valencia, Valencia, Spain
| | - Guillermo Jimenez-Perez
- Physense, BCN Medtech, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, Spain
| | - Lluis Mont
- Arrhythmia Section, Cardiology Department, Cardiovascular Clinical Institute, Hospital Clínic, Universitat de Barcelona - IDIBAPS, Barcelona, Spain
| | - Antonio Berruezo
- Cardiology Department, Heart Institute, Teknon Medical Center, Barcelona, Spain
| | - Diego Penela
- Cardiology Department, Heart Institute, Teknon Medical Center, Barcelona, Spain
| | - Oscar Camara
- Physense, BCN Medtech, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, Spain
| | - Rafael Sebastian
- Computational Multiscale Simulation Lab (CoMMLab), Department of Computer Science, Universitat de Valencia, Valencia, Spain
| |
Collapse
|
20
|
Nezamabadi K, Mayfield J, Li P, Greenland GV, Rodriguez S, Simsek B, Mousavi P, Shatkay H, Abraham MR. Toward ECG-based analysis of hypertrophic cardiomyopathy: a novel ECG segmentation method for handling abnormalities. J Am Med Inform Assoc 2022; 29:1879-1889. [PMID: 35923089 PMCID: PMC9552290 DOI: 10.1093/jamia/ocac122] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/08/2022] [Revised: 06/22/2022] [Accepted: 07/21/2022] [Indexed: 11/13/2022] Open
Abstract
OBJECTIVE Abnormalities in impulse propagation and cardiac repolarization are frequent in hypertrophic cardiomyopathy (HCM), leading to abnormalities in 12-lead electrocardiograms (ECGs). Computational ECG analysis can identify electrophysiological and structural remodeling and predict arrhythmias. This requires accurate ECG segmentation. It is unknown whether current segmentation methods developed using datasets containing annotations for mostly normal heartbeats perform well in HCM. Here, we present a segmentation method to effectively identify ECG waves across 12-lead HCM ECGs. METHODS We develop (1) a web-based tool that permits manual annotations of P, P', QRS, R', S', T, T', U, J, epsilon waves, QRS complex slurring, and atrial fibrillation by 3 experts and (2) an easy-to-implement segmentation method that effectively identifies ECG waves in normal and abnormal heartbeats. Our method was tested on 131 12-lead HCM ECGs and 2 public ECG sets to evaluate its performance in non-HCM ECGs. RESULTS Over the HCM dataset, our method obtained a sensitivity of 99.2% and 98.1% and a positive predictive value of 92% and 95.3% when detecting QRS complex and T-offset, respectively, significantly outperforming a state-of-the-art segmentation method previously employed for HCM analysis. Over public ECG sets, it significantly outperformed 3 state-of-the-art methods when detecting P-onset and peak, T-offset, and QRS-onset and peak regarding the positive predictive value and segmentation error. It performed at a level similar to other methods in other tasks. CONCLUSION Our method accurately identified ECG waves in the HCM dataset, outperforming a state-of-the-art method, and demonstrated similar good performance as other methods in normal/non-HCM ECG sets.
Collapse
Affiliation(s)
- Kasra Nezamabadi
- Computational Biomedicine Lab, Computer and Information Sciences, University of Delaware, Newark, Delaware, USA
| | - Jacob Mayfield
- Division of Cardiology, Hypertrophic Cardiomyopathy Center of Excellence, University of California San Francisco, San Francisco, California, USA
| | - Pengyuan Li
- Computational Biomedicine Lab, Computer and Information Sciences, University of Delaware, Newark, Delaware, USA
| | - Gabriela V Greenland
- Division of Cardiology, Hypertrophic Cardiomyopathy Center of Excellence, University of California San Francisco, San Francisco, California, USA
| | - Sebastian Rodriguez
- Division of Cardiology, Hypertrophic Cardiomyopathy Center of Excellence, University of California San Francisco, San Francisco, California, USA
| | - Bahadir Simsek
- Division of Cardiology, Hypertrophic Cardiomyopathy Center of Excellence, University of California San Francisco, San Francisco, California, USA
| | - Parvin Mousavi
- School of Computing, Queen's University, Kingston, Ontario, Canada
| | - Hagit Shatkay
- Computational Biomedicine Lab, Computer and Information Sciences, University of Delaware, Newark, Delaware, USA
| | - M Roselle Abraham
- Hypertrophic Cardiomyopathy Center of Excellence, Division of Cardiology, University of California San Francisco, San Francisco, USA
| |
Collapse
|
21
|
Enhanced Pre-Processing for Deep Learning in MRI Whole Brain Segmentation using Orthogonal Moments. BRAIN MULTIPHYSICS 2022. [DOI: 10.1016/j.brain.2022.100049] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/18/2023] Open
|
22
|
Petmezas G, Stefanopoulos L, Kilintzis V, Tzavelis A, Rogers JA, Katsaggelos AK, Maglaveras N. State-of-the-art Deep Learning Methods on Electrocardiogram Data: A Systematic Review (Preprint). JMIR Med Inform 2022; 10:e38454. [PMID: 35969441 PMCID: PMC9425174 DOI: 10.2196/38454] [Citation(s) in RCA: 10] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/03/2022] [Revised: 06/03/2022] [Accepted: 07/03/2022] [Indexed: 11/13/2022] Open
Abstract
Background Electrocardiogram (ECG) is one of the most common noninvasive diagnostic tools that can provide useful information regarding a patient’s health status. Deep learning (DL) is an area of intense exploration that leads the way in most attempts to create powerful diagnostic models based on physiological signals. Objective This study aimed to provide a systematic review of DL methods applied to ECG data for various clinical applications. Methods The PubMed search engine was systematically searched by combining “deep learning” and keywords such as “ecg,” “ekg,” “electrocardiogram,” “electrocardiography,” and “electrocardiology.” Irrelevant articles were excluded from the study after screening titles and abstracts, and the remaining articles were further reviewed. The reasons for article exclusion were manuscripts written in any language other than English, absence of ECG data or DL methods involved in the study, and absence of a quantitative evaluation of the proposed approaches. Results We identified 230 relevant articles published between January 2020 and December 2021 and grouped them into 6 distinct medical applications, namely, blood pressure estimation, cardiovascular disease diagnosis, ECG analysis, biometric recognition, sleep analysis, and other clinical analyses. We provide a complete account of the state-of-the-art DL strategies per the field of application, as well as major ECG data sources. We also present open research problems, such as the lack of attempts to address the issue of blood pressure variability in training data sets, and point out potential gaps in the design and implementation of DL models. Conclusions We expect that this review will provide insights into state-of-the-art DL methods applied to ECG data and point to future directions for research on DL to create robust models that can assist medical experts in clinical decision-making.
Collapse
Affiliation(s)
- Georgios Petmezas
- Lab of Computing, Medical Informatics and Biomedical-Imaging Technologies, The Medical School, Aristotle University of Thessaloniki, Thessaloniki, Greece
| | - Leandros Stefanopoulos
- Lab of Computing, Medical Informatics and Biomedical-Imaging Technologies, The Medical School, Aristotle University of Thessaloniki, Thessaloniki, Greece
| | - Vassilis Kilintzis
- Lab of Computing, Medical Informatics and Biomedical-Imaging Technologies, The Medical School, Aristotle University of Thessaloniki, Thessaloniki, Greece
| | - Andreas Tzavelis
- Department of Biomedical Engineering, Northwestern University, Evanston, IL, United States
| | - John A Rogers
- Department of Material Science, Northwestern University, Evanston, IL, United States
| | - Aggelos K Katsaggelos
- Department of Electrical and Computer Engineering, Northwestern University, Evanston, IL, United States
| | - Nicos Maglaveras
- Lab of Computing, Medical Informatics and Biomedical-Imaging Technologies, The Medical School, Aristotle University of Thessaloniki, Thessaloniki, Greece
| |
Collapse
|
23
|
Short Single-Lead ECG Signal Delineation-Based Deep Learning: Implementation in Automatic Atrial Fibrillation Identification. SENSORS 2022; 22:s22062329. [PMID: 35336500 PMCID: PMC8953093 DOI: 10.3390/s22062329] [Citation(s) in RCA: 4] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 01/28/2022] [Revised: 03/09/2022] [Accepted: 03/14/2022] [Indexed: 01/27/2023]
Abstract
Physicians manually interpret an electrocardiogram (ECG) signal morphology in routine clinical practice. This activity is a monotonous and abstract task that relies on the experience of understanding ECG waveform meaning, including P-wave, QRS-complex, and T-wave. Such a manual process depends on signal quality and the number of leads. ECG signal classification based on deep learning (DL) has produced an automatic interpretation; however, the proposed method is used for specific abnormality conditions. When the ECG signal morphology change to other abnormalities, it cannot proceed automatically. To generalize the automatic interpretation, we aim to delineate ECG waveform. However, the output of delineation process only ECG waveform duration classes for P-wave, QRS-complex, and T-wave. It should be combined with a medical knowledge rule to produce the abnormality interpretation. The proposed model is applied for atrial fibrillation (AF) identification. This study meets the AF criteria with RR irregularities and the absence of P-waves in essential oscillations for even more accurate identification. The QT database by Physionet is utilized for developing the delineation model, and it validates with The Lobachevsky University Database. The results show that our delineation model works properly, with 98.91% sensitivity, 99.01% precision, 99.79% specificity, 99.79% accuracy, and a 98.96% F1 score. We use about 4058 normal sinus rhythm records and 1804 AF records from the experiment to identify AF conditions that are taken from three datasets. The comprehensive testing has produced higher negative predictive value and positive predictive value. This means that the proposed model can identify AF conditions from ECG signal delineation. Our approach can considerably contribute to AF diagnosis with these results.
Collapse
|
24
|
Tao R, Zhang S, Wang Y, Mi X, Ma J, Shen C, Zheng G. MCG-Net: End-to-end Fine-grained Delineation and Diagnostic Classification of Cardiac Events from Magnetocardiographs. IEEE J Biomed Health Inform 2021; 26:1057-1067. [PMID: 34780340 DOI: 10.1109/jbhi.2021.3128169] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
Abstract
In this paper, we propose an end-to-end deep learning architecture, referred as MCG-Net, integrating convolutional neural network (CNN) with transformer-based global context block for fine-grained delineation and diagnostic classification of four cardiac events from magnetocardiogram (MCG) data, namely Q-, R-, S- and T-waves. MCG-Net} takes advantage of a multi-resolution CNN backbone as well as the state-of-the-art (SOTA) transformer encoders that facilitate global temporal feature aggregation. Besides the novel network architecture, we introduce a multi-task learning scheme to achieve simultaneous delineation and classification. Specifically, the problem of MCG delineation is formulated as multi-class heatmap regression. Meanwhile, a binary diagnostic classification label as well as a duration are jointly estimated for each cardiac event using features that are temporally aligned by event heatmaps. The framework is evaluated on a clinical MCG dataset, containing data collected from 270 subjects with cardiac anomalies and 108 control subjects. We designed and conducted a two-fold cross-validation study to validate the proposed method and to compare its performance with the SOTA methods. Experimental results demonstrated that our method outperformed counterparts on both event delineation and diagnostic classification tasks, achieving respectively an average ECG-F1 of 0.987 and an average Event-F1 of 0.975 for MCG delineation, and an average accuracy of 0.870, an average sensitivity of 0.732, an average specificity of 0.914 and an average AUC of 0.903 for diagnostic classification. Comprehensive ablation experiments are additionally performed to investigate effectiveness of different network components.
Collapse
|