Zuzana Koscova, R. Smíšek, P. Nejedly, J. Halámek, P. Jurák, P. Leinveber, K. Čurila, F. Plesinger
Background: QRS duration is a common measure linked to conduction abnormalities in heart ventricles. Aim: We propose a QRS detector, further able to locate QRS onset and offset in one inference step. Method: A 3-second window from 12 leads of UHF ECG signal (5 kHz) is standardized and processed with the UNet network. The output is an array of QRS probabilities, further processed with probability and distance criterion, allowing us to determine duration and final location of QRSs. Results: The model was trained on 2,250 ECG recordings from the FNUSA-ICRC hospital (Brno, Czechia). The model was tested on 5 different datasets: FNUSA, a dataset from FNKV hospital (Prague, Czechia), and three public datasets (Cipa, Strict LBBB, LUDB). Regarding QRS duration, results showed a mean absolute error of 13.99 ± 4.29 ms between annotated durations and the output of the proposed model. A QRS detection F-score was 0.98 ± 0.01. Conclusion: Our results indicate high QRS detection performance on both spontaneous and paced UHF ECG data. We also showed that QRS detection and duration could be combined in one deep learning algorithm.
{"title":"Ultra-High Frequency ECG Deep-Learning Beat Detector Delivering QRS Onsets and Offsets","authors":"Zuzana Koscova, R. Smíšek, P. Nejedly, J. Halámek, P. Jurák, P. Leinveber, K. Čurila, F. Plesinger","doi":"10.22489/CinC.2022.230","DOIUrl":"https://doi.org/10.22489/CinC.2022.230","url":null,"abstract":"Background: QRS duration is a common measure linked to conduction abnormalities in heart ventricles. Aim: We propose a QRS detector, further able to locate QRS onset and offset in one inference step. Method: A 3-second window from 12 leads of UHF ECG signal (5 kHz) is standardized and processed with the UNet network. The output is an array of QRS probabilities, further processed with probability and distance criterion, allowing us to determine duration and final location of QRSs. Results: The model was trained on 2,250 ECG recordings from the FNUSA-ICRC hospital (Brno, Czechia). The model was tested on 5 different datasets: FNUSA, a dataset from FNKV hospital (Prague, Czechia), and three public datasets (Cipa, Strict LBBB, LUDB). Regarding QRS duration, results showed a mean absolute error of 13.99 ± 4.29 ms between annotated durations and the output of the proposed model. A QRS detection F-score was 0.98 ± 0.01. Conclusion: Our results indicate high QRS detection performance on both spontaneous and paced UHF ECG data. We also showed that QRS detection and duration could be combined in one deep learning algorithm.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"498 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128936763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. McCullum, Hasan Saeed, Benjamin Moody, D. Perry, Eric Gottlieb, T. Pollard, Xavier Borrat Frigola, Qiao Li, Gari D. Clifford, R. Mark, Li-wei H. Lehman
To develop robust algorithms for automated diagnosis of medical conditions such as cardiac arrhythmias, researchers require large collections of data with human expert annotations. Currently, there is a lack of accessible, open-source platforms for human experts to collaboratively develop these annotated datasets through a web interface. In this work, we developed a flexible, generalizable, web-based framework to enable multiple users to create and share annotations on multi-channel physiological waveforms. Using the developed annotation platform, we carried out a pilot study to assess the validity of ventricular tachycardia (VT) alarms from multiple commercial monitors. Thus far, four clinical experts have used this annotation tool to annotate a total of 5,658 VT alarm events, among which approximately 44%(N=2,468) have been labeled by two independent annotators.
{"title":"PhysioTag: An Open-Source Platform for Collaborative Annotation of Physiological Waveforms","authors":"L. McCullum, Hasan Saeed, Benjamin Moody, D. Perry, Eric Gottlieb, T. Pollard, Xavier Borrat Frigola, Qiao Li, Gari D. Clifford, R. Mark, Li-wei H. Lehman","doi":"10.22489/CinC.2022.335","DOIUrl":"https://doi.org/10.22489/CinC.2022.335","url":null,"abstract":"To develop robust algorithms for automated diagnosis of medical conditions such as cardiac arrhythmias, researchers require large collections of data with human expert annotations. Currently, there is a lack of accessible, open-source platforms for human experts to collaboratively develop these annotated datasets through a web interface. In this work, we developed a flexible, generalizable, web-based framework to enable multiple users to create and share annotations on multi-channel physiological waveforms. Using the developed annotation platform, we carried out a pilot study to assess the validity of ventricular tachycardia (VT) alarms from multiple commercial monitors. Thus far, four clinical experts have used this annotation tool to annotate a total of 5,658 VT alarm events, among which approximately 44%(N=2,468) have been labeled by two independent annotators.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130417986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Parvaneh, Zaniar Ardalan, Joomyung Song, Kathan Vyas, C. Potes
A digital phonocardiogram (PCG) provides an opportunity for automated screening in resource-constrained environments. As part of the George B. Moody PhysioNet Challenge 2022, our team, Life_Is _Now, developed a computational approach using an ensemble of deep learning classifiers for identifying abnormal cardiac function from PCG. A stratified 5-fold cross-validation was used for model development and evaluation for murmur and clinical outcome identification. The backbone of our trained classifiers is a modified pre-trained deep convolutional neural network on AudioSet-Youtube corpus (YAMNet) and transfer learning. The YAMNet model is modified and finetuned on the publicly available PhysioNet dataset. Our murmur and clinical outcome classifiers received a weighted accuracy score of 0.831 and a Challenge cost score of 14,850 from cross-validation on the public training set. Our murmur scores were 0.678 and outcome score were 10,518 on the hidden validation set. However, we did not receive the official score for the hidden test set as our entry crashed in evaluation on the test set.
数字心音图(PCG)为资源受限环境下的自动筛查提供了机会。作为2022年George B. Moody PhysioNet挑战赛的一部分，我们的团队Life_Is _Now开发了一种使用深度学习分类器集合的计算方法，用于从PCG中识别异常心功能。分层5重交叉验证用于模型开发和评估杂音和临床结果识别。我们训练的分类器的主干是基于AudioSet-Youtube语料库(YAMNet)和迁移学习的改进预训练深度卷积神经网络。YAMNet模型在公开可用的PhysioNet数据集上进行修改和微调。我们的杂音和临床结果分类器在公共训练集的交叉验证中获得了0.831的加权准确率分数和14850的挑战成本分数。在隐藏验证集上，我们的杂音得分为0.678，结果得分为10,518。但是，我们没有收到隐藏测试集的官方分数，因为我们的条目在测试集的评估中崩溃了。
{"title":"Heart Murmur Detection Using Ensemble of Deep Learning Classifiers for Phonocardiograms Recorded from Multiple Auscultation Locations","authors":"S. Parvaneh, Zaniar Ardalan, Joomyung Song, Kathan Vyas, C. Potes","doi":"10.22489/CinC.2022.241","DOIUrl":"https://doi.org/10.22489/CinC.2022.241","url":null,"abstract":"A digital phonocardiogram (PCG) provides an opportunity for automated screening in resource-constrained environments. As part of the George B. Moody PhysioNet Challenge 2022, our team, Life_Is _Now, developed a computational approach using an ensemble of deep learning classifiers for identifying abnormal cardiac function from PCG. A stratified 5-fold cross-validation was used for model development and evaluation for murmur and clinical outcome identification. The backbone of our trained classifiers is a modified pre-trained deep convolutional neural network on AudioSet-Youtube corpus (YAMNet) and transfer learning. The YAMNet model is modified and finetuned on the publicly available PhysioNet dataset. Our murmur and clinical outcome classifiers received a weighted accuracy score of 0.831 and a Challenge cost score of 14,850 from cross-validation on the public training set. Our murmur scores were 0.678 and outcome score were 10,518 on the hidden validation set. However, we did not receive the official score for the hidden test set as our entry crashed in evaluation on the test set.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130490383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
"Fazeelat Mazhar, Francesco Regazzoni, C. Bartolucci, C. Corsi, L. Dede’, A. Quarteroni, S. Severi
Electromechanical coupling is crucial for modeling a realistic representation of $Ca^{+2}$ transient and $Ca^{+2}$ cycling. Cellular $Ca^{+2}$ dynamics in atria differ fundamentally from the ventricles. A biophysically detailed electrophysiology model is hence necessary to reproduce the experimentally observed phenomena like $Ca^{+2}$ wave propagation in human atrial myocytes. In this work, we present a novel detailed and yet computationally efficient electrophysiology model, its coupling with a contraction myofilament model and the effect of mechano-calcium feedback on coupling. This novel electromechanical model was calibrated for a collection of human atrial data and was evaluated by reproducing the rate adaptation property of action potential, $Ca^{+2}$ transient and the active force. The aim of this article is to present a new electromechanical model for human atrial myocyte and to analyse the mechanism behind the rate adaptation.
{"title":"A Novel Human Atrial Electromechanical Cardiomyocyte Model with Mechano-Calcium Feedback Effect","authors":"\"Fazeelat Mazhar, Francesco Regazzoni, C. Bartolucci, C. Corsi, L. Dede’, A. Quarteroni, S. Severi","doi":"10.22489/CinC.2022.195","DOIUrl":"https://doi.org/10.22489/CinC.2022.195","url":null,"abstract":"Electromechanical coupling is crucial for modeling a realistic representation of $Ca^{+2}$ transient and $Ca^{+2}$ cycling. Cellular $Ca^{+2}$ dynamics in atria differ fundamentally from the ventricles. A biophysically detailed electrophysiology model is hence necessary to reproduce the experimentally observed phenomena like $Ca^{+2}$ wave propagation in human atrial myocytes. In this work, we present a novel detailed and yet computationally efficient electrophysiology model, its coupling with a contraction myofilament model and the effect of mechano-calcium feedback on coupling. This novel electromechanical model was calibrated for a collection of human atrial data and was evaluated by reproducing the rate adaptation property of action potential, $Ca^{+2}$ transient and the active force. The aim of this article is to present a new electromechanical model for human atrial myocyte and to analyse the mechanism behind the rate adaptation.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131323002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Katerina Iscra, A. Miladinović, M. Ajčević, Luca Restivo, Simone Kresevic, M. Merlo, G. Sinagra, A. Accardo
Early non-invasive diagnosis of Ischemic Heart Disease (IHD) can often be challenging. HRV features have a potentially important role in risk stratification for subjects with suspected heart disease. However, there is no consensus on the HRV preprocessing steps, particularly on how to properly treat ectopic beats. We aimed to investigate the performance of the models for classification of early IHD versus healthy subjects (HC) based on HRV features extracted from signals excluding ectopic beats and based on the same features extracted from the signals that contain both ectopic and normal heartbeats. This study encompassed 385 subjects (170 IHD and 215 HC). The models were produced by logistic regression method considering two sets of HRV features obtained by two preprocessing approaches. The results showed that the model with the input features from HRV signals including normal and ectopic beats presented a higher classification accuracy (72.7%) than the model based on features extracted only from normal heart beats (67.8%). In addition, the evaluation of the feature importance by analysis of produced nomograms and observed significant differences between features extracted with two preprocessing approaches, showed also that the exclusion of the ectopic beats modifies the features' discriminatory power between HC and IHD.
{"title":"Does Ectopic Beats Bring More Discriminatory Information to Diagnose Ischemic Heart Disease?","authors":"Katerina Iscra, A. Miladinović, M. Ajčević, Luca Restivo, Simone Kresevic, M. Merlo, G. Sinagra, A. Accardo","doi":"10.22489/CinC.2022.199","DOIUrl":"https://doi.org/10.22489/CinC.2022.199","url":null,"abstract":"Early non-invasive diagnosis of Ischemic Heart Disease (IHD) can often be challenging. HRV features have a potentially important role in risk stratification for subjects with suspected heart disease. However, there is no consensus on the HRV preprocessing steps, particularly on how to properly treat ectopic beats. We aimed to investigate the performance of the models for classification of early IHD versus healthy subjects (HC) based on HRV features extracted from signals excluding ectopic beats and based on the same features extracted from the signals that contain both ectopic and normal heartbeats. This study encompassed 385 subjects (170 IHD and 215 HC). The models were produced by logistic regression method considering two sets of HRV features obtained by two preprocessing approaches. The results showed that the model with the input features from HRV signals including normal and ectopic beats presented a higher classification accuracy (72.7%) than the model based on features extracted only from normal heart beats (67.8%). In addition, the evaluation of the feature importance by analysis of produced nomograms and observed significant differences between features extracted with two preprocessing approaches, showed also that the exclusion of the ectopic beats modifies the features' discriminatory power between HC and IHD.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126453914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This work aimed to perform a numerical study of aortic hemodynamics and evaluate both Newtonian and non-Newtonian blood flow parameters in an ascending aortic aneurysm model. An aortic model was reconstructed from a medical computed tomography (CT) image, and finite element method laminar blood flow modelling was performed using different blood parameters. The inflow boundary conditions were defined as a flow profile, and the outlet boundary conditions were defined as the pressure at each outlet. The first simulation was calculated by considering blood as a Newtonian fluid, while in the second simulation, using the Carreau model, blood was assumed to be a non-Newtonian fluid. The results showed that average systolic and diastolic velocities were 2% and 9% higher, respectively, for the non-Newtonian fluid. In addition, the wall shear stress (WSS) values on the surface of the aneurysm were 30% higher during systole in the non-Newtonian simulation, while the average WSS on the artery surface in diastole was 20% higher for the Newtonian fluid.
{"title":"Comparison of Newtonian and Non-Newtonian Blood Flow in an Ascending Aortic Aneurysm","authors":"A. Petuchova, A. Maknickas","doi":"10.22489/CinC.2022.085","DOIUrl":"https://doi.org/10.22489/CinC.2022.085","url":null,"abstract":"This work aimed to perform a numerical study of aortic hemodynamics and evaluate both Newtonian and non-Newtonian blood flow parameters in an ascending aortic aneurysm model. An aortic model was reconstructed from a medical computed tomography (CT) image, and finite element method laminar blood flow modelling was performed using different blood parameters. The inflow boundary conditions were defined as a flow profile, and the outlet boundary conditions were defined as the pressure at each outlet. The first simulation was calculated by considering blood as a Newtonian fluid, while in the second simulation, using the Carreau model, blood was assumed to be a non-Newtonian fluid. The results showed that average systolic and diastolic velocities were 2% and 9% higher, respectively, for the non-Newtonian fluid. In addition, the wall shear stress (WSS) values on the surface of the aneurysm were 30% higher during systole in the non-Newtonian simulation, while the average WSS on the artery surface in diastole was 20% higher for the Newtonian fluid.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126270992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Elliott, Daniele Cinque, L. Mainardi, J. F. R. Matas
In-silico modelling is increasingly relied upon to gain new insights into the underlying mechanisms of atrial fibrillation. Due to the complex nature of the atria, insilico models typically exclude cellular heterogeneity. One question that remains unanswered is the impact of cellular heterogeneity on reentrant mechanisms and in the vulnerable window (VW). This study aims to present the impact of cellular heterogeneity on the AF mechanisms and susceptibility to re-entry behaviour. Cellular heterogeneity was introduced into the whole atrial model using the population of models approach and regionally specific node assignment. Each atrial model was stimulated from the SA node, followed by a series of rapid-paced ectopic beats at one of three locations in the left atria. Results showed a small, insignificant increase in reentrant frequency as a result of cellular heterogeneity, with only minor changes to the re-entrant circuit. However, the vulnerable window was significantly impacted through the introduction of cellular heterogeneity. The results suggest that cellular heterogeneity in the atrial model resulted in an increased VW for reentry depending on EB location. This suggests that local cellular heterogeneity plays a significant role in the susceptibility to re-entries, but does not significantly impact the path or frequency of re-entries.
{"title":"Cellular Heterogeneity in the Atria: An In Silico Study in the Impact on Reentries","authors":"J. Elliott, Daniele Cinque, L. Mainardi, J. F. R. Matas","doi":"10.22489/CinC.2022.296","DOIUrl":"https://doi.org/10.22489/CinC.2022.296","url":null,"abstract":"In-silico modelling is increasingly relied upon to gain new insights into the underlying mechanisms of atrial fibrillation. Due to the complex nature of the atria, insilico models typically exclude cellular heterogeneity. One question that remains unanswered is the impact of cellular heterogeneity on reentrant mechanisms and in the vulnerable window (VW). This study aims to present the impact of cellular heterogeneity on the AF mechanisms and susceptibility to re-entry behaviour. Cellular heterogeneity was introduced into the whole atrial model using the population of models approach and regionally specific node assignment. Each atrial model was stimulated from the SA node, followed by a series of rapid-paced ectopic beats at one of three locations in the left atria. Results showed a small, insignificant increase in reentrant frequency as a result of cellular heterogeneity, with only minor changes to the re-entrant circuit. However, the vulnerable window was significantly impacted through the introduction of cellular heterogeneity. The results suggest that cellular heterogeneity in the atrial model resulted in an increased VW for reentry depending on EB location. This suggests that local cellular heterogeneity plays a significant role in the susceptibility to re-entries, but does not significantly impact the path or frequency of re-entries.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122224322","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
"Michele Orini, S. van Duijvenboden, A. Tinker, P. Munroe, P. Lambiase
Photoplethysmography (PPG) is used in many wearable devices and it is becoming the most commonly measured cardiovascular signal, but its association with cardiovascular events is undetermined. This study uses data from the UK Biobank to assess the association between PPG morphological features and risk of cardiovascular (CV) events. N=175,284 individuals without CV disease were included (44.6% male, $56.4pm 8.1$ years old). A single finger PPG waveform of 101 data points, evenly sampled over the cycle length was available. The PPG waveforms were normalized between 0 and 1 and the maximum of the first derivative during the pulse's upslope was measured $(x_{MAX}^{prime})$. Cox regressions were used to assess the association between $x_{MAX}^{prime}$ and mortality and cardiovascular events. After a median follow-up period of 11.2 years, incidence of all-cause mortality (ACM), myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF) and stroke (STR), ranged between 2.1% and 5.2%. A reduction of 1 standard deviation in $x_{MAX}^{prime}$ was associated with increased risk of all outcomes with hazard ratio between 1.20 and 1.30. After adjusting for sex, age, and body mass index, associations remained significant for all outcomes except AF.
{"title":"Association Between Photoplethysmography Pulse Upslope and Cardiovascular Events in over 170,000 UK Biobank Participants","authors":"\"Michele Orini, S. van Duijvenboden, A. Tinker, P. Munroe, P. Lambiase","doi":"10.22489/CinC.2022.226","DOIUrl":"https://doi.org/10.22489/CinC.2022.226","url":null,"abstract":"Photoplethysmography (PPG) is used in many wearable devices and it is becoming the most commonly measured cardiovascular signal, but its association with cardiovascular events is undetermined. This study uses data from the UK Biobank to assess the association between PPG morphological features and risk of cardiovascular (CV) events. N=175,284 individuals without CV disease were included (44.6% male, $56.4pm 8.1$ years old). A single finger PPG waveform of 101 data points, evenly sampled over the cycle length was available. The PPG waveforms were normalized between 0 and 1 and the maximum of the first derivative during the pulse's upslope was measured $(x_{MAX}^{prime})$. Cox regressions were used to assess the association between $x_{MAX}^{prime}$ and mortality and cardiovascular events. After a median follow-up period of 11.2 years, incidence of all-cause mortality (ACM), myocardial infarction (MI), heart failure (HF), atrial fibrillation (AF) and stroke (STR), ranged between 2.1% and 5.2%. A reduction of 1 standard deviation in $x_{MAX}^{prime}$ was associated with increased risk of all outcomes with hazard ratio between 1.20 and 1.30. After adjusting for sex, age, and body mass index, associations remained significant for all outcomes except AF.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122290021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alan Kennedy, D. Finlay, R. Bond, D. Guldenring, J. Mclaughlin, Chris Crockford"
This study assessed the performance of a deep neural network (PulseAI, Belfast, United Kingdom) used in conjunction with a dry-electrode ECG sensor device (RhythmPad, D&FT, United Kingdom) to detect AF automatically. Simultaneous pairs of 12-lead ECGs and single-lead dry-electrode ECGs were collected from 622 patients. The 12-lead ECGs were manually overread and used as reference diagnoses. Twenty-two patients were confirmed with AF and had an interpretable 12-lead and single-lead dry-electrode ECG recording. The deep neural network analysed the dry-electrode ECGs, and performance was compared to the 12-lead interpretation. Overall, the deep neural network algorithm yielded a sensitivity of 96% (95% CI, 87%-100%), specificity of 99% (95% CI, 98%-100%) and positive predictive value of 81% (95% CI, 66%-96%) for detection of AF episodes. When coupled with dry-electrode ECG sensors, the PulseAI neural network allows for large-scale and low-cost screening for AF. Widespread implementation of this technology may allow for earlier detection, treatment, and management of patients with AF.
{"title":"AI-Enabled ECG Combined with Dry Electrode Sensors for Population-Based Screening of Atrial Fibrillation","authors":"Alan Kennedy, D. Finlay, R. Bond, D. Guldenring, J. Mclaughlin, Chris Crockford\"","doi":"10.22489/CinC.2022.312","DOIUrl":"https://doi.org/10.22489/CinC.2022.312","url":null,"abstract":"This study assessed the performance of a deep neural network (PulseAI, Belfast, United Kingdom) used in conjunction with a dry-electrode ECG sensor device (RhythmPad, D&FT, United Kingdom) to detect AF automatically. Simultaneous pairs of 12-lead ECGs and single-lead dry-electrode ECGs were collected from 622 patients. The 12-lead ECGs were manually overread and used as reference diagnoses. Twenty-two patients were confirmed with AF and had an interpretable 12-lead and single-lead dry-electrode ECG recording. The deep neural network analysed the dry-electrode ECGs, and performance was compared to the 12-lead interpretation. Overall, the deep neural network algorithm yielded a sensitivity of 96% (95% CI, 87%-100%), specificity of 99% (95% CI, 98%-100%) and positive predictive value of 81% (95% CI, 66%-96%) for detection of AF episodes. When coupled with dry-electrode ECG sensors, the PulseAI neural network allows for large-scale and low-cost screening for AF. Widespread implementation of this technology may allow for earlier detection, treatment, and management of patients with AF.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121240415","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Loïc Jeanningros, F. Braun, J. V. Zaen, M. L. Bloa, A. Porretta, C. Teres, C. Herrera, G. Domenichini, Patrice Carroz, D. Graf, P. Pascale, J. Vesin, J. Thiran, E. Pruvot, M. Lemay
Photoplethysmography (PPG) has recently gained increasing interest for less obtrusive long-term cardiovascular monitoring. As for cardiac arrhythmia (CA), most research and available PPG devices have focused on the detection of atrial fibrillation (AF), the most common CA. However, other less studied CAs can induce errors in standard AF detectors. To address the PPG-based detection of both AF and non-AF CAs, we investigate novel features, extracted by pulse wave analysis (PWA), that provide insight into the morphology of individual pulses. Their discriminative power was evaluated based on the RELIEFF algorithm for feature selection, and we compared performance metrics for CA classification with and without PWA features. The classification accuracy using ridge regression was increased by 0.4%, from 75.6% to 76.0%, when using PWA features on top of temporal and spectral features. Likewise, the classification of non-AF CAs was globally improved. These results show the potential of extracting measures about individual pulse morphologies to improve detection of various CAs.
{"title":"Pulse Wave Analysis of Photoplethysmography Signals to Enhance Classification of Cardiac Arrhythmias","authors":"Loïc Jeanningros, F. Braun, J. V. Zaen, M. L. Bloa, A. Porretta, C. Teres, C. Herrera, G. Domenichini, Patrice Carroz, D. Graf, P. Pascale, J. Vesin, J. Thiran, E. Pruvot, M. Lemay","doi":"10.22489/CinC.2022.023","DOIUrl":"https://doi.org/10.22489/CinC.2022.023","url":null,"abstract":"Photoplethysmography (PPG) has recently gained increasing interest for less obtrusive long-term cardiovascular monitoring. As for cardiac arrhythmia (CA), most research and available PPG devices have focused on the detection of atrial fibrillation (AF), the most common CA. However, other less studied CAs can induce errors in standard AF detectors. To address the PPG-based detection of both AF and non-AF CAs, we investigate novel features, extracted by pulse wave analysis (PWA), that provide insight into the morphology of individual pulses. Their discriminative power was evaluated based on the RELIEFF algorithm for feature selection, and we compared performance metrics for CA classification with and without PWA features. The classification accuracy using ridge regression was increased by 0.4%, from 75.6% to 76.0%, when using PWA features on top of temporal and spectral features. Likewise, the classification of non-AF CAs was globally improved. These results show the potential of extracting measures about individual pulse morphologies to improve detection of various CAs.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121363541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：