"Ricardo Maximiliano Rosales, Konstantinos A. Mountris, M. Doblaré, M. Mazo, Emilio L. Pueyo
Depolarization sequences triggering mechanical contraction of the heart are largely determined by the cardiac conduction system $(CS)$. Many biophysical models of cardiac electrophysiology still have poor representations of the $CS$. This work proposes a semiautomatic method for the generation of an anatomically-realistic porcine $CS$ that reproduces ventricular activation properties in swine computational models. Personalized swine biventricular models were built from magnetic resonance images. Electrical propagation was described by the monodomain model. The $CS$ was defined from manually-determined anatomic landmarks using geodesic paths and a fractal tree algorithm. Two $CS$ distributions were defined, one restricted to the subendocardium and another one by performing a subendo-to-intramyocardium projection based on histological porcine data. Depolarization patterns as well as left ventricular transmural and inter-ventricular delays were assessed to describe ventricular activation by the two $CS$ distributions. The electrical excitations calculated using the two $CS$ distributions were in good agreement with reported activation patterns. The pig-specific subendo-intramyocardial $CS$ led to improved reproduction of experimental activation delays in ventricular endocardium and epicardium.
{"title":"Ventricular Conduction System Modeling for Electrophysiological Simulation of the Porcine Heart","authors":"\"Ricardo Maximiliano Rosales, Konstantinos A. Mountris, M. Doblaré, M. Mazo, Emilio L. Pueyo","doi":"10.22489/CinC.2022.030","DOIUrl":"https://doi.org/10.22489/CinC.2022.030","url":null,"abstract":"Depolarization sequences triggering mechanical contraction of the heart are largely determined by the cardiac conduction system $(CS)$. Many biophysical models of cardiac electrophysiology still have poor representations of the $CS$. This work proposes a semiautomatic method for the generation of an anatomically-realistic porcine $CS$ that reproduces ventricular activation properties in swine computational models. Personalized swine biventricular models were built from magnetic resonance images. Electrical propagation was described by the monodomain model. The $CS$ was defined from manually-determined anatomic landmarks using geodesic paths and a fractal tree algorithm. Two $CS$ distributions were defined, one restricted to the subendocardium and another one by performing a subendo-to-intramyocardium projection based on histological porcine data. Depolarization patterns as well as left ventricular transmural and inter-ventricular delays were assessed to describe ventricular activation by the two $CS$ distributions. The electrical excitations calculated using the two $CS$ distributions were in good agreement with reported activation patterns. The pig-specific subendo-intramyocardial $CS$ led to improved reproduction of experimental activation delays in ventricular endocardium and epicardium.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"595 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":"127517342","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}
Mostafa M. Moussa, Yahya Alzaabi, Ahsan H. Khandoker
Obstructive Sleep Apnea Syndrome (OSAS) and Major Depressive Disorder (MDD) are both common conditions associated with poor quality of life. We seek to classify OSAS and depression in OSAS patients, as well as sleep stages using multiple machine learning techniques. We have extracted features from 5-minute intervals of electrocardiograms (ECG), breathing signals, and electroen-cephalograms (EEG) recorded from a total of 118 subjects, of which 89 are used for training and 10-fold cross-validation and 29 are used for testing or a 75/25% split. The best classification performance of OSAS was obtained with light sleep and deep sleep with ReliefF using random forest and boosted trees, respectively. It has yielded an accuracy of 100.00%, F1-Score of 100.00%, Cohen's k Coefficient of 1.00, and a Matthews correlation coefficient (MCC) of 1.00. All sleep stages with 10 principal components using random forest yielded an accuracy of 77.50%, F1-Score of 78.05%, Cohen's k of 0.571, and an MCC of 0.632 for classification of depression in OSAS patients. Sleep staging was best done using bagged trees with features selected via sequential backward feature selection, yielding an accuracy of 76.90%, F1-Score of 75.90%, Cohen's k of 0.480, and an MCC of 0.634. These results show promise in detecting OSAS and depression in OSAS patients, particularly using light and deep sleep data.
{"title":"ECG, EEG, Breathing Signals, and Machine Learning: Computer-Aided Detection of Obstructive Sleep Apnea Syndrome and Depression","authors":"Mostafa M. Moussa, Yahya Alzaabi, Ahsan H. Khandoker","doi":"10.22489/CinC.2022.082","DOIUrl":"https://doi.org/10.22489/CinC.2022.082","url":null,"abstract":"Obstructive Sleep Apnea Syndrome (OSAS) and Major Depressive Disorder (MDD) are both common conditions associated with poor quality of life. We seek to classify OSAS and depression in OSAS patients, as well as sleep stages using multiple machine learning techniques. We have extracted features from 5-minute intervals of electrocardiograms (ECG), breathing signals, and electroen-cephalograms (EEG) recorded from a total of 118 subjects, of which 89 are used for training and 10-fold cross-validation and 29 are used for testing or a 75/25% split. The best classification performance of OSAS was obtained with light sleep and deep sleep with ReliefF using random forest and boosted trees, respectively. It has yielded an accuracy of 100.00%, F1-Score of 100.00%, Cohen's k Coefficient of 1.00, and a Matthews correlation coefficient (MCC) of 1.00. All sleep stages with 10 principal components using random forest yielded an accuracy of 77.50%, F1-Score of 78.05%, Cohen's k of 0.571, and an MCC of 0.632 for classification of depression in OSAS patients. Sleep staging was best done using bagged trees with features selected via sequential backward feature selection, yielding an accuracy of 76.90%, F1-Score of 75.90%, Cohen's k of 0.480, and an MCC of 0.634. These results show promise in detecting OSAS and depression in OSAS patients, particularly using light and deep sleep data.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"65 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":"128772042","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}
Ismail Elnaggar, Jouni Pykäri, Tero Hurnanen, O. Lahdenoja, A. Airola, M. Kaisti, T. Vasankari, M. Savontaus, T. Koivisto
Differences in cardiac time intervals (CTIs) have previously been shown in different patient groups with varying levels of cardiac function. These studies relied on methods such as conventional echocardiography or tissue doppler imaging performed by a specialist to extract CTIs. The goal of this study was to evaluate the ability of using a combination of single lead ECG and 3-axis seismocardiography (SCG) from a sensor placed on a subject's sternum to automatically extract CTIs. For each subject, pre-ejection period (PEP), left ventricular ejection time ($L$ VET), total systolic time $(TST)$, and total diastolic time $(TDT)$, which were normalized by the mean heart rate representing the entire recording were extracted using a custom developed algorithm. LVET was on average 20.5 % shorter in the NKHCD group $vs$ PRE-TAVI $(p< 0.05)$) and 5.9% shorter in the $HCD$ group $vs$ PRE-TAVI $(p> 0.05)$). Comparing CTIs between the subjects who had data recorded before and after receiving a TAVI procedure, $a$ 12.6% postoperative reduction in LVET $(p < 0.05)$ was found on average as well as a 30.2% increase in $PEP/L$ VET $(p < 0.05)$. These results are in line with literature where LVET increases with age and severe aortic stenosis and decreases after TAVI procedures when echocardiography was the main methodology used to extract CTIs.
{"title":"Cardiac Time Intervals Derived from Electrocardiography and Seismocardiography in Different Patient Groups","authors":"Ismail Elnaggar, Jouni Pykäri, Tero Hurnanen, O. Lahdenoja, A. Airola, M. Kaisti, T. Vasankari, M. Savontaus, T. Koivisto","doi":"10.22489/CinC.2022.370","DOIUrl":"https://doi.org/10.22489/CinC.2022.370","url":null,"abstract":"Differences in cardiac time intervals (CTIs) have previously been shown in different patient groups with varying levels of cardiac function. These studies relied on methods such as conventional echocardiography or tissue doppler imaging performed by a specialist to extract CTIs. The goal of this study was to evaluate the ability of using a combination of single lead ECG and 3-axis seismocardiography (SCG) from a sensor placed on a subject's sternum to automatically extract CTIs. For each subject, pre-ejection period (PEP), left ventricular ejection time ($L$ VET), total systolic time $(TST)$, and total diastolic time $(TDT)$, which were normalized by the mean heart rate representing the entire recording were extracted using a custom developed algorithm. LVET was on average 20.5 % shorter in the NKHCD group $vs$ PRE-TAVI $(p< 0.05)$) and 5.9% shorter in the $HCD$ group $vs$ PRE-TAVI $(p> 0.05)$). Comparing CTIs between the subjects who had data recorded before and after receiving a TAVI procedure, $a$ 12.6% postoperative reduction in LVET $(p < 0.05)$ was found on average as well as a 30.2% increase in $PEP/L$ VET $(p < 0.05)$. These results are in line with literature where LVET increases with age and severe aortic stenosis and decreases after TAVI procedures when echocardiography was the main methodology used to extract CTIs.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"22 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":"130679575","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. Coveney, C. Corrado, C. Roney, Richard D. Wilkinson, J. Oakley, S. Niederer, R. Clayton
Atrial fibrillation is an increasingly common condition. Computational models that describe left atrial electrophysiology have the potential to be used to guide interventions such as catheter ablation. Calibration of these models to faithfully represent left atrial structure and function in a particular patient is challenging because electrophysiology observations obtained in the clinical setting are typically sparse and noisy, and can be difficult to register to a mesh obtained from imaging. Probabilistic approaches show promise as a way to obtain personalised models while taking account of noise, sparseness, and uncertainty. We have developed a workflow in which parameter fields are represented as Gaussian processes, and the posterior distribution is inferred using MCMC. Our workflow has been tested using synthetic data, generated from simulations where the spatial variation in model parameters is known, and we have shown that both features and parameters can be recovered from simulated sparse measurements.
{"title":"A Workflow for Probabilistic Calibration of Models of Left Atrial Electrophysiology","authors":"S. Coveney, C. Corrado, C. Roney, Richard D. Wilkinson, J. Oakley, S. Niederer, R. Clayton","doi":"10.22489/CinC.2022.283","DOIUrl":"https://doi.org/10.22489/CinC.2022.283","url":null,"abstract":"Atrial fibrillation is an increasingly common condition. Computational models that describe left atrial electrophysiology have the potential to be used to guide interventions such as catheter ablation. Calibration of these models to faithfully represent left atrial structure and function in a particular patient is challenging because electrophysiology observations obtained in the clinical setting are typically sparse and noisy, and can be difficult to register to a mesh obtained from imaging. Probabilistic approaches show promise as a way to obtain personalised models while taking account of noise, sparseness, and uncertainty. We have developed a workflow in which parameter fields are represented as Gaussian processes, and the posterior distribution is inferred using MCMC. Our workflow has been tested using synthetic data, generated from simulations where the spatial variation in model parameters is known, and we have shown that both features and parameters can be recovered from simulated sparse measurements.","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":"130757644","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}
Jakub Hejc, Richard Redina, Tomas Kulik, M. Pešl, Z. Stárek
Freedom from atrial fibrillation at 1 year is estimated to be between 55–80 % of patients undergoing catheter ablation. A significant number of them would require repeat procedures due to recurrent $AF$. Patients at higher risk for developing recurrent $AF$ could benefit from different ablation strategies and post-ablation rhythm control therapy. We aim to identify the exercise-based risk factors associated with the first recurrence of $AF$ between 3 and 36 months following the ablation. Patients $(n=98$, 69.4 % men) referred for catheter ablation of paroxysmal $AF$ underwent simultaneous arm ergometry, exercise echocardiography and invasive left atrial pressure measurements. After the index ablation procedure, follow-up visits were scheduled. The observed freedom from $AF$ ecurrence during the follow-up was 81 %. Multivariable-adjusted $Cox$ regression revealed the peak $VO_{2}$ as the most significant predictor of late $AF$ reccurence (hazard ratio 0.53, $p < 0.005)$. Among analyzed parameters, the lowest prediction error was achieved by including left atrial vol{###}- $ume$ index, left atrial pressure and peak $VO_{2}$ into age and sex adjusted $Cox$ model ($AIC=132.02$, C-statistics $=0.83$ ). Presence of either decreased exercise capacity or elevated left atrial pressure is able to identify patients with potentially impaired left atrial function and different clinical outcome after conventional pulmonary vein isolation.
{"title":"Exercise-based Predictors of Late Recurrence of Atrial Fibrillation After Catheter Ablation","authors":"Jakub Hejc, Richard Redina, Tomas Kulik, M. Pešl, Z. Stárek","doi":"10.22489/CinC.2022.106","DOIUrl":"https://doi.org/10.22489/CinC.2022.106","url":null,"abstract":"Freedom from atrial fibrillation at 1 year is estimated to be between 55–80 % of patients undergoing catheter ablation. A significant number of them would require repeat procedures due to recurrent <tex>$AF$</tex>. Patients at higher risk for developing recurrent <tex>$AF$</tex> could benefit from different ablation strategies and post-ablation rhythm control therapy. We aim to identify the exercise-based risk factors associated with the first recurrence of <tex>$AF$</tex> between 3 and 36 months following the ablation. Patients <tex>$(n=98$</tex>, 69.4 % men) referred for catheter ablation of paroxysmal <tex>$AF$</tex> underwent simultaneous arm ergometry, exercise echocardiography and invasive left atrial pressure measurements. After the index ablation procedure, follow-up visits were scheduled. The observed freedom from <tex>$AF$</tex> ecurrence during the follow-up was 81 %. Multivariable-adjusted <tex>$Cox$</tex> regression revealed the peak <tex>$VO_{2}$</tex> as the most significant predictor of late <tex>$AF$</tex> reccurence (hazard ratio 0.53, <tex>$p < 0.005)$</tex>. Among analyzed parameters, the lowest prediction error was achieved by including left atrial vol{###}- <tex>$ume$</tex> index, left atrial pressure and peak <tex>$VO_{2}$</tex> into age and sex adjusted <tex>$Cox$</tex> model (<tex>$AIC=132.02$</tex>, C-statistics <tex>$=0.83$</tex> ). Presence of either decreased exercise capacity or elevated left atrial pressure is able to identify patients with potentially impaired left atrial function and different clinical outcome after conventional pulmonary vein isolation.","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":"130815563","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}
O. Duport, V. Rolle, Gustavo Guerrero, A. Beuchée, Alfredo I. Hernández
In preterm infants, recurrent episodes of apnea, bradycardia and severe intermittent hypoxia are mainly related to cardiorespiratory immaturity. These episodes are associated with major risks during the first weeks of life. Cardiorespiratory data consisting of a continuous 12 hours recording of transthoracic impedance and ECG signals were acquired in 18 preterm neonates. 106 isolated apnea events (>10 sec) were manually annotated from the database, of which 19 apneas with bradycardia. A system-level physiological model of cardio-respiratory interactions in the newborn is proposed and used to reproduce simulations of mixed apneas with and without bradycardia, by modifying the functional residual capacity. A first qualitative comparison between the simulations and the clinical data shows a close match between the experimental and simulated heart rate series during apnea with bradycardia (RMSE 4.96 bpm) and without (RMSE 2.02 bpm).
{"title":"Model-Based Analysis of Apnea-Bradycardia events in Newborns","authors":"O. Duport, V. Rolle, Gustavo Guerrero, A. Beuchée, Alfredo I. Hernández","doi":"10.22489/CinC.2022.305","DOIUrl":"https://doi.org/10.22489/CinC.2022.305","url":null,"abstract":"In preterm infants, recurrent episodes of apnea, bradycardia and severe intermittent hypoxia are mainly related to cardiorespiratory immaturity. These episodes are associated with major risks during the first weeks of life. Cardiorespiratory data consisting of a continuous 12 hours recording of transthoracic impedance and ECG signals were acquired in 18 preterm neonates. 106 isolated apnea events (>10 sec) were manually annotated from the database, of which 19 apneas with bradycardia. A system-level physiological model of cardio-respiratory interactions in the newborn is proposed and used to reproduce simulations of mixed apneas with and without bradycardia, by modifying the functional residual capacity. A first qualitative comparison between the simulations and the clinical data shows a close match between the experimental and simulated heart rate series during apnea with bradycardia (RMSE 4.96 bpm) and without (RMSE 2.02 bpm).","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"18 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":"116678008","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}
Cardiovascular disease (CVD) patients were targeted from cardiology department in this study to segregate who had stenosis and also identify the principal diseased coronary artery using PPG and ECG signals. After pre-processing these signals, dicrotic notch region of PPG and S-T segment of ECG, within each cardiac cycle was extracted as templates. A new fused segment was generated from two templates by a proposed algorithm. Utilizing statistics on three templates we defined the term Coronary Health Index (CHI) to evaluate the status of coronary arteries. Setting CHI thresholding values, healthy and stenosed artery were differentiated. Using CHI values from patients with stenosis, the classification of arteries (LAD, RCA, and LCx) was performed using Graph Attentive Convolution Network. Among 408 CVD patients 256 had occlusion in either LAD or RCA or LCx. Binary classification among presence and absence of stenosis was carried out with 0.92 accuracy, 0.91 recall, 0.91 precision, 0.90 specificity, and 0.92 F-score. Identification of stenosed artery was measured with Kappa score (0.89) and Youden's J statistic value (0.84). AUC(0.93) and AP(0.92) values from ROC and PRC curves, respectively. This derived CHI could be able to study stenosis in non-invasive, easy and cost-effective manner.
{"title":"Coronary Health Index (CHI) as A Determinant for Arterial Stenosis, Derived Using PPG and ECG Signals","authors":"Poulomi Pal, M. Mahadevappa","doi":"10.22489/CinC.2022.316","DOIUrl":"https://doi.org/10.22489/CinC.2022.316","url":null,"abstract":"Cardiovascular disease (CVD) patients were targeted from cardiology department in this study to segregate who had stenosis and also identify the principal diseased coronary artery using PPG and ECG signals. After pre-processing these signals, dicrotic notch region of PPG and S-T segment of ECG, within each cardiac cycle was extracted as templates. A new fused segment was generated from two templates by a proposed algorithm. Utilizing statistics on three templates we defined the term Coronary Health Index (CHI) to evaluate the status of coronary arteries. Setting CHI thresholding values, healthy and stenosed artery were differentiated. Using CHI values from patients with stenosis, the classification of arteries (LAD, RCA, and LCx) was performed using Graph Attentive Convolution Network. Among 408 CVD patients 256 had occlusion in either LAD or RCA or LCx. Binary classification among presence and absence of stenosis was carried out with 0.92 accuracy, 0.91 recall, 0.91 precision, 0.90 specificity, and 0.92 F-score. Identification of stenosed artery was measured with Kappa score (0.89) and Youden's J statistic value (0.84). AUC(0.93) and AP(0.92) values from ROC and PRC curves, respectively. This derived CHI could be able to study stenosis in non-invasive, easy and cost-effective manner.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"194 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":"131405815","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}
M. Kloosterman, M. Boonstra, F. Asselbergs, P. Loh, T. Oostendorp, P. V. Dam
The relation between abnormal ventricular activation and corresponding ECGs still requires additional understanding. The presence of disease breaks the equivalence in equivalent dipole layer-based $ECG$ simulations. In this study, endocardial and epicardial patches were introduced to simulate abnormal wave propagation in different types of substrates. The effect of these different types of substrates on the $QRS$ complex was assessed using a boundary element method forward $heart/torso$ and a 64-lead body surface potential map (BSPM). Activation was simulated using the fastest route algorithm with six endocardial foci and $QRS$ complexes corresponding to abnormal patch activation were compared to the $QRS$ complexes of normal ventricular activation using correlation coefficient $(CC)$. Abnormal patch activation affected both $QRS$ morphology and duration. These $QRS$ changes were observed in different leads, depending on substrate location. With insights obtained in such simulations, risk-stratification and understanding of disease progression may be further enhanced.
{"title":"Modeling Structural Abnormalities in Equivalent Dipole Layer Based ECG Simulations","authors":"M. Kloosterman, M. Boonstra, F. Asselbergs, P. Loh, T. Oostendorp, P. V. Dam","doi":"10.22489/CinC.2022.160","DOIUrl":"https://doi.org/10.22489/CinC.2022.160","url":null,"abstract":"The relation between abnormal ventricular activation and corresponding ECGs still requires additional understanding. The presence of disease breaks the equivalence in equivalent dipole layer-based $ECG$ simulations. In this study, endocardial and epicardial patches were introduced to simulate abnormal wave propagation in different types of substrates. The effect of these different types of substrates on the $QRS$ complex was assessed using a boundary element method forward $heart/torso$ and a 64-lead body surface potential map (BSPM). Activation was simulated using the fastest route algorithm with six endocardial foci and $QRS$ complexes corresponding to abnormal patch activation were compared to the $QRS$ complexes of normal ventricular activation using correlation coefficient $(CC)$. Abnormal patch activation affected both $QRS$ morphology and duration. These $QRS$ changes were observed in different leads, depending on substrate location. With insights obtained in such simulations, risk-stratification and understanding of disease progression may be further enhanced.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"57 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":"128408341","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}
Camila R Restivo, Gabriel V Costa, I. Sandoval, M. Guillem, J. Salinet
Atrial fibrillation (AF) is a common supraventricular arrhythmia (SVA) in clinical practice and is characterized by uncoordinated electrical activity of the atria. This study aims to evaluate the influence on the forward solution of AF torso biomarkers under different levels of noise, 3D cardiorespiratory torso/atria morphologies, and number of atria electrodes. 2,048 atrial epicardium electrograms (AEGs) from 5 AF mathematical models were used to estimate 771 body surface potentials (BSPs). The BSPs and respective frequency/phase maps of are obtained after: (i) introduction of noise in the AEGs, (ii) 3D geometry torso/atria modification, and (iii) reduction in electrodes (from 2,048 to 256, 128, 64 e 32; interpolation methods: Linear/Laplacian). To reduce biomarkers disparity, a Butterworth bandpass filter (BPF) at different cut-off frequencies (0.5-30, 3–30 and HDF±1 Hz) is applied on the AEGs prior BSPs estimation. The above methodology is extended to two AF patients (EDGAR database). The estimation of AF BSPs, in different noise ranges, limits the effectiveness of the forward solution. Phase biomarkers are sensitive to the AEGs' pre-processing strategy. The BPF around HDF showed the best agreement between the different SNR levels. Due to the 3D morphological changes, HDF areas variability increased.
心房颤动(AF)是临床上常见的室上性心律失常(SVA)，其特点是心房电活动不协调。本研究旨在评估不同水平的噪声、三维心肺躯干/心房形态和心房电极数量对房颤躯干生物标志物正向溶液的影响。利用5个AF数学模型的2,048张心房心外膜电图(aeg)估计771个体表电位(BSPs)。bsp和各自的频率/相位图是在以下情况下获得的:(i)在aeg中引入噪声，(ii) 3D几何躯干/心房修改，以及(iii)减少电极(从2,048减少到256、128、64 e 32;插值方法:线性/拉普拉斯)。为了减少生物标志物的差异，在不同的截止频率(0.5- 30,3 -30和HDF±1hz)下应用巴特沃斯带通滤波器(BPF)对aaegs先验BSPs估计。上述方法扩展到2例房颤患者(EDGAR数据库)。在不同噪声范围下对自动对焦BSPs的估计限制了正演解的有效性。相生物标志物对aeg的预处理策略敏感。HDF附近的BPF在不同信噪比水平之间表现出最好的一致性。由于三维形态的变化，HDF面积变异性增加。
{"title":"Validation of a Customized Method for Estimating Electrical Potentials in the Torso From Atrial Signals: a Computational-Clinical Study","authors":"Camila R Restivo, Gabriel V Costa, I. Sandoval, M. Guillem, J. Salinet","doi":"10.22489/CinC.2022.369","DOIUrl":"https://doi.org/10.22489/CinC.2022.369","url":null,"abstract":"Atrial fibrillation (AF) is a common supraventricular arrhythmia (SVA) in clinical practice and is characterized by uncoordinated electrical activity of the atria. This study aims to evaluate the influence on the forward solution of AF torso biomarkers under different levels of noise, 3D cardiorespiratory torso/atria morphologies, and number of atria electrodes. 2,048 atrial epicardium electrograms (AEGs) from 5 AF mathematical models were used to estimate 771 body surface potentials (BSPs). The BSPs and respective frequency/phase maps of are obtained after: (i) introduction of noise in the AEGs, (ii) 3D geometry torso/atria modification, and (iii) reduction in electrodes (from 2,048 to 256, 128, 64 e 32; interpolation methods: Linear/Laplacian). To reduce biomarkers disparity, a Butterworth bandpass filter (BPF) at different cut-off frequencies (0.5-30, 3–30 and HDF±1 Hz) is applied on the AEGs prior BSPs estimation. The above methodology is extended to two AF patients (EDGAR database). The estimation of AF BSPs, in different noise ranges, limits the effectiveness of the forward solution. Phase biomarkers are sensitive to the AEGs' pre-processing strategy. The BPF around HDF showed the best agreement between the different SNR levels. Due to the 3D morphological changes, HDF areas variability increased.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"107 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":"134258352","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}
Heart sound recordings are a key non-invasive tool to detect both congenital and acquired heart conditions. As part of the George B. Moody PhysioNet Challenge 2022, we present an approach based on Bidirectional Long Short-Term Memory (BiLSTM) neural networks for the detection of murmurs and prediction of clinical outcome from Phonocardiograms (PCGs). We used the homomorphic, Hilbert, power spectral density, and wavelet envelopes as signal features, from which we extracted fixed-length segments of 4 seconds to train the network. Using the official challenge scoring metrics, our team SmartBeatIT achieved a murmur weighted accuracy score of 0.757 on the hidden test set (ranked 6th out of 40 teams), and an outcome cost score of 13815 (ranked 25th out of 39 teams). With 5-fold cross-validation on the training set, in the murmur detection task we obtained sensitivities of 0.827 and 0.312 for the Present and Unknown classes and a specificity of 0.801; and a sensitivity of 0.676 and a specificity of 0.544 in the outcome prediction task.
心音记录是检测先天性和后天性心脏病的一种关键的非侵入性工具。作为George B. Moody PhysioNet Challenge 2022的一部分，我们提出了一种基于双向长短期记忆(BiLSTM)神经网络的方法，用于检测杂音并预测心音图(pcg)的临床结果。我们使用同态、希尔伯特、功率谱密度和小波包络作为信号特征，从中提取固定长度的4秒片段来训练网络。使用官方挑战得分指标，我们的团队SmartBeatIT在隐藏测试集上获得了0.757的低加权准确率得分(在40支团队中排名第6)，以及13815的结果成本得分(在39支团队中排名第25)。通过对训练集进行5倍交叉验证，在杂音检测任务中，我们获得了Present和Unknown类别的灵敏度分别为0.827和0.312，特异性为0.801;结果预测任务的敏感性为0.676，特异性为0.544。
{"title":"Detection of Heart Sound Murmurs and Clinical Outcome with Bidirectional Long Short-Term Memory Networks","authors":"S. Monteiro, A. Fred, H. Silva","doi":"10.22489/CinC.2022.153","DOIUrl":"https://doi.org/10.22489/CinC.2022.153","url":null,"abstract":"Heart sound recordings are a key non-invasive tool to detect both congenital and acquired heart conditions. As part of the George B. Moody PhysioNet Challenge 2022, we present an approach based on Bidirectional Long Short-Term Memory (BiLSTM) neural networks for the detection of murmurs and prediction of clinical outcome from Phonocardiograms (PCGs). We used the homomorphic, Hilbert, power spectral density, and wavelet envelopes as signal features, from which we extracted fixed-length segments of 4 seconds to train the network. Using the official challenge scoring metrics, our team SmartBeatIT achieved a murmur weighted accuracy score of 0.757 on the hidden test set (ranked 6th out of 40 teams), and an outcome cost score of 13815 (ranked 25th out of 39 teams). With 5-fold cross-validation on the training set, in the murmur detection task we obtained sensitivities of 0.827 and 0.312 for the Present and Unknown classes and a specificity of 0.801; and a sensitivity of 0.676 and a specificity of 0.544 in the outcome prediction task.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"40 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":"130297799","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：