T. Baptiste, Angela W. C. Lee, M. Strocchi, Charles Sillett, D. Ennis, U. Haberland, R. Rajani, A. Rinaldi, S. Niederer
{"title":"The Influence of Left Atrial Wall Thickness and Curvature on Wall Strain in Patient-Specific Atrium Models","authors":"T. Baptiste, Angela W. C. Lee, M. Strocchi, Charles Sillett, D. Ennis, U. Haberland, R. Rajani, A. Rinaldi, S. Niederer","doi":"10.22489/CinC.2022.243","DOIUrl":"https://doi.org/10.22489/CinC.2022.243","url":null,"abstract":"Fibrosis is thought to be a major contributor to atrial fibrillation. Strain is a potential signal for fibrosis in the left atrium (LA). Local strain can be impacted by local anatomy. This study investigated correlation of local strain magnitude with local anatomy described by curvature and wall thickness. We created $3D$ motion models of the LA from retrospective gated computed tomography images from 8 patients. We calculated wall thickness and endocardial curvature across the LA at end-diastole $(ED)$ then calculated LA endocardial area strain throughout the cardiac cycle, using the $ED$ frame as the reference. The average Pearson's correlation of end-systolic strain with inverse wall thickness and curvature was - $0.076pm0.095$ and 0.01 $7pm0.81$ respectively. The correlations between inverse wall thickness, curvature and the first four principal components of strain showed no greater dependence of strain on wall thickness or curvature. The LA was divided into 18 regions and correlation was calculated regionally. Regionally, the range of correlation of strain at ES with thickness and curvature was $(-0.58-0.43)$ and $(-0.49-0.47)$ respectively. Neither wall thickness nor curvature appear to strongly influence strain. This is consistent with either boundary forces acting on the atria or variations in regional stiffness impacting regional differences in strain.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"23 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":"114498163","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}
{"title":"Depressed Patients Identification Using Cardiovascular Signals","authors":"Mohammad Sami Zitouni, A. Khandoker","doi":"10.22489/CinC.2022.308","DOIUrl":"https://doi.org/10.22489/CinC.2022.308","url":null,"abstract":"In this study, we present a deep learning based frame-work for the identification of Major Depressive disorder (MDD) patients from cardiovascular signals. In this work, multi-modal cardiovascular signals, including electrocar-diogram (ECG) and finger photoplethysmography (PPG), are used. The signals were collected from 60 subjects for 10 minutes, out of whom 30 were diagnosed with MDD by a psychiatric, and 30 were healthy. The signals are pre-processed and segmented into 30 seconds segments to be able to perform the identification in half a minute window, which proved to be sufficient in this work. Then, time-frequency analysis is performed on the signals for feature extraction and then a recurrent neural network architecture based on Long Short-Term Memory (LSTM) networks is utilized for the identification of the MDD patients. The results demonstrated a robust performance with an accuracy of 85.7%. This study can be considered an advancement towards the involvement of artificial intelligence tools in the assisted diagnosis and monitoring of mental diseases, and reducing their risk and impact on human daily life.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"106 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":"114502227","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}
{"title":"Equivalent Dipole Trajectories Assessed from the 12-Lead ECG Using a Tailored Human Torso Model","authors":"V. Starc","doi":"10.22489/CinC.2022.289","DOIUrl":"https://doi.org/10.22489/CinC.2022.289","url":null,"abstract":"We extended our application for the assessment of moving equivalent dipoles (ED) from the surface ECG by incorporating a BEM method to calculate potentials on the surface of a tailored human torso model and explored whether it could provide reliable ED trajectories from the 12-lead ECGs compared to those from the body surface potential map (BSPM) data. We used 17 recordings of the Dalhousie BSPM data (EDGAR) with ECG signals arising from different pacing sites in the same patient and tested for the congruency of the derived ED trajectory patterns of the 12-lead and BSPM data sets. We found that the ED trajectories from these two sets are mutually shifted or rotated by less than the median offset of 1.5 cm and deviation angle of 15°. We believe that assessing the ED trajectory with this accuracy may help improve the detection of depolarization abnormalities in the clinical setting.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"122 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":"114708864","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. Ogbomo-Harmitt, A. Qureshi, A. King, O. Aslanidi
{"title":"Impact of Fibrosis Border Zone Characterisation on Fibrosis-Substrate Isolation Ablation Outcome for Atrial Fibrillation","authors":"S. Ogbomo-Harmitt, A. Qureshi, A. King, O. Aslanidi","doi":"10.22489/CinC.2022.218","DOIUrl":"https://doi.org/10.22489/CinC.2022.218","url":null,"abstract":"Atrial fibrillation (AF) is globally the most common type of cardiac arrhythmia and is a precursor for serious conditions such as stroke. The success rate of AF treatments, such as catheter ablation (including the current gold standard, pulmonary vein isolation), is suboptimal, warranting better strategies. Fibrosis-substrate isolation ablation (FISA) is a promising new ablation strategy currently showing success in clinical trials. However, to perform FISA, the left atrial (LA) fibrosis border zone (FBZ) needs to be characterised. This study investigates the impact of FBZ characterisation on FISA outcomes for AF simulated using 10 patient-specific 3D LA models. Simulations show that (i) including a large amount of FBZ tissue within FISA lesions can increase the success of AF termination, and (ii) FISA is more effective for patients with Utah fibrosis stages III and IV. These results can help clinicians to improve the stratification of AF patients and the implementation of the FISA strategy.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"35 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":"115003288","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}
"Carlos Fambuena Santos, I. Hernández-Romero, C. Herrero Martín, Jana Reventós Presmanes, Eric Invers Rubio, L. Mont, Andreu M. Climent, Maria de la Salud Guillem Sánchez"
{"title":"Probabilistic Dominant Frequency Estimation in AF From ECGI","authors":"\"Carlos Fambuena Santos, I. Hernández-Romero, C. Herrero Martín, Jana Reventós Presmanes, Eric Invers Rubio, L. Mont, Andreu M. Climent, Maria de la Salud Guillem Sánchez\"","doi":"10.22489/CinC.2022.362","DOIUrl":"https://doi.org/10.22489/CinC.2022.362","url":null,"abstract":"Non-invasive estimation of high frequency activation regions in atrial fibrillation (AF) may have an important role in patient stratification and ablation guidance. This work presents a methodology to robustly estimate DF maps in ECGI, where the uncertainty associated to the estimates is modelled making use of a set of ECGI solutions from a range of different lambda parameters (DF-LR) in Tikhonov O-order regularization. The proposed DF-LR method was compared to the $DFs$ obtained from the standard L-curve (DF-LC) optimization. Specifically, the highest dominant frequency (HDF) found with both methods was tested in 2 AF simulations. In addition, the reproducibility of the DF maps was studied in a clinical case using ECGI signals from a persistent AF patient. DF-LR method overcame the DF-LC in terms of HDF sensitivity. Furthermore, the mean absolute difference between consecutive DF maps was lower in DF-LR method $(0.64pm 0.34Hzquad vs quad 1.38pm 0.11 quad Hz)$ showing higher reproducibility.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"83 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":"116958683","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}
C. B. Espinosa, Jorge Sánchez, O. Doessel, A. Loewe
{"title":"Diffusion Reaction Eikonal Alternant Model: Towards Fast Simulations of Complex Cardiac Arrhythmias","authors":"C. B. Espinosa, Jorge Sánchez, O. Doessel, A. Loewe","doi":"10.22489/CinC.2022.054","DOIUrl":"https://doi.org/10.22489/CinC.2022.054","url":null,"abstract":"Reaction-diffusion (RD) computer models are suitable to investigate the mechanisms of cardiac arrthymias but not directly applicable in clinical settings due to their computational cost. On the other hand, alternative faster eikonal models are incapable of reproducing reentrant activation when solved by iterative methods. The diffusion reaction eikonal alternant model (DREAM) is a new method in which eikonal and RD models are alternated to allow for reactivation. To solve the eikonal equation, the fast iterative method was modified and embedded into DREAM. Obtained activation times control transmembrane voltage courses in the RD model computing, while repolarization times are provided back to the eikonal model. For a planar wave-front in the center of a 2D patch, DREAM action potentials (APs) have a small overshoot in the upstroke compared to pure RD simulations (monodomain) but similar AP duration. DREAM conduction velocity does not increase near boundaries or stimulated areas as it occurs in RD. Anatomical reentry was reproduced with the S1-S2 protocol. This is the first time that an iterative method is used to solve the eikonal model in a version that admits reactivation. This method can facilitate uptake of computer models in clinical settings. Further improvements will allow to accurately represent even more complex patterns of arrhythmia.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"15 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":"124073595","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}
"Miriam Gutiérrez Fernández-Calvillo, Miguel Ángel Cámara-Vázquez, I. Hernández-Romero, Maria de la Salud Guillem Sánchez", Andreu M. Climent, Ó. Barquero-Pérez
{"title":"Non-Invasive Atrial Fibrillation Driver Localization Using Recurrent Neural Networks and Body Surface Potentials","authors":"\"Miriam Gutiérrez Fernández-Calvillo, Miguel Ángel Cámara-Vázquez, I. Hernández-Romero, Maria de la Salud Guillem Sánchez\", Andreu M. Climent, Ó. Barquero-Pérez","doi":"10.22489/CinC.2022.163","DOIUrl":"https://doi.org/10.22489/CinC.2022.163","url":null,"abstract":"Ablation is the main therapy to control Atrial Fibrillation (AF). However, the underlying mechanism for AF initiation and maintenance remains mostly unknown and represent a major challenge. ECG Imaging (ECGI) has been presented to address this issue, but it is an ill-posed problem and presents several limitations. Many Deep Learning methods have been proposed for AF characterization, but few provide a solution involving the location of the AF driver. In this work, we propose finding the location of AF drivers using Body Surface Potentials (BSPs) and CNN-LSTM with an attention layer networks as a supervised classification problem. The AF driver was correctly located the 94.42% of the time with an average Cohen's Kappa of 0.87. Hence, the proposed model could provide an effective solution for identifying AF driver location for ablation procedures as a non-invasive approach.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"45 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":"125737201","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}
Zhongrui Bai, Baiju Yan, Xiang-Xiang Chen, Yirong Wu, Peng Wang
{"title":"Murmur Detection and Clinical Outcome Classification Using a VGG-like Network and Combined Time-Frequency Representations of PCG Signals","authors":"Zhongrui Bai, Baiju Yan, Xiang-Xiang Chen, Yirong Wu, Peng Wang","doi":"10.22489/CinC.2022.318","DOIUrl":"https://doi.org/10.22489/CinC.2022.318","url":null,"abstract":"For the George B. Moody PhysioNet Challenge 2022, our team, PhysioDreamfly, developed a deep neural network approach for detecting murmurs and identifying abnormal clinical outcomes from phonocardiograms (PCGs). In our approach, a VGG-like CNN model is used as the classifier. Images consisting of Log-Mel spectrograms and wavelet scalogram that transformed from unsegmented PCGs are used as model inputs. We combined the murmur and outcome labels to address the two tasks as one multi-label task, and introduced a weighted focal loss function to optimize the model. Our murmur detection classifier received a weighted accuracy score of 0.752 (ranked 11th out of 40 teams) and Challenge cost score of 12831(ranked 18th out of 39 teams) on the hidden test set.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"37 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":"125028430","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}
Aikaterini Vraka, V. Bertomeu-González, L. Sörnmo, Roberto Zangróniz, R. Alcaraz, J. J. Rieta
{"title":"The P-Wave Time-Domain Significant Features to Evaluate Substrate Modification After Catheter Ablation of Paroxysmal Atrial Fibrillation","authors":"Aikaterini Vraka, V. Bertomeu-González, L. Sörnmo, Roberto Zangróniz, R. Alcaraz, J. J. Rieta","doi":"10.22489/CinC.2022.011","DOIUrl":"https://doi.org/10.22489/CinC.2022.011","url":null,"abstract":"The outcome of catheter ablation (CA) of atrial fibrillation (AF) is vastly analyzed by the entire P-wave duration (PWD). However, the first and second P-wave parts, corresponding to right (RA) and left atrial (LA) wavefront propagation, may be unequally modified. Five-minute lead II recordings before and after the first-ever CA of 40 parox-ysmal AF patients were analyzed and P-wave features were calculated: $PWD_{on-off}$ of the entire P-wave and each P-wave part $(RA:PWD_{on-peak}, LA:PWD_{peak-off})$ and the time from P-wave onset or offset to the R-peak $(PWD_{on-R}$ and $PWD_{off-R}$, respectively). Heart-rate (HR) adjustment $(HRA)$ mitigated the HR fluctuations. Prelpost-CA comparison was performed with Mann-Whitney U-test and median values were calculated. Pearson's correlation was calculated between PWD and the remaining features. The effect of CA with $(Delta$: −17.96%) or without HRA $(Delta$: −9.84%) was significant at the entire $PWD_{on-off}$ and at the $PWD_{peak-off}(HRA:Delta$: −27.77%, no HRA: $Delta$: −22.03%). $PWD_{on-off}$ showed a stronger correlation with RA than $LA(rho_{max}=0.805 vs rho_{max}=0.541)$. P-wave features corresponding to RA are more strongly related to the entire P-wave. Nevertheless, only the P-wave part associated with LA is significantly affected by CA. That being so, studies are encouraged to incorporate part-time P-wave analysis.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"30 5 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":"129503099","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, J. Ramírez, A. Tinker, P. Munroe, P. Lambiase
{"title":"Uncoupling Between Heart Rate Variability and Heart Rate During Exercise and Recovery as a Predictor of Cardiovascular Events","authors":"\"Michele Orini, S. van Duijvenboden, J. Ramírez, A. Tinker, P. Munroe, P. Lambiase","doi":"10.22489/CinC.2022.229","DOIUrl":"https://doi.org/10.22489/CinC.2022.229","url":null,"abstract":"Heart rate (HR) variability (HRV) is a non-invasive cardiac autonomic marker, which, in normal conditions, is inversely associated with the underlying HR. This study investigates the hypothesis that uncoupling between HRV and HR during exercise and recovery may indicate increased cardiovascular risk. UK Biobank participants without underlying cardiovascular disease (n =48,671, 46.3% male 56.3±8.2 years old) underwent an ECG exercise stress test. Uncoupling between HR and HRV was measured as v = 1 - rHR,HRV, where r indicates the Spearman's correlation coefficients between the HR profile and the instantaneous HRV power. Cox regressions were used to assess the association between the uncoupling index, v, and major adverse cardiovascular events (MACE). Models were adjusted for age, sex, body mass index, blood pressure, resting HR, HR increase and decrease during exercise and recovery, respectively. During a median follow-up of 10 years, incidence of MACE was 2.9%. In the adjusted model, 1 standard deviation increase in log-transformed v was associated with MACE, with hazard ratio (95% confidence interval) = 1.09 (1.03, 1.15), p=0.004. In conclusion, in middle-aged man and women without underlying cardiovascular disease, the uncoupling between HR and HRV during exercise and recovery was associated with MACE.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"31 3 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":"129954621","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}