Atrial fibrillation ablation outcome prediction with a machine learning fusion framework incorporating cardiac computed tomography.

BACKGROUND: Structural changes in the left atrium (LA) modestly predict outcomes in patients undergoing catheter ablation for atrial fibrillation (AF). Machine learning (ML) is a promising approach to personalize AF management strategies and improve predictive risk models after catheter ablation by integrating atrial geometry from cardiac computed tomography (CT) scans and patient-specific clinical data. We hypothesized that ML approaches based on a patient's specific data can identify responders to AF ablation. METHODS: Consecutive patients undergoing AF ablation, who had preprocedural CT scans, demographics, and 1-year follow-up data, were included in the study for a retrospective analysis. The inputs of models were CT-derived morphological features from left atrial segmentation (including the shape, volume of the LA, LA appendage, and pulmonary vein ostia) along with deep features learned directly from raw CT images, and clinical data. These were merged intelligently in a framework to learn their individual importance and produce the optimal classification. RESULTS: Three hundred twenty-one patients (64.2 ± 10.6 years, 69% male, 40% paroxysmal AF) were analyzed. Post 10-fold nested cross-validation, the model trained to intelligently merge and learn appropriate weights for clinical, morphological, and imaging data (AUC 0.821) outperformed those trained solely on clinical data (AUC 0.626), morphological (AUC 0.659), or imaging data (AUC 0.764). CONCLUSION: Our ML approach provides an end-to-end automated technique to predict AF ablation outcomes using deep learning from CT images, derived structural properties of LA, augmented by incorporation of clinical data in a merged ML framework. This can help develop personalized strategies for patient selection in invasive management of AF.

Feasibility of intraprocedural integration of cardiac CT to guide left ventricular lead implantation for CRT upgrades.

BACKGROUND: Optimal positioning of the left ventricular (LV) lead is an important determinant of cardiac resynchronization therapy (CRT) response. OBJECTIVE: Evaluate the feasibility of intraprocedural integration of cardiac computed tomography (CT) to guide LV lead implantation for CRT upgrades. METHODS: Patients undergoing LV lead upgrade underwent ECG-gated cardiac CT dyssynchrony and LV scar assessment. Target American Heart Association segment selection was determined using latest non-scarred mechanically activating segments overlaid onto real-time fluoroscopy with image co-registration to guide optimal LV lead implantation. Hemodynamic validation was performed using a pressure wire in the LV cavity (dP/dtmax) ). RESULTS: 18 patients (male 94%, 55.6% ischemic cardiomyopathy) with RV pacing burden 60.0 ± 43.7% and mean QRS duration 154 ± 30 ms underwent cardiac CT. 10/10 ischemic patients had CT evidence of scar and these segments were excluded as targets. Seventeen out of 18 (94%) patients underwent successful LV lead implantation with delivery to the CT target segment in 15 out of 18 (83%) of patients. Acute hemodynamic response (dP/dtmax ≥ 10%) was superior with LV stimulation in CT target versus nontarget segments (83.3% vs. 25.0%; p = .012). Reverse remodeling at 6 months (LV end-systolic volume improvement ≥15%) occurred in 60% of subjects (4/8 [50.0%] ischemic cardiomyopathy vs. 5/7 [71.4%] nonischemic cardiomyopathy, p = .608). CONCLUSION: Intraprocedural integration of cardiac CT to guide optimal LV lead placement is feasible with superior hemodynamics when pacing in CT target segments and favorable volumetric response rates, despite a high proportion of patients with ischemic cardiomyopathy. Multicentre, randomized controlled studies are needed to evaluate whether intraprocedural integration of cardiac CT is superior to standard care.

Tracking the motion of intracardiac structures aids the development of future leadless pacing systems.

BACKGROUND: Leadless pacemakers preclude the need for permanent leads to pace endocardium. However, it is yet to be determined whether a leadless pacemaker of a similar design to those manufactured for the right ventricle (RV) fits within the left ventricle (LV), without interfering with intracardiac structures. METHODS: Cardiac computed tomography scans were obtained from 30 patients indicated for cardiac resynchronisation therapy upgrade. The mitral valve annulus, chordae tendineae, papillary muscles and LV endocardial wall were marked in the end-diastolic frame. Intracardiac structures motions were tracked through the cardiac cycle. Two pacemaker designs similar to commercially manufactured leadless systems (Abbott's Nanostim LCP and Medtronic's Micra TPS) as well as theoretical designs with calculated optimal dimensions were evaluated. Pacemakers were virtually placed across the LV endocardial surface and collisions between them and intracardiac structures were detected throughout the cycle. RESULTS: Probability maps of LV intracardiac structures collisions on a 16-segment AHA model indicated possible placement for the Nanostim LCP, Micra TPS, and theoretical designs. Thresholding these maps at a 20% chance of collision revealed only about 36% of the endocardial surface remained collision-free with the deployment of Micra TPS design. The same threshold left no collision-free surface in the case of the Nanostim LCP. To reach at least half of the LV endocardium, the volume of Micra TPS, which is the smaller design, needed to be decreased by 41%. CONCLUSION: Due to the presence of intracardiac structures, placement of leadless pacemakers with dimensions similar to commercially manufactured RV systems would be limited to apical regions.

Improved co-registration of ex-vivo and in-vivo cardiovascular magnetic resonance images using heart-specific flexible 3D printed acrylic scaffold combined with non-rigid registration.

BACKGROUND: Ex-vivo cardiovascular magnetic resonance (CMR) imaging has played an important role in the validation of in-vivo CMR characterization of pathological processes. However, comparison between in-vivo and ex-vivo imaging remains challenging due to shape changes occurring between the two states, which may be non-uniform across the diseased heart. A novel two-step process to facilitate registration between ex-vivo and in-vivo CMR was developed and evaluated in a porcine model of chronic myocardial infarction (MI). METHODS: Seven weeks after ischemia-reperfusion MI, 12 swine underwent in-vivo CMR imaging with late gadolinium enhancement followed by ex-vivo CMR 1 week later. Five animals comprised the control group, in which ex-vivo imaging was undertaken without any support in the LV cavity, 7 animals comprised the experimental group, in which a two-step registration optimization process was undertaken. The first step involved a heart specific flexible 3D printed scaffold generated from in-vivo CMR, which was used to maintain left ventricular (LV) shape during ex-vivo imaging. In the second step, a non-rigid co-registration algorithm was applied to align in-vivo and ex-vivo data. Tissue dimension changes between in-vivo and ex-vivo imaging were compared between the experimental and control group. In the experimental group, tissue compartment volumes and thickness were compared between in-vivo and ex-vivo data before and after non-rigid registration. The effectiveness of the alignment was assessed quantitatively using the DICE similarity coefficient. RESULTS: LV cavity volume changed more in the control group (ratio of cavity volume between ex-vivo and in-vivo imaging in control and experimental group 0.14 vs 0.56, p < 0.0001) and there was a significantly greater change in the short axis dimensions in the control group (ratio of short axis dimensions in control and experimental group 0.38 vs 0.79, p < 0.001). In the experimental group, prior to non-rigid co-registration the LV cavity contracted isotropically in the ex-vivo condition by less than 20% in each dimension. There was a significant proportional change in tissue thickness in the healthy myocardium (change = 29 ± 21%), but not in dense scar (change = - 2 ± 2%, p = 0.034). Following the non-rigid co-registration step of the process, the DICE similarity coefficients for the myocardium, LV cavity and scar were 0.93 (±0.02), 0.89 (±0.01) and 0.77 (±0.07) respectively and the myocardial tissue and LV cavity volumes had a ratio of 1.03 and 1.00 respectively. CONCLUSIONS: The pattern of the morphological changes seen between the in-vivo and the ex-vivo LV differs between scar and healthy myocardium. A 3D printed flexible scaffold based on the in-vivo shape of the LV cavity is an effective strategy to minimize morphological changes in the ex-vivo LV. The subsequent non-rigid registration step further improved the co-registration and local comparison between in-vivo and ex-vivo data.

Pulmonary vein encirclement using an Ablation Index-guided point-by-point workflow: cardiovascular magnetic resonance assessment of left atrial scar formation.

AIMS: A point-by-point workflow for pulmonary vein isolation (PVI) targeting pre-defined Ablation Index values (a composite of contact force, time, and power) and minimizing interlesion distance may optimize the creation of contiguous ablation lesions whilst minimizing scar formation. We aimed to compare ablation scar formation in patients undergoing PVI using this workflow to patients undergoing a continuous catheter drag workflow. METHODS AND RESULTS: Post-ablation cardiovascular magnetic resonance imaging was performed in patients undergoing 1st-time PVI using a parameter-guided point-by-point workflow (n = 26). Total left atrial scar burden and the width and continuity of the pulmonary vein encirclement were determined on analysis of atrial late gadolinium enhancement sequences. Comparison was made with a cohort of patients (n = 20) undergoing PVI using continuous drag lesions. Mean post-ablation scar burden and scar width were significantly lower in the point-by-point group than in the control group (6.6 ± 6.8% vs. 9.6 ± 5.0%, P = 0.03 and 7.9 ± 3.6 mm vs. 10.7 ± 2.3 mm, P = 0.003). More complete bilateral pulmonary vein encirclements were seen in the point-by-point group (P = 0.038). All patients achieved acute PVI. CONCLUSION: Pulmonary vein isolation using a point-by-point workflow is feasible and results in a lower scar burden and scar width with more complete pulmonary vein encirclements than a conventional drag lesion approach.

Optimization of late gadolinium enhancement cardiovascular magnetic resonance imaging of post-ablation atrial scar: a cross-over study.

BACKGROUND: Cardiovascular magnetic resonance (CMR) imaging may be used to visualize post-ablation atrial scar (PAAS), and three-dimensional late gadolinium enhancement (3D LGE) is the most widely employed technique for imaging of chronic scar. Detection of PAAS provides a unique non-invasive insight into the effects of the ablation and may help guide further ablation procedures. However, there is evidence that PAAS is often not detected by CMR, implying a significant sensitivity problem, and imaging parameters vary between leading centres. Therefore, there is a need to establish the optimal imaging parameters to detect PAAS. METHODS: Forty subjects undergoing their first pulmonary vein isolation procedure for AF had detailed CMR assessment of atrial scar: one scan pre-ablation, and two scans post-ablation at 3 months (separated by 48 h). Each scan session included ECG- and respiratory-navigated 3D LGE acquisition at 10, 20 and 30 min post injection of a gadolinium-based contrast agent (GBCA). The first post-procedural scan was performed on a 1.5 T scanner with standard acquisition parameters, including double dose (0.2 mmol/kg) Gadovist and 4 mm slice thickness. Ten patients subsequently underwent identical scan as controls, and the other 30 underwent imaging with a reduced, single, dose GBCA (n = 10), half slice thickness (n = 10) or on a 3 T scanner (n = 10). Apparent signal-to-noise (aSNR), contrast-to-noise (aCNR) and imaging quality (Likert Scale, 3 independent observers) were assessed. PAAS location and area (%PAAS scar) were assessed following manual segmentation. Atrial shells with standardised %PAAS at each timepoint were then compared to ablation lesion locations to assess quality of scar delineation. RESULTS: A total of 271 3D acquisitions (out of maximum 280, 96.7%) were acquired. Likert scale of imaging quality had high interobserver and intraobserver intraclass correlation coefficients (0.89 and 0.96 respectively), and showed lower overall imaging quality on 3 T and at half-slice thickness. aCNR, and quality of scar delineation increased significantly with time. aCNR was higher with reduced, single, dose of GBCA (p = 0.005). CONCLUSION: 3D LGE CMR atrial scar imaging, as assessed qualitatively and quantitatively, improves with time from GBCA administration, with some indices continuing to improve from 20 to 30 min. Imaging should be performed at least 20 min post-GBCA injection, and a single dose of contrast should be considered. TRIAL REGISTRATION: Trial registry- United Kingdom National Research Ethics Service 08/H0802/68 - 30th September 2008.

Patient-specific simulations predict efficacy of ablation of interatrial connections for treatment of persistent atrial fibrillation.

AIMS: Treatments for persistent atrial fibrillation (AF) offer limited efficacy. One potential strategy aims to return the right atrium (RA) to sinus rhythm (SR) by ablating interatrial connections (IAC) to isolate the atria, but there is limited clinical data to evaluate this ablation approach. We aimed to use simulation to evaluate and predict patient-specific suitability for ablation of IAC to treat AF. METHODS AND RESULTS: Persistent AF was simulated in 12 patient-specific geometries, incorporating electrophysiological heterogeneity and fibres, with IAC at Bachmann's bundle, the coronary sinus, and fossa ovalis. Simulations were performed to test the effect of left atrial (LA)-to-RA frequency gradient and fibrotic remodelling on IAC ablation efficacy. During AF, we simulated ablation of one, two, or all three IAC, with or without pulmonary vein isolation and determined if this altered or terminated the arrhythmia. For models without structural remodelling, ablating all IAC terminated RA arrhythmia in 83% of cases. Models with the LA-to-RA frequency gradient removed had an increased success rate (100% success). Ablation of IACs is less effective in cases with fibrotic remodelling (interstitial fibrosis 50% success rate; combination remodelling 67%). Mean number of phase singularities in the RA was higher pre-ablation for IAC failure (success 0.6 ± 0.8 vs. failure 3.2 ± 2.5, P < 0.001). CONCLUSION: This simulation study predicts that IAC ablation is effective in returning the RA to SR for many cases. Patient-specific modelling approaches have the potential to stratify patients prior to ablation by predicting if drivers are located in the LA or RA. We present a platform for predicting efficacy and informing patient selection for speculative treatments.

A three-dimensional left atrial motion estimation from retrospective gated computed tomography: application in heart failure patients with atrial fibrillation.

Background: A reduced left atrial (LA) strain correlates with the presence of atrial fibrillation (AF). Conventional atrial strain analysis uses two-dimensional (2D) imaging, which is, however, limited by atrial foreshortening and an underestimation of through-plane motion. Retrospective gated computed tomography (RGCT) produces high-fidelity three-dimensional (3D) images of the cardiac anatomy throughout the cardiac cycle that can be used for estimating 3D mechanics. Its feasibility for LA strain measurement, however, is understudied. Aim: The aim of this study is to develop and apply a novel workflow to estimate 3D LA motion and calculate the strain from RGCT imaging. The utility of global and regional strains to separate heart failure in patients with reduced ejection fraction (HFrEF) with and without AF is investigated. Methods: A cohort of 30 HFrEF patients with (n = 9) and without (n = 21) AF underwent RGCT prior to cardiac resynchronisation therapy. The temporal sparse free form deformation image registration method was optimised for LA feature tracking in RGCT images and used to estimate 3D LA endocardial motion. The area and fibre reservoir strains were calculated over the LA body. Universal atrial coordinates and a human atrial fibre atlas enabled the regional strain calculation and the fibre strain calculation along the local myofibre orientation, respectively. Results: It was found that global reservoir strains were significantly reduced in the HFrEF + AF group patients compared with the HFrEF-only group patients (area strain: 11.2 ± 4.8% vs. 25.3 ± 12.6%, P = 0.001; fibre strain: 4.5 ± 2.0% vs. 15.2 ± 8.8%, P = 0.001), with HFrEF + AF patients having a greater regional reservoir strain dyssynchrony. All regional reservoir strains were reduced in the HFrEF + AF patient group, in whom the inferior wall strains exhibited the most significant differences. The global reservoir fibre strain and LA volume + posterior wall reservoir fibre strain exceeded LA volume alone and 2D global longitudinal strain (GLS) for AF classification (area-under-the-curve: global reservoir fibre strain: 0.94 ± 0.02, LA volume + posterior wall reservoir fibre strain: 0.95 ± 0.02, LA volume: 0.89 ± 0.03, 2D GLS: 0.90 ± 0.03). Conclusion: RGCT enables 3D LA motion estimation and strain calculation that outperforms 2D strain metrics and LA enlargement for AF classification. Differences in regional LA strain could reflect regional myocardial properties such as atrial fibrosis burden.

Quantifying the impact of shape uncertainty on predicted arrhythmias.

BACKGROUND: Personalised computer models are increasingly used to diagnose cardiac arrhythmias and tailor treatment. Patient-specific models of the left atrium are often derived from pre-procedural imaging of anatomy and fibrosis. These images contain noise that can affect simulation predictions. There are few computationally tractable methods for propagating uncertainties from images to clinical predictions. METHOD: We describe the left atrium anatomy using our Bayesian shape model that captures anatomical uncertainty in medical images and has been validated on 63 independent clinical images. This algorithm describes the left atrium anatomy using Nmodes=15 principal components, capturing 95% of the shape variance and calculated from 70 clinical cardiac magnetic resonance (CMR) images. Latent variables encode shape uncertainty: we evaluate their posterior distribution for each new anatomy. We assume a normally distributed prior. We use the unscented transform to sample from the posterior shape distribution. For each sample, we assign the local material properties of the tissue using the projection of late gadolinium enhancement CMR (LGE-CMR) onto the anatomy to estimate local fibrosis. To test which activation patterns an atrium can sustain, we perform an arrhythmia simulation for each sample. We consider 34 possible outcomes (31 macro-re-entries, functional re-entry, atrial fibrillation, and non-sustained arrhythmia). For each sample, we determine the outcome by comparing pre- and post-ablation activation patterns following a cross-field stimulus. RESULTS: We create patient-specific atrial electrophysiology models of ten patients. We validate the mean and standard deviation maps from the unscented transform with the same statistics obtained with 12,000 Monte Carlo (ground truth) samples. We found discrepancies <3% and <2% for the mean and standard deviation for fibrosis burden and activation time, respectively. For each patient case, we then compare the predicted outcome from a model built on the clinical data (deterministic approach) with the probability distribution obtained from the simulated samples. We found that the deterministic approach did not predict the most likely outcome in 80% of the cases. Finally, we estimate the influence of each source of uncertainty independently. Fixing the anatomy to the posterior mean and maintaining uncertainty in fibrosis reduced the prediction of self-terminating arrhythmias from ≃14% to ≃7%. Keeping the fibrosis fixed to the sample mean while retaining uncertainty in shape decreased the prediction of substrate-driven arrhythmias from ≃33% to ≃18% and increased the prediction of macro-re-entries from ≃54% to ≃68%. CONCLUSIONS: We presented a novel method for propagating shape uncertainty in atrial models through to uncertainty in numerical simulations. The algorithm takes advantage of the unscented transform to compute the output distribution of the outcomes. We validated the unscented transform as a viable sampling strategy to deal with anatomy uncertainty. We then showed that the prediction computed with a deterministic model does not always coincide with the most likely outcome. Finally, we found that shape uncertainty affects the predictions of macro-re-entries, while fibrosis uncertainty affects the predictions of functional re-entries.

Right atrial function and fibrosis in relation to successful atrial fibrillation ablation.

AIMS: Bi-atrial remodelling in patients with atrial fibrillation (AF) is rarely assessed and data on the presence of right atrial (RA) fibrosis, the relationship between RA and left atrial (LA) fibrosis, and possible association of RA remodelling with AF recurrence after ablation in patients with AF is limited. METHODS AND RESULTS: A total of 110 patients with AF undergoing initial pulmonary vein isolation (PVI) were included in the present study. All patients were in sinus rhythm during cardiac magnetic resonance (CMR) imaging performed prior to ablation. LA and RA volumes and function (volumetric and feature tracking strain) were derived from cine CMR images. The extent of LA and RA fibrosis was assessed from 3D late gadolinium enhancement images. AF recurrence was followed up for 12 months after PVI using either 12-lead electrocardiograms or Holter monitoring. Arrhythmia recurrence was observed in 39 patients (36%) after the 90-day blanking period, occurring at a median of 181 (interquartile range: 122-286) days. RA remodelling parameters were not significantly different between patients with and without AF recurrence after ablation, whereas LA remodelling parameters were different (volume, emptying fraction, and strain indices). LA fibrosis had a strong correlation with RA fibrosis (r = 0.88, P < 0.001). Both LA and RA fibrosis were not different between patients with and without AF recurrence. CONCLUSIONS: This study shows that RA remodelling parameters were not predictive of AF recurrence after AF ablation. Bi-atrial fibrotic remodelling is present in patients with AF and moreover, the amount of LA fibrosis had a strong correlation with the amount of RA fibrosis.

Evaluation of an open-source pipeline to create patient-specific left atrial models: A reproducibility study.

This work presents an open-source software pipeline to create patient-specific left atrial models with fibre orientations and a fibrDEFAULTosis map, suitable for electrophysiology simulations, and quantifies the intra and inter observer reproducibility of the model creation. The semi-automatic pipeline takes as input a contrast enhanced magnetic resonance angiogram, and a late gadolinium enhanced (LGE) contrast magnetic resonance (CMR). Five operators were allocated 20 cases each from a set of 50 CMR datasets to create a total of 100 models to evaluate inter and intra-operator variability. Each output model consisted of: (1) a labelled surface mesh open at the pulmonary veins and mitral valve, (2) fibre orientations mapped from a diffusion tensor MRI (DTMRI) human atlas, (3) fibrosis map extracted from the LGE-CMR scan, and (4) simulation of local activation time (LAT) and phase singularity (PS) mapping. Reproducibility in our pipeline was evaluated by comparing agreement in shape of the output meshes, fibrosis distribution in the left atrial body, and fibre orientations. Reproducibility in simulations outputs was evaluated in the LAT maps by comparing the total activation times, and the mean conduction velocity (CV). PS maps were compared with the structural similarity index measure (SSIM). The users processed in total 60 cases for inter and 40 cases for intra-operator variability. Our workflow allows a single model to be created in 16.72 ± 12.25 min. Similarity was measured with shape, percentage of fibres oriented in the same direction, and intra-class correlation coefficient (ICC) for the fibrosis calculation. Shape differed noticeably only with users' selection of the mitral valve and the length of the pulmonary veins from the ostia to the distal end; fibrosis agreement was high, with ICC of 0.909 (inter) and 0.999 (intra); fibre orientation agreement was high with 60.63% (inter) and 71.77% (intra). The LAT showed good agreement, where the median ± IQR of the absolute difference of the total activation times was 2.02 ± 2.45 ms for inter, and 1.37 ± 2.45 ms for intra. Also, the average ± sd of the mean CV difference was -0.00404 ± 0.0155 m/s for inter, and 0.0021 ± 0.0115 m/s for intra. Finally, the PS maps showed a moderately good agreement in SSIM for inter and intra, where the mean ± sd SSIM for inter and intra were 0.648 ± 0.21 and 0.608 ± 0.15, respectively. Although we found notable differences in the models, as a consequence of user input, our tests show that the uncertainty caused by both inter and intra-operator variability is comparable with uncertainty due to estimated fibres, and image resolution accuracy of segmentation tools.

Global Sensitivity Analysis of Four Chamber Heart Hemodynamics Using Surrogate Models.

Computational Fluid Dynamics (CFD) is used to assist in designing artificial valves and planning procedures, focusing on local flow features. However, assessing the impact on overall cardiovascular function or predicting longer-term outcomes may requires more comprehensive whole heart CFD models. Fitting such models to patient data requires numerous computationally expensive simulations, and depends on specific clinical measurements to constrain model parameters, hampering clinical adoption. Surrogate models can help to accelerate the fitting process while accounting for the added uncertainty. We create a validated patient-specific four-chamber heart CFD model based on the Navier-Stokes-Brinkman (NSB) equations and test Gaussian Process Emulators (GPEs) as a surrogate model for performing a variance-based global sensitivity analysis (GSA). GSA identified preload as the dominant driver of flow in both the right and left side of the heart, respectively. Left-right differences were seen in terms of vascular outflow resistances, with pulmonary artery resistance having a much larger impact on flow than aortic resistance. Our results suggest that GPEs can be used to identify parameters in personalized whole heart CFD models, and highlight the importance of accurate preload measurements.

Machine Learning-Enabled Multimodal Fusion of Intra-Atrial and Body Surface Signals in Prediction of Atrial Fibrillation Ablation Outcomes.

BACKGROUND: Machine learning is a promising approach to personalize atrial fibrillation management strategies for patients after catheter ablation. Prior atrial fibrillation ablation outcome prediction studies applied classical machine learning methods to hand-crafted clinical scores, and none have leveraged intracardiac electrograms or 12-lead surface electrocardiograms for outcome prediction. We hypothesized that (1) machine learning models trained on electrograms or electrocardiogram (ECG) signals can perform better at predicting patient outcomes after atrial fibrillation ablation than existing clinical scores and (2) multimodal fusion of electrogram, ECG, and clinical features can further improve the prediction of patient outcomes. METHODS: Consecutive patients who underwent catheter ablation between 2015 and 2017 with panoramic left atrial electrogram before ablation and clinical follow-up for at least 1 year following ablation were included. Convolutional neural network and a novel multimodal fusion framework were developed for predicting 1-year atrial fibrillation recurrence after catheter ablation from electrogram, ECG signals, and clinical features. The models were trained and validated using 10-fold cross-validation on patient-level splits. RESULTS: One hundred fifty-six patients (64.5±10.5 years, 74% male, 42% paroxysmal) were analyzed. Using electrogram signals alone, the convolutional neural network achieved an area under the receiver operating characteristics curve (AUROC) of 0.731, outperforming the existing APPLE scores (AUROC=0.644) and CHA2DS2-VASc scores (AUROC=0.650). Similarly using 12-lead ECG alone, the convolutional neural network achieved an AUROC of 0.767. Combining electrogram, ECG, and clinical features, the fusion model achieved an AUROC of 0.859, outperforming single and dual modality models. CONCLUSIONS: Deep neural networks trained on electrogram or ECG signals improved the prediction of catheter ablation outcome compared with existing clinical scores, and fusion of electrogram, ECG, and clinical features further improved the prediction. This suggests the promise of using machine learning to help treatment planning for patients after catheter ablation.

CArdiac MagnEtic resonance assessment of bi-Atrial fibrosis in secundum atrial septal defects patients: CAMERA-ASD study.

AIMS: Atrial septal defects (ASD) are associated with atrial arrhythmias, but the arrhythmia substrate in these patients is poorly defined. We hypothesized that bi-atrial fibrosis is present and that right atrial fibrosis is associated with atrial arrhythmias in ASD patients. We aimed to evaluate the extent of bi-atrial fibrosis in ASD patients and to investigate the relationships between bi-atrial fibrosis, atrial arrhythmias, shunt fraction, and age. METHODS AND RESULTS: Patients with uncorrected secundum ASDs (n = 36; 50.4 ± 13.6 years) underwent cardiac magnetic resonance imaging with atrial late gadolinium enhancement. Comparison was made to non-congenital heart disease patients (n = 36; 60.3 ± 10.5 years) with paroxysmal atrial fibrillation (AF). Cardiac magnetic resonance parameters associated with atrial arrhythmias were identified and the relationship between bi-atrial structure, age, and shunt fraction studied. Bi-atrial fibrosis burden was greater in ASD patients than paroxysmal AF patients (20.7 ± 14% vs. 10.1 ± 8.6% and 14.8 ± 8.5% vs. 8.6 ± 6.1% for right and left atria respectively, P = 0.001 for both). In ASD patients, right atrial fibrosis burden was greater in those with than without atrial arrhythmias (33.4 ± 18.7% vs. 16.8 ± 10.3%, P = 0.034). On receiver operating characteristic analysis, a right atrial fibrosis burden of 32% had a 92% specificity and 71% sensitivity for predicting the presence of atrial arrhythmias. Neither age nor shunt fraction was associated with bi-atrial fibrosis burden. CONCLUSION: Bi-atrial fibrosis burden is greater in ASD patients than non-congenital heart disease patients with paroxysmal AF. Right atrial fibrosis is associated with the presence of atrial arrhythmias in ASD patients. These findings highlight the importance of right atrial fibrosis to atrial arrhythmogenesis in ASD patients.

Predicting Atrial Fibrillation Recurrence by Combining Population Data and Virtual Cohorts of Patient-Specific Left Atrial Models.

BACKGROUND: Current ablation therapy for atrial fibrillation is suboptimal, and long-term response is challenging to predict. Clinical trials identify bedside properties that provide only modest prediction of long-term response in populations, while patient-specific models in small cohorts primarily explain acute response to ablation. We aimed to predict long-term atrial fibrillation recurrence after ablation in large cohorts, by using machine learning to complement biophysical simulations by encoding more interindividual variability. METHODS: Patient-specific models were constructed for 100 atrial fibrillation patients (43 paroxysmal, 41 persistent, and 16 long-standing persistent), undergoing first ablation. Patients were followed for 1 year using ambulatory ECG monitoring. Each patient-specific biophysical model combined differing fibrosis patterns, fiber orientation maps, electrical properties, and ablation patterns to capture uncertainty in atrial properties and to test the ability of the tissue to sustain fibrillation. These simulation stress tests of different model variants were postprocessed to calculate atrial fibrillation simulation metrics. Machine learning classifiers were trained to predict atrial fibrillation recurrence using features from the patient history, imaging, and atrial fibrillation simulation metrics. RESULTS: We performed 1100 atrial fibrillation ablation simulations across 100 patient-specific models. Models based on simulation stress tests alone showed a maximum accuracy of 0.63 for predicting long-term fibrillation recurrence. Classifiers trained to history, imaging, and simulation stress tests (average 10-fold cross-validation area under the curve, 0.85±0.09; recall, 0.80±0.13; precision, 0.74±0.13) outperformed those trained to history and imaging (area under the curve, 0.66±0.17) or history alone (area under the curve, 0.61±0.14). CONCLUSION: A novel computational pipeline accurately predicted long-term atrial fibrillation recurrence in individual patients by combining outcome data with patient-specific acute simulation response. This technique could help to personalize selection for atrial fibrillation ablation.

Hyperparameter optimisation and validation of registration algorithms for measuring regional ventricular deformation using retrospective gated computed tomography images.

Recent dose reduction techniques have made retrospective computed tomography (CT) scans more applicable and extracting myocardial function from cardiac computed tomography (CCT) images feasible. However, hyperparameters of generic image intensity-based registration techniques, which are used for tracking motion, have not been systematically optimised for this modality. There is limited work on their validation for measuring regional strains from retrospective gated CCT images and open-source software for motion analysis is not widely available. We calculated strain using our open-source platform by applying an image registration warping field to a triangulated mesh of the left ventricular endocardium. We optimised hyperparameters of two registration methods to track the wall motion. Both methods required a single semi-automated segmentation of the left ventricle cavity at end-diastolic phase. The motion was characterised by the circumferential and longitudinal strains, as well as local area change throughout the cardiac cycle from a dataset of 24 patients. The derived motion was validated against manually annotated anatomical landmarks and the calculation of strains were verified using idealised problems. Optimising hyperparameters of registration methods allowed tracking of anatomical measurements with a mean error of 6.63% across frames, landmarks, and patients, comparable to an intra-observer error of 7.98%. Both registration methods differentiated between normal and dyssynchronous contraction patterns based on circumferential strain ([Formula: see text], [Formula: see text]). To test whether a typical 10 temporal frames sampling of retrospective gated CCT datasets affects measuring cardiac mechanics, we compared motion tracking results from 10 and 20 frames datasets and found a maximum error of [Formula: see text]. Our findings show that intensity-based registration techniques with optimal hyperparameters are able to accurately measure regional strains from CCT in a very short amount of time. Furthermore, sufficient sensitivity can be achieved to identify heart failure patients and left ventricle mechanics can be quantified with 10 reconstructed temporal frames. Our open-source platform will support increased use of CCT for quantifying cardiac mechanics.

Non-invasive simulated electrical and measured mechanical indices predict response to cardiac resynchronization therapy.

BACKGROUND: Cardiac Resynchronization Therapy (CRT) in dyssynchronous heart failure patients is ineffective in 20-30% of cases. Sub-optimal left ventricular (LV) pacing location can lead to non-response, thus there is interest in LV lead location optimization. Invasive acute haemodynamic response (AHR) measurements have been used to optimize the LV pacing location during CRT implantation. In this manuscript, we aim to predict the optimal lead location (AHR>10%) with non-invasive computed tomography (CT) based measures of cardiac anatomical and mechanical properties, and simulated electrical activation times. METHODS: Non-invasive measurements from CT images and ECG were acquired from 34 patients indicated for CRT upgrade. The LV lead was implanted and AHR was measured at different pacing sites. Computer models of the ventricles were used to simulate the electrical activation of the heart, track the mechanical motion throughout the cardiac cycle and measure the wall thickness of the LV on a patient specific basis. RESULTS: We tested the ability of electrical, mechanical and anatomical indices to predict the optimal LV location. Electrical (RV-LV delay) and mechanical (time to peak contraction) indices were correlated with an improved AHR, while wall thickness was not predictive. A logistic regression model combining RV-LV delay and time to peak contraction was able to predict positive response with 70 ± 11% accuracy and AUROC curve of 0.73. CONCLUSION: Non-invasive electrical and mechanical indices can predict optimal epicardial lead location. Prospective analysis of these indices could allow clinicians to test the AHR at fewer pacing sites and reduce time, costs and risks to patients.

Optimisation of Left Atrial Feature Tracking Using Retrospective Gated Computed Tomography Images.

Retrospective gated cardiac computed tomography (CCT) images can provide high contrast and resolution images of the heart throughout the cardiac cycle. Feature tracking in retrospective CCT images using the temporal sparse free-form deformations (TSFFDs) registration method has previously been optimised for the left ventricle (LV). However, there is limited work on optimising nonrigid registration methods for feature tracking in the left atria (LA). This paper systematically optimises the sparsity weight (SW) and bending energy (BE) as two hyperparameters of the TSFFD method to track the LA endocardium from end-diastole (ED) to end-systole (ES) using 10-frame retrospective gated CCT images. The effect of two different control point (CP) grid resolutions was also investigated. TSFFD optimisation was achieved using the average surface distance (ASD), directed Hausdorff distance (DHD) and Dice score between the registered and ground truth surface meshes and segmentations at ES. For baseline comparison, the configuration optimised for LV feature tracking gave errors across the cohort of 0.826 ± 0.172mm ASD, 5.882 ± 1.524mm DHD, and 0.912 ± 0.033 Dice score. Optimising the SW and BE hyperparameters improved the TSFFD performance in tracking LA features, with case specific optimisations giving errors across the cohort of 0.750 ± 0.144mm ASD, 5.096 ± 1.246mm DHD, and 0.919 ± 0.029 Dice score. Increasing the CP resolution and optimising the SW and BE further improved tracking performance, with case specific optimisation errors of 0.372 ± 0.051mm ASD, 2.739 ± 0.843mm DHD and 0.949 ± 0.018 Dice score across the cohort. We therefore show LA feature tracking using TSFFDs is improved through a chamber-specific optimised configuration.

Standardised computed tomographic assessment of left atrial morphology and tissue thickness in humans.

AIMS: Left atrial (LA) remodelling is a common feature of many cardiovascular pathologies and is a sensitive marker of adverse cardiovascular outcomes. The aim of this study was to establish normal ranges for LA parameters derived from coronary computed tomographic angiography (CCTA) imaging using a standardised image processing pipeline to establish normal ranges in a previously described cohort. METHODS: CCTA imaging from 193 subjects recruited to the Budapest GLOBAL twin study was analysed. Indexed LA cavity volume (LACVi), LA surface area (LASAi), wall thickness and LA tissue volume (LATVi) were calculated. Wall thickness maps were combined into an atlas. Indexed LA parameters were compared with clinical variables to identify early markers of pathological remodelling. RESULTS: LACVi is similar between sexes (31 ml/m2 v 30 ml/m2) and increased in hypertension (33 ml/m2 v 29 ml/m2, p = 0.009). LASAi is greater in females than males (47.8 ml/m2 v 45.8 ml/m2 male, p = 0.031). Median LAWT was 1.45 mm. LAWT was lowest at the inferior portion of the posterior LA wall (1.14 mm) and greatest in the septum (median = 2.0 mm) (p < 0.001). Conditions known to predispose to the development of AF were not associated with differences in tissue thickness. CONCLUSIONS: The reported LACVi, LASAi, LATVi and tissue thickness derived from CCTA may serve as reference values for this age group and clinical characteristics for future studies. Increased LASAi in females in the absence of differences in LACVi or LATVi may indicate differential LA shape changes between the sexes. AF predisposing conditions, other than sex, were not associated with detectable changes in LAWT.Clinical trial registration:http://www.ClinicalTrials.gov/NCT01738828.

CemrgApp: An interactive medical imaging application with image processing, computer vision, and machine learning toolkits for cardiovascular research.

Personalised medicine is based on the principle that each body is unique and will respond to therapies differently. In cardiology, characterising patient specific cardiovascular properties would help in personalising care. One promising approach for characterising these properties relies on performing computational analysis of multimodal imaging data. An interactive cardiac imaging environment, which can seamlessly render, manipulate, derive calculations, and otherwise prototype research activities, is therefore sought-after. We developed the Cardiac Electro-Mechanics Research Group Application (CemrgApp) as a platform with custom image processing and computer vision toolkits for applying statistical, machine learning and simulation approaches to study physiology, pathology, diagnosis and treatment of the cardiovascular system. CemrgApp provides an integrated environment, where cardiac data visualisation and workflow prototyping are presented through a common graphical user interface.