Left Atrial Volume as a Biomarker of Atrial Fibrillation at Routine Chest CT: Deep Learning Approach

Bratt, Alex; Guenther, Z.; Hahn, Lewis D.; Kadoch, Michel; Adams, P.L.; Leung, Ann N.; Guo, H. Henry

doi:10.1148/ryct.2019190057

Cited by 9 publications

(10 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For anatomic metric analysis, Varela et al 2 analyzed left atrial anatomy from MRI across a cohort of 144 patients to predict atrial fibrillation recurrence using vertical asymmetry together with left atrial sphericity to give an area under the ROC curve of 0.71. Bratt et al 3 demonstrated that atrial volume is a good predictor of atrial fibrillation recurrence, with an ROC AUC of 0.77. They automatically segmented the left atrial body from computed tomography scans using deep learning and showed that atrial volume is an independent predictor of atrial fibrillation, with an age-adjusted relative risk of 2.9.…”

Section: Discussionmentioning

confidence: 99%

“…They automatically segmented the left atrial body from computed tomography scans using deep learning and showed that atrial volume is an independent predictor of atrial fibrillation, with an age-adjusted relative risk of 2.9. 3 Costa et al 4 showed that left atrial volume is more important than atrial fibrillation type for predicting atrial fibrillation recurrence following pulmonary vein isolation. In contrast to these studies, Ebersberger et al 13 showed no association between pulmonary vein properties or left atrial anatomic or functional properties measured on computed tomography and early atrial fibrillation recurrence at 3 to 4 months post-ablation.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

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

Roney

Sim

et al. 2022

Circ: Arrhythmia and Electrophysiology

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

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

Roney

Sim

et al. 2022

Circ: Arrhythmia and Electrophysiology

View full text Add to dashboard Cite

show abstract

“…As main segmentation network, we opted for an end‐to‐end seven labels 3D U‐Net++ 24 illustrated in Figure 6, an improvement to a classic U‐Net 23 architecture. In the later years, U‐Net derivative networks have shown convincing performances in segmentation tasks 25–31 . Though being more memory‐consuming, 3D geometry was chosen over 2D 28 or multiview 2D 26,32 for its improved performances, enabling to feed better contextual information to the network 25,27,29–31,33,34 .…”

Section: Methodsmentioning

confidence: 99%

A conventional‐to‐spectral CT image translation augmentation workflow for robust contrast injection‐independent organ segmentation

et al. 2021

View full text Add to dashboard Cite

In computed tomography (CT) cardiovascular imaging, the numerous contrast injection protocols used to enhance structures make it difficult to gather training datasets for deep learning applications supporting diverse protocols. Moreover, creating annotations on noncontrast scans is extremely tedious. Recently, spectral CT's virtual-noncontrast images (VNC) have been used as data augmentation to train segmentation networks performing on enhanced and true-noncontrast (TNC) scans alike, while improving results on protocols absent of their training dataset. However, spectral data are not widely available, making it difficult to gather specific datasets for each task. As a solution, we present a data augmentation workflow based on a trained image translation network, to bring spectral-like augmentation to any conventional CT dataset. Method: The conventional CT-to-spectral image translation network (HUSpect-Net) was first trained to generate VNC from conventional housnfied units images (HU), using an unannotated spectral dataset of 1830 patients. It was then tested on a second dataset of 300 spectral CT scans by comparing VNC generated through deep learning (VNC DL ) to their true counterparts. To illustrate and compare our workflow's efficiency with true spectral augmentation,HUSpectNet was applied to a third dataset of 112 spectral scans to generate VNC DL along HU and VNC images. Three different three-dimensional (3D) networks (U-Net, X-Net, and U-Net++) were trained for multilabel heart segmentation, following four augmentation strategies. As baselines, trainings were performed on contrasted images without (HUonly) and with conventional gray-values augmentation (HUaug). Then, the same networks were trained using a proportion of contrasted and VNC/VNC DL images (TrueSpec/GenSpec). Each training strategy applied to each architecture was evaluated using Dice coefficients on a fourth multicentric multivendor single-energy CT dataset of 121 patients, including different contrast injection protocols and unenhanced scans. The U-Net++ results were further explored with distance metrics on every label. Results: Tested on 300 full scans, our HUSpectNet translation network shows a mean absolute error of 6.70 ± 2.83 HU between VNC DL and VNC, while peak signal-to-noise ratio reaches 43.89 dB. GenSpec and TrueSpec show very close results regardless of the protocol and used architecture: mean Dice coefficients (DSC mean ) are equal with a margin of 0.006, ranging from 0.879 to 0.938. Their performances significantly increase on TNC scans (p-values < 0.017 for all architectures) compared to HUonly and HUaug, with DSC mean of 1108

show abstract

“…Machine learning and statistical techniques may be used across populations of patients to predict, from imaging data, how likely it is that AF will recur after catheter ablation therapy. Bratt et al 120 found that atrial volume is an independent predictor of AF, where volume was calculated from computed tomography scans that were automatically segmented using deep learning approaches. Varela et al 121 built a statistical shape model from 144 AF magnetic resonance angiography images and showed that using vertical asymmetry together with left atrial sphericity is predictive of AF recurrence.…”

Section: Section 3: Using Advanced Imaging Techniques To Guide Ablation Proceduresmentioning

confidence: 99%

Applications of multimodality imaging for left atrial catheter ablation

Roney

Sillett

Whitaker

et al. 2021

European Heart Journal - Cardiovascular Imaging

View full text Add to dashboard Cite

Atrial arrhythmias, including atrial fibrillation and atrial flutter, may be treated through catheter ablation. The process of atrial arrhythmia catheter ablation, which includes patient selection, pre-procedural planning, intra-procedural guidance, and post-procedural assessment, is typically characterized by the use of several imaging modalities to sequentially inform key clinical decisions. Increasingly, advanced imaging modalities are processed via specialized image analysis techniques and combined with intra-procedural electrical measurements to inform treatment approaches. Here, we review the use of multimodality imaging for left atrial ablation procedures. The article first outlines how imaging modalities are routinely used in the peri-ablation period. We then describe how advanced imaging techniques may inform patient selection for ablation and ablation targets themselves. Ongoing research directions for improving catheter ablation outcomes by using imaging combined with advanced analyses for personalization of ablation targets are discussed, together with approaches for their integration in the standard clinical environment. Finally, we describe future research areas with the potential to improve catheter ablation outcomes.

show abstract

Left Atrial Volume as a Biomarker of Atrial Fibrillation at Routine Chest CT: Deep Learning Approach

Cited by 9 publications

References 29 publications

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

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

A conventional‐to‐spectral CT image translation augmentation workflow for robust contrast injection‐independent organ segmentation

Applications of multimodality imaging for left atrial catheter ablation

Contact Info

Product

Resources

About