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

Abstract

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.

In-silico drug trials for precision medicine in atrial fibrillation: From ionic mechanisms to electrocardiogram-based predictions in structurally-healthy human atria

“…This is important as it points towards the possibility of predicting the response of pharmacological treatment using both non-invasive markers and ionic current properties. Moreover, it highlights the integral part of multi-scale modeling and simulation for AF treatment personalization (Roney et al, 2022).…”

“…Accordingly, in-silico trials constitute a central paradigm for drug safety (Passini et al, 2021), with unexploited capabilities for drug efficacy evaluation at higher dimensional levels (i.e., whole-organ scale) (Margara et al, 2022). The latter could support the introduction of humanbased multiscale modeling and simulation into precision medicine, as shown by Roney et al, 2022 for the prediction of atrial fibrillation (AF) recurrence after catheter ablation. Similarly, in-silico trials offer the possibility of identifying key modulators of successful pharmacological treatment, that could guide tailored AF therapies.…”

Elevated tissue inhibitor of metalloproteinase-1 along with left atrium hypertrophy predict atrial fibrillation recurrence after catheter ablation

“…Previous AF recurrence predictive models have incorporated demographic, medical history, electrocardiographic, echocardiographic, and cardiac images ( 14 – 16 ). These predictive models focused solely on clinical parameters, while the pathophysiologic biomarkers were not evaluated.…”

How synergy between mechanistic and statistical models is impacting research in atrial fibrillation

“…Moreover, a digital twin heart may indicate a CA strategy is appropriate for a patient by predicting the likelihood of AF recurrence before a specific therapy is selected ( Muffoletto et al, 2019 ; Shade et al, 2020 ; Seno et al, 2021 ; Roney et al, 2022 ). For example, in the study of Roney et al, AF patient-specific models incorporating fibrotic remodeling from LGE-MRI scans were constructed to test six different ablation approaches.…”