Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models
<p><strong>Background:</strong> 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 pat...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Lippincott, Williams and Wilkins
2022
|
| _version_ | 1797113321578561536 |
|---|---|
| author | Roney, CH Sim, I Yu, J Beach, M Mehta, A Alonso Solis-Lemus, J Kotadia, I Whitaker, J Corrado, C Razeghi, O Vigmond, E Narayan, SM O’Neill, M Williams, SE Niederer, SA |
| author_facet | Roney, CH Sim, I Yu, J Beach, M Mehta, A Alonso Solis-Lemus, J Kotadia, I Whitaker, J Corrado, C Razeghi, O Vigmond, E Narayan, SM O’Neill, M Williams, SE Niederer, SA |
| author_sort | Roney, CH |
| collection | OXFORD |
| description | <p><strong>Background:</strong> 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.</p>
<p><strong>Methods:</strong> 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.</p>
<p><strong>Results:</strong> 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).</p>
<p><strong>Conclusion:</strong> 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.</p> |
| first_indexed | 2024-04-23T08:26:58Z |
| format | Journal article |
| id | oxford-uuid:6b61510a-eb82-4e14-b616-4cf80c77bf0f |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-04-23T08:26:58Z |
| publishDate | 2022 |
| publisher | Lippincott, Williams and Wilkins |
| record_format | dspace |
| spelling | oxford-uuid:6b61510a-eb82-4e14-b616-4cf80c77bf0f2024-04-19T14:52:16ZPredicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial modelsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:6b61510a-eb82-4e14-b616-4cf80c77bf0fEnglishSymplectic Elements Lippincott, Williams and Wilkins2022Roney, CHSim, IYu, JBeach, MMehta, AAlonso Solis-Lemus, JKotadia, IWhitaker, JCorrado, CRazeghi, OVigmond, ENarayan, SMO’Neill, MWilliams, SENiederer, SA<p><strong>Background:</strong> 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.</p> <p><strong>Methods:</strong> 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.</p> <p><strong>Results:</strong> 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).</p> <p><strong>Conclusion:</strong> 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.</p> |
| spellingShingle | Roney, CH Sim, I Yu, J Beach, M Mehta, A Alonso Solis-Lemus, J Kotadia, I Whitaker, J Corrado, C Razeghi, O Vigmond, E Narayan, SM O’Neill, M Williams, SE Niederer, SA Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models |
| title | Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models |
| title_full | Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models |
| title_fullStr | Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models |
| title_full_unstemmed | Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models |
| title_short | Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models |
| title_sort | predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient specific left atrial models |
| work_keys_str_mv | AT roneych predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT simi predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT yuj predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT beachm predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT mehtaa predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT alonsosolislemusj predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT kotadiai predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT whitakerj predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT corradoc predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT razeghio predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT vigmonde predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT narayansm predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT oneillm predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT williamsse predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels AT niederersa predictingatrialfibrillationrecurrencebycombiningpopulationdataandvirtualcohortsofpatientspecificleftatrialmodels |