A new machine learning approach for predicting likelihood of recurrence following ablation for atrial fibrillation from CT

Thomas Atta-Fosu,Anant Madabhushi,Walid Saliba,Mina K Chung,Michael Labarbera,Paul Schoenhagen,Milind Y Desai,Deborah Kwon,Patrick J Tchou,Soumya Ghose,Bruce D Lindsay

doi:10.1186/s12880-021-00578-4

Abstract

ObjectiveTo investigate left atrial shape differences on CT scans of atrial fibrillation (AF) patients with (AF+) versus without (AF−) post-ablation recurrence and whether these shape differences predict AF recurrence.MethodsThis retrospective study included 68 AF patients who had pre-catheter ablation cardiac CT scans with contrast. AF recurrence was defined at 1 year, excluding a 3-month post-ablation blanking period. After creating atlases of atrial models from segmented AF+ and AF− CT images, an atlas-based implicit shape differentiation method was used to identify surface of interest (SOI). After registering the SOI to each patient model, statistics of the deformation on the SOI were used to create shape descriptors. The performance in predicting AF recurrence using shape features at and outside the SOI and eight clinical factors (age, sex, left atrial volume, left ventricular ejection fraction, body mass index, sinus rhythm, and AF type [persistent vs paroxysmal], catheter-ablation type [Cryoablation vs Irrigated RF]) were compared using 100 runs of fivefold cross validation.ResultsDifferences in atrial shape were found surrounding the pulmonary vein ostia and the base of the left atrial appendage. In the prediction of AF recurrence, the area under the receiver-operating characteristics curve (AUC) was 0.67 for shape features from the SOI, 0.58 for shape features outside the SOI, 0.71 for the clinical parameters, and 0.78 combining shape and clinical features.ConclusionDifferences in left atrial shape were identified between AF recurrent and non-recurrent patients using pre-procedure CT scans. New radiomic features corresponding to the differences in shape were found to predict post-ablation AF recurrence.

Highlights

Atrial fibrillation (AF) is a common cardiac arrhythmia in which rapid and irregular electrical atrial activation causes loss of synchronized contraction of the atria
In this paper we present a new statistical and machine learning approach, the Differential Atlas for Identifying Sites Predictive of Recurrence, (DiSRn) to address the two questions of (1) whether and where there exist differences between left atrium (LA) shape of post-ablation AF recurrent (AF+) and non-recurrence (AF−) patients, and (2) how these shape differences may be transformed into quantitative descriptors for predicting recurrence using pre-ablation contrast-enhanced computed tomography angiogram (CE-CTA) scans
Univariate analyses showed that patients who had AF recurrence had higher left ventricular ejection fraction (LVEF) and older age with a trend toward less sinus rhythm at the time of ablation

Summary

Introduction

Atrial fibrillation (AF) is a common cardiac arrhythmia in which rapid and irregular electrical atrial activation causes loss of synchronized contraction of the atria. Potential consequences include symptoms from the rapid and irregular conduction to the ventricle, loss of atrioventricular synchrony, and risk of thromboembolic complications, such as stroke. Atta‐Fosu et al BMC Med Imaging (2021) 21:45 catheter ablation, the latter primarily directed toward isolation of the pulmonary vein ostia, where initiating triggers have been observed. For persistent or long-standing persistent AF, the success of ablation can be limited: up to 80% may recur within a year [1], and for this population there is controversy over whether additional substrate ablation should be performed beyond pulmonary vein (PV) isolation. There is interest in predicting the likelihood of recurrence from pre-ablation contrast-enhanced computed tomography angiogram (CE-CTA) scans, which may aid in patient selection for ablation and in procedure and post-procedure planning. Interrogation of quantifiable effects of structural parameters through radiologic biomarkers from pre-ablation CE-CTA scans may improve patient and procedural stratification

Methods

Results

Discussion

Conclusion