Search

Assessment of left atrial (LA) fibrosis from late gadolinium enhancement (LGE) magnetic resonance imaging (MRI) adds to the management of patients with atrial fibrillation. However, accurate assessment of fibrosis in the LA wall remains challenging. Excluding anatomical structures in the LA proximity using clipping techniques can reduce misclassification of LA fibrosis. A novel FK-means approach for combined automatic clipping and automatic fibrosis segmentation was developed. This approach combines a feature-based Voronoi diagram with a hierarchical 3D K-means fractal-based method. The proposed automatic Voronoi clipping method was applied on LGE-MRI data and achieved a Dice score of 0.75, similar to the score obtained by a deep learning method (3D UNet) for clipping (0.74). The automatic fibrosis segmentation method, which uses the Voronoi clipping method, achieved a Dice score of 0.76. This outperformed a 3D UNet method for clipping and fibrosis classification, which had a Dice score of 0.69. Moreover, the proposed automatic fibrosis segmentation method achieved a Dice score of 0.90, using manual clipping of anatomical structures. The findings suggest that the automatic FK-means analysis approach enables reliable LA fibrosis segmentation and that clipping of anatomical structures in the atrial proximity can add to the assessment of atrial fibrosis.

Introduction:Catheter ablation for atrial fibrillation (AF) can improve symptoms in this common arrhythmia, but AF recurs in up to 45% of patients within a year after ablation. Morphological characteristics of the left atrium (LA) on CT scans may be predictive of AF recurrence. We aimed to study the relationship between variations of the LA, pulmonary veins (PVs) and myocardium with AF recurrence post-ablation using machine derived fractal measurements.Methods:Pre-ablation contrast CT scans were obtained from patients undergoing AF ablation. AF recurrence was defined at 3 months to 1 year. Three segmentation models were created from each CT scan. The LA lumen model consisted of the volume defined by the LA lumen extending to the PV ostia. The PV model consisted of the LA lumen and the full structure of PVs visible on CT. The LA myocardial tissue model consisted of the LA wall. 3D fractal analysis was applied to these models to quantify morphological variation. 7 features for each model were extracted using power low and spectral density function to characterize self-similarity and heterogeneity of the models. 119 patients (67 with and 52 without AF recurrence) were used to train random forest classifiers that were tested on a separate dataset of 66 patients (20 with and 46 without AF recurrence). Prediction performance was quantified by Area Under the Receiver Operating Characteristic Curve (AUC).Results:3D fractal features characterizing LA and PV morphology predicted AF recurrence after ablation. On the test set, The PV model had best prediction (AUC 0.80), followed by the LA volume model (AUC 0.75), the myocardial tissue model (AUC 0.71).Conclusions:Fractal analysis of morphological variations in LA and PV anatomy can be used to predict AF recurrence following ablation.