Three-dimensional Rotational Angiography as a Periprocedural Imaging Tool in Atrial Fibrillation Ablation

High-throughput 3DRA segmentation of brain vasculature and aneurysms using deep learning

BACKGROUND AND OBJECTIVES

Automatic segmentation of the cerebral vasculature and aneurysms facilitates incidental detection of aneurysms. The assessment of aneurysm rupture risk assists with pre-operative treatment planning and enables in-silico investigation of cerebral hemodynamics within and in the vicinity of aneurysms. However, ensuring precise and robust segmentation of cerebral vessels and aneurysms in neuroimaging modalities such as three-dimensional rotational angiography (3DRA) is challenging. The vasculature constitutes a small proportion of the image volume, resulting in a large class imbalance (relative to surrounding brain tissue). Additionally, aneurysms and vessels have similar image/appearance characteristics, making it challenging to distinguish the aneurysm sac from the vessel lumen.

METHODS

We propose a novel multi-class convolutional neural network to tackle these challenges and facilitate the automatic segmentation of cerebral vessels and aneurysms in 3DRA images. The proposed model is trained and evaluated on an internal multi-center dataset and an external publicly available challenge dataset.

RESULTS

On the internal clinical dataset, our method consistently outperformed several state-of-the-art approaches for vessel and aneurysm segmentation, achieving an average Dice score of 0.81 (0.15 higher than nnUNet) and an average surface-to-surface error of 0.20 mm (less than the in-plane resolution (0.35 mm/pixel)) for aneurysm segmentation; and an average Dice score of 0.91 and average surface-to-surface error of 0.25 mm for vessel segmentation. In 223 cases of a clinical dataset, our method accurately segmented 190 aneurysm cases.

CONCLUSIONS

The proposed approach can help address class imbalance problems and inter-class interference problems in multi-class segmentation. Besides, this method performs consistently on clinical datasets from four different sources and the generated results are qualified for hemodynamic simulation. Code available at https://github.com/cistib/vessel-aneurysm-segmentation.

Intra-procedural three-dimensional rotational angiography in cryoballoon ablation for atrial fibrillation

Cryoballoon (CB) is an established technology for atrial fibrillation (AF) ablation and is usually performed using solely fluoroscopy. We aimed to study the feasibility of three-dimensional rotational angiography (3DRA) as intra-procedural imaging in CB ablation. Analyzed data were retrospectively collected from patients that underwent second generation CB ablation from February 2015 to August 2017. We studied 68 consecutive patients that received 3DRA (3DRA group). Sixty-six patients who received conventional X-ray imaging served as a control group. 3DRA was performed via an introducer placed in the left atrium. Angiographic images were segmented and fused with live fluoroscopy to guide the ablation. We have analyzed 134 CB patients (73.8% male, 56.9 ± 11.4 years). Paroxysmal AF was present in 77.6% of patients. 3DRA was successfully performed in all 3DRA group patients. The mean procedure time was significantly shorter in the control group (82.4 ± 26.3 min) than in the 3DRA group (121.1 ± 21.4 min) (p < 0.0001). Total radiation dose (419.3 ± 317.9 vs 998.3 ± 673 mGy, p < 0.0001) and contrast administration (83.2 ± 22.3 mL vs 191.6 ± 33.4 mL, p < 0.0001) were significantly lower in control group. There was no significant difference in 2-year success rate, 35.2% of patients had AF recurrence in the 3DRA group and 30.3% in the control group (p = 0.584). Major complications occurred in 2.9% and 1.5% of patients in 3DRA group and control group, respectively (p = 1.000). 3DRA is a feasible method of intra-procedural imaging to guide CB ablation. However, it prolongs procedure time, increases radiation dose and contrast administration with no significant effect on procedure outcomes and complication rates.

The industrialization of ablation: a highly standardized and reproducible workflow for radiofrequency ablation of atrial fibrillation

BACKGROUND OR PURPOSE

The purpose of this analysis was to report on efficacy of a standardized workflow for atrial fibrillation (AF) ablation using technology advances such as 3D imaging and contact force sensing in a real-world setting.

METHODS

Consecutive AF ablations from 2014 to 2015 at a high-volume site in Belgium were included. The workflow consisted of a pre-specified procedure sequence including 3D modeling followed by radiofrequency encircling of the pulmonary veins (25 W posterior wall, 35 W anterior wall) with a THERMOCOOL SMARTTOUCH® Catheter guided by CARTO VISITAG™ Module (2.5 mm/5 s stability, 50% > 7 g) and ablation index (targets: 550 anterior wall, 400 posterior wall). Efficiency endpoints were procedure time, fluoroscopy time, and radiation dose. The primary effectiveness endpoint was freedom from atrial arrhythmia recurrence.

RESULTS

A total of 605 paroxysmal AF (PAF) and 182 persistent AF (PsAF) patients were followed for 436 ± 199 days. Mean procedure times were short (PAF: 96.1 ± 26.2 min; PsAF: 109.2 ± 35.6 min) with most procedures (90.6% PAF; 81.3% PsAF) completed in ≤ 120 min. Minimal fluoroscopy was utilized (PAF: 6.1 ± 3.8 min, 5.9 ± 3.4 Gy*cm²; PsAF: 6.9 ± 4.7 min, 7.4 ± 4.9 Gy*cm²). Freedom from atrial arrhythmia recurrence was higher for PAF than PsAF patients (OR: 2.0, 95% CI: 1.4-2.9, p = 0.0003), but adjusted mean rates were high in both groups (81.0% vs. 67.9%). Rates were adjusted for prior ablation and age (at 65 years).

CONCLUSION

AF ablation using a standardized workflow resulted in low procedure times and variability, with minimal fluoroscopy exposure. Long-term freedom from atrial arrhythmia recurrence was high in both PAF and PsAF populations.

Mapping and Imaging in Non-paroxysmal AF

Despite intense research efforts, maintenance of sinus rhythm in patients with non-paroxysmal AF remains challenging with suboptimal outcomes. A major limitation to the success of current ablation-based treatments is that our understanding of AF pathophysiology is incomplete. Advances in imaging and mapping tools have been reported to improve ablation outcomes. However, the role of these new approaches on the clinical care of patients with AF remains to be validated and better understood before wide adoption can occur. This article reviews the current techniques of imaging and mapping that can be applied in the management of patients with non-paroxysmal AF with a focus on their relevance to catheter ablation. Future applications and opportunities for new knowledge are also discussed.

Pulmonary vein anatomy variants as a biomarker of atrial fibrillation - CT angiography evaluation

Background

It has been suggested that changes in pulmonary veins (PV) and left atrium (LA) anatomy may have an influence on initiating atrial fibrillation (AF) and the effectiveness of pulmonary vein isolation (PVI) in patients (pts) with atrial fibrillation.

The aim of the study was to assess anatomy abnormalities of the PV and LA in the patients with the history of AF and compare it with the control group(CG).

Methods

The multi-slice tomography (MSCT) scans were performed in 224 AF pts. before PVI (129 males, mean age 59 ± 9 yrs). The CG consisted of 40 pts. without AF (26 males, age 45 ± 9 yrs). LA and PV anatomy were evaluated. Diameters of PV ostia were measured in two directions: anterior-posterior (AP) and superior-inferior (SI) automatically using Vitrea 4.0.

Results

Pulmonary veins anatomy variants were observed more frequently in the atrial fibrillation group - 83 pts. (37%) vs 6 pts. (15%) in CG; 9% (21 pts) left common ostia (CO), 2% (5 pts) right CO, 19% (42 pts) additional right PV (APV), (1.8%) 4 pts. APV left, 8% right early branching (EB) and 3.5% left EB. The LA diameter differed significantly in AF vs CG group (41.2 ± 6 mm vs 35 ± 4.2 mm, p < 0.0001) respectively.

Conclusions

The anomalies of pulmonary vein anatomy occurred more often in pts. with AF. They can be defined as an image biomarkers of atrial fibrillation. Right additional (middle) pulmonary vein was the most important anomaly detected in AF patients as well as enlargered diameters of the LA and PV ostia.