The CryoAI 3D deep learning model predicted the need for radiofrequency touch-up ablation in atrial fibrillation patients undergoing cryoablation with an AUC of 84.07% and a PPV of 96.15%.
Observational (n=289)
No
Does the CryoAI 3D deep learning model accurately predict the need for radiofrequency touch-up ablation in atrial fibrillation patients undergoing cryoballoon ablation?
A novel 3D deep learning model (CryoAI) using pre-procedural CT scans can accurately predict the need for radiofrequency touch-up during cryoballoon ablation for atrial fibrillation, potentially improving procedural planning and reducing procedure times.
Effect estimate: AUC 84.07%
• Developed a 3D deep learning model (CryoAI) to predict touch-up need in cryoablation. • CryoAI (VoxNet++) outperformed PointNet and VoxNet in both internal and external datasets. • Grad-CAM-based 3D visualization provided an interpretable model that focuses on atrial structures. • Gaussian Curvature analysis identified high-risk regions for incomplete ablation contact. • CryoAI enables preoperative planning to possibly reduce procedural time and improve outcomes. Atrial fibrillation (AF) is a common heart rhythm disorder that can be treated with cryoballoon ablation (CBA). CBA occasionally requires additional radiofrequency-based touch-up ablation due to anatomical challenges. This study developed a 3D deep learning model to predict the complexity of CBA procedures and potentially reduce subsequent interventions, minimizing increased procedure times, costs, risks, and patient discomfort. We included 190 AF patients who underwent computed tomography (CT) scans at Taipei Veterans General Hospital from November 2014 to October 2020, divided into touch-up and non-touch-up groups. An 80:20 ratio was used to allocate patients to training and test sets, with an independent external validation set comprising 99 patients from October 2020 to August 2023. Three artificial intelligence (AI) models, PointNet, VoxNet, and an advanced version, CryoAI (VoxNet++), were developed to predict the need for touch-up ablation from 3D voxel-reconstructed CT images. CryoAI demonstrated the best overall discriminative performance among the tested models, achieving an area under the curve (AUC) of 84.07% in the internal test set, with a high positive predictive value (PPV) of 96.15%. In external validation, CryoAI maintained high performance with a PPV of 95.77%. Using a 60° curvature cutoff, all touch-up sites were localized to above-threshold regions in both the internal (n = 8) and external (n = 9) cohorts. Integrating Grad-CAM and a Gaussian Curvature module within our 3D Activation Visualization highlights critical zones for cryoballoon positioning and potential touch-up lesions, enhancing pre-procedural planning. The CryoAI model demonstrates promising discriminative ability for predicting the need for RF touch-up ablation in patients undergoing cryoballoon ablation for atrial fibrillation.
Liu et al. (Fri,) conducted a observational in Atrial fibrillation (n=289). CryoAI (VoxNet++) 3D deep learning model vs. PointNet and VoxNet models was evaluated on Prediction of the need for touch-up ablation from 3D voxel-reconstructed CT images (AUC 84.07%). The CryoAI 3D deep learning model predicted the need for radiofrequency touch-up ablation in atrial fibrillation patients undergoing cryoablation with an AUC of 84.07% and a PPV of 96.15%.