Clinical implementation of 3D deep learning techniques in predicting touch-up lesions for atrial fibrillation patients undergoing cryoablation
Year: 2026
Type: article
Abstract: Background: 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. Methods: 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. Results: 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. Conclusions: The CryoAI model demonstrates promising discriminative ability for predicting the need for RF touch-up ablation in patients undergoing cryoballoon ablation for atrial fibrillation.
Source: IJC Heart & Vasculature
Authors: Chih-Min Liu, Wei-Wen Chen, Shih-Lin Chang, Chien-Chao Tseng, Ching-Chun Huang, Yenn-Jiang Lin, Li-Wei Lo, Yu-Feng Hu, Fa-Po Chung, Ting-Yung Chang, Chin-Yu Lin, Tze-Fan Chao, Ta-Chuan Tuan, Jo-Nan Liao, Ling Kuo, Cheng-I Wu, Shin-Huei Liu, Jacky Chung-Hao Wu, Henry Horng-Shing Lu, Shih-Ann Chen
Institutions: National Yang Ming Chiao Tung University, Taipei Veterans General Hospital, Taichung Veterans General Hospital, Institute of Biomedical Sciences, Academia Sinica, Academia Sinica, Kaohsiung Medical University, Cornell University, Kaohsiung Medical University Chung-Ho Memorial Hospital, National Chung Hsing University, China Medical University, China Medical University Hospital
Language: English
Subfield: Cardiology and Cardiovascular Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $3,330
Awards: Generation of Human Pacemaker Cells by Silk Fibroin to Model and Treat Cardiac Conduction Disease.( I ), Implementation of P4 Medicine Models in the Identification and Management of High-Risk Features for Subjects with Cardiovascular Aging( III ), G2058598248, G2240004855, Implementation of P4 Medicine Models in the Identification and Management of High-Risk Features for Subjects with Cardiovascular Aging, Development and Application of Intelligent Fabric Defect Detection System, Smart Medical AI and Repository Taskforce Center(1/2), Combining Online Learning and Deep Learning for Streaming Predictions( I ), G346155331, Metabolic Regulation of Autonomic Nerve Responses in Pacemaker Cells, G3633983566, G4017571536, 文 Aldehyde Dehydrogenase 2 Mutation in the Pathogenesis of Atrial Fibrillation: a Potential Target for Upstream Therapies, G426526813, G4406126668, G4535980199, G6040490157, Portable Nmr System Augmented by Ai/Ml for Biomedical Applications, The Creation of Biological Pacemaker Cells by Small Molecule-Induced Somatic Reprogramming( III ), G6389194954, G6519941372, Smart Medical Ai and Repository Taskforce Center( II ), Artificial Intelligence to Precision Health: Integrating Dynamic Physiological Signals and Emr to Build a Medical Digital Twin Platform( II ), The Creation of Biological Pacemaker Cells by Small Molecule-Induced Somatic Reprogramming( II ), Intelligent Cardiovascular Disease Diagnosis and Management Platform( I ), Artificial Intelligence to Precision Health: Integrating Dynamic Physiological Signals and Emr to Build a Medical Digital Twin Platform( I ), G6751117809, Applications of Federated Learning in Smart Healthcare, G7084014123, G7158638617, Image Segmentation Modeling with Fairness Based on I-Score and Deep Learning( I ), Smart Medical Ai and Repository Taskforce Center (II)( I ), G7984216634, G850950385, G8697452720, Development of the Ai Detection Model for Fabric Defects, Application of Big Data and Ai-Based Approaches to Construct Personalized Disease Risk Models for Predicting or Preventing Childhood Allergic Diseases( I )
OpenAlex ID: W7148630645