Prediction of difficulty in cryoballoon ablation with a three-dimensional deep learning model using polygonal mesh representation

IF 2.2 Q2 CARDIAC & CARDIOVASCULAR SYSTEMS

Journal of Arrhythmia Pub Date : 2025-04-25 DOI:10.1002/joa3.70078

Kazutaka Nakasone MD, PhD, Makoto Nishimori MD, PhD, Masakazu Shinohara MD, PhD, Kunihiko Kiuchi MD, PhD, Mitsuru Takami MD, PhD, Kimitake Imamura MD, PhD, Yu Izawa MD, PhD, Toshihiro Nakamura MD, PhD, Yusuke Sonoda MD, PhD, Hiroyuki Takahara MD, PhD, Kyoko Yamamoto MD, PhD, Yuya Suzuki MD, PhD, Kenichi Tani MD, Hidehiro Iwai MD, Yusuke Nakanishi MD, Ken-ichi Hirata MD, PhD, Koji Fukuzawa MD, PhD

{"title":"Prediction of difficulty in cryoballoon ablation with a three-dimensional deep learning model using polygonal mesh representation","authors":"Kazutaka Nakasone MD, PhD, Makoto Nishimori MD, PhD, Masakazu Shinohara MD, PhD, Kunihiko Kiuchi MD, PhD, Mitsuru Takami MD, PhD, Kimitake Imamura MD, PhD, Yu Izawa MD, PhD, Toshihiro Nakamura MD, PhD, Yusuke Sonoda MD, PhD, Hiroyuki Takahara MD, PhD, Kyoko Yamamoto MD, PhD, Yuya Suzuki MD, PhD, Kenichi Tani MD, Hidehiro Iwai MD, Yusuke Nakanishi MD, Ken-ichi Hirata MD, PhD, Koji Fukuzawa MD, PhD","doi":"10.1002/joa3.70078","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Background</h3>\n \n <p>Cryoballoon ablation (CBA) is useful for pulmonary vein (PV) isolation. However, some cases are challenging, requiring multiple applications and/or touch-up ablations. Although several predictors of CBA difficulty have been reported, none have assessed the spatial location and morphology of the left atrium and PVs. This study aimed to develop a three-dimensional (3D) deep learning (DL) model to predict CBA difficulty and compare its accuracy with conventional manual measurement.</p>\n </section>\n \n <section>\n \n <h3> Methods</h3>\n \n <p>A 28-mm cryoballoon (Arctic Front Advance, Medtronic) was used in all cases. CBA difficulty was defined as requiring touch-up ablation and/or more than three applications per PV. We developed a DL model that can learn polygonal meshes and predict CBA difficulty. In the conventional method, predictors included a thinner left lateral ridge, higher left superior PV (LSPV) ovality index, longer LSPV ostium-bifurcation distance, and shorter right inferior PV ostium-bifurcation distance.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>A total of 189 patients who underwent CBA for drug-resistant atrial fibrillation between January 2015 and January 2022 were included. The DL model was superior to the conventional method in accuracy (0.793 vs. 0.630, <i>p</i> = .042) and specificity (0.796 vs. 0.609, <i>p</i> = .022), with the AUC-ROC of 0.821.</p>\n </section>\n \n <section>\n \n <h3> Conclusions</h3>\n \n <p>We developed a 3D DL model that can detect CBA difficulty using a polygonal mesh representation. By predicting difficult cases in advance, strategies can be developed to increase success rates.</p>\n </section>\n </div>","PeriodicalId":15174,"journal":{"name":"Journal of Arrhythmia","volume":"41 2","pages":""},"PeriodicalIF":2.2000,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/joa3.70078","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Arrhythmia","FirstCategoryId":"1085","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/joa3.70078","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CARDIAC & CARDIOVASCULAR SYSTEMS","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Cryoballoon ablation (CBA) is useful for pulmonary vein (PV) isolation. However, some cases are challenging, requiring multiple applications and/or touch-up ablations. Although several predictors of CBA difficulty have been reported, none have assessed the spatial location and morphology of the left atrium and PVs. This study aimed to develop a three-dimensional (3D) deep learning (DL) model to predict CBA difficulty and compare its accuracy with conventional manual measurement.

Methods

A 28-mm cryoballoon (Arctic Front Advance, Medtronic) was used in all cases. CBA difficulty was defined as requiring touch-up ablation and/or more than three applications per PV. We developed a DL model that can learn polygonal meshes and predict CBA difficulty. In the conventional method, predictors included a thinner left lateral ridge, higher left superior PV (LSPV) ovality index, longer LSPV ostium-bifurcation distance, and shorter right inferior PV ostium-bifurcation distance.

Results

A total of 189 patients who underwent CBA for drug-resistant atrial fibrillation between January 2015 and January 2022 were included. The DL model was superior to the conventional method in accuracy (0.793 vs. 0.630, p = .042) and specificity (0.796 vs. 0.609, p = .022), with the AUC-ROC of 0.821.

Conclusions

We developed a 3D DL model that can detect CBA difficulty using a polygonal mesh representation. By predicting difficult cases in advance, strategies can be developed to increase success rates.

Abstract Image

查看原文本刊更多论文

用多边形网格表示的三维深度学习模型预测低温球囊消融难度

低温球囊消融（CBA）是肺静脉（PV）分离的有效方法。然而，有些情况具有挑战性，需要多次应用和/或补强消融。虽然已经报道了一些CBA困难的预测因素，但没有一个评估左心房和pv的空间位置和形态。本研究旨在建立一个三维（3D）深度学习（DL）模型来预测CBA难度，并将其与传统人工测量的准确性进行比较。方法采用美敦力公司Arctic Front Advance公司的28 mm冷冻球囊。CBA难度定义为每个PV需要补片消融和/或超过三次应用。我们开发了一个可以学习多边形网格并预测CBA难度的深度学习模型。在常规方法中，预测因子包括左侧侧脊较薄、左侧上PV （LSPV）卵形指数较高、LSPV口分岔距离较长、右侧下PV口分岔距离较短。结果纳入2015年1月至2022年1月期间接受CBA治疗的耐药房颤患者189例。DL模型的准确率（0.793比0.630,p = 0.042）和特异性（0.796比0.609,p = 0.022）均优于常规方法，AUC-ROC为0.821。我们开发了一个3D DL模型，可以使用多边形网格表示来检测CBA难度。通过提前预测困难病例，可以制定策略来提高成功率。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊