Feasibility of Auto-Quantified Epicardial Adipose Tissue in Predicting Atrial Fibrillation Recurrence After Catheter Ablation

Ling Kuo; Guan Jie Wang; Shih Ling Chang; Yenn Jiang Lin; Fa Po Chung; Li Wei Lo; Yu Feng Hu; Tze Fan Chao; Ta Chuan Tuan; Jo Nan Liao; Ting Yung Chang; Chin Yu Lin; Chih Min Liu; Shin Huei Liu; Ming Ren Kuo; Guan Yi Li; Yu Shan Huang; Cheng I. Wu; Shih Ann Chen; Chia Feng Lu

doi:10.1253/circj.CJ-23-0808

Feasibility of Auto-Quantified Epicardial Adipose Tissue in Predicting Atrial Fibrillation Recurrence After Catheter Ablation

Ling Kuo, Guan Jie Wang, Shih Ling Chang, Yenn Jiang Lin, Fa Po Chung, Li Wei Lo, Yu Feng Hu, Tze Fan Chao, Ta Chuan Tuan, Jo Nan Liao, Ting Yung Chang, Chin Yu Lin, Chih Min Liu, Shin Huei Liu, Ming Ren Kuo, Guan Yi Li, Yu Shan Huang, Cheng I. Wu, Shih Ann Chen^*, Chia Feng Lu^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

Background: The aim of this study was to build an auto-segmented artificial intelligence model of the atria and epicardial adipose tissue (EAT) on computed tomography (CT) images, and examine the prognostic significance of auto-quantified left atrium (LA) and EAT volumes for AF. Methods and Results: This retrospective study included 334 patients with AF who were referred for catheter ablation (CA) between 2015 and 2017. Atria and EAT volumes were auto-quantified using a pre-trained 3-dimensional (3D) U-Net model from pre-ablation CT images. After adjusting for factors associated with AF, Cox regression analysis was used to examine predictors of AF recurrence. The mean (±SD) age of patients was 56±11 years; 251 (75%) were men, and 79 (24%) had non-paroxysmal AF. Over 2 years of follow-up, 139 (42%) patients experienced recurrence. Diabetes, non-paroxysmal AF, non-pulmonary vein triggers, mitral line ablation, and larger LA, right atrium, and EAT volume indices were linked to increased hazards of AF recurrence. After multivariate adjustment, non-paroxysmal AF (hazard ratio [HR] 0.6; 95% confidence interval [CI] 0.4-0.8; P=0.003) and larger LA-EAT volume index (HR 1.1; 95% CI 1.0-1.2; P=0.009) remained independent predictors of AF recurrence. Conclusions: LA-EAT volume measured using the auto-quantified 3D U-Net model is feasible for predicting AF recurrence after CA, regardless of AF type.

Original language	English
Pages (from-to)	1089-1098
Number of pages	10
Journal	Circulation Journal
Volume	88
Issue number	7
DOIs	https://doi.org/10.1253/circj.CJ-23-0808
State	Published - Jul 2024

Keywords

Atrial fibrillation
Catheter ablation
Computed tomography
Epicardial adipose tissue
Machine learning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1253/circj.CJ-23-0808

Cite this

Kuo, L., Wang, G. J., Chang, S. L., Lin, Y. J., Chung, F. P., Lo, L. W., Hu, Y. F., Chao, T. F., Tuan, T. C., Liao, J. N., Chang, T. Y., Lin, C. Y., Liu, C. M., Liu, S. H., Kuo, M. R., Li, G. Y., Huang, Y. S., Wu, C. I., Chen, S. A., & Lu, C. F. (2024). Feasibility of Auto-Quantified Epicardial Adipose Tissue in Predicting Atrial Fibrillation Recurrence After Catheter Ablation. Circulation Journal, 88(7), 1089-1098. https://doi.org/10.1253/circj.CJ-23-0808

@article{0d28a702470e4526be8a0affbaddab7b,

title = "Feasibility of Auto-Quantified Epicardial Adipose Tissue in Predicting Atrial Fibrillation Recurrence After Catheter Ablation",

abstract = "Background: The aim of this study was to build an auto-segmented artificial intelligence model of the atria and epicardial adipose tissue (EAT) on computed tomography (CT) images, and examine the prognostic significance of auto-quantified left atrium (LA) and EAT volumes for AF. Methods and Results: This retrospective study included 334 patients with AF who were referred for catheter ablation (CA) between 2015 and 2017. Atria and EAT volumes were auto-quantified using a pre-trained 3-dimensional (3D) U-Net model from pre-ablation CT images. After adjusting for factors associated with AF, Cox regression analysis was used to examine predictors of AF recurrence. The mean (±SD) age of patients was 56±11 years; 251 (75%) were men, and 79 (24%) had non-paroxysmal AF. Over 2 years of follow-up, 139 (42%) patients experienced recurrence. Diabetes, non-paroxysmal AF, non-pulmonary vein triggers, mitral line ablation, and larger LA, right atrium, and EAT volume indices were linked to increased hazards of AF recurrence. After multivariate adjustment, non-paroxysmal AF (hazard ratio [HR] 0.6; 95% confidence interval [CI] 0.4-0.8; P=0.003) and larger LA-EAT volume index (HR 1.1; 95% CI 1.0-1.2; P=0.009) remained independent predictors of AF recurrence. Conclusions: LA-EAT volume measured using the auto-quantified 3D U-Net model is feasible for predicting AF recurrence after CA, regardless of AF type.",

keywords = "Atrial fibrillation, Catheter ablation, Computed tomography, Epicardial adipose tissue, Machine learning",

author = "Ling Kuo and Wang, {Guan Jie} and Chang, {Shih Ling} and Lin, {Yenn Jiang} and Chung, {Fa Po} and Lo, {Li Wei} and Hu, {Yu Feng} and Chao, {Tze Fan} and Tuan, {Ta Chuan} and Liao, {Jo Nan} and Chang, {Ting Yung} and Lin, {Chin Yu} and Liu, {Chih Min} and Liu, {Shin Huei} and Kuo, {Ming Ren} and Li, {Guan Yi} and Huang, {Yu Shan} and Wu, {Cheng I.} and Chen, {Shih Ann} and Lu, {Chia Feng}",

year = "2024",

month = jul,

doi = "10.1253/circj.CJ-23-0808",

language = "English",

volume = "88",

pages = "1089--1098",

journal = "Circulation Journal",

issn = "1346-9843",

publisher = "Japanese Circulation Society",

number = "7",

}

Kuo, L, Wang, GJ, Chang, SL , Lin, YJ, Chung, FP, Lo, LW , Hu, YF, Chao, TF, Tuan, TC, Liao, JN, Chang, TY, Lin, CY, Liu, CM, Liu, SH, Kuo, MR, Li, GY, Huang, YS, Wu, CI, Chen, SA & Lu, CF 2024, 'Feasibility of Auto-Quantified Epicardial Adipose Tissue in Predicting Atrial Fibrillation Recurrence After Catheter Ablation', Circulation Journal, vol. 88, no. 7, pp. 1089-1098. https://doi.org/10.1253/circj.CJ-23-0808

TY - JOUR

T1 - Feasibility of Auto-Quantified Epicardial Adipose Tissue in Predicting Atrial Fibrillation Recurrence After Catheter Ablation

AU - Kuo, Ling

AU - Wang, Guan Jie

AU - Chang, Shih Ling

AU - Lin, Yenn Jiang

AU - Chung, Fa Po

AU - Lo, Li Wei

AU - Hu, Yu Feng

AU - Chao, Tze Fan

AU - Tuan, Ta Chuan

AU - Liao, Jo Nan

AU - Chang, Ting Yung

AU - Lin, Chin Yu

AU - Liu, Chih Min

AU - Liu, Shin Huei

AU - Kuo, Ming Ren

AU - Li, Guan Yi

AU - Huang, Yu Shan

AU - Wu, Cheng I.

AU - Chen, Shih Ann

AU - Lu, Chia Feng

PY - 2024/7

Y1 - 2024/7

N2 - Background: The aim of this study was to build an auto-segmented artificial intelligence model of the atria and epicardial adipose tissue (EAT) on computed tomography (CT) images, and examine the prognostic significance of auto-quantified left atrium (LA) and EAT volumes for AF. Methods and Results: This retrospective study included 334 patients with AF who were referred for catheter ablation (CA) between 2015 and 2017. Atria and EAT volumes were auto-quantified using a pre-trained 3-dimensional (3D) U-Net model from pre-ablation CT images. After adjusting for factors associated with AF, Cox regression analysis was used to examine predictors of AF recurrence. The mean (±SD) age of patients was 56±11 years; 251 (75%) were men, and 79 (24%) had non-paroxysmal AF. Over 2 years of follow-up, 139 (42%) patients experienced recurrence. Diabetes, non-paroxysmal AF, non-pulmonary vein triggers, mitral line ablation, and larger LA, right atrium, and EAT volume indices were linked to increased hazards of AF recurrence. After multivariate adjustment, non-paroxysmal AF (hazard ratio [HR] 0.6; 95% confidence interval [CI] 0.4-0.8; P=0.003) and larger LA-EAT volume index (HR 1.1; 95% CI 1.0-1.2; P=0.009) remained independent predictors of AF recurrence. Conclusions: LA-EAT volume measured using the auto-quantified 3D U-Net model is feasible for predicting AF recurrence after CA, regardless of AF type.

AB - Background: The aim of this study was to build an auto-segmented artificial intelligence model of the atria and epicardial adipose tissue (EAT) on computed tomography (CT) images, and examine the prognostic significance of auto-quantified left atrium (LA) and EAT volumes for AF. Methods and Results: This retrospective study included 334 patients with AF who were referred for catheter ablation (CA) between 2015 and 2017. Atria and EAT volumes were auto-quantified using a pre-trained 3-dimensional (3D) U-Net model from pre-ablation CT images. After adjusting for factors associated with AF, Cox regression analysis was used to examine predictors of AF recurrence. The mean (±SD) age of patients was 56±11 years; 251 (75%) were men, and 79 (24%) had non-paroxysmal AF. Over 2 years of follow-up, 139 (42%) patients experienced recurrence. Diabetes, non-paroxysmal AF, non-pulmonary vein triggers, mitral line ablation, and larger LA, right atrium, and EAT volume indices were linked to increased hazards of AF recurrence. After multivariate adjustment, non-paroxysmal AF (hazard ratio [HR] 0.6; 95% confidence interval [CI] 0.4-0.8; P=0.003) and larger LA-EAT volume index (HR 1.1; 95% CI 1.0-1.2; P=0.009) remained independent predictors of AF recurrence. Conclusions: LA-EAT volume measured using the auto-quantified 3D U-Net model is feasible for predicting AF recurrence after CA, regardless of AF type.

KW - Atrial fibrillation

KW - Catheter ablation

KW - Computed tomography

KW - Epicardial adipose tissue

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85196174345&partnerID=8YFLogxK

U2 - 10.1253/circj.CJ-23-0808

DO - 10.1253/circj.CJ-23-0808

M3 - Article

C2 - 38355108

AN - SCOPUS:85196174345

SN - 1346-9843

VL - 88

SP - 1089

EP - 1098

JO - Circulation Journal

JF - Circulation Journal

IS - 7

ER -

Feasibility of Auto-Quantified Epicardial Adipose Tissue in Predicting Atrial Fibrillation Recurrence After Catheter Ablation

Abstract

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this