基于机器学习的房颤病人左心耳血流动力学研究

刘笑语; 林洪涛; 钱姜宏; 蔡声泽; 樊红光; 高琪

doi:10.7638/kqdlxxb-2023.0223

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于机器学习的房颤病人左心耳血流动力学研究

研究论文 | 更新时间：2026-07-27

- 基于机器学习的房颤病人左心耳血流动力学研究
- Machine learning-based study on left atrial appendage hemodynamics in atrial fibrillation patients
- 空气动力学学报 2024年42卷第9期页码：14-20
- 作者机构：
  
  1. 浙江大学航天航空学院,杭州,310058
  2. 杭州晟视科技有限公司,杭州,311100
  3. 浙江大学控制学院,杭州,310058
  4. 中国医学科学院阜外医院,北京,100037
  5. 浙江大学经血管植入器械全国重点实验室,杭州,310009
- 作者简介：
  
  刘笑语（1998—），女，山东人，博士生，研究方向：生物流体力学. E-mail：liu.xiaoyu@zju.edn.cn
  高琪*，教授，研究方向：流体力学. E-mail：qigao@zju.edu.cn
- 基金信息：
  
  国家自然科学基金（12072320）
- DOI：10.7638/kqdlxxb-2023.0223
  中图分类号： O357;Q954.56
- 纸质出版：2024
- 稿件说明：
移动端阅览
刘笑语, 林洪涛, 钱姜宏, 等. 基于机器学习的房颤病人左心耳血流动力学研究[J]. 空气动力学学报, 2024, 42(9): 14−20.

LIU Xiaoyu, LIN Hongtao, QIAN Jianghong, et al. Machine learning-based study on left atrial appendage hemodynamics in atrial fibrillation patients[J]. 2024, 42(9): 14-20.
刘笑语, 林洪涛, 钱姜宏, 等. 基于机器学习的房颤病人左心耳血流动力学研究[J]. 空气动力学学报, 2024, 42(9): 14−20. DOI： 10.7638/kqdlxxb-2023.0223

LIU Xiaoyu, LIN Hongtao, QIAN Jianghong, et al. Machine learning-based study on left atrial appendage hemodynamics in atrial fibrillation patients[J]. 2024, 42(9): 14-20. DOI： 10.7638/kqdlxxb-2023.0223.

摘要

房颤是一种常见的心血管疾病，其发病率在全球范围内呈现出快速增长的趋势。房颤病人的左心房及左心耳内的血流动力学规律，比如速度场和压力场，是评估血栓形成风险的重要依据之一。目前对于心血管问题的传统研究方法是计算流体力学，尽管其精度被认可，但高昂的计算成本限制了其在临床快速诊断方面的应用。为了解决这个问题，机器学习方法开始被引入，本文提出了一种基于边卷积层和双通道的神经网络架构。经过训练，该架构可以由左心耳的点云模型，预测出内部的速度场和压力场。该架构对测试集中样本的速度场预测误差约为10%，压力场预测的误差约为5%。应用本文方法对314例房颤患者的左心房及左心耳的血流动力学特性进行研究，发现房颤患者的左心耳内流速低于正常人，而左心房平均压力高于正常人，男性相比于女性更容易患有房颤。

Abstract

Atrial fibrillation is a common cardiovascular disease with an increasing incidence worldwide. The hemodynamic patterns within the left atrium and left atrial appendage in atrial fibrillation patients

such as velocity and pressure fields

are important indicators for assessing the risk of thrombus formation. The conventional research method for cardiovascular issues is computational fluid dynamics

which

although recognized for its accuracy

is limited in its clinical application due to high computational costs. To address this issue

machine learning methods have been introduced

proposing a neural network architecture based on edge convolutional layers and dual channels. After training

this architecture can predict the internal velocity and pressure fields from a point cloud model of the left atrial appendage. The architecture achieves prediction errors of around 10% for velocity fields and 5% for pressure fields on the test dataset. Hemodynamic analysis reveals that atrial fibrillation patients have lower flow velocities in the left atrial appendage and higher average pressures in the left atrium compared to healthy individuals

and males are more prone to atrial fibrillation than females.

关键词

Keywords

references

KRIJTHE B P, KUNST A, BENJAMIN E J, et al . Projections on the number of individuals with atrial fibrillation in the European Union, from 2000 to 2060[J ] . European Heart Journal, 2013 , 34 ( 35 ): 2746 − 2751 .

PATEL N J, DESHMUKH A, PANT S, et al . Contemporary trends of hospitalization for atrial fibrillation in the United States, 2000 through 2010[J ] . Circulation, 2014 , 129 ( 23 ): 2371 − 2379 .

LEE J M, SEO J, UHM J S, et al . Why is left atrial appendage morphology related to strokes? An analysis of the flow velocity and orifice size of the left atrial appendage[J ] . Journal of Cardiovascular Electrophysiology, 2015 , 26 ( 9 ): 922 − 927 .

ETLI M, CANBOLAT G, KARAHAN O, et al . Numerical investigation of patient-specific thoracic aortic aneurysms and comparison with normal subject via computational fluid dynamics (CFD)[J ] . Medical Biological Engineering Computing, 2021 , 59 ( 1 ): 71 − 84 .

CILLA M, CASALES M, PEÑA E, et al . A parametric model for study ing the aorta hemodynamics by means of the computational fluid dynamics[J ] . Journal of Biomechanics, 2020 , 103 : 109691 .

XIE J, CHENG Z H, GU L H, et al . Evaluation of cerebrovascular hemodynamics in vascular dementia patients with a new individual computational fluid dynamics algorithm[J ] . Computer Methods and Programs in Biomedicine, 2022 , 213 : 106497 .

NAGAYAMA Y, EMOTO T, HAYASHI H, et al . Coronary stent evaluation by CTA: Image quality comparison between super-resolution deep learning reconstruction and other reconstruction algorithms[J ] . AJR American Journal of Roentgenology, 2023 , 221 ( 5 ): 599 − 610 .

ZHAO Y Y, CHEN T, WANG H, et al . Influence of three-dimensional visual reconstruction technology combined with virtual surgical planning of CTA images on precise resection of liver cancer in hepatobiliary surgery[J ] . Computational and Mathematical Methods in Medicine, 2022 , 2022 : 4376654 .

LEHMAN C D, MERCALDO S, LAMB L R, et al . Deep learning vs traditional breast cancer risk models to support risk-based mammography screening[J ] . Journal of the National Cancer Institute, 2022 , 114 ( 10 ): 1355 − 1363 .

XIE W S, REDER N P, KOYUNCU C, et al . Prostate cancer risk stratification via nondestructive 3D pathology with deep learning-assisted gland analysis[J ] . Cancer Research, 2022 , 82 ( 2 ): 334 − 345 .

FOERSCH S, GLASNER C, WOERL A C, et al . Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer[J ] . Nature Medicine, 2023 , 29 : 430 − 439 .

ESTEVA A, FENG J, VAN DER WAL D, et al . Prostate cancer therapy personalization via multi-modal deep learning on randomized phase III clinical trials[J ] . NPJ Digital Medicine, 2022 , 5 : 71 .

LI G Y, WANG H R, ZHANG M Z, et al . Prediction of 3D cardiovascular hemodynamics before and after coronary artery bypass surgery via deep learning[J ] . Communications Biology, 2021 , 4 : 99 .

ZHANG X L, MAO B Y, CHE Y, et al . Physics-informed neural networks (PINNs) for 4D hemodynamics prediction: An investigation of optimal framework based on vascular morphology[J ] . Computers in Biology and Medicine, 2023 , 164 : 107287 .

FEREZ X M, MILL J, JUHL K A, et al . Deep learning framework for real-time estimation of in-silico thrombotic risk indices in the left atrial appendage[J ] . Frontiers in Physiology, 2021 , 12 : 694945 .

GAO Q, LIN H T, QIAN J H, et al . A deep learning model for efficient end-to-end stratification of thrombotic risk in left atrial appendage[J ] . Engineering Applications of Artificial Intelligence, 2023 , 126 : 107187 .

DEMIRKIRAN A, AMIER R P, HOFMAN M B M, et al . Altered left atrial 4D flow characteristics in patients with paroxysmal atrial fibrillation in the absence of apparent remodeling[J ] . Scientific Reports, 2021 , 11 : 5965 .

LEI C X, GAO Q, WEI R J, et al . Fractal geometry illustrated left atrial appendage morphology that predicted thrombosis and stroke in patients with atrial fibrillation[J ] . Frontiers in Cardiovascular Medicine, 2022 , 9 : 779528 .

JASAK H, JEMCOV A, TUKOVIC Z. OpenFOAM: A C++ library for complex physics simulations[C]//International workshop on coupled methods in numerical dynamics. 2007, 1000: 1-20.

International Organization for Standardization. Cardiovascular implants—cardiac valve prostheses—part 1: General requirements: ISO 5840-1: 2015[S]. ANSI/AAMI: 2015.

FANNI B M, CAPELLINI K, DI LEONARDO M, et al . Correlation between LAA morphological features and computational fluid dynamics analysis for non-valvular atrial fibrillation patients[J ] . Applied Sciences, 2020 , 10 ( 4 ): 1448 .

KU D N . Blood flow in arteries[J ] . Annual Review of Fluid Mechanics, 1997 , 29 : 399 − 434 .

CHARLES R Q, HAO S, MO K C, et al. PointNet: Deep learning on point sets for 3D classification and segmentation[C]//Proc of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA. IEEE, 2017: 77−85.

PHAN A V, LE NGUYEN M, NGUYEN Y L H, et al . DGCNN: A convolutional neural network over large-scale labeled graphs[J ] . Neural Networks, 2018 , 108 : 533 − 543 .

AGMON Y, KHANDHERIA B K, MEISSNER I, et al . Left atrial appendage flow velocities in subjects with normal left ventricular function[J ] . The American Journal of Cardiology, 2000 , 86 ( 7 ): 769 − 773 .

YOSHIDA K, ULFARSSON M, ORAL H, et al . Left atrial pressure and dominant frequency of atrial fibrillation in humans[J ] . Heart Rhythm, 2011 , 8 ( 2 ): 181 − 187 .

KUECHERER H F, MUHIUDEEN I A, KUSUMOTO F M, et al . Estimation of mean left atrial pressure from transesophageal pulsed doppler echocardiography of pulmonary venous flow[J ] . Circulation, 1990 , 82 ( 4 ): 1127 − 1139 .

WONG G R, NALLIAH C J, LEE G, et al . Sex-related differences in atrial remodeling in patients with atrial fibrillation: Relationship to ablation outcomes[J ] . Circulation Arrhythmia and Electrophysiolo gy, 2022 , 15 ( 1 ): e009925 .

KOIZUMI R, FUNAMOTO K, HAYASE T, et al . Numerical analysis of hemodynamic changes in the left atrium due to atrial fibrillation[J ] . Journal of Biomechanics, 2015 , 48 ( 3 ): 472 − 478 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于深度神经网络的高阶非线性激波判别式

基于多支微射流的超声速射流主动控制

无人机气动/风能获取一体化设计

基于稀疏神经核的多保真度代理模型

基于条件扩散模型的火箭气动外形快速设计方法