|本期目录/Table of Contents|

[1]张昊,童基均,吴金玉,等.基于自适应正则化光流模型的超声心动图心肌室壁边界识别[J].浙江理工大学学报,2023,49-50(自科三):293-299.
 ZHANG Hao,TONG Jijun,WU Jinyu,et al.Ventricular wall boundary recognition based on adaptive regularized optical flow model in echocardiography[J].Journal of Zhejiang Sci-Tech University,2023,49-50(自科三):293-299.
点击复制

基于自适应正则化光流模型的超声心动图心肌室壁边界识别()
分享到:

浙江理工大学学报[ISSN:1673-3851/CN:33-1338/TS]

卷:
第49-50卷
期数:
2023年自科第三期
页码:
293-299
栏目:
出版日期:
2023-05-31

文章信息/Info

Title:
Ventricular wall boundary recognition based on adaptive regularized optical flow model in echocardiography
文章编号:
1673-3851 (2023) 05-0293-07
作者:
张昊童基均吴金玉朱新建刘姗娜张华青
1.浙江理工大学计算机科学与技术学院,杭州 310018;2.浙江大学医学院附属第四医院,浙江义乌 322000;3.浙江大学医学院附属第二医院,杭州 310003
Author(s):
ZHANG Hao TONG Jijun WU Jinyu ZHU Xinjian LIU Shanna ZHANG Huaqing
1.School of Computer Science and Technology, Zhejiang Sci-Tech University,  Hangzhou 310018, China;2.The Fourth Affiliated Hospital Zhejiang University School of  Medicine, Yiwu 322000, China;3.The Second Affiliated Hospital Zhejiang University School of  Medicine, Hangzhou 310003, China
关键词:
自适应正则化光流模型超声心动图心肌室壁边界识别
分类号:
TS195-644
文献标志码:
A
摘要:
超声心动图的心肌室壁运动估计是心脏病诊疗的重要技术,但常用的光流模型估计方法采用恒定的正则化平滑系数,识别的室壁边界误差大,运动边界不清晰。针对上述问题,采用自适应正则化光流模型,并将小波分析融入模型,提出了一种超声心动图室壁边界识别的方法。先对相邻两帧图像进行预处理,然后利用Brox光流法对相邻图像进行运动估计得到初始光流矢量,接着光流失量进行二维小波分析得到运动场相对突变的空间分布信息,随后将突变分布信息经过函数映射反馈至光流计算模型中以修正平滑项正则化系数并重新计算光流场,最后重复光流计算过程直到光流场小波高频分量收敛得到最终的计算结果。仿真实验和临床实验的结果表明:该方法的均方根误差和光流角度误差均小于其他方法,平均对比度噪声比高于其他方法,表现出良好的光流计算精度和边界识别能力,可为心脏病临床诊疗提供一种量化评估的手段。

参考文献/References:

1 ]于潇 . 左心室超声心动图降噪和分割方法研究[ D . 哈尔滨 : 哈尔滨工业大学 ,2012:10 - 11.

2]江业慧,张平洋,冉红,等. 基于AFI的无室壁运动异常冠心病患者左室心肌运动及其与冠脉狭窄程度间关系的研究[J. 中国超声医学杂志, 2020, 36(2): 125-128.

3]向君彦,罗艺,韩小容. 3种超声造影方法评估冠状动脉粥样硬化性心脏病患者室壁运动异常的临床价值[J.检验医学与临床,2018,15(14)2096-2098.

4Ge J Y, Chen T S, Ma C Z, et al. Can intracardiac echocardiography completely replace transesophageal echocardiography to guide left atrial appendage closure?:The comparisons of intracardiac echocardiography with transesophageal echocardiographyJ. Journal of Cardiac Surgery, 2022, 37(9): 2766-2775.

5Alsharqi M, Woodward W J, Mumith J A, et al. Artificial intelligence and echocardiographyJ. Echo Research and Practice, 2018, 5(4): R115-R125.

6Horn B K P, Schunck B G. Determining optical flowC]∥Proceedings of International Society for Optics and Photonics Conference on Techniques and Applications of Image Understanding. Washington D. C., 1981: 319-331.

7Lucas B D , Kanade T . An iterative image registration technique with an application to stereo visionC]∥Proceedings of the 7th International Joint Conference on Artificial Intelligence. Vancouver, Canada: 1981: 674-679.

8Zhang C W, Zhou X D, Zhuge X Y, et al. Learnable optical flow network for radar echo extrapolationJ. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2021,14:1260-1266.

9Zhong Y R, Ji P, Wang J Y, et al. Unsupervised deep epipolar flow for stationary or dynamic scenesC]∥2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Long Beach, CA, USAIEEE, 2019:12087-12096.

10Shah S T H, Xiang X Z, Ahmed W. Optical flow estimation with convolutional neural netsJ. Pattern Recognition and Image Analysis,2021,31(4):656-670.

undefined

相似文献/References:

[1]陈风雷,马正义.热方程边界值决定反问题的数值方法[J].浙江理工大学学报,2016,35-36(自科6):945.
 CHEN Fenglei,MA Zhengyi.Numerical Methods of Inverse Problem of Boundary Value Determination of Heat Equation[J].Journal of Zhejiang Sci-Tech University,2016,35-36(自科三):945.
[2]杨亮亮,胡鑫杰,王秀锋.基于直曲线判断的曲线拟合研究[J].浙江理工大学学报,2017,37-38(自科4):551.
 YANG Liangliang,HU Xinjie,WANG Xiufeng.Research on Curve Fitting Based on Straight Curve Judgment[J].Journal of Zhejiang Sci-Tech University,2017,37-38(自科三):551.
[3]汪亚明,李斌权,韩永华,等.基于改进ELM的撑杆跳起跳过杆轨迹模型研究[J].浙江理工大学学报,2018,39-40(自科1):75.
 WANG Yaming,LI Binquan,HAN Yonghua,et al.Research on trajectory model of pole vault based on improved extreme learning machine[J].Journal of Zhejiang Sci-Tech University,2018,39-40(自科三):75.

备注/Memo

备注/Memo:
收稿日期: 2022-10-12
网络出版日期:2023-01-16
基金项目: 浙江省医药卫生科技计划(2023KY841);义乌市科技项目(20-3-239);浙江省基础公益研究计划项目(LGF20H150006)
作者简介: 张昊(1998-),男,浙江温州人,硕士研究生,主要从事计算机视觉方面研究
通信作者: 童基均,E-mail:jijuntong@zstu.deu.cn
更新日期/Last Update: 2023-09-08