基于总变差正则化算法的三维肺部呼吸过程电阻抗成像方法  

Three-dimensional electrical impedance imaging method of pulmonary respiration based on total variation regularization algorithm

作　　者：王琦[1,2] 陆纪璇 李秀艳[2,3] 段晓杰[2,3] WANG Qi;LU Ji-xuan;LI Xiu-yan;DUAN Xiao-jie(School of Life Sciences,Tiangong University,Tianjin 300387,China;Tianjin Key Laboratory of Photoelectric Detection Technology and System,Tiangong University,Tianjin 300387,China;School of Electronic and Information Engineering,Tiangong University,Tianjin 300387,China)

机构地区：[1]天津工业大学生命科学学院,天津300387 [2]天津工业大学天津市光电检测技术与系统重点实验室,天津300387 [3]天津工业大学电子与信息工程学院,天津300387

出　　处：《天津工业大学学报》2021年第6期53-60,共8页Journal of Tiangong University

基　　金：国家自然科学基金资助项目(61872269,61601324,61903273);天津市自然科学基金资助项目(18JCYBJC85300);天津市企业科级特派员项目(18JCTPJC61600)。

摘　　要：为了提高对肺部呼吸运动进行连续监测的精确性,针对二维成像不能满足人体结构成像需求的问题,利用三维电阻抗断层成像技术(electrical impedance tomography,EIT)对肺部呼吸过程进行成像,采用总变差正则化(total variation,TV)算法与Tikhonov正则化算法实现肺部三维EIT成像,并对2种算法在不同呼吸状态下的图像重建结果的误差与肺区域比(lung regional ratio,LRR)进行对比与评价。结果表明:与传统Tikhonov正则化算法相比,仿真数据中,TV算法使三维肺部EIT成像误差降低了58.86%,肺区域比(LRR)准确率提高了14.58%;实验数据中,TV算法使成像误差降低了53.81%,LRR准确率提高了16.33%;说明TV算法比Tikhonov正则化算法更适合进行三维EIT图像重建工作,证明了三维EIT用于人体呼吸运动监测的可行性。In order to improve the accuracy of continuous monitoring of pulmonary respiration movement, aiming at the problem of two-dimensional imaging cannot meet the needs of human structure imaging, three-dimensional electrical impedance tomography(EIT) technology was used to image the pulmonary respiration process. The 3 D EIT images of lung were reconstructed by total variation(TV) algorithm and the Tikhonov regularization algorithm, and the lung regional ratio(LRR) and the errors of image reconstruction results of the two algorithms under different breathing states were compared and evaluated. The results show that compared with the traditional Tikhonov regularization algorithm, the simulation data from the TV algorithm reduces the imaging error of 3 D lung EIT by 58.86% and improves the accuracy of lung area ratio(LRR) by 14.58%;the experimental data from the TV algorithm reduces the imaging error of 3 D lung EIT by 53.81% and improves the accuracy of LRR by 16.33%.It can be concluded that compared with Tikhonov regularization algorithm, TV algorithm is more suitable for reconstruction of 3 D EIT image, which proves the feasibility of 3 D EIT for human respiratory movement monitoring.

关 键 词：总变差正则化算法 肺部三维成像 电阻抗断层成像技术 肺部呼吸 连续监测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]