一种基于加权频差阻尼最小二乘的准静态EIT算法  

Quasi-static EIT algorithm based on weighted frequency difference damped least square

作　　者：杨琳[1] 徐灿华[1] 付峰[1] 代萌[1] 董秀珍[1] 

机构地区：[1]第四军医大学生物医学工程学院,西安710032

出　　处：《仪器仪表学报》2013年第8期1879-1885,共7页Chinese Journal of Scientific Instrument

基　　金：国家科技支撑计划项目(2011BAI08B13);国家自然科学基金(51207161);国家自然科学基金重点项目(50937005);国家自然科学基金面上项目(61071033);陕西省自然科学基金(2012JQ4012)资助项目

摘　　要：传统的准静态电阻抗成像算法不能减少背景在重构结果中造成的伪影,导致无法进行有效成像。针对这一问题,提出一种基于加权频差阻尼最小二乘法的成像算法。该算法对成像数据采用加权频差的处理方式,克服了背景对成像结果的影响;同时结合阻尼最小二乘重构算法的正则化方法,改善了EIT逆问题求解的病态性。算法功能在Visual C++环境下完成,并通过仿真实验和物理模型实验对该算法的有效性进行验证。结果表明,无论模型的背景阻抗分布是否随频率变化,该算法均能够准确地检测出扰动目标。因此,该算法是一种有效的准静态电阻抗成像算法。Traditional quasi-static electrical impedance tomography algorithm can not reduce the artifacts caused by the background, so the effective imaging can not be achieved. Aiming at this problem, an algorithm of quasi-static electrical impedance tomography （QS-EIT） based on weighted frequency difference damped least-square （WFD-DLS） is pro- posed. The method adopts the weighted frequency difference approach to process the imaging data,which overcomes the influence of the background on reconstruction results. Combined with the damped least-square reconstruction regulariza- tion method proposed by our group, this new algorithm improves the ill-posed property of the solution for the inverse EIT problem. The algorithm was implemented in Microsoft Visual C ＋ ＋ environment ; and the simulation experiments and physical model experiments verified the effectiveness of the algorithm. The results show that this algorithm could accurately detect the perturbation target in background no mater whether the model background impedance distribution varies with frequency or not. Therefore,the proposed algorithm is an effective QS-EIT algorithm.

关 键 词：准静态电阻抗成像算法 仿真实验 物理模型实验 

分 类 号：R318.0[医药卫生—生物医学工程]