脑功能光学成像的迭代广义指示函数分析法  

Recursive Generalized Indicator Functions Method for Analyzing the Optical Imaging of Functional Brain

作　　者：黄晓斌[1] 胡德文[2] 刘亚东[2] 周宗潭[2] 万建伟[1] 董国华[2] 

机构地区：[1]国防科技大学电子科学与工程学院,湖南长沙410073 [2]国防科技大学机电工程与自动化学院,湖南长沙410073

出　　处：《电子学报》2006年第4期664-669,共6页Acta Electronica Sinica

基　　金：国家自然科学基金(No.30370416);国家杰出青年科学基金(No.60225015);高等学校优秀青年教师教学科研奖励计划

摘　　要：在脑光学功能成像领域,由T.Yokoo等人提出的广义指示函数法(Generalized Indicator Functions;G IF)能够在极低信噪比下有效地提取大脑行为模式图.但进一步研究发现,该算法在处理脑功能光学图像序列时存在计算量大的缺点.为解决一问题,本文将W eng等在处理FERET人脸数据库时提出的一种迭代算法与G IF算法相结合,给出了一种迭代格式的G IF算法———RG IF(Recursive G IF),RG IF算法利用迭代计算的特点能大幅削减计算量.利用仿真和实验数据对G IF和RG IF算法进行了对比分析,结果表明RG IF不仅能够大大节省计算时间,同时检测效果与G IF相当.In the analysis of optical imaging of functional brain,the generalized indicator functions （GIF） algorithm presented by T. Yokoo,etc. is an efficient method to extract the brain activity map. But further study shows that this algorithm has the shortage of heavy computation in dealing with brain image series. In order to resolve this problem, a recursive GIF （RGIF） algorithm is presented,which is the combination of Weng＇s recursive algorithm in dealing with the FERET face database and the GIF algorithm, the RGIF algorithm can sharply reduce the computation utilizing the characteristic of recursive algorithm. We compare the GIF/and RGIF algorithms using the simulated and experimental datum, the results show that the RGIF algorithm can relieve the computational burden substantially with at almost the same computing precision as that with GIF algorithm.

关 键 词：脑光学功能成像 广义指示函数法 迭代广义指示函数法 

分 类 号：Q632[生物学—生物物理学]