基于MATLAB与VC混合编程的脑电偶极子源定位.pdf

25 5 河 北 工 业 科 技 Vol. 25, No . 5 2008 9 H ebei Journal of Industrial Science and Technology Se t. 2008 : 1008-1534( 2008) 05-0287- 04 MAT LAB VC+ + 1 2 1 , , ( 1. 沧州职业技术学院信息工程系, 河北沧州 061001; 2. 天津大学仁爱学院计算机科学与技术系, 天津 301636) : 脑电活动源可 用一个或几个偶极子等效脑电偶极子源定位就是通过改变等效偶极子 的参数使正向计算所得头皮电位与测量电位拟合, 在数学上表现为最小均方误差的非线性代价函 数, 这是一个非常复杂的优化问题, 求解通常依赖于非线性优化算法提出了基于MAT LA B 与 VC+ + 混合编程的仿真计算利用了MAT LAB 强大的遗传算法工具箱和VC+ + 界面编程容易 和代码执行效率高的优点实验结果表明, 采用混合编程的仿真计算极大地提高了编程效率, 减少 了系统资源的消耗 : 偶极子源定位; 遗传算法; MAT LAB; VC+ + ; 混合编程 : T P391 : A Simulation of EEG di ole source localization based on mixed rogramming of MAT LAB and VC+ + 1 2 1 YU Guo-li , ZHANG Lei , SANG Jin-ge ( 1. Information Engineering De artment, Cangzhou Vocational College of Technology , Cangzhou Hebei 061001, China; 2. De artment of Com uter Science and T echnology, Renai College of T ianjin University, Tianjin 301636, China) Abstract: Active neurons w ithin small regions of the brain are often modeled as one or several current di oles. The roblem of di ole source localization is solved by changing the di oles arameters so that the ot ential of scal calculated by the forw ard roblem is equal to measured otential. It is re resented by the least-square-error non-linear ex ensive function in mat hematics. It is a very com lex roblem and its solution de ends on the algorithm of non-linear o timization. This a er ro oses the sim- ulat ion based on mixed rogramming of MATLAB and VC+ + . It takes advantage of genetic algorithm of MATLABs too-l box and makes use of the easy interface rogramming and the high