摘要
目的:验证独立成分分析(ICA)方法在处理视觉运动核磁共振数据中的有效性。方法:将ICA方法应用于视觉运动任务态的功能磁共振的数据处理。选用Fast ICA算法,根据有效的筛选标准选择最佳的独立成分,并将独立成分与功能模板数据进行比较。结果:选用Fast ICA算法进行数据的ICA处理,并选取成分8与功能数据进行对比。结果显示成分8显示的脑部活跃区域与功能数据较为相符。结论:采用Fast ICA方法所分离出来的独立成分,能够比较准确地显示脑部与运动视觉相关的活跃区域,同时也验证了ICA方法在分离视觉信息处理的背侧通路的有效性。
Objective To verify the effectiveness of independent component analysis(ICA) in visuomotor functional magnetic resonance imaging(f MRI) data processing. Methods ICA was applied in the visuomotor f MRI data processing. According to the effective screening criteria, the proper independent components were obtained with Fast ICA algorithm. The obtained independent components were compared with the function template data. Results Fast ICA algorithm was used for the ICA processing of data. Component 8 was selected to compare with the functional data, and the comparison showed that the brain activation areas revealed with component 8 were close to that obtained with functional data. Conclusion Components obtained with Fast ICA method clearly and accurately reveal the brain activation areas related to the visuomotor, which also verifys the effectiveness of ICA in the data processing for dorsal stream.
引文
[1]CHEN T,JIANG C,DING J,et al.The analysis of visual motion tacking[J].Adv Psychol Sci,2012,20(3):354-364.
[2]STRIGARO G,RUGE D,CHEN J C,et al.Interaction between visual and motor cortex:a transcranial magnetic stimulation study[J].Physiology,2015,593(10):2365-2377.
[3]ANTHONY J B,TERRANCE J S.An information-maximization approach to blind deconvolution[J].Neural Comput,1995,7(6):1129-1159.
[4]AAPO H.Fast and robust fixed-point algorithms for independent component analysis[J].IEEE Trans Neural Networks,1999,10(3):626-634.
[5]CHEN M,ZHOU P.A novel framework based on Fast ICA for high density surface EMG decomposition[J].IEEE Trans Neural Syst Rehabil Eng,2016,24(1):117-127.
[6]SAMI S,ROBERTSON E M,MIALL R,et al.The time course of task-specific memory consolidation effects in resting state networks[J].J Neurosci,2014,34(11):3982-3992.
[7]MAB,CHEN J J.Independent component analysis in brain image data[J].Comput Eng,2014,40(3):205-207.
[8]吴微,彭华.Fast ICA和Robust ICA算法在盲源分离中的性能分析[J].计算机应用研究,2014,31(1):31-32.WU W,PENG H.Performance analysis of Fast ICA and Robust ICA on blind sources separation[J].Computer Engineering and Applications,2014,31(1):31-32.
[9]CALHOUN V D,POTLURU V K,PHLYPO R,et al.Independent component analysis for brain f MRI does indeed select for maximal independence[J].PLo S One,2013,8(8):73309.
[10]ZHANG M,CHEN Y,SHEN Y,et al.Classification prediction of Duchenne muscular dystrophy with a machine learning method[J].Journal of University of Shanghai for Science and Technology,2016(2):154-159.
[11]张杰,刘辉,欧伦伟.改进的Fast ICA算法研究[J].计算机工程与应用,2014,50(6):210-212.ZHANG J,LIU H,OU L W.Research on improved Fast ICA algorithms[J].Computer Engineering and Applications,2014,50(6):210-212.
[12]季策,胡祥楠,朱丽春,等.改进的高阶收敛Fast ICA算法[J].东北大学学报,2011,32(10):1390-1393.JI C,HU X N,ZHU L C,et al.Improved higher order convergent Fast ICA algorithm[J].Journal of Northeastern University,2011,32(10):1390-1393.
[13]JI B,ZHANG R,ZHANG J.Gender difference in dynamic thalamocortical functional connections[J].Journal of University of Shanghai for Science and Technology,2016(2):160-167.
[14]DU Y H,GUI Z G,LIU Y J,et al.Review of brain function network analysis methods based on independent component analysis[J].Acta Biophysica Sinica,2013,29(4):266-275.