DTI数据分析及应用.ppt

单击此处编辑母版标题样式,单击此处编辑母版文本样式,第二级,第三级,第四级,第五级,*,精选课件ppt,*,DTI,数据分析及应用,1,精选课件ppt,Page 2,内容提纲,DTI,的研究内容,DTI,数据处理流程,DTI Studio,FSL:FMRIBs Diffusion Toolbox,2,精选课件ppt,Page 3,扩散张量成像的研究内容,纤维跟踪算法,基于,DTI,的应,用研究,3,精选课件ppt,D,xy,xy,D,yy,D,yz,=,(,v,1,v,2,v,3,)0,2,0,扩散张量的数学描述,D=,特征分解,特征值,:,1,2,3,0,特征向量,:,v,i,v,j,i jPage 4,D,xx,D,xy,D,xz,1,0 0,v,1,v,2,D,xz,D,yz,D,zz,0 0,3,v,3,4,精选课件ppt,Page 5,确定性跟踪算法,跟踪终止条件,Mori et al.,Ann Neurol,1999,5,精选课件ppt,Page 6,确,定,性,跟,踪,结,果,Catani et al,Brain,2005,粗大的白质纤维束,6,精选课件ppt,Uncertainty,Page 7,纤维走向的不确定性,Jones,MRM,2003,Linearity,Bootstrap,方法,7,精选课件ppt,Page 8,概率跟踪算法,Direction Uncertainty,DTI Noise,Partial Volume Effects,Slide from Tri Ngo,8,精选课件ppt,Page 9,概率跟踪的方法,Non-parametric(model free)approaches,Bootstrap method,HARDI:Q-ball,DSI,Parametric approaches,Prior knowledge and models:Bayesian framework,Probability density function(PDF):local,global,How to estimate the distribution of fiber,orientations within a voxel?,9,精选课件ppt,Page 10,概率跟踪的思想,Reference:,Behrens,T.E.et al.Characterization and propagation of uncertainty,in diffusion-weighted MR imaging.,Magn Reson Med,50,1077-88(2003).,10,精选课件ppt,Page 11,概,率,跟,踪,结,果,Friman et al,IEEE TMI,2006,11,精选课件ppt,Page 12,概率跟踪的优点：,估计纤维走向的不确定性,一定程度上解决纤,维交叉问题,研究,FA,较低的灰质脑区之间的解剖连接,跟踪结果对噪声更稳定,定量描述空间任意两个体素之间的连接概率,概率跟踪的缺点：,需要采集较多梯度方向的,DTI,图像,计算量大，耗时,12,精选课件ppt,Page 13,Connectivity-based classification of,thalamic voxels produces clusters,Behrens et al,Nature Neuroscience,2003,13,精选课件ppt,Page 14,Improvements on the diffusion tensor model,single fibre,multiple fibres,Slide from Saad Jbabdi,14,精选课件ppt,Page 15,确定性跟踪常用软件,:,DTI Studio,MedINRIA,3D Slicer,等,概率跟踪常用软件：,FSL,15,精选课件ppt,Page 16,扩散张量成像的研究内容,纤维跟踪算法,基于,DTI,的应,用研究,16,精选课件ppt,Page 17,扩散属性测度,以上三种情况的,ADC=0.7 x 10,-3,mm,2,/s,17,精选课件ppt,Page 18,18,精选课件ppt,Page 19,19,精选课件ppt,Page 2020,精选课件ppt,Page 21,基于扩散属性测度的临床研究,基于全脑配准的分析方法,基于体素的统计分析（,VBA,）,基于白质骨架的空间统计分析（,TBSS,）,基于感兴趣区的分析方法,手工画感兴趣区的方法,基于纤维重建的定量分析,21,精选课件ppt,Page 22,Voxel-Based Analysis(VBA),VBM on FA(Ashburner,2000;Rugg-Gunn,2001),Strengths,Fully automated&quick,Investigation whole brain,Implementation steps,Preprocessing,Normalization of FA images,Smooth,Voxelwise statistics(e.g.controls patients),Issues,Alignment difficult;smallest systematic shifts between,groups can be incorrectly interpreted as FA change,No objective way to choose smoothing extent(6,8 or 10,mm?),22,精选课件ppt,Page 23,23,精选课件ppt,Page 24,24,精选课件ppt,Page 25,25,精选课件ppt,Page 26,26,精选课件ppt,Page 27,27,精选课件ppt,Page 28,28,精选课件ppt,Page 29,29,精选课件ppt,Page 30,30,精选课件ppt,精神分裂症患者的,VBA,分析,FA,降低的脑区：,cerebral peduncle;,frontal regions;,inferior temporal,gyrus;,medial parietal lobes;,hippocampal gyrus;,Insula;,right anterior,cingulum bundle;,right corona radiata,Page 31,Hao et al.Neuroreport 2006,31,精选课件ppt,Page 32,Tract-Based Spatial Statistics,(TBSS),Part of FSL software,(),Overcome the drawbacks in VBA method,such as,alignment issue and smoothing issue,Flowchart,32,精选课件ppt,Page 33,TBSS steps in detail:,preprocessing,-create FA images from your diffusion,study data,tbss_1_preproc,-prepare your FA data in your TBSS,working directory in the right format,tbss_2_reg,-apply nonlinear registration of all FA,images into standard space,tbss_3_postreg,-create the mean FA image and,skeletonise it,tbss_4_prestats,-project all subjects FA data onto the,mean FA skeleton,stats,(e.g.,randomise),-feed the 4D projected FA data,into GLM modelling and thresholding in order to find,voxels which correlate with your model.,33,精选课件ppt,Page 34,Do cross-subject voxelwise stats on,skeleton-projected FA,34,精选课件ppt,Page 35,Fig.TBSS results from,15 MS patients.A,B:,3D surface renderings,of the mean FA,skeleton.C:Yellow,shows the where FA,correlates negatively,with EDSS disability,score.D:Red as above.,In C and D,green,shows the mean FA,skeleton,blue shows,the group mean lesion,distribution,and the,background image is,the MNI152.,Smith et al.,NeuroImage,2006,35,精选课件ppt,Page 36,Scholz et al.,Nature Neuroscience 2009,36,精选课件ppt,Page 37,TBSS data acquisition requirement:,Voxel size should be less than 3 3 3 mm,3,.,At least one b=0 should be acquired;ideally one b=0,image for every eight diffusion-weighted images.,b-value should be at least 800 s/mm,2,.,At least six-gradient directions must be acquired.it is,better to use more unique sampling directions(with,isotropic angular density18)than to obtain repeat,samples of the same set of directions.,SNR in the diffusion-weighted images should be,maximized.An example protocol that should lead to,sufficiently high SNR is having b=1,000 s mm,2,24,diffusion-weighted images and SNR greater than 15 in,the b=0 image.,37,精选课件ppt,Page 38,The data should not be upsampled(e.g.,through unfiltered,zero-padding during reconstruction)if this is done in such a,way as to introduce ringing into the data.,If multiple repeats of b=0 or diffusion-weighted images are to,be acquired,they must not be averaged on the scanner(as,theymust be coregistered before averaging,and any risk of,averaging the complex data should be avoided).,Fat saturation should be used whenever possible to remove,signal from the scalp,which can disrupt signal in the brain,owing to chemical shift or ghosting artifacts.,A vitamin capsule leftright marker(oil,not water)should be,attached to the right side of the head to avoid any leftright,ambiguities during data conversion and analysis.,38,精选课件ppt,Page 39,Computing equipment:,Unix-based computers.AppleMac(running Mac OS X,version 10.4 or higher)and PCs(running Linux flavors,RedHat 9,Enterprise,FC4,Suse 9.0-9.3 or Debian),High RAM requirements(particularly if tens of,subjects are used in a study),it is likely that the,computer will need to be 64 bit.The computer should,have at least a 1 GHz CPU clock,1 GB RAM,5 GB,s 20 GB free hard disk space.,If multiple networked computers(or a computer cluster),are available,the registration steps can be parallelized,greatly reducing the total computation time.,39,精选课件ppt,Page 40,白质纤维束的定量分析,FA:Left Cingulum(Red)Right Cingulum(Blue),Parameterization process,Gong et al,Human Brain Mapping,2005,40,精选课件ppt,Page 41,同正常人相比，早期盲人的视放射白质扩散异常，,FA,值显著,降低，,ADC,和,23,显著升高。,早期盲人大脑白质扩散异常研究,Shu et al,Human Brain Mapping,2009,41,精选课件ppt,Page 42,单张量模型的假设无法解决,纤维交叉问题,纤维跟踪技术的准确性缺乏严格的评价体系,扩散张量成像的局限性,42,精选课件ppt,Page 43,内容提纲,DTI,的研究内容,DTI,数据处理流程,DTI Studio,FSL:FMRIBs Diffusion Toolbox,43,精选课件ppt,Page 44,数据处理的,基本流程,44,精选课件ppt,Page 45,DWI from Scanner,S,0,S,1,S,2,S,3,S,4,S,5,S,6,45,精选课件ppt,Page 46,Preprocessing,DICOM data conversion,Image quality check,Eddy current correction,46,精选课件ppt,Page 47,内容提纲,DTI,的研究内容,DTI,数据处理流程,DTI Studio,FSL:FMRIBs Diffusion Toolbox,47,精选课件ppt,Page 48,DTI Studio,Download&Install,User Manual,Mailing list,48,精选课件ppt,Page 49,Launching the Program and Hardware,Requirement,DtiStudio-latest-x86.exe,for Windows,system,More than 1GB RAM is recommended,49,精选课件ppt,Page 50,Main Functions,Image Viewer,Diffusion Tensor Calculation,Fiber Tracking and Editing,3D Visualization,Image,ROI Drawing and Statistics,50,精选课件ppt,Page 51,How to do tensor calculation,and fiber tracking?,51,精选课件ppt,Page 52,E:workTrainingExampleData,Raw data:MRIcroN dcm2nii.exe,(.img,.hdr,.bvec,.bval),Eddy current correction:AIR,Tensor,FA,MD calculation:DTIstudio,Fiber tractography:DTIstudio,ROI selection,52,精选课件ppt,Page 53,第一步：对原始,DICOM,数据进行格式转换。利用,MRIcroN,软件,中的,dcm2nii.exe,工具，将,DTI,原始数据文件夹拖入，即可得到,DTI,扫描的梯度编码文件,.bval,和,.bvec,，以及转换后的,NIFTI,格式,的图像文件（,Output Format,选择,4D NIfTI hdr/img,）。,53,精选课件ppt,Page 54,第二步：对,DTI,图像进行头动和涡流校正。打开,DTI studio,File-MRI View3D,选中上一步得到的,4D.img,文件，,Image Parameters,中选择,Image,为,Analyze,，点击,OK,，然后在,Image,面板,Image Processing,区域选择,Automatic,Image Registration(AIR),，按图,3,进行设置，然后点击,OK,，等图像配准完成后，,在,Image,面板的,Orthogonal Views,区域的文件下拉框中看到,Air_,开头的一系列文,件，为校正后的,DTI,图像文件，点击,Save,，将,Air_,开头的所有文件选中，选择,Raw Data,，保存为一个,4D,的,.dat,文件。,MRI View,3D,参数,54,精选课件ppt,Page 55,AIR,的参数,设置,55,精选课件ppt,Page 56,头动和涡流校正后的,DTI,图像保存,56,精选课件ppt,Page 57,第三步：张量解算以及,FA,ADC,等扩散指标的计算。打开,DTI studio,File-DTI Mapping,选择,Philips REC,格式，,Continue,，按图,5,进行参数,设置，,Add a file,中选中上一步保存的,4D.dat,文件，点击,OK,，在,DtiMap,面板的,Calculation,区域选中,Tensor,Color Map etc.,（计算,ADC,值,选择,ADC-Map,）根据图像选择噪声水平，点击,OK,，然后等,DTI,Studio,算完后在,Image,面板的,Orthogonal View,区域可看到计算出来的各,种扩散属性文件。对于,想要保存的文件，如,FA,EigenVector-0,，,Color,Map-0,等可以分别进行,Save,（,.dat,格式），便于下一次查看和使用。,57,精选课件ppt,Page 58,DtiMap,面板进行张量解算,58,精选课件ppt,Page 59,各种扩散属性的显示,59,精选课件ppt,Page 60,第四步：纤维跟踪及可视化。第一种方法：基于前面步骤，在,DtiMap,面板的,Fiber Tracking,区域点击,Fiber Tracking,，然后进行参数设置，点击,OK,，就会进,行基于全脑体素（,FA0.2,）的纤维重建；第二种方法：如果上一步已保存,FA,和,Eigen Vector-0,文件，可重新打开,DTI Studio,File-Fiber-Tracking,，选上,FA-,Map,文件和,Principle Vector,文件，并进行参数设置，点击,OK,，就会进行基于全,脑体素（,FA0.2,）的纤维重建。通过任何一种方法，算完后右下角会出现,Fiber,面板，再此面板中可以对特定纤维束进行显示和编辑，并可以对纤维属性进行统,计分析。,60,精选课件ppt,Page 61,纤维跟踪方法,2,61,精选课件ppt,Page 62,纤维跟踪的参数设置,62,精选课件ppt,Page 63,重建纤维束的可视化,63,精选课件ppt,Page 64,64,精选课件ppt,Page 65,Major white matter tracts,Reference:,Wakana S,Caprihan A,Panzenboeck MM,et al.,Reproducibility of,quantitative tractography methods applied to cerebral white matter,.,Neuroimage,2007,36(3):630-644.,65,精选课件ppt,Page 66,纤维属性,的统计分,析,66,精选课件ppt,Page 67,New Modules,ROIEditor,ROI drawing,selection and manipulation,User defined ROI statistics,Atlas&WMPM based statistics,Tract specific probabilistic quantification,DiffeoMap,Image registration,67,精选课件ppt,Page 68,内容提纲,DTI,的研究内容,DTI,数据处理流程,DTI StudioFSL:FMRIBs Diffusion Toolbox,68,精选课件ppt,Page 69,FSL,FDT:FMRIBs Diffusion Toolbox-DWI,Analysis and Tractography,eddy_correct,dtifit,bedpostx,probtrackx,DTI data with,30 gradient directions,69,精选课件ppt,Page 70,70,精选课件ppt,Page 71,Processing steps:,cd/fsl_course_data/fdt/subj1,dcm2nii,Bring up fslview and open data,Extracting nodif and generating a brain mask,Correcting for eddy current and head motion,Dtifit-fitting diffusion tensors to the data,Bedpostx(2530h/person),Probtrackx,71,精选课件ppt,Page 72,Commands:,eddy_correct sub001.nii data 0,fslroi data nodif 0 1,bet nodif nodif_mask-m-f 0.3,dtifit-k data-m nodif_brain_mask-r,bvecs-b bvals-o dti,bedpostx,可参考,72,精选课件ppt,Page 73,Reference Book,Email:,73,精选课件ppt,Page 74,谢谢大家！,74,精选课件ppt,此课件下载可自行编辑修改，供参考！,感谢您的支持，我们努力做得更好！,75,精选课件ppt,