[转载]ROI analysis 感兴趣脑区分析 资料汇总
2016-07-10 15:29阅读：
Marsbar,FSL是操作ROI分析的软件：
关于ROI的制作介绍：
http://miykael.github.io/nipype-beginner-s-guide/regionOfInterest.html
ROI的讨论：
http://www.brainvoyager.com/ubb/Forum4/HTML/000380.html
感兴趣脑区的两篇重要文献：
Brett, M., Anton, J., Valabregue, R., & Poline, J. (2002).
Region of interest analysis using the MarsBar toolbox for SPM 99.
Neuroimage, 16, S497
MarsBar toolbox
的介绍。
Poldrack, R. A. (2007). Region of interest analysis for fMRI.
Social cognitive and affective neuroscience, 2(1), 67-70
这篇文献的作者也是：
《Handbook of fMRI data Analysis》by Rusell A. Poldrack
et al ；该书有章节专门介绍ROI，思路与该片文献一致。但文献似乎更加清晰些。为此，我对文献进行总结吧：
看特殊脑区；
ROIs for exploration
ROIs for statistical control：
a.界定ROI信号有这里提及了两种方法：
其中之一，{regions specified very
large--->signal from small voxels with noises;}--->(to
threshold statistical map&count number)--->sensitive to
specific threshold& unreliable measures of
acitivition.(选一个的脑区界定特别大，那么就对阈值非常敏感，对激活脑区的测量不够稳定)
另外一种方法：ROI used for statistical
control(restrict voxel-wise analyses to a set of ROIs and control
for multiple comparisons in those voxels)--->no signal
specification;but no provide same level of insight into pattern of
activation.
（统计上控制ROI，具体就是对ROIs的集合进行voxel-wise分析的限制，并控制这些voxels的多重比较。但这样的不足就是：no
signal specification;but no provide same level of insight into
pattern of activation.）
ROIs for functional
specification：用于检验Voxels的集合其在功能上的同步性，进而看这集合对某项任务的敏感性。这样的研究用于视觉加工中比较多；这样用“localizer”scan来决定ROIs{这里需要界定：localizer,
Localizer:
All our scan sessions begin with a single-slice, three-axis
localizer scan that gives a view of the subject’s head in the three
scanner-frame axes.源自：http://cbs.fas.harvard.edu/science/core-facilities/neuroimaging/information-investigators/MRphysicsfaq#dropout}
很多情况下，定义有ROI的scan是不是来自要分析的scan，是从其它的SCAN里界定的。有研究者认为，最好把的localizer的facotrial
design的分析里面包括 感兴趣的对比。-----这一段读的比较虐心，逻辑上我没弄透。待来日再弄明白。
Defining ROIs：解剖上，或功能上来界定ROI。
其中
解剖上，需要界定基于每个被试自己解剖坐标（e.g.ALL
atlas; talairach atlas）来定义每个被试的ROI。作者建议，做好用atlas-based
ROI;且定义ROI基于probabilistic atlases of macrosopic antatomy or
probalilistic atlases of Brodmann's areas which are as part of SPM
anatomy toolbox.
功能上，一种方法：
use
a separate ‘localizer’ scan to
identify voxels in a particular anatomical region that show a
particular response； these voxels
are then explored to examine their response to some other
manipulation.（用单独 localizer
scan来确定在特定解剖脑区来显示特定的反应的）。另外一种方法：Alternatively,functional ROIs can be created using orthogonal contrasts in a factorial design
(Friston et al.,2006). Exploratory
ROIs are often created by placing small spheres at local
maxima in the statistical map; this provides a set of ROIs that
span the clusters of interest. Because the goal of exploratory ROIs
is not statistical control, it is also acceptable to place ROIs in
anatomical areas of interest (using one's best judgment about the
placement), particularly for examining null results; however, it is
again critical to note that although this kind of analysis can be
useful for exploration it must not be used for inference since it
is heavily
biased.(以统计图最大的定位激活来定一个球，这个要的ROI不需要统计控制；也可以在检验虚无结果时，在感兴趣的解剖区域定义ROI。但需注意的是：尽管这样的分析可以
用于探索，但由于其存在很大偏侧，故不能进行inference。)
此外，还有基于前人研究来定义ROI，但是这样的定义为了避免偏差，最好是from
meta-analyses of the domain or task of interest. 这里有很大关于这样的方法 for
meta-analysis of functional imaging studies (e.g. Turkeltaub
et al., 2002; Wager and Smith,
2003), and these methods can be used to
generate ROIs that will be less sensitive to noise than those based
on single-study activations.
取ROI时，需要界定阈值。
从marsbar的使用说明来说，定义ROI不能拿需要分析的任务数据中激活脑区来定义ROI，需要那单独（independent）data来定义ROI，之后再去任务激活脑区提取ROI数据进行统计分析。
关于ROI界定，Vul, E., Harris, C., Winkielman, P., & Pashler, H.
(2009). Puzzlingly high correlations in fMRI studies of emotion,
personality, and social cognition. Perspectives on psychological
science, 4(3), 274-290 这篇文章批判了一篇文章，一石激起千层浪。有很多关于这篇文献的反对意见的。
============
DPARSFA里的坐标出处
1. AAL Atlas
The AAL atlas is the same as provided in MRIcroN. You can find the
details of each region from
http://www.restfmri.net/forum/DPARSF_V1_0#comment-373
Center of Mass
2. Harvard Oxford Atlas
The atlas is based on the Harvard Oxford atlas files (thresholded
at probability 25%) provided in FSL:
HarvardOxford-cort-maxprob-thr25-2mm.nii.gz (cortical)
HarvardOxford-sub-maxprob-thr25-2mm.nii.gz (subcortical)
Given the order in the original files were not easy to use in
extracting ROI time series (e.g., the left right ROI share the same
index), the ROIs were re-indexed as detailed in the
{DPARSF}TemplatesHarvardOxford_Atlas_NewIndex_YCG.xlsx
Center of Mass
3. Dosenbach 160 ROIs
The ROI center is extracted from Table S6 in Dosenbach et al.,
2010. Prediction of individual brain maturity using fMRI. Science
329, 1358-1361. The ROIs were summarized from meta-analysis.
ROI Cennter
4. Andrews-Hanna’s default mode network ROIs
The ROI center is extracted from Table S1 in Andrews-Hanna et al.,
2010. Functional-anatomic fractionation of the brain's default
network. Neuron 65, 550-562.
5. Craddock’s clustering ROIs
The ROIs were using the ROI files provided in
http://www.nitrc.org/projects/cluster_roi/ as described in Craddock
et al., 2012. A whole brain fMRI atlas generated via spatially
constrained spectral clustering. Hum Brain Mapp 33, 1914-1928. The
ROI names were not provided yet from that website.
Center
of Mass
