看了DPARSFA 2.1的新增功能后有以下疑问，望严老师指点。
主要是关于这一段话：
Normalize by DARTEL has been added. Details: (1) "T1 Coreg to Fun": the individual structural T1 image is coregistered to the mean functional image after motion correction. (2) "New Segment + DARTEL": New Segment -- The transformed structural image is then segmented into gray matter, white matter and cerebrospinal fluid by using "New Segment" in SPM8. (3) "New Segment + DARTEL": DARTEL -- Create Template, and DARTEL -- Normalize to MNI space (Many Subjects) for GM, WM, CSF and T1 Images (unmodulated, modulated and smoothed [8 8 8] kernel versions). (4) "Normalize by DARTEL": DARTEL Normalize to MNI space (Few Subjects) for functional images. (5) "Smooth by DARTEL": DARTEL Normalize to MNI space (Few Subjects) for functinal images but with smooth kernel as specified, the smoothing is part of the normalisation to MNI space computes these average intensities from the original data, rather than the warped versions.
Questions：
关于Checkbox "Reorient Fun*"模块： 在dcm2nii DICOM to NIfTI conversion官方网站的介绍为“you will only want to orient anatomical scans to canonical space - reorienting fMRI data can disrupt slice timing correction (as the software assumes that the slice order of the stored data is correlated with the time of acquisition”。那么DPARSFA中的这一功能是否会有这个问题？新版DPARSFA是否建议在功能像的预处理中加入这一步以改善后续处理？
Normalize by DARTEL较normalized by unified segmentation有什么优点呢，严老师更推荐用哪一种标准化呢？界面中的New Segment + DARTEL较Segment有什么优势呢？Google搜到有人说New Segment更稳定，提示用DARTEL标准化可能更好吗？
"New Segment + DARTEL": DARTEL -- Create Template, and DARTEL -- Normalize to MNI space (Many Subjects) for GM, WM, CSF and T1 Images (unmodulated, modulated and smoothed [8 8 8] kernel versions). (4) "Normalize by DARTEL": DARTEL Normalize to MNI space (Few Subjects) for functional images. 这两个模块中的Many和Few是什么意思呢？
“Reorient after coreg interactively”，为什么coreg后还有必要Reorient呢？
New Segment + DARTEL的平滑核默认为8 mm吗？这一步平滑核既然已经设置为8 mm，为什么在"Smooth by DARTEL"这一步可以自由设置平滑核？该处设置的平滑核是针对功能像还是结构像？"Normalize by DARTEL" 这一步是用的默认的平滑核吗，与"Smooth by DARTEL"区别仅仅在于一个可以设平滑核，一个不可以？
Submitted by YAN Chao-Gan on Wed, 07/04/2012 - 05:01 Permalink
Re
1. 不会有这个问题。因为Reorient Fun是在Slice timing以及realignment之后的步骤，不会影响slice timing correction. 将功能像和结构像分别手动reorient到一个较好位置，有助于提高coregistration和normalization的准确度。
2. 推荐用New Segment+DARTEL。目前SPM组比较推荐这种方式。另外，也有一篇文章进行了评估：
Klein, A., Andersson, J., Ardekani, B.A., Ashburner, J., Avants, B., Chiang, M.C., Christensen, G.E., Collins, D.L., Gee, J., Hellier, P., Song, J.H., Jenkinson, M., Lepage, C., Rueckert, D., Thompson, P., Vercauteren, T., Woods, R.P., Mann, J.J., Parsey, R.V., 2009. Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration. Neuroimage 46, 786-802.
3. 我的理解：Many Subjects是在DARTEL中指结构像，每个被试只有少数图像。Few Subjects是每个被试都有很多个时间点（很多图像）。
4. Reorient after coreg interactively. 在coreg之后，如果图像没有一个好的orientation，则后续的segment和DARTEL有可能效果不好。当然，如果你已经分别对功能像和结构像进行了reorient，则不用选这一步。
5. 8mm的平滑核是针对结构像的。这个参数我没有留出来自由设置。后面可以自由设置的平滑核都是针对功能像的。