老师您好，关于ALFF two-sample t test有四个问题向您请教

Submitted by cappuccinoxt on Mon, 03/12/2012 - 09:38

老师您好！
在臧老师07年发表的论文中（Altered baseline brain activity in children with ADHD revealed by resting-state functional MRI.Brain Development, 2007)，统计部分two-sample t test采用了uncorrected的校正方式（A two-sample t-test was performed to see the ALFF difference between the two groups. Voxels with a P value <0.01 (uncorrected) and cluster size >270 mm3 (10 voxels) were considered to show significant difference between the two groups.）查阅其它ALFF相关文章后，发现关于ALFF的two-sample t test多采用此种分析方式，e.g:For ALFF comparison, a combined threshold of p < 0.005 and a minimum cluster size of 486 mm3 was used, which resulted into a corrected threshold of p < 0.05 determined by AlphaSim program in AFNI.（Abnormal spontaneous brain activity in medication-naïve ADHD children: A resting state fMRI study.Neuroscience Letters,2011).
我的问题是：1.采用此种校正方式，而不是FWE或FDR，是不是因为ALFF双样本T检验得到的结果通常在[-5,5]之间，无法进行FWE或FDR
2.这种校正方法和FWE/FDR有什么区别
3.虽然您07年的文献没有注明，文中采用的P value <0.01 (uncorrected) and cluster size >270 mm3 (10 voxels) 校正方式，是不是也对应着一个
corrected p value？
4.这种对p值uncorrected，又对体素个数进行限定的校正方式，是否和corrected p value有一一对应的关系？如果有，有公式可循吗？
不知道我是否描述清楚了问题，谢谢老师的耐心解答。
学生

Submitted by YAN Chao-Gan on Mon, 03/12/2012 - 22:21

我的理解：
1. Zang et al., 2007不是AlphaSim校正。
2. FDR只考虑高度，不考虑宽度，SPM的FWE会考虑宽度，但校正非常严。AlphaSim则用随机模拟的方式计算宽度和高度，相对较松一些。详见restfmri.net/forum/course第二部分。
3. 另外一个校正程序Gaussian Random Field theory multiple comparison correction （easythresh in FSL, 来源同SPM的FWE），计算公式见Flitney, D.E., & Jenkinson, M. 2000. Cluster Analysis Revisited. Tech. rept. Oxford Centre for Functional Magnetic Resonance Imaging of the Brain, Department of Clinical Neurology, Oxford University, Oxford, UK. TR00DF1. 会在下一版的REST中出现。