Hi, I have a question about the nilearn correction. Do I need to transform spm t map into z map then conduct correction in nilearn?
the function nilearn.glm.threshold_stats_img document say: in stat_img Niimg-like object or None, optional. Statistical image (presumably in z scale) whenever height_control is ‘fpr’ or None, stat_img=None is acceptable. If it is ‘fdr’ or ‘bonferroni’, an error is raised if stat_img is None.
In my case, I ran a 2nd lvele two-sample t test in spm and got two t maps: spmT001.nii and spmT002.nii. I want to use nilearn to do a fdr/fwer correction and visualize them.
imgG = image.load_img('spmT_0001.nii')
fdrmap,fdrvalue=threshold_stats_img(imgG,alpha=0.05, height_control='fdr')
plotting.plot_stat_map(fdrmap,
threshold=fdrvalue,
cut_coords=(-13,39,43),
cmap='hot',
black_bg=True,
title='group interaction,fdr=0.05')
print(f'the fdr=0.05 threshold is {fdrvalue:.3g}')