search

MICA-MNI / BrainSpace

BrainSpace is an open-access toolbox that allows for the identification and analysis of gradients from neuroimaging and connectomics datasets | available in both Python and Matlab |
http://brainspace.readthedocs.io
BSD 3-Clause "New" or "Revised" License
180 stars 72 forks source link

The results for gradients 1 and 2 are reversed #56

Closed HuXinUESTC closed 2 years ago

HuXinUESTC commented 2 years ago

Hi, Thank you for noticing this issue. When I used the BrainSpace to compute the gradeint, I got and visualized the gradient 1 and gradient 2. Contrary to the canonical gradients established by Margulies(2016), it seems that my Gradient 1 is similar to canonical gradient2, while my Gradient 2 is similar to canonical gradient1. Has anything like that ever happened to you? Could you provide some suggestion to me ? My computing environment is Matlab 2018a and BranSpace 0.1.1

HuXinUESTC commented 2 years ago

PS: all my data are normal adults.

ReinderVosDeWael commented 2 years ago

This is a fairly common issue that may occur when the eigenvalue of the sensory gradient is higher than the transmodal gradient's eigenvalue. In general, I wouldn't worry about this and just continue your analysis with these gradients.

Processing strategies may make some difference in this matter. Anecdotally, I've seen the normalized angle kernel change gradient position more often than the cosine similarity kernel.

HuXinUESTC commented 2 years ago

Got it, thanks a lot. Maybe we should define every gradient according to the extreme location of the gradient axis, rather than the propotion of explaned variance.

AstonshisL commented 8 months ago

胡学长，请问是否可以跟您联系？我现在就读于CQUPT，关于梯度的问题研究了许久但无法解决，不知道可否请教您