search

KamitaniLab / GenericObjectDecoding

Demo code for Horikawa and Kamitani (2017) Generic decoding of seen and imagined objects using hierarchical visual features. Nat Commun https://www.nature.com/articles/ncomms15037.
150 stars 46 forks source link

Why does the length of data in 'ROI_VC' is different in two files? #24

Closed dlmdlmacy closed 2 years ago

dlmdlmacy commented 2 years ago

Hello! Firstly, thank you for your excellent work! I have one question about the dataset.

I downloaded the data at figshare. There are two kinds of files: Subject1.h5 and Subject1_ImageNetTraining/ImageNetTest/Imagery.h5 (take subject1 for example). I use bdpy to select 'ROI_VC' data, but the length is 4466 and 3444 respectively. Why does this happen? Is this because one was using SPM to preprocess and the other was using fMRIPrep?

Thank you!

ShuntaroAoki commented 2 years ago

Yes, the difference is due to different preprocessing. SPM was used in the original study (Horikawa & Kamitani, 2017), and recently we switched our lab's preprocessing pipeline to fmriprep. We confirmed that the results were well consistent between SPM and fmriprep in our in-house study.

dlmdlmacy commented 2 years ago

Thanks a lot for your help! I'll close this issue.