TrangeTung
commented
1 year ago Hi Rastakhan,
Typically, I chose these voxels as the preliminary “non-brain tissue” mask, which are (1) outside the mouse brain; (2) with high spatial SNR (larger than 15); (3) with high temporal SNR (larger than 15). Considering the different (t)SNR and scanning parameters, you could adjust the threshold as your need. By the way, we did not directly set these non-brain tissue fMRI signals as the noise regressors. As mentioned in our recently published article, we subsequently applied the PCA analysis to extract the PCs, and then set the PCs as the nuisance regressors. e.g. Sleep fMRI with simultaneous electrophysiology at 9.4 T in male mice | Nature Communications Through the PCA analysis, we could obtain stable PCs “regressors”, even the tissue mask with different thresholds behaves differently. In addition, it would be better to manually check of the tissue mask and ensure the voxels of “mussle and jaw” were included.
-- Sincerely, Trange (Chuanjun Tong) Insititute of Neuroscience, Shanghai, China
发送自 Windows 10 版邮件应用
发件人: Rastakhan 发送时间: 2023年6月16日 12:26 收件人: TrangeTung/Mouse_Sleep_fMRI 抄送: Subscribed 主题: [TrangeTung/Mouse_Sleep_fMRI] Threshold of regression of non-braintissue in preprocessing (Issue #2)
Hi, During the preprocessing steps, you’ve set a threshold to determine non-brain tissue in the mouse and regress that part of signal. But I found that this threshold behaves differently in several other data. So I wonder how to set this threshold, by manually checking the tissue mask or by some other specific standards? And whether the whole non-brain tissue should be included? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>
Yiyun-Qi
Hi, During the preprocessing steps, you’ve set a threshold to determine non-brain tissue in the mouse and regress that part of signal. But I found that this threshold behaves differently in several other data. So I wonder how to set this threshold, by manually checking the tissue mask or by some other specific standards? And whether the whole non-brain tissue should be included?