Sorry, I forgot to add something.

In fact, I've used preprocessed-data of HCP (b-value=1000, 2000, 3000) to do QSDR in DSI-STUDIO.

In this case, nothing was wrong, as far as I see...

Best,

Ikko 

2015-03-17 22:03 GMT+09:00 Matt Glasser <[log in to unmask]>:

Well the eddy current distortions will be there regardless of if the SNR is too poor to see them.  We generally prefer analyzing data that gives viewable and artifact correctable results.

Peace,

Matt.

From: 木村一皓 <[log in to unmask]>
Reply-To: FSL - FMRIB's Software Library <[log in to unmask]>
Date: Tuesday, March 17, 2015 at 7:06 AM
To: <[log in to unmask]>
Subject: Re: [FSL] EDDY tool for DSI dataset

Dear Jess,

(Sorry I'm not expert, just your colleague)

 According to "DSI-STUDIO," doing eddy current correction is not recommended for the DSI data for its high b-value.

You might want to see here.

http://dsi-studio.labsolver.org/discussion-group ("DSI-Preprocessing")

And the thesis I've read dealing with DSI-data also didn't do motion-correct, nor eddy-current correction steps.

http://www.sciencedirect.com/science/article/pii/S105381191101130X

So I  think, for DSI-data, skipping these steps would be correct...

Best,

Ikko

Medicine / Medicine

Osaka University / Japan

2015-03-17 20:26 GMT+09:00 Jesper Andersson <[log in to unmask]>:

Dear Jess,

you are right that the current release doesn’t support DSI data. You could try eddy_correct which is also part of the FSL distribution.

I am afraid that I know too little about non-FSL software for eddy current-correction, but maybe someone else has suggestions for something the works with DSI data?

Jesper

On 13 Mar 2015, at 18:18, Jiaqi Sun <[log in to unmask]> wrote:

Hi all,

As far as I know the current release of eddy doesn't work for DSI data. Please correct me if I am wrong. Does anyone happen to know about any alternative method I could use to perform eddy current correction on DSI data?

Any help is highly appreciated!!!

Thanks!

Jess