Dear Jay,
as Vasilis has pointed out eddy will output as many volumes as you put in by default. If you have paired diffusion directions (i.e. each diffusion gradient is acquired twice with opposing PE) you have the option to specify --resamp=lsr (and probably you need --fep) which uses a model similar to applytopup to combine the two images into a single image with some of the resolution from the severly distorted areas recovered. SO, in that case you would get 31 volumes.
I am a little curious about your line
> I see that the output image file 'eddy_corrected_data' still has 62 uncorrected volumes.
Does it really look like eddy hasn’t corrected them?
Jesper
On 19 Oct 2015, at 13:55, SUBSCRIBE FSL Jay <[log in to unmask]> wrote:
> Hello,
>
> I have two DTI scans acquired in opposite phase-encoding directions (one b0 and 30 diffusion-weighted directions). I am trying to perform eddy to correct for eddy current distortions, movement and susceptibility artifacts. EDDY does not output the expected corrected and merged DTI data. The following are the steps that i have followed as instructed in the manual
>
> 1. fslroi data_raw_dtiA.nii.gz b0_blip_up 0 1
>
> 2. fslroi data_raw_dtiP.nii.gz b0_blip_down 0 1
>
> 3. fslmerge -t b0_blip_up b0_up_down b0_up_down
> fslmerge -t merged_data_AP data_raw_dtiA data_raw_dtiP
>
> 4. topup --imain=b0_up_down --datain=template_list --config=b02b0.cnf --out=my_topup_output --iout=my_hifi_b0
>
> 5. fslmaths my_hifi_b0 -Tmean my_hifi_b0_avg
> 6. bet2 my_hifi_b0_avg my_hifi_b0_avg_BET -m -n -g -0.125
>
> 7. indx=""
> for ((i=1; i<=31; i+=1)); do indx="$indx 1"; done
> for ((i=32; i<=62; i+=1)); do indx="$indx 2"; done
> echo $indx > index.txt
>
> 8. eddy --imain=merged_data_AP --mask=my_hifi_b0_avg_BET_mask --acqp=template_list --index=index.txt --bvecs=bvecs --bvals=bvals --topup=my_topup_output --out=eddy_corrected_data
>
> template_list consists of
> 0 -1 0 0.1
> 0 1 0 0.1
> bvecs (contains 62 rows : one b0 gradient followed by 30 diffusion-weighted gradients (AP acquisition) + one b0 gradient followed by 30 diffusion-weighted gradients (PA acquisition))
> bvals (contains 62 rows corresponding to bvecs)
>
> I see that the output image file 'eddy_corrected_data' still has 62 uncorrected volumes.
> Can you please tell me where in the processing steps am i doing wrong?
>
> best wishes
> Jay
|