Hi all, I used a standard code to compute fwdtimodel = fwdti.FreeWaterTensorModel(gtab) fwdtifit = fwdtimodel.fit(data, mask) FW = fwdtifit.fw FA = fwdtifit.fa MD = fwdtifit.md The MD image has few bins and most of the brain has the same intensity of like 0.000569. Can you pls help me why this would happen? I did check on mailing list, but didnt find any question related to this. [image: image.png]
Hi Madhura, Indeed something went wrong with the free water imaging fit in your data. Can you give more information about your acquisition parameters, e.g. number gradient directions and exact b-values that are you using. Note that this model should only be used in multi b-value acquisition, was that the case? Best, Rafael madhura ingalhalikar <mingalhalikar@gmail.com> escreveu (quarta, 17/01/2024 à(s) 15:33):
Hi all, Thanks for your response. The acquisition is multishell 800 and 2000 bvalues. After using the correct mask and using a threshold for visualization, the image looks better, but not as great as I would expect. I also tried the code on your sample image and get a beautiful output, so nothing is wrong with the code. However, for our images, I haven't corrected for distortion. How much does that affect? Regards, Madhura On Mon, Jan 15, 2024 at 6:38 PM madhura ingalhalikar < mingalhalikar@gmail.com> wrote:
Hi Madhura, Indeed something went wrong with the free water imaging fit in your data. Can you give more information about your acquisition parameters, e.g. number gradient directions and exact b-values that are you using. Note that this model should only be used in multi b-value acquisition, was that the case? Best, Rafael madhura ingalhalikar <mingalhalikar@gmail.com> escreveu (quarta, 17/01/2024 à(s) 15:33):
Hi all, Thanks for your response. The acquisition is multishell 800 and 2000 bvalues. After using the correct mask and using a threshold for visualization, the image looks better, but not as great as I would expect. I also tried the code on your sample image and get a beautiful output, so nothing is wrong with the code. However, for our images, I haven't corrected for distortion. How much does that affect? Regards, Madhura On Mon, Jan 15, 2024 at 6:38 PM madhura ingalhalikar < mingalhalikar@gmail.com> wrote:
participants (2)
-
madhura ingalhalikar -
Rafael Henriques