[Neuroimaging] Has anyone tried using dipy's piesno function?

[Samuel St-Jean]

It's the opposite, piesno is made for diffusion data and estimate_sigma is
not. Both will give weird result if the scanner changes the intensity of
the background. And finally, I am totally biased toward piesno since the
math behind it makes sense, provided you have stationnary background on
each slice. Scanners don't always respect that one for esthetic purposes
apparently.
On Sep 12, 2015 3:00 AM, "Bago" <mrbago at gmail.com> wrote:

> Hi all,
>
>   I've been playing around with dipy's piesno function and wanted to know
> if anyone else had much experience with it. I've found that most of our
> data has very low intensities in the background and that piesno tends to
> grossly underestimate the noise in these data sets. I've tried estimating
> the noise a few other ways including the estimate_noise function from the
> same module.
>
> I know that piesno was not developed for diffusion data, but wanted to
> know if anyone is regularly using it for diffusion data and how well it's
> working for them.
>
> Thanks,
> Bago
>
> _______________________________________________
> Neuroimaging mailing list
> Neuroimaging at python.org
> https://mail.python.org/mailman/listinfo/neuroimaging
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/neuroimaging/attachments/20150912/93cf937b/attachment.html>