Outputting auxiliary information

[davecash75]

Hi Pawel- Since the Nifti format cannot hold information about time frames or the tracer and many modelling software needs that information, would it be possible to create a BIDS sidecar or a SIF file as part of the output to mmrchain so that it would be easier to use these files as inputs into tracer kinetics programs? Many thanks David Cash

[jjleewustledu]

Pawel and David:

To gather some information regarding time frames and tracer characteristics, I’ve used pydicom and interfile packages in my own methods that call NIPET. Working codes are at:

https://github.com/jjleewustledu/NiftyPETy/blob/master/respet/recon/reconstruction.py <https://github.com/jjleewustledu/NiftyPETy/blob/master/respet/recon/reconstruction.py>
https://github.com/jjleewustledu/NiftyPETy/blob/master/tests/test_reconstruction.py <https://github.com/jjleewustledu/NiftyPETy/blob/master/tests/test_reconstruction.py>

Method createDynamic demonstrates the essential ideas. Please note that these methods are embedded in a rather elaborate scheme for aligning NAC PET frames, registering mu-maps to each NAC frame and then creating AC frames. Only the subset of methods called by createDynamic are likely to be of interest.

John

On May 2, 2019, at 9:56 AM, davecash75 [email protected] wrote:

Hi Pawel- Since the Nifti format cannot hold information about time frames or the tracer and many modelling software needs that information, would it be possible to create a BIDS sidecar or a SIF file as part of the output to mmrchain so that it would be easier to use these files as inputs into tracer kinetics programs? Many thanks David Cash

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/pjmark/NIPET/issues/6, or mute the thread https://github.com/notifications/unsubscribe-auth/AD4SZZ2M6KUYSOGIAQSAIYLPTL6IHANCNFSM4HJ7EQPQ.

[pjmark]

Dave, this is my plan to use BIDS for this, among other reasons.

John, is your pipeline compatible with BIDS?

Cheers, Pawel

[jjleewustledu]

Pawel,

I’ve not seen many specifications for PET in BIDS and I may be missing significant BIDS requirements. Have you any requirements lists that you advocate?

I’ve tried to follow BIDS except for circumstances for which there is a conflict with idioms from XNAT, in which case I use the XNAT idioms. The json information I record is specific to my scheme for registering mu-maps to time-frames. I’ll do a more thorough comparison with BIDS specifications and get back to you.

John

On May 16, 2019, at 1:04 PM, Pawel [email protected] wrote:

Dave, this is my plan to use BIDS for this, among other reasons.

John, is your pipeline compatible with BIDS?

Cheers, Pawel

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/pjmark/NIPET/issues/6?email_source=notifications&email_token=AD4SZZ2PFOREC2YSTQ3UDADPVWO2VA5CNFSM4HJ7EQP2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVSTPWA#issuecomment-493172696, or mute the thread https://github.com/notifications/unsubscribe-auth/AD4SZZ4UWUAKXPEESDEK5ATPVWO2VANCNFSM4HJ7EQPQ.

[davecash75]

Hi John- The BIDS format for PET is indeed a work in progress, so I think some interim decisions about how best to fill in the auxiliary metadata will have to be taken. I think APPIAN has a reasonable looking first stab at what should be included and that is what I have been working to in terms of the information I have been filling in. @pjmark I see some software that stores the prompts, randoms and trues in order to generate weights for non-linear fitting in kinetic modelling. Is that something that also be added to BIDS?

[jjleewustledu]

Some of my projects may benefit from APPIAN. I’ll fork NIPET and see what I can do.

John

On May 17, 2019, at 05:58, davecash75 [email protected] wrote:

Hi John- The BIDS format for PET is indeed a work in progress, so I think some interim decisions about how best to fill in the auxiliary metadata will have to be taken. I think APPIAN has a reasonable looking first stab at what should be included and that is what I have been working to in terms of the information I have been filling in. @pjmark I see some software that stores the prompts, randoms and trues in order to generate weights for non-linear fitting in kinetic modelling. Is that something that also be added to BIDS?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.