B1+ and flip-angle mapping
wex...@googlemail.com
Dear All,
We are currently conducting a multi-centre study on the UK's MRI scanners. We intend to publicly release the data of a traveling heads study conducted at each site. The data includes a number of structural, task fMRI and field-mapping sequences. Whilst we will be releasing the data as a collection of DICOM files, we are also keen to release it concurrently in BIDS format.
As the data is collected at 7T the scans are affected by strong B1+ (RF transmit) inhomogeneity. We therefore measure B1+ as part of the protocol. However, as far as I can see the current BIDS specification does not support B1+ or flip-angle mapping. I believe, that as the amount of ultra-high field scanning increases, supporting B1+/flip-angle data will be important for the specification.
Is there any support for this type of data in the upcoming specification releases?
If not, could I propose that the specification adopts a syntax for storing this type of data?
I would propose that it would either fit in the “anat” or “fieldmap” categories. Whilst it would traditionally be seen as fieldmapping I can foresee some confusion with B0 maps.
There are many different ways of measuring flip-angles and b1 maps. I do not suggest the specification tries to encode all of these different methods, but provides a tag, or tags, for the derived fieldmap/flip-angle map.
There are several ways that the information can be encoded (i.e. different units).
1. Maps can be expressed as a percentage, or fraction, of the target flip-angle. This is constant across acquisitions as long as the reference level is not changed (reference voltage on Siemens, drive scale on Philips).
2. Maps can be a direct measurement of the flip-angle achieved by a certain pulse at a certain amplitude.
3. B1 maps express the field produced by a coil in an object for a given instantaneous drive power or voltage. This can be expressed in many different units, e.g. tesla, hertz, radians/s etc. The maps might also rarely be expressed “per-volt” or “per-watt”.
It is possible to move between any of these options given knowledge of the pulse in the sequence.
I hope what I’ve suggested seems vaguely sensible and relevant. And I look forward to hearing some thoughts on this.
Best wishes,
Will Clarke
FMRIB Centre
University of Oxford
Foreseeable issues:
· Many possible units. E.g. (m/u)T or Hz or degrees or percent. “Per-volt”, “per-watt”.
· Must clarify what sequences the B1/flip-angle map is relevant to.
· Clarity on the term fieldmap. Both B0 and B1 maps are called fieldmaps, with context providing differentiation.
· Many different measurement sequences. Store only final derived map?
Chris Gorgolewski
Dear All,
We are currently conducting a multi-centre study on the UK's MRI scanners. We intend to publicly release the data of a traveling heads study conducted at each site. The data includes a number of structural, task fMRI and field-mapping sequences. Whilst we will be releasing the data as a collection of DICOM files, we are also keen to release it concurrently in BIDS format.
As the data is collected at 7T the scans are affected by strong B1+ (RF transmit) inhomogeneity. We therefore measure B1+ as part of the protocol. However, as far as I can see the current BIDS specification does not support B1+ or flip-angle mapping. I believe, that as the amount of ultra-high field scanning increases, supporting B1+/flip-angle data will be important for the specification.
Is there any support for this type of data in the upcoming specification releases?
If not, could I propose that the specification adopts a syntax for storing this type of data?
I would propose that it would either fit in the “anat” or “fieldmap” categories. Whilst it would traditionally be seen as fieldmapping I can foresee some confusion with B0 maps.
There are many different ways of measuring flip-angles and b1 maps. I do not suggest the specification tries to encode all of these different methods, but provides a tag, or tags, for the derived fieldmap/flip-angle map.
There are several ways that the information can be encoded (i.e. different units).
1. Maps can be expressed as a percentage, or fraction, of the target flip-angle. This is constant across acquisitions as long as the reference level is not changed (reference voltage on Siemens, drive scale on Philips).
2. Maps can be a direct measurement of the flip-angle achieved by a certain pulse at a certain amplitude.
3. B1 maps express the field produced by a coil in an object for a given instantaneous drive power or voltage. This can be expressed in many different units, e.g. tesla, hertz, radians/s etc. The maps might also rarely be expressed “per-volt” or “per-watt”.
It is possible to move between any of these options given knowledge of the pulse in the sequence.
I hope what I’ve suggested seems vaguely sensible and relevant. And I look forward to hearing some thoughts on this.
Best wishes,
Will Clarke
FMRIB Centre
University of Oxford
Foreseeable issues:
· Many possible units. E.g. (m/u)T or Hz or degrees or percent. “Per-volt”, “per-watt”.
· Must clarify what sequences the B1/flip-angle map is relevant to.
· Clarity on the term fieldmap. Both B0 and B1 maps are called fieldmaps, with context providing differentiation.
· Many different measurement sequences. Store only final derived map?
--
You received this message because you are subscribed to the Google Groups "bids-discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bids-discussion+unsubscribe@googlegroups.com.
To post to this group, send email to bids-discussion@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bids-discussion/b9f94887-e083-4a7a-866c-562eb104ca5e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
wex...@googlemail.com
To unsubscribe from this group and stop receiving emails from it, send an email to bids-discussi...@googlegroups.com.
To post to this group, send email to bids-di...@googlegroups.com.
Chris Gorgolewski
Hi Chris,Sorry for the long delay in answering these questions."fmap" with a distinct suffix does sound best. _flipmap or _b1map?
I'm afraid there is no standard across the platforms I am familiar with (Siemens and some Philips). I would suggest that the most common and immediately usable is percentage of the target flip-angle. This can be applied to any sequence in a study as long as the target flip-angle is known for any of those sequences. It is also equivalent to percentage of target B1 (in tesla) for all constant frequency RF pulses.I think that I would recommend starting with a fairly restricted definition. I.e. the measurement relative to the target. If there is a need for more the units field can be used to extend the definition.
To unsubscribe from this group and stop receiving emails from it, send an email to bids-discussion+unsubscribe@googlegroups.com.
To post to this group, send email to bids-discussion@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bids-discussion/da1bd51c-191a-4c5f-b52b-0da4a79ca9a2%40googlegroups.com.
gilles.de...@gmail.com
Op donderdag 21 december 2017 01:59:25 UTC+1 schreef Chris Gorgolewski: