Upgrade to 1.5.1

[sairamgeethanath]

Hi @btasdelen: I noticed you are working on porting to 1.5.0. Please let me know if you have or are planning to merge soon. I see that there is a 1.5.1 out, so I was planning to spend some time on porting as well, but will do so after you have done your PR.

[mcencini]

Hi @sairamgeethanath, I did port some of the 1.5.1 features (rotation and rf shimming extensions). When 1.5.0 is ready, we could start from there if you want, to avoid duplicated effort :D

[sairamgeethanath]

Hi @mcencini - Maybe I got confused or missed something. @btasdelen below probably indicates you might be porting both versions. Please let me know if I have any gaps in understanding, and I'd be happy to start porting if needed. I want to ensure we have it up to date around January for an upcoming project report 😉

[mcencini]

Hi @sairamgeethanath, I do apologize if I caused confusion. I indeed worked on both versions - I opened #328, which is the complete v1.5.0 version last week, just a couple of days after the last message. PR is now waiting for review to be merged - it looks large, but most of the sub-PRs have already been reviewed by @btasdelen and the last changes only regards plotting. It would be maybe best to at least check that it actually works on a scanner running Siemens v1.5.x interpreter - I do not expect major bugs, but you never know.

As for 1.5.1, I did port the two major extensions, namely rotations and rf shimming. To enable rotations, I did have to refactor the gradient continuity check, so I think that after #328 is tested on a scanner by someone and merged I could open a PR with this refactor first, and then open another one with rotations and rf shimming. After that, I think the major missing pieces compared to MATLAB toolbox are the transformFOV class, calcMomentsBtensor() and mr.calcRfPower() / seq.calcRfPower() (and maybe seq.sound()?) but these should be relatively less important compared to rotation. So we could maybe release a v1.5.1 when rotation is in, and then add the other features in v1.5.1post.

[sairamgeethanath]

This sounds like a great plan, @mcencini! I am now unfortunately not at a Siemens site so will focus on sorting that soon. For the rest, I will start focusing on the merges from Monday. Many thanks for all your time and effort!!

[sairamgeethanath]

I will be happy to review 328 on Monday and contribute! Thanks @mcencini!

[sairamgeethanath]

That sounds great, @mcencini! Thanks so much for this wonderful effort. I was getting asked a bit on the timeline so thought I will clear some time for it. Please also checkout this blurb for potential future directions here

[btasdelen]

@mcencini undertook the entirety of the effort for 1.5.0 and 1.5.1. We just need to test/review several relevant PRs and merge them.

[sairamgeethanath]

@mcencini - please let me know if you need any help. My student (@LeoMcBills) and I are happy to help if required, so we can put this out fairly quickly. Second, we also want to switch the license to something closer to MIT so downstream folks (read - interpreter implementers and vendors) dont have issues with adoption.

[mcencini]

Hi! I think PR #308 will introduce some conflicting changes to the plotting tools, which I also updated to support v1.5.x and align with MATLAB toolbox. Hence, it is maybe best to wait until it is merged (@schuenke), then fix the conflict with the v1.5.0_dev branch and then we can convert #319 draft to PR.

#319 should probably need some testing on the scanner, as I do not have a Siemens platform to test on. Then, we can move to 1.5.1, starting from rotations which I think is the most anticipated new feature (@wtclarke). I did split the 1.5.1 rotation implementation in two parts - one updates the gradient continuity check to align with MATLAB toolbox and enable the check in presence of rotation, and the second one implements the actual rotation extension, including plotting and waveforms_export() -like support.

It is not directly related to rotation extension, but maybe #302 could be merged together with these 1.5.1 updates. Does that sounds like a good plan (@btasdelen, @schuenke, @FrankZijlstra)?

EDIT: I am totally in favor of switching to MIT license - I already commented on #324

[schuenke]

#308 should be ready to merge now and afterwards I am happy to help with any potential merge conflicts in the plot function.

[wtclarke]

Hi All, Its great to see how much progress there has been. If you need any help testing things for 1.5.0/1.5.1 then I and a couple of members of my group would be very keen to help (we are very keen to get thing like the rotation matrix interface working).

[sairamgeethanath]

@wtclarke - Pypulseq is community-sustained. Please feel free to help us improve code coverage on both releases.

[wtclarke]

Indeed, I guess I was asking if there was anything specific that you could point us towards where effort would be most appreciated.

[sairamgeethanath]

We are discussing with @mcencini about 1.5.0 (porting and testing). Once he clarifies, we can quickly move. I'd love to chat with you about 1) Sequence library - expansion, testing and distribution; 2) Github spark, FAIR and CARE reviews for Pypulseq as potential areas for contribution. Please let me know if any of these sound interesting to you.

[mcencini]

@wtclarke, @sairamgeethanath: sorry again if I was not clear and did not highlight that #328 was ready to review