Monday night and all’s well…
So it’s that time of year again. The various tentacles of the MRI research community have once again come together to share and borrow ideas. This year we’re in Stockholm, a town that’s close to my heart, and so I thought I’d blog a few stray observations about the conference and my experiences of it. One caveat, though, before I begin: ISMRM is gigantic, and one person can’t possibly see it all. So I’m going to cover my own meandering without regard for the larger picture. In the interests of brevity, I’ll try to stick to the positives.
Today started as a set-up day. I went to check on my poster (1688, for the terminally interested) and found that Hubert had done a sterling job of pinning it up for me. At least, that was what I though until I got within 10 feet of hte board, when both my poster and the one that it shares the board with unceremoniously fell off. It took about four attempts to get it hung up again — apparently one of the “features” of ISMRM this year is posters boards that don’t really like pins very much. Hey-ho. Whilst there I ran into Cagatay Demiralp, who also had a poster, and chatted to him for a while as well as to a PhD student from Cambridge who showed us his e-poster which used Catagay’s tract-colouring scheme.
The first session I made it to was Abdominal and whole-body diffusion imaging. This isn’t a field I know very well, and it was interesting to see what was going on. The session was very clinical, with lots of applications of DWI for cancer. Since actually using diffusion imaging in a clinical setting isn’t something that I see very often, it was interesting to see all the application. In particluar there were a couple of talks on DTI in the kidney which were very interesting. Quite a diversity of approaches — from simple (apparently unnormalised) DW images through to some nice tensor and tractography stuff.
Next up was Brain anatomy and diffusion imaging. More of the usual gang here, with talks from Geoff Parker and Derek Jones. Derek’s talk was about an interesting niggle to do with TBSS to do with orientational dependence in the statistics, and it developed quite a bit of discussion. One of the things that Derek is very good at is identifying important, practical issues in imaging and coming up with useful solutions. This was very much in that vein, although Steve JOnes and Co were obviously very involved in the work and had come up with a fix to the issue to do with a correction from non-stationary statistics. Given that this problem affects just about every TBSS study ever done, this was clearly something that people were interested in. Good stuff.
There was also a couple of talks about validation — this is good as there aren’t enough of these. It’s good to see some numbers put on things like tractography connectivities, the whole field needs it. Couldn’t help but feel a little stung when he dismissed software phantoms as “not taking the imaging process into account in any meaningful way”, though. ouch!
What was good about this session was that there were a lot of good and novel ideas on show. Too often talks can be incremental, or a small variation on some other technique that may or may not work. Today we saw some genuinely imaginative thinking, and some new approaches to interesting problems. Hopefully this bodes well for the rest of the meeting.