High-resolution postmortem human brain MRI at 7 tesla | Hacker News Reader

High-resolution postmortem human brain MRI at 7 tesla

pulkit-khandelwal.github.io

122 pointspulks17 comments

This tool provides performs segmentation, parcellation and registration of ultra high-resolution (< 300 microns) postmortem human brain hemisphere at 7 tesla t2w MRI. This pipeline leverages advances in both deep learning and classical surface-based modeling techniques. The developed method allows us to perform vertex-wise analysis in the template space and thereby link morphometry measures with pathology measurements derived from histology.

Open articleView on HN

For anyone whose brain is still in their head and/or doesn't have 7T data, there is also FreeSurfer[1] to play with.

I only really know about the physics of MRI, so I admit I don't fully understand what the advantage of this project is against other work; I assume it would have a lot to do with the fact that the data is so much higher resolution and can identify finer structure within the brain.

[1] https://surfer.nmr.mgh.harvard.edu/

> admit I don't fully understand what the advantage of this project is against other work

Replying to myself here; seems that the combination of being ex vivo and 7T means that the resolution increase over a typical in vivo patient scan is really really tremendous since your dead brain is happy to remain completely still through multi-hour sequences. The resolution you can get is primarily dependent on field strength but like many things in signal processing, you can trade integration time for higher resolution to a point. So this dataset is at about 100micron while 7T in vivo can only get down to maybe 300-1000micron (best case depending on patient movement)

Dang this sucks, I had all the files from my brain MRI when I got Dengue Fever but my laptop recently got stolen out of my car and the CD they gave it to me on was in it. This would have been fun to mess with. The MRI facility is in another country so not sure how easy it would be to get them to send me the data again (although they should I gave them 3k dollars).

Just ask them. I don’t know which country it’s in but they may be legally obligated to send it to you (maybe for a modest fee), regardless of which country you’re from. They usually are in developed countries but YMMV everywhere else.

Every country is different but I'd be surprised if they don't still have it.

The irony is they will probably still have to burn it on a CD and mail it to you. Good old DICOM!

Yeah I'm going to try and get it now. That sucks though, because that laptop was the last device I had that had a cd drive! Maybe an excuse to finally get one of those modded vintage Thinkpads with modern mobo/cpu.

There's always USB CD/DVD drives you can use, or I've even seen external USB enclosures for Desktop and Laptop CD/DVD drive form factors which you can just shove one into.

Ironically an external USB DVD reader will probably be cheaper than getting your hands on that CD...

Almost certainly you can probably get it from them again.

We changed the url from https://github.com/Pulkit-Khandelwal/purple-mri to the project page, which is a bit more explanatory. Readers may want to look at both, of course.

There is something a bit spooky about looking at a dead brain. To think that a person lived in that their entire life, and now it's just lying on a table not doing anything.

I guess that's true for whole bodies and all of their parts too, but something about the abstraction of removing the skin, bones, face, etc that we are used to interacting and distilling to the seat of consciousness and emotions makes it spookier.

7T is so rad.

edit: I know it was a lazy comment. but I did truly mean it. I did high-resolution imaging of the hippocampal subfields in my PhD dissertation, and I wished so much to have access to a 7T for structural imaging of those subfields...So powerful.

The Biospec 15.2 Tesla MRI is now sold for more than 10 years…

It has an 11cm bore. You can’t fit a brain inside it. 7T is the cutting edge of useful MRI.

We can cut it in 8 brain cubes.

For those using a smaller bore diameter, 7 T is 300 MHz proton frequency, and the engineering challenges to maintain a homogeneous field over 600 mm are immense.

I find myself dwelling on Karaoke/Cold Lazarus more and more as technology progresses.

It seems to exist on Youtube. https://www.youtube.com/watch?v=IwYlzSQTL2A&list=PLmRu2axUu2...

It's hard to say how it's relevant without spoiling it too much though.

Reply on news.ycombinator.com