search

joshbrew / DIY_DOT

DIY Diffuse Optical Tomography board plans. Wiring not finished yet, just publishing meow and updating later.
MIT License
3 stars 0 forks source link

Not so far future (in about 3-7 years from now) #1

Open dumblob opened 3 years ago

dumblob commented 3 years ago

In 2018 European Union launched one of the biggest programs ever (in terms of money at least; lasting 10 years) - namely Quantum Flagship. And one of their goals is Sensing brain activity in a way which overcomes the (extremely) low resolution current technologies allow.

In 7 years from now, there will be commercial products offering these new "quantum" technology guts (first 3 years are for research finalization, next 4 years for commercial analyses and prototyping, last 3 years for finalization of commercially sellable "better-prototypes").

If you want to pursue the brain imaging etc. further, then this might be the only (as of now) known way how to make any serious advancement. So feel free to keep an eye on the gradual outcomes of the program.

And when EU (and its citizens) already paid for this, it'd be "dumb" to not use the open... stuff it produced.

dumblob commented 3 years ago

Speaking about future, there is also a current state of the art technology for moderately fast but very precise voxel (brain, tissue, eye, ...) imaging - called OCT.

And I'd like to point out a DVD/CD drive based DYI version https://spie.org/news/5807-multiple-reference-optical-coherence-tomography-for-smartphone-applications?SSO=1 . This DYI stuff might be of high interest to you.

joshbrew commented 3 years ago

Quantum, huh? I have read about some of the ways they might achieve this, but interferometry is still underdeveloped and more within my scope (ala OCT). Problem with quantum stuff is you generally need the latest materials science and fancy clean room labs. I've want to play with ultrasound too but there are not really cost effective small piezo arrays or these newer capacitive arrays they make (forgot what they're called, MEMs I think).

And yes OCT is the dream, I've thought about how to do it for a while but this DIY version might just be what I needed, the lenses are the biggest problem for cost but if a CD is enough then damn I was overthinking it haha. I've seen CD spectrometers but not CD OCT so I am thoroughly impressed. My idea was that you could probably scale up to get coarse grained FNIRs images in 3D if you were a little extra clever about it, would kill resolution but the idea is just to get something better than DOT. There is 3D angiography with OCT that is really wicked and the brain lends itself well to it if you can filter the skull out somehow. There's something called gaussian windows you can use to do that to some degree and squeeze extra information out from deeper volumes, and I know of a handful of RF and astronomy tricks that might also help. It'll take me a while to replicate any of this myself, though, one step at a time!

dumblob commented 3 years ago

Quantum, huh? I have read about some of the ways they might achieve this, but interferometry is still underdeveloped and more within my scope (ala OCT). Problem with quantum stuff is you generally need the latest materials science and fancy clean room labs. I've want to play with ultrasound too but there are not really cost effective small piezo arrays or these newer capacitive arrays they make (forgot what they're called, MEMs I think).

That's all totally true - I've given it some time yesterday and found out the Quantum Flagship overestimates the achievements in the upcoming years regarding brain sensing. So yeah, no need to hurry as I thought originally :wink:.

And yes OCT is the dream, I've thought about how to do it for a while but this DIY version might just be what I needed, the lenses are the biggest problem for cost but if a CD is enough then damn I was overthinking it haha. I've seen CD spectrometers but not CD OCT so I am thoroughly impressed. My idea was that you could probably scale up to get coarse grained FNIRs images in 3D if you were a little extra clever about it, would kill resolution but the idea is just to get something better than DOT. There is 3D angiography with OCT that is really wicked and the brain lends itself well to it if you can filter the skull out somehow. There's something called gaussian windows you can use to do that to some degree and squeeze extra information out from deeper volumes, and I know of a handful of RF and astronomy tricks that might also help. It'll take me a while to replicate any of this myself, though, one step at a time!

This sounds much more feasible. I'll follow your GitHub activity to see where your steps are going as a DYI OCT for brain would be, how to put that, a dream? damn cool? awesome? Maybe all that and more :wink:.