Slashdot Mirror
Monitoring Brain Activity With Mesh Electronics
An anonymous reader writes: Medical researchers have long known that bioelectronics could substantially improve patient diagnosis and treatment, but the difficulty in putting that circuitry into place kept more traditional options at the forefront. Now, a team of scientists has found a clever way to deliver flexible electronic meshes via syringe, which could make it easier to monitor complex brain activity without dangerous surgery. "The scientists demonstrated they could inject a 2mm wide sample of the mesh through a glass needle with an inner diameter of only 95m. During injection, the mesh structure continuously unfolds as it exits the needle. Injection of the mesh through a needle with a 600m inner diameter produced similar results." The team has already tested the technique on rodents, and found minimal response from astrocytes, cells involved in repairing damaged brain tissue. They were able to record the rodents's brain activity as well.
95 meters is kinda big for a needle
The system, Tiny, Angstrom-level Regional Dermal Injection System, or TARDIS, benefis both medicine and research.
(-1: Post disagrees with my already-settled worldview) is not a valid mod option.
95m is right about 300 feet. I don't thing such a "needle" could inject anything. Squash the target maybe, but only if the rim hits it.
In unrelated news: 'm' is not an abbreviation of 'micron'.
Have slashdot editors reached a new low? Or do they want us to make "your momma" jokes?
"The scientists demonstrated they could inject a 2mm wide sample of the mesh through a glass needle with an inner diameter of only 95m.
The more significant achievement seems to be, at least to me, creating a syringe with an inner diameter of 95 meters.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
...had the "micro" prefixes where appropriate. Thus:
"The scientists demonstrated they could inject a 2mm wide sample of the mesh through a glass needle with an inner diameter of only 95m. During injection, the mesh structure continuously unfolds as it exits the needle. Injection of the mesh through a needle with a 600m inner diameter produced similar results."
... ideally what we really need to get to eventually is the point where we can read all neurons at the same time. Injectable meshes aren't going to cut it for that.
The best I can envision is injecting bioluminescent proteins into the brain that flash when different types of activation or chemical concentration are achieved in different neurons. Ideally they'd flash at different frequencies for different cells by having the color adjusted by various local concentrations of chemicals that vary between cells, but it'd be workable even if you only have one color for all neurons. You'd then need numerous 360-degree digital photosensors (wired or wireless - ideally the latter) scattered throughout the brain, recording the flashes of light and correlating the different directions the light comes from with corresponding neuron activations.
The key difference between that and a mesh is that a mesh has to touch and individually measure each cell - nearly a hundred billion perfectly positioned connections, each connection with the hardware to encode the activity into packets to transmit out of the brain. But if you have one optical micro sensor that can read the flashes from a million nearby cells, you don't need nearly 100 billion individually wired connections, you need 100.000 microsensors, each with the processing power of a modern digital camera. No, we're not to that point with today's technology. But we're sure a heck of a lot closer to that than to being able to wire up nearly 100 billion neurons individually.
Of course, such a situation could be run in reverse: adding photosensitive proteins to neurons to cause or suppress activation in them, and having the ability to project light at high resolution in 360 degrees to trigger or suppress individual neurons. However, I expect that'd be more challenging without having neuron-specific activation frequencies. It's one thing to see a flash passing through a dozen neurons on the way to the sensor and figure which neuron was activated based on the shape of the flash, but if you're projecting through parts of a dozen different neurons to try to activate one further away, you risk activating them as well.
To me, the ultimate goal of all of this would be to enable a gentle end-of-life transition to a digital mind. If we reach the point where we can accurately simulate brain function given sufficient input data, then we can one by one: A) pick a neuron, B) begin to simulate it, C) suppress its actual firing, D) activate its neighbors based on its simulated firing, E) trigger apoptosis in the no-longer-needed neuron. Then move on to its neighbors, one by one, until there's nothing left of the brain but the simulation.
"99 dead duelists of Dios on the wall. 99 dead duelists of Dios! Take one's ring, pass it around..."
Shades of John Crichton's implant ...
"Ahh! I see you're in that indeterminate Schrodinger state where - oh, uh
Don't like the "girlie" part, limited cognition is not gender based.
However, perhaps they suffer from microcephaly https://en.wikipedia.org/wiki/Microcephaly which we would just render as cephaly.
The missing micron is quite a bit funnier if you've just skimmed another recent story submission:
We don't even need to bring up Tepco, which is just as well since plutonium is a different beast. We are talking plutonium, aren't we?
Mars Climate Orbiter
No worries. Better theirs than ours.
If News for Nerds still can't handle Unicode in 2015, I think the human race needs to pull in their horns, and stick to long baths and the companionship of bright-yellow rubber duckies.
I'm not usually scared of needles, but I'll be steering well clear of that one.