I’m very excited to see how this progresses and how A.I. can help solve medical issues that have up to now stumped us.
One really interesting thing they’ve reported about this is that it requires a head set to connect the brain implants to the spine electrodes, but after a while the patient was able to walk using a stroller without the headset because the body regrew nerve tissue and Reenabled some connection to the brain.
It's very cool, and uses the same principles as lots of other neurorehab electrical devices out there. I used an FES ('Functional Electrical Stimulation') device for a while that has a little accelerometer in it to figure out when I intended to take a step, and then stimulate (via conductive pads placed atop the skin) the correct muscles in the lower leg to lift the ankle up (dorsiflexion). Over time, when the brain correlates X input with Y result, even if Y is very very weak, bruised nerves can recover (in the case of recent brain/spinal injuries) and new nerves can grow. I was advised to have the device on its 'training' setting (where it would periodically activate a 'step') even when lounging on the sofa. Hebb said: "What fires together, wires together"...
My suspicion (tho I'd love to be educated on this) is that the reference patient did not have a truly 'complete' spinal cord injury, but rather one where there were at least some nervecells remaining/bruised. If there was a complete section of nerve wiring missing, I don't see how re-growth could occur (but perhaps there's some stem-cell wizardry that can help with that???). Anyway, it wouldn't matter much; if these devices become small and convenient enough, it's not much burden to wear them forever.
This is very exciting! One possibility that is, I think, covered up by all the hype around "AI works like our brains" (which I disagree with) is that: our brains are too complex for us to understand and ML may allow us to build more capable interfaces that we still don't understand - but which are easier to understand than the brain. It's an extremely promising new avenue for incremental progress.
>"AI works like our brains" (which I disagree with)
While AI works like our brains is patently false it might just be analogous enough to not matter in the end if the details are the same to get similar emergent behavior.
"AI works like our brains" is indeed generally not so, but CNNs do work like the visual cortex, somewhat--it's where they got the idea for them in the first place.
As a student I was taught that some day stem cells would allow sci patients to recover. Instead, hard tech seems to be the more apparent and workable solution.
What is the practical reason to favor electrical implants over electromagnetic ones? One would conclude that using electromagnetic induction would be a no-brainer when it is very clear the body works against electrical probes (by fighting the implants). Could someone more knowledgeable than myself clarify this.
It's very difficult (ie: impossible) to target individual nerve fibres with induction. Direct stimulation is the only way to achieve the required resolution.
Bio-compatibility issues are not intractable. We know about many biomaterials that the body doesn't have a strong adverse reaction to. I'm sure people are developing materials that are both conductive and bio-compatible.
The amount of current required for magnetic stimulation is much higher and not compatible with implantable devices. TMS requires large coils and currents relative to electrical stimulation.
Speak for yourself there, I've been in favor of brain augments essentially ever since I was aware of them as a concept as a teen and hoping they'd arrive in my lifetime. Personally it's because I don't hold most of my identity as sacred and would like to have full control over my automatic behaviors and thought patterns.
Always good to see novel uses of ML. However, let's not pretend this will ever offer any benefit to people with spinal cord injuries.
Any research that does not involve repairing the physical damage to the spinal cord itself is a distraction. Anyone who claims otherwise does not understand the physiology involved or is being deliberately disingenuous.
It's hard to think of an apt analogy, but perhaps consider a patient who walks into a doctor's office with a broken arm. Instead of setting the break and applying a cast, the doctor offers them a referral to a psychologist to help them adjust to life with a broken arm.
> Moreover, neurorehabilitation supported by the BSI improved neurological recovery. The participant regained the ability to walk with crutches overground even when the BSI was switched off. This digital bridge establishes a framework to restore natural control of movement after paralysis.
So, not only did the brain–spine interface ("BSI") itself allow the patent to walk when they couldn't before (which strikes me as an absolute win, even if you call it a distraction) but it also facilitated neurological recovery over the longer term.
> Always good to see novel uses of ML. However, let's not pretend this will ever offer any benefit to people with spinal cord injuries.
Apart from the person in their study who can now walk, that had the spinal injury (with the use of a crawler)...
In relation to your second point, you are aware regenereative science and this implant approach can progress in parallel? This offers something tangible, plus with more refining can be rolled out en masse. What regenerative science can be applied now that would give such a quality of life improvement?
[+] [-] phobotics|2 years ago|reply
One really interesting thing they’ve reported about this is that it requires a head set to connect the brain implants to the spine electrodes, but after a while the patient was able to walk using a stroller without the headset because the body regrew nerve tissue and Reenabled some connection to the brain.
[+] [-] padolsey|2 years ago|reply
My suspicion (tho I'd love to be educated on this) is that the reference patient did not have a truly 'complete' spinal cord injury, but rather one where there were at least some nervecells remaining/bruised. If there was a complete section of nerve wiring missing, I don't see how re-growth could occur (but perhaps there's some stem-cell wizardry that can help with that???). Anyway, it wouldn't matter much; if these devices become small and convenient enough, it's not much burden to wear them forever.
[+] [-] aeturnum|2 years ago|reply
[+] [-] carlmr|2 years ago|reply
While AI works like our brains is patently false it might just be analogous enough to not matter in the end if the details are the same to get similar emergent behavior.
[+] [-] ElfinTrousers|2 years ago|reply
[+] [-] belter|2 years ago|reply
[+] [-] dr_|2 years ago|reply
[+] [-] barelyauser|2 years ago|reply
[+] [-] 0xbadc0de5|2 years ago|reply
[+] [-] jupp0r|2 years ago|reply
[+] [-] __loam|2 years ago|reply
[+] [-] etrautmann|2 years ago|reply
[+] [-] __loam|2 years ago|reply
[+] [-] _joel|2 years ago|reply
[+] [-] verisimi|2 years ago|reply
"Please give us a neural lace!"
Says no one... yet. But the priming is underway!
[+] [-] ninjanomnom|2 years ago|reply
[+] [-] 0xbadc0de5|2 years ago|reply
Any research that does not involve repairing the physical damage to the spinal cord itself is a distraction. Anyone who claims otherwise does not understand the physiology involved or is being deliberately disingenuous.
It's hard to think of an apt analogy, but perhaps consider a patient who walks into a doctor's office with a broken arm. Instead of setting the break and applying a cast, the doctor offers them a referral to a psychologist to help them adjust to life with a broken arm.
[+] [-] pdabbadabba|2 years ago|reply
> Moreover, neurorehabilitation supported by the BSI improved neurological recovery. The participant regained the ability to walk with crutches overground even when the BSI was switched off. This digital bridge establishes a framework to restore natural control of movement after paralysis.
So, not only did the brain–spine interface ("BSI") itself allow the patent to walk when they couldn't before (which strikes me as an absolute win, even if you call it a distraction) but it also facilitated neurological recovery over the longer term.
[+] [-] _joel|2 years ago|reply
Apart from the person in their study who can now walk, that had the spinal injury (with the use of a crawler)...
In relation to your second point, you are aware regenereative science and this implant approach can progress in parallel? This offers something tangible, plus with more refining can be rolled out en masse. What regenerative science can be applied now that would give such a quality of life improvement?
[+] [-] germinalphrase|2 years ago|reply
[+] [-] snapcaster|2 years ago|reply
[+] [-] SpliffnCola|2 years ago|reply