4 ms·
The real problem with brain implants is that they are still high risk. You do not want to do them unless you have no other options. There's no way to put them
by XorNot 13y ago
The real problem with brain implants is that they are still high risk. You do not want to do them unless you have no other options.
There's no way to put them in without inherently causing some amount of tissue damage - so you're committed to brain damage occurring at the moment.
And once in, the good ones, you can't take out - neurons grow into them and trying to pull them out will take a chunk of tissue with them.
If there's ever going to be a significant uptake in such technology, these are all issues we have to solve - i.e. we need a way of getting them wired in non-destructively, and a way to undo that non-destructively. That's well into science fiction realms - nanotechnological wiring which carefully weaves between cells and the like.
- jessriedel 13y ago> neurons grow into them and trying to pull them out will take a chunk of tissue with them. Do you have a cite for this?
- XorNot 13y agoI could dig one up, but basically it was something which happened when they tried to - and succeeded at - solving the immune rejection problem with the glass ones. Modifying the surface to prevent a layer of scar tissue building up made the implants much more capable, but the consequence was that they were now bonded to the brain permanently essentially. Never used in people - but the point stands - making a good connection to the brain involves something similar.
- dm2 13y agoIt wouldn't be necessary to pull one out unless you needed to increase the bandwidth of the connection. A BCI would probably be able to be unplugged. There is also no reason that it couldn't be wireless. http://www.extremetech.com/extreme/149879-brown-university-creates-first-wireless-implanted-brain-computer-interface http://www.extremetech.com/extreme/149879-brown-university-c... As far as putting them in it isn't 100% necessary to damage the brain when installing one. The BCI connection could "grow" around and into a part of the brain to create the connection. Safety would be the #1 concern when building the implants, so even if something went wrong hopefully the damage would be limited or non-existent.
- caycep 13y agoThe old way they used to do DBS implants in France was to use for the pre-implantation mapping this horrific looking 5-electrode array, the "Ben-Gun", that they'd just jam through the frontal lobe. A lot of surgeons this side of the atlantic viewed it (with horror and disdain) as sort of boring a nice 5 cm wide tunnel through the brain...fortunately being used less often these days.
- caycep 13y agoActually the DBS implants for Parkinson's disease are relatively safe, and are being used more and more often, frequently to correct adverse effects from medications...it's about a 1-2% chance of a minor hemorrhage with infection rates differing among centers. The VA had something terrible, like 9% but the series out of UCSF did better. These are all treatable, or at worst, you just remove the device. In terms of removal, it's actually easier with age, for some odd reason. Surgeons report that after a few years, they slide out. A lot of handwaving as to why, but some thought is that there's some sort of encapsulation layer that builds between the surface and the neural tissue, or some layer of fluid space or whatnot. It's a matter of perfecting the tech. The experimental BCI stuff - i.e. Utah arrays, are still externalized and are risky, and yes they do have a ways to go before ready for population-wide deployment..
- malandrew 13y agoHow complex is the surgery for these? For example, do we already have some sort of skull/brain endoscope whereby you drill a little whole and do the surgery by wire, or do you still have to open up the skull to install the implants?
- caycep 13y agoIt's not too bad. More like the "brain endoscope" route, except it's just a think micro electrode to confirm the target, then the actual wire itself. Nothing optical (although our surgeon does use an optical endoscope for a different type of surgery, pituitary tumor removal). Pretty standard stereotactic stuff - metal rigid frame (google Leksell or CRW/Luminant for examples) gets bolted into the outermost bone layer (w/ local anesthesia) that puts the head into an XYZ cartesian coordinate system. The complex stuff is the planning - MRI images have to be taken w/ the frame on, and there's some linear image registration that happens via software. Once the surgeon plans the trajectory (basically a straight line to the targeted brain structure), it's relatively easy - the incision/craniotomy is basically a ~1 cm or less burr hole drilled into the skull, and a micro drive advances first a micro electrode (for confirmation via single unit nerve firing recordings every ~ 0.5 mm or so), then the actual electrode, which is a little over 1 mm in diameter. The surgeon may adjust the trajectory by a few mm either way if he decides he doesn't like the recordings along the length of the trajectory. The battery/onboard electronics is in a pacemaker-like device that is just put in under the skin in the chest...the tunneling of the wire from the skull to to chest is arguably the hardest part surgically. Epilepsy brain mapping, on the other hand, is a whole lot more complex, with much more signals processing and data acquisition involved (a bunch of 8x8, 16x16 or more grids of ~ 0.5-1cm electrodes recording local field potentials, and a open craniotomy where you take off a large chunk of skull)