Better Brain Implants With Ultrathin Carbon Fiber Electrodes 82
An anonymous reader writes "A new neural interface delicate enough not to damage nerve tissue, but resilient enough to last decades has been made. Made from a single carbon fiber and coated with chemicals, the technology is believed to be fully resistant to proteins in the brain. From the article: 'The new microthread electrode, designed to pick up signals from a single neuron as it fires, is only about 7 micrometers in diameter. That is the thinnest yet developed, and about 100 times as thin as the conventional metal electrodes widely used to study animal brains.
“We wanted to see if we could radically change implant technology,” says Takashi Kozai, a researcher at the University of Pittsburgh and the lead author on the paper, published today in the journal Nature Materials. “We want to see an electrode that lasts 70 years.”'"
The brain moves. (Score:5, Interesting)
The brain shifts in the skull, especially during impact.
Any rigid strong wire risks being ripped out, as the brain stretches, or doing the cheese wire thing.
Cheesewired brain is bad.
I wouldn't get too excited... (Score:4, Interesting)
Not to burst anyone's sci-fi bubble, but 7 microns is really more of an incremental improvement in terms of size. In our lab we already use 12 micron wire on a regular basis, but honestly we use 25 or 50 micron wire more often: larger diameter wire equals better signal quality from lower resistance. Impedance of neural electrodes is usually on the order of 10^4 ohms, you don't want to go much higher unless you really enjoy getting miserable signal-to-noise ratios. And if you can't get a signal, it doesn't matter how good your coating is or how much residual damage might be caused.
In terms of the long-term argument, they're going to need prove recording durability for longer than 6 weeks if this is really going to work. 6 weeks is impressive, but nowhere near the decades-long durability the summary is talking about.
Still A Long Way Off (Score:5, Interesting)
In order to listen to a neuron for long, or help people control a prosthetic as they do a natural limb, the electrodes need to be able to survive for years in the brain without doing significant damage. With only six weeks of testing, the team couldn’t say for sure how the electrode would fare in the long term, but the results were promising. "Typically, we saw a peak in immune response at two weeks, then by three weeks it subsided, and by six weeks it had already stabilized."
The electrode has to last for years (the summary says they're shooting for 70), but they only have six weeks of successful testing. The acute rejection subsided, but it could become a chronic, repeated rejection. With artificial hearts, acute rejection is most likely to occur in the first 3 to 6 months. Six weeks seems like a short time for this. Obviously the brain is a very different organ, but part of the reason they're pursuing this is because science knows far less about the brain than it does the heart.
Sensor, not stimulator technology (Score:5, Interesting)
As lab animal sensor technology, it's interesting. As a *stimulator* technology, it's fairly pointless. The current spread has to be enough to actually trigger nearby nodes of Ranvier for myelinated, or to stimulate significant physical areas of myelinated nerves. And stimulating them directly, electrically, requires enough charge deposited to cause hydrolisis around sych fine electrodes. Unless you can magically get the electrodes by the nodes of Ranvier, and *keep* them there or encourage the nodes to keep reforming there for the life of the electrode, you're screwed.
Oh, and *forget* ever doing an MRI on someone with these in their nerves. The likelihood of forming loops in such fine fibers is very high, and they *will* couple electromagnetically to the MRI, with big pulses of current going around the loops and both thermally cooking and mechanically yanking their way around the brain tissue when the MRI pulses.
Re:Sensor, not stimulator technology (Score:2, Interesting)
Oh, and *forget* ever doing an MRI on someone with these in their nerves. The likelihood of forming loops in such fine fibers is very high, and they *will* couple electromagnetically to the MRI, with big pulses of current going around the loops and both thermally cooking and mechanically yanking their way around the brain tissue when the MRI pulses.
Holy damn shit.
That's a whole new nightmare I'd never heard or thought of for cortical implants, especially during the phase when they're common enough that people outside medical studies have them, but rare enough nobody at Podunk County Hospital knows to check for them before a NMR scan.