Artificial arms get closer to the real thing | Hacker News Reader

Artificial arms get closer to the real thing

nature.com

47 pointsColinWright13 comments

You know, I've seen all these neural interfaces that connect electrical impulses to nerve cells, and I wonder: why aren't we using optogenetics to connect them instead? it would seem to me that using light to penetrate the skin rather than metal would make for a much longer lived interface, since you wouldn't have to worry about the body rejecting the metal. Is there some reason that's not used?

I work in the field, so I thought I'd chime in here. Our lab (and others) have published a few papers on using optogenetics to control peripheral nervous system targets, but as some of the other commenters have mentioned, it's very much in the early stages right now, and is likely to be a long way away from human trials.

A few points:

1. There are two relevant ways to make motor neurons light sensitive. As another commenter mentioned, you can create a transgenic line of animals (this is necessarily restricted to mice/rats). That's been done, and it's been shown that you can control muscle movement in such mice [1]. In that work, a rather interesting, and now replicated [2] observation was made, which is that if you use optogenetics to do this (muscle control) you actually induce less fatigue than if you use electrical stimulation to do this. Why this is the case is not yet understood.

2. The other (more relevant) way to do this would be to use viral vectors to transduce the motor neurons you want to control with the light-sensitive ion channels. We published a paper last year showing this was possible in rats [3]. The advantage here is that you can get control over individual muscles without any effect on unwanted muscles, through a subtlety of the injection method used. There are naturally many caveats here, the most important being that human gene therapy is still in its infancy, and is typically used only for very life-threatening disorders. It's also important to note that in both these cases, you still needed a way to deliver light to the relevant nerve. While there's lots of work going on to make light-sensitive ion channels even more light-sensitive, it's likely that preliminary optogenetic control of muscle will still require implants.

3. One of the applications talked about in the link here goes the other way - it's focused on delivering touch information back to the brain. There's less published here with optogenetics than there is in motor control. We've published a paper using viral methods to control sensation earlier this year (in mice) [4], but the type of sensation we were trying to control was pain, not simple touch.

That said, I know that many groups are trying to develop ways to specifically make touch-transducing neurons light sensitive - I'd be surprised if we didn't see quite a few papers on this over the next year or two.

[1] http://www.nature.com/nm/journal/v16/n10/abs/nm.2228.html [2] http://www.sciencemag.org/content/344/6179/94.short [3] http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjourna... [4] http://www.nature.com/nbt/journal/v32/n3/full/nbt.2834.html

That's incredibly interesting! If you can treat cells to receive signals on one end, and transmit them on the other, then can you then use some sort of optical fiber to possibly branch around damage along a nerve path? As they say, the internet routes around damage. Can we do the same with our nervous system?

And this is why I like HN!

It requires modifying DNA to make neurons sensitive to light. Which is easy to do for a mouse, alter some egg cells do IVF and collect the few that survive and breed from there. But, that's not going to work for people. There is some research into using a virus to modify people, but that's vary early in testing and more focused on genetic diseases.

Hypothetically, I suppose we might start doing such DNA modifications to people. But, better neural interfaces seems low on the list.

Optogenetics only works if you can genetically engineer the nerve cells to respond to light. Gene therapy is still... iffy, at this point.

One reason is that optogenetics is pretty new research that would have to rely on other pretty new research, human genetic modification is not very advanced (like, there aren't reliable methods that are well understood to be safe). Genetic modification is also really controversial.

Looks more functional than my real hand. I had an arm reattached which I believe was the wrong choice. (Doctors argued against, parents argued for) Now I'm stuck with a useless hand that no doctor will remove. I'd rather just have a stump and optionally a prosthetic. But I don't understand why so many prosthetics try and appear real and end up in the uncanny valley. They are robot arms, make them look like robots.

Within the past two decades we have gone from prostheses that do nothing to simple mechanical arms with very limited functionality to artificial arms that now provide much of the basic functionality as the organic ones with which we are born.

Given the accelerating pace of innovation in this field, before long artificial arms are bound to become better -- faster, stronger, more precise, etc. -- than the real thing!

I'm also watching the 3d printer > sugar printing > test tube blood vessel creation / organ structure creation > implantation. That research, blended with adult stem cell biology shows amazing progress in recreating organs at will.

It is funny, with all this new technology I fear grow older less and less.

I wonder when we will replace "healthy" organs because the natural organs are inferier.

We remove them already - tonsils, appendix, wisdom teeth, cataracts, vestigial tails

The tech is amazing, can't believe how fast it's progressing but as someone that is missing a hand I'd never use one I don't think. I can see how they are incredibly useful for people that have more difficulties.

Reply on news.ycombinator.com