Slashdot Mirror
Rewiring a Damaged Brain
An anonymous reader writes "Researchers in the Midwest are developing microelectronic circuitry to guide the growth of axons in a brain damaged by trauma. The goal is to rewire the brain connectivity and bypass the damaged region in order to restore normal behavior and movement. 'The device, which [professor Pedram Mohseni] calls a brain-machine-brain interface, includes a microchip on a circuit board smaller than a quarter. The microchip amplifies signals, called neural action potentials, produced by the neurons in one part of the brain and uses an algorithm to separate these signals — brain spike activity — from noise and other artifacts. Upon spike discrimination, the microchip sends a current pulse to stimulate neurons in another part of the brain, artificially connecting the two brain regions.'"
Too bad it's only for physical trauma. Emotional trauma is yours to keep!
I would be curious to see if this could eventually be used to offset the effects of Brain Surgery as well.
Having gone through a waking craniotomy to remove a benign tumor from my left temporal cortex in 2006, I'd quickly come to realize that certain things I was capable of before surgery were very difficult, if not nearly impossible without re learning the process all over again. For me, the issue was a loss of linguistics. I was unable to pronunciate "B" or "V" for over a week and had to re-teach myself. Beyond that, I was no longer fluent at speaking both Spanish and English where I was before. None the less, I still consider myself lucky that I have my life and my memories still.
In any case, work such as this can only help foster a shorter recovery time for brain trauma patients or better yet, recover capabilities that could have been completely lost. Did I mention being a cyborg sounds cool too?
Nice to see treatments being developed. At least when I took that Philosophy of Mind course in the early 90's, most of what we knew about the brain came from trauma... specifically, bicycle accidents. Basically, case studies looked at where the trauma was located, and built hypothesis about what that area did based on what no longer worked correctly in the patient. Oliver Sacks' The Man Who Mistook His Wife for a Hat was one of our texts, a collection of interesting studies along these lines. Brain damaged patients didn't get treatment regarding their brain truama, per se, they got studied. Sacks was a pure researcher... but somehow got involved with studying patients, and subsequently got fed up with the established idea that there were no treatments. see Awakenings.
We knew then that the brain tries to reroute things. I met someone recently that suffered from trauma induced skitzophrenia. He said it had been explained to him that a head trauma caused damage to a part of his brain that was between his eyes and that which interprets what he sees... and over time his brain rerouted the signals through other parts that were not damaged, such as memory centers... so he constantly is seeing people that aren't there, but are part of his memory. He claims there is a seamless interaction between these memory people, and the empirical environment... they are not ghosts, so he has to watch closely in crowds to see the reactions of people, and that's how he tells them apart... the memory people only react to movement, avoiding the real people and solid objects, but real people react to what is happening, what is being said, what they are watching.
Stem cell research appears to hold a lot of promise for brain trauma patients such as the man I met.
The Admin and the Engineer
The brain already does this itself. It's called neural plasticity. If they brain can do it, it will. If it can't, sticking wires into it and applying shocks and other intrusions and insults is not going to make it happen. Not properly anyway.
TFA is about neural jumper cables that can focus on only the signals they want, bypass damage and send the signal to another location. Fine idea except you kill the target quickly. But it specifically states "artificially". That makes the stuff about guiding axonal growth complete bullshit.
Neural connection is guided by glial cells, which are half the brain. If a region is damaged, both kinds of cells are damaged -- there's nothing to guide the growth of neural cells which are also damaged anyway. If you stimulate growth without the guiding mechanism, the cells form a tangle called a neuroma. The best outcome would be no result. Such neuromas caused by severed nerves, such as in amputations ('stump neuromas') are one of the causes of phantom limb pain. Neuromas in the cortex may not cause pain, but if they produce any result other than none, it'll be wrong and potentially interfering with function in the undamaged areas. Plus, stimulating growth where it can't happen properly is an excellent way to stimulate excessive, unguided, pathological growth -- tumors.
"I may be synthetic, but I'm not stupid." -- Bishop 341-B
"researchers foresee the possibility of using the approach in patients 10 years from now."
How can medical research move so fast and so slow at the SAME TIME?
That's easy. It's all about the funding. Now some VC will sink a ton of money into this, after which the pace will slow to glacial while they await regulatory approval. Right about when they need more money, they'll announce another breakthrough, or something favorable enough to secure more funding. Eventually some newer idea will knock this one off its pedestal, or they'll ship a product and get bought out by a large pharmaceutical or medical device company. The doctors and engineers will be free to repeat the cycle once their options have vested.
The potential of this is incredible. If this technology is ever fully developed, it would allow you to do something much more interesting than connecting 2 portions of damaged brain. There's no reason a powerful computer cluster couldn't simulate a portion of brain tissue and "stand in" as fake neurons on the other side of the link.
If the simulation were accurate enough, it would be possible for the patient to train the simulated brain tissue to mimic the original. Recovering stroke patients do this all the time. In the human brain, somehow one portion of the brain can train another portion and can smoothly distribute information around. So in principle, the computer simulation's neurons could gradually be coded with some of the skills of the person connected.
This would put us a LOT closer to real artificial intelligence, because we would now be able to see what is actually going on in a working area of human neural tissue. Do this on enough patients, and you'd have electronic analogues of most of the brain.
And the cool part : it might be possible someday to gradually replace a person's brain entirely through a series of surgeries and installing more and more microchips followed by a recovery and training period. You might be able to capture enough of a person's memories, personalities, and skills that the computer simulation would be capable of learning new abilities like the original person and passing the turing test.
The regulatory aspects in particular are why I never get too excited by things like this.
"Regulatory aspects?" It's Rewiring a Damaged Brain - literally brain surgery with some chip-building tossed in. Yes there are regulations, but progress is slow because it is hard to find brains to screw around with. This is not a process you take lightly.
I live in constant fear of the Coming of the Red Spiders.
Or instead, the people helping the millions of people who suffer brain damage from accidents and strokes could be, maybe, the millions of people who suffer brain damage from accidents and strokes, thereby obviating the need for a fucking expensive and pointless war entirely.
To have a right to do a thing is not at all the same as to be right in doing it