Nothing makes you appreciate your vision more than getting a little older and realizing that it used to be better and that it will probably get worse. But imagine how much more difficult it would be if you were totally blind. That was what happened to [Berna Gomez] when, at 42, she developed a medical condition that destroyed her optic nerves leaving her blind in a matter of days and ending her career as a science teacher. But thanks to science [Gomez] can now see, at least to some extent. She volunteered after 16 years to have a penny-sized device with 96 electrodes implanted in her visual cortex. The research is in the Journal of Clinical Investigation and while it is a crude first step, it shows lots of promise and uses some very novel techniques to overcome certain limitations.
The 96 electrodes were in a 10×10 grid with the four corner electrodes missing. The resolution, of course, is lacking, but the project turned to a glasses-mounted camera to acquire images and process them, reducing them to signals for the electrodes that may not directly map to the image.
After a few weeks to settle, the electrodes could cause bright spots to appear to the patient. However, she also randomly saw bright spots before the surgery, so the first task was to teach her to distinguish between the normal flashes and the ones from the electrodes. In the end, the patient was able to differentiate horizontal from vertical lines and identify some letters.
The processing is not only necessary to accommodate the low resolution, but also to overcome some limitations in the electrodes. For example, activating two neighboring electrodes at once tended to create a single spot a little brighter than the normal spot. In addition, only 88 of the electrodes caused visualization in the patient, so any scheme had to map around the “dead pixels.”
While we don’t think brain prosthesis design is a wise area for hacking, it does point to future possibilities as science learns more about how to feed input right to our brains. Imagine a VR headset that could just draw images on your visual cortex, bypassing lenses and your eyeballs.
There have been many attempts to provide some level of sight to the blind. If you don’t have the stomach for brain surgery, you can always focus on accessibility devices.
I just watched a video about this technology the other day. This looks like a very promising tech, especially with the prospect of future improvements.
What an amazing time to be alive!
Geordi’s Visor.
Regarding the future: Nothing like having a computer virus draw ads directly on your visual cortex
We already have medical implants, like pacemakers, that are not secured from intrusion. Cochlear implants probably could be hacked to play ads via programming/configuration interface.
One way to make such devices safe is to store the program in EPROM…
I’d suggest MROM… less likely to be confused with EEPROM.
Mask ROM makes more sense if you are making custom ICs… which is often the case with medical devices.
But the reason these medical devices use flash ROM is for storing custom configurations made for a patient. So instead of using flash ROM the device could store settings in RAM for setting up the patient, and once optimal configuration is done, the settings could be burned in EPROM. Once burned RAM settings are ignored…
Wait till Elon reads this
(way) too late
https://www.cnet.com/news/elon-musk-shows-neuralink-brain-implant-working-in-a-pig/
I remember seeing something like this in a documentary about 20 years ago, direct stimulation of the visual cortex with electrodes, why hasn’t this advanced further? Anyway how to they send the signals, could modulating the signals improve “sensing”, also wouldn’t this have a learning curve, the longer it’s used the better you get understanding it?
I did a (1-page) paper on various sight technologies in ~2004 or ’05? Hrmm… ah, found it: https://en.wikipedia.org/wiki/Visual_prosthesis#Dobelle_Eye
This is the part I recall most clearly (from the William H Dobelle wiki page):
> 2002, 38-year-old Jens Naumann, a blind man, was able to use the device to drive a car in the parking lot of the Dobelle Institute
The eyes, or more specifically the retinas are part of the brain. People tend to overlook that and just see them as dumb cameras, but that is far from the truth as they do a lot of image processing work before the compressed and interpreted sensory data is passed to the visual cortex. The problem with all inorganic implant technology is that it is inorganic and that it tends to kill off cells eventually, neuronal dieback. The next big issue is electromagnetic interference, or worse the induction of currents strong enough to trigger seizures, even potentially fatal ones. Truly effective and permanent implants will probably be developed from something really creepy such as genetically engineered parasites.