Brain Machine Interface Learns Along wth Patient
A Brain Machine Interface is a device that interprets signals from the brain and then routes them to a robotic hand or limb. Up until recently, these devices have only been able to interpret brain signals and serve as a bridge between the biological and the mechanical. But researchers at the University of Florida have taken this concept further and developed Brain Machine Interfaces that not only translate brain signals into mechanical movement, but also evolve with the brain as it learns.
These devices would have the ability to adapt to a person's behavior over time, and with that knowledge, would assist the user in completing a task more efficiently. So rather than simply taking orders from the brain, the device is actually aware of the goal that is trying to be accomplished, and will assist the brain in accomplishing the task more efficiently.
To test these devices, the UF researchers put three of them into three different lab rats. The rats were then taught how to move a robotic arm towards a target using just their thoughts. Whenever the robotic arm hit the target (which was the rat's goal), the rat was rewarded with water.
The device's goal on the other hand was to earn as many points as possible. The closer the robotic arm moved to the target, the more points it would receive. Thus the device was able to "learn" which brain signals from the rat led to the most points, which made the process more efficient for the rat.
Justin Sanchez, the UF study's lead author, said that he and the other researchers think that this dialogue involving a goal is how they are going to be able to make these systems that evolve over time.
I found this very interesting because I had no idea that research was going on that would allow a mechanical device to "dialogue" with the brain and actually assist the brain in accomplishing tasks. Obviously, this kind of technology has the potential to work wonders for amputee patients.
The article can be found at http://www.physorg.com/news133535377.html
Daniel Verona
These devices would have the ability to adapt to a person's behavior over time, and with that knowledge, would assist the user in completing a task more efficiently. So rather than simply taking orders from the brain, the device is actually aware of the goal that is trying to be accomplished, and will assist the brain in accomplishing the task more efficiently.
To test these devices, the UF researchers put three of them into three different lab rats. The rats were then taught how to move a robotic arm towards a target using just their thoughts. Whenever the robotic arm hit the target (which was the rat's goal), the rat was rewarded with water.
The device's goal on the other hand was to earn as many points as possible. The closer the robotic arm moved to the target, the more points it would receive. Thus the device was able to "learn" which brain signals from the rat led to the most points, which made the process more efficient for the rat.
Justin Sanchez, the UF study's lead author, said that he and the other researchers think that this dialogue involving a goal is how they are going to be able to make these systems that evolve over time.
I found this very interesting because I had no idea that research was going on that would allow a mechanical device to "dialogue" with the brain and actually assist the brain in accomplishing tasks. Obviously, this kind of technology has the potential to work wonders for amputee patients.
The article can be found at http://www.physorg.com/news133535377.html
Daniel Verona
posted by Daniel Verona at 8:03 AM
0 Comments:
Post a Comment
<< Home