Scientists are making significant strides when it comes to modeling computers on the power and efficiency of your brain, an ongoing project that could transform both health and computing.
The latest advancement comes out of Stanford University, where researchers have created a circuit board that is modeled after the functions of the brain.
It’s called Neurogrid, and the entire device is roughly the size of an iPad. It is comprised of 16 “neurocore” chips that can act like one million neurons in the brain, simulating billions of synapses, or connections in your body that perform a myriad of functions from moving muscles to making sense of sights and sounds.
As advanced as that sounds, Neurogrid pales in comparison to the human brain, according to Kwabena Boahen an associate professor of bioengineering at Stanford who was involved with the National Institutes of Health-funded project.
“The human brain, with 80,000 times more neurons than Neurogrid, consumes only three times as much power,” Boahen writes in a paper published in Proceedings of the IEEE. “Achieving this level of energy efficiency while offering greater configurability and scale is the ultimate challenge neuromorphic engineers face.”
Added to Boahen’s challenge is the sheer cost of producing this kind of technology– $40,000 to be exact.
But that’s an estimate of what it takes to build Neurogrid in its prototypical form. Using the most up-to-date manufacturing technology could knock the cost down to $400.
The cost of your, more efficient version: $0.
The remarkable creation could be utilized in many ways, but researchers hope that it will help power prosthetic limbs efficiently and intelligently.
They envision a circuit board that could be implanted in the brain that can interpret its signals and power prosthetic limbs—all without external power or overheating the brain.