Silk brain implants, developed by Brian Litt at the University of Pennsylvania in 2010, are in the news again, this time as part of an NIH-funded study where they're being used to stop epilepsy in rats.
What are silk brain implants? They're silk membranes just 2.5 microns thick which support a network of flexible electrodes for neural interfacing. The membranes are designed to dissolve, leaving the electrodes behind.
A couple of technologies have been developed since 2010 which could be used with the silk membranes to make them more useful. The first are flexible microchips, just 30 microns thick, developed in Belgium and announced in October 2012. The second are nano-scale carbon nanotube neural harpoons, a millimeter long and nanometers wide, developed at Duke University and announced in July 2013.
The neural harpoons could hook up to neurons up to a millimeter in the brain, while the flexible microchips do localized processing. This approach would provide much clearer signals than electrodes which just sit on the surface of the brain. The harpoons, crafted through ion beam sharpening, are capable of snagging individual neurons and measuring their input. It would take some work to mechanically embed them in the silk brain implant and provide them with the capability of launching themselves into a specific location in neural tissue.
What sorts of applications could this sort of device be used for? It could teach us about the functions of specific neurons in live brains. That information, in turn, could be used to build interfaces that implant false memories, or real knowledge. There's even the possibility they could eventually be used to record or create dreams.
Before these carbon nanotube harpoons, brain implants were made of glass or silicon, substances which can break or damage tissue. The nanotubes are flexible, and too small to seriously damage tissue. They are fantastic conductors, sending high-fidelity electrical signals from one end to the other.
Experiments found that the nanotube harpoons were less likely to break off in cells than silicon probes, though there is still a risk of breakage. Insulation and the geometry of the device also need to be improved. Nonetheless, it's clear this is a significant step forward for brain-computer interfaces.