A paraplegic patient controlled home appliances with the power of his mind using a neural implant
With a device implanted in his brain, a patient with ALS has learned to control TV, lights and other home devices.Researchers at Johns Hopkins Medical Center have published the results of a three-month clinical trial of an implantable brain-computer interface in patient Tim Evans with amyotrophic lateral sclerosis (ALS). Results show sustained accuracy of 90% over a three-month period without the need to retrain or calibrate the neuroimplant algorithm.
ALS is a progressive disease of the nervous system that causes muscle weakness and loss of motor and speech functions. Evans currently has severe speech and swallowing problems. He can speak slowly, but is difficult for most people to understand.
Researchers surgically implanted a Cortical Communication (CortiCom) neurointerface on Evans' brain regions responsible for speech and upper extremity function. The device provides a direct channel of communication between the brain and external intelligent devices.
The patient had two electrocorticographic (ECoG) meshes placed on the surface of the brain. An ECoG grid is a thin sheet of postage stamp-sized electrodes placed in the human brain to record electrical signals produced by thousands of neurons.
Tim Evans worked with researchers for weeks to teach the interface to recognize unique brain signals by repeating aloud each of six commands as they appeared on the screen. Once the BCI deep learning algorithm was trained, Evans was asked to give the same verbal commands to control the communication panel in real time, usually for about five minutes each day for the next three months.
"Although Tim's speech was difficult for most listeners to understand, BCI was able to accurately translate his brain activity into computer commands, allowing him to move around and select items on the communication board at a comfortable pace," Shiyu Luo, co-author of the study from Johns Hopkins University.
In addition to expressing through the communication board how he felt and what he wanted, Tim was able to use the BCI to turn lights on and off and to select videos to watch on YouTube.
The scientists note that unlike many other neural interface studies they used electrodes that do not penetrate the brain. This allows them to record large populations of neurons from the surface of the brain, rather than individual cells. This allows them to develop a stable system that doesn't need to be constantly retrained. And so the approach potentially gives participants the freedom to use BCIs anytime and anywhere without constant intervention from a researcher.