Northwestern Events Calendar


"Extending BCI use across tasks and over time: Development of a robust, cortically-controlled FES prosthesis for spinal cord injury"

When: Wednesday, May 27, 2020
1:00 PM - 2:00 PM  

Where: Online

Audience: Faculty/Staff - Student - Post Docs/Docs - Graduate Students

Contact: Donna Daviston   312.503.1687

Group: Physiology Roundtables

Category: Lectures & Meetings


Dr. Lee Miller, professor, will give a presentation.

Spinal cord injury is devastating, and currently has no real treatment. Ultimately, we’d want to regrow the damaged axons, but in the meantime, it is now possible to literally reconnect the brain and muscles electronically. My group has pioneered the development of a novel brain computer interface (BCI) that decodes muscle activity (EMG) from signals recorded from the motor cortex of monkeys. We use these synthetic EMG signals to control electrical stimulation of paralyzed muscles, a procedure called Functional Electrical Stimulation (FES). FES causes the muscles to contract and thereby restores voluntary control of the paralyzed limb. Following initial proof of concept several years ago, we have been working to develop more complex decoders that are applicable over long periods of time, and across a broad range of behaviors, including well-learned, stereotyped movements in the lab, as well as more natural movements in the monkey’s home cage. We are examining the representation within the cortex of these different behaviors, and the corresponding relation between cortical activity, EMG, and movement. In the past few years, there has been much interest in the fact that information from the millions of neurons active during movement can be reduced to a small number of “latent” signals in a low-dimensional manifold. However, even within this manifold, there are significant nonlinearities that make linear decoders fail to generalize across behaviors. We are developing deep neural networks that appear to have better generalization capacity. Another concern is that the predictions made by fixed decoders become inaccurate as recorded neurons change over days and weeks. We are also developing decoders, based on the latent signals, which appear to remain stable even for one or two years. Through this approach, we intend to develop an FES-based BMI that will restore voluntary movement across a broad range of motor tasks without need for intermittent recalibration.

Add to Calendar

Add Event To My Group:

Please sign-in