Quantcast

One of the world’s leading researchers in how to mitigate brain deficits that accrue with age will be a featured speaker at next month’s GPU Technology Conference, in San Jose, Calif.

Dr. Adam Gazzaley, director of the Neuroscience Imaging Center at the University of California, San Francisco, will use his GTC keynote on Thursday, March 27, to discuss his work using GPUs to create closed-loop systems of brain activity and video game mechanics to improve brain function.

Gazzaley is one of a number of brain researchers who are using GPUs to explore the functioning of the brain.

GPUs + Video Gaming Helps Mind Stay Sharper with Age

Dr. Adam Gazzaley, director of the Neuroscience Imaging Center at UCSF.
Dr. Adam Gazzaley, director of the Neuroscience Imaging Center at UCSF.

Gazzaley’s research on custom-designed video games that target different neural processes has revealed that not only do his research subjects — particularly adults between 60 to 85 — perform better in the game, but after a period of time this training provided cognitive benefits in other areas that are typically impaired as we age. (To learn more, check out a paper on Gazzaley’s research in the September 2013 issue of Nature.)

His work also involves the use of electroencephalography (EEG) to map areas of the brain that are activated when individuals play the games designed in his lab.

In collaboration with colleagues at the Swartz Center for Computational Research at UCSF, he is now building a processing pathway to use NVIDIA GPUs to send instantaneous feedback on this brain activity back into the game, which is adapted on the fly to “exercise” brain areas focused on task performance.

During his GTC keynote, Gazzaley will detail the next phase in his research: Harnessing additional GPU power to make brain exercises faster and more efficient.

Mapping the Brain … and Playing Baseball?

Gazzaley’s work is just one of several recent examples of research, powered by GPU-accelerated computing, that provide new visibility into the workings of the human brain:

  • Simulated Cerebellum Helps Robot Swing for the Fences — Japanese researchers have used GPUs and the CUDA parallel programming model to create a 100,000-neuron simulation of the human cerebellum, one of the largest simulations of its kind in the world. Based on work initiated at the Riken Brain Science Institute near Tokyo, Japan, they applied the data from this model to teach a robot to learn to hit a ball with a racket-shaped bat.
  • Modeling the Human Brain — Researchers at Germany’s Jülich Institute of Neuroscience and Medicine are using Tesla GPUs to accelerate the reconstruction of histological brain sections necessary for rendering a high-definition, structurally accurate and realistic model of the human brain by as much as 50x. Once developed, the model will give them a new level of visibility into brain architecture, function and interconnections.

You can view the entire list of GTC sessions and other happenings on the GTC website.