[CCoE Notice] Seminar: Toward a Fully Bidirectional and Autonomous Brain Machine Interface

[Grayson, Audrey A]

 SEMINAR
MONDAY, MARCH 23, 2015 SEC 204:12:00 – 1:00 PM

Joseph ThachilFrancis, PhD
Associate Professorof Physiology and Pharmacology
State University of New York Downstate MedicalCenter

Toward a Fully Bidirectional and Autonomous Brain Machine Interface


Recently several groups have been able to decode intentions directly from the brainin order to control computercursors as well as roboticarms and hands.Such systems are calledbrain machine, or brain computerinterfaces (BMI/BCI). It has become clear that somatosensory feedbackfrom such neuroprosthetic limbs to the usershould allow for more naturalintegration with the user and help make such systems become a part of one self,rather than beingseen as a tool, or attachment. Over the past decade my lab has been workingon both the movement intention decoding and sensoryfeedback problems. Recentlywe have discovered that there is rewardmodulation of mirrorlike neurons in the primarysensorimotor cortex, and wehave coupled this information with a reinforcement learning (RL) architecture to produce an autonomous BMI, that senseswhen the systemeither needs to be corrected for future use, or is operating as the individual would like. On the sensory feedback side of the BMI we have utilized model predictive controland microstimulation of the sensory thalamusto produce naturallike cortical responses in S1. We acomplished this by generating template S1 responses to natural touch and then minimized the difference betweenour prosthetically induceS1 responses and these templates. I will introduce and discuss both of these sides of our BMI development.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://Bug.EGR.UH.EDU/pipermail/engi-dist/attachments/20150323/f96f6166/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SEMINAR Dr. Joe Francis.docx
Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document
Size: 53544 bytes
Desc: SEMINAR Dr. Joe Francis.docx
Url : http://Bug.EGR.UH.EDU/pipermail/engi-dist/attachments/20150323/f96f6166/attachment-0001.bin