Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Development and characterization of an intracortical closed-loop brain-computer interface

Weiss, Jeffrey M. (2018) Development and characterization of an intracortical closed-loop brain-computer interface. Master's Thesis, University of Pittsburgh. (Unpublished)

Primary Text

Download (2MB) | Preview


Intracortical brain-computer interfaces (BCI) have the potential to restore motor function to people with paralysis by extracting movement intent signals directly from motor cortex. While current technology has allowed individuals to perform simple object interactions with robotic arms, such demonstrations have depended exclusively on visual feedback. Additional forms of sensory feedback may lessen the dependence on vision and allow for more dexterous control. Intracortical microstimulation (ICMS) has been proposed as a method of adding somatosensory feedback to BCI by directly stimulating somatosensory cortex to evoke tactile sensations referred to the hand. Our lab recently demonstrated that ICMS can elicit graded and focal tactile sensations in an individual with spinal cord injury (SCI). However, several challenges must be resolved to demonstrate the viability of ICMS as a technique for incorporating sensory feedback in a closed-loop BCI.

First, microstimulation generates large voltage transients that appear as artifacts in the neural recordings used for BCI control. These artifacts can corrupt the recorded signal throughout the entire stimulus train, and must be eliminated to allow for continuous BCI decoding. Second, it is unknown whether the sensations elicited by ICMS can be perceived quickly enough for use as a feedback signal.

Here, I present several aspects of the development of a closed-loop BCI system, including a method for artifact rejection and the characterization of simple reaction times to ICMS of human somatosensory cortex. A human participant with tetraplegia due to SCI was implanted with four microelectrode arrays in primary motor and somatosensory cortices. I implemented a robust method of artifact rejection that preserves neural data as soon as 750 microseconds after each stimulus pulse by applying signal blanking and an appropriate digital filter. I validated this method by comparing BCI performance with and without ICMS and found that performance was maintained with ICMS and artifact rejection. Next, I characterized simple reaction times to single-channel ICMS, and found that responses to ICMS were comparable, and often faster, than responses to electrical stimulation on the hand. These findings suggest that ICMS is a viable method to provide feedback in a closed-loop BCI.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Weiss, Jeffrey M.jmw182@pitt.edujmw1820000-0003-1332-674X
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairCollinger, Jennifer
Committee MemberGaunt, Robert
Committee MemberLoughlin, Patrick
Date: 11 June 2018
Date Type: Publication
Defense Date: 19 March 2018
Approval Date: 11 June 2018
Submission Date: 2 April 2018
Access Restriction: 2 year -- Restrict access to University of Pittsburgh for a period of 2 years.
Number of Pages: 104
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Bioengineering
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: ICMS, BCI, Artifact Rejection, Reaction Times
Date Deposited: 11 Jun 2018 17:13
Last Modified: 11 Jun 2020 05:15


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item