Combining firing rate and spike-train synchrony features in the decoding of motor cortical activity

Combining firing rate and spike-train synchrony features in the decoding of motor cortical activity
Title:
Combining firing rate and spike-train synchrony features in the decoding of motor cortical activity
Other Titles:
2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)
Keywords:
Publication Date:
25 August 2015
Citation:
G. Chew, K. K. Ang, R. Q. So, Z. Xu and C. Guan, "Combining firing rate and spike-train synchrony features in the decoding of motor cortical activity," 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Milan, 2015, pp. 1091-1094.
Abstract:
Decoding of directional information in the motor cortex traditionally utilizes only firing rate information. However, information from other features could be extracted and combined with firing rate in order to increase classification accuracy. This study proposes the combination of firing rate and spike-train synchrony information in the decoding of motor cortical activity. Synchrony measures used are Event Synchronization (ES), SPIKE-Distance, and ISI-Distance. All data used for analyses were obtained from implanted electrode recordings of the primary motor cortex of a monkey that was trained to manipulate a motorized vehicle with 4 degrees of freedom (left, right, front and stop) via joystick control. Firstly, synchrony features could decode time periods, which were otherwise incorrectly decoded by firing rate alone, above chance levels. Secondly, using an ensemble classifier design for offline analysis, combining firing rate and ISI-distance information increases overall decoding accuracy by 1.1%. These results show that synchrony features in spike-trains do contain information not carried in firing rate. In addition, these results also demonstrate the feasibility of combining synchrony and firing rate for improving the classification accuracy of invasive brain-machine interface (BMI) in the control of neural prosthetics.
License type:
PublisherCopyrights
Funding Info:
Description:
(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
ISSN:
1094-687X
1558-4615
ISBN:
978-1-4244-9271-8
978-1-4244-9270-1
Files uploaded:

File Size Format Action
07318555.pdf 1.71 MB PDF Open