Back to the articles list | Back to browse issues page

XML Persian Abstract Print

A Neural Decoding Method Based on Neurons Activities Pattern Recognition for Implantable Intra-cortical BMIs
Danial Katoozian1 , Hossein Hosseini-Nejad * 1, M.Reza Abolghasemi2
1- Faculty of Electrical Engineering of K.N. Toosi University of Technology
2- Faculty of Electrical and computer of Tehran Uinversity
Abstract:   (1260 Views)
Converting motor intention to a machine command is called decoding in Brain Machine Interface (BMI) field. Despite recent advances, decoding remains among the most challenging steps in BMI. Furthermore, the majority of algorithms currently used in decoding require a computer which is not practical for implantable BMI systems. To address this issue, this paper proposes a novel approach based on hyperdimensional computing. This approach involves the conversion of the input space to binary, followed by the selection of the most similar vector to the answer. The proposed method is evaluated using a real dataset and demonstrates a reasonable accuracy rate with very low computational complexity. Furthermore, the proposed algorithm is implemented on a field-programmable gate array, indicating that it is a practical choice for real-time implantable BMI applications requiring a low computational cost method with a medium level of accuracy.

 
Keywords: Intra-cortical neural decoding Neurons activities patterns Implantable brain machine interface Hyperdimensional computing Hardware implementation Real time processing
     
Type of Article: Research paper | Subject: Special
Received: 2023/12/10 | Accepted: 2024/03/13 | ePublished ahead of print: 2024/05/4
Add your comments about this article : Your username or Email:
CAPTCHA
Send email to the article author

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

© 2024 CC BY-NC 4.0 | Journal of Control

Designed & Developed by : Yektaweb