Skip to main navigation Skip to search Skip to main content

Graph Network Modeling of Brain Connectivity: An Exploration of Word and Object Recognition Tasks

  • Faculty of Computing, Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Understanding the neural mechanisms underlying cognitive processes, such as word and object recognition, is crucial for advancing cognitive neuroscience. This study explores the neural mechanisms underlying word and object recognition using a graph-based approach, BrainGNN, applied to fMRI data. By focusing on the fMRI scans related to these cognitive tasks, we constructed ROI-wise functional connectivity matrices based on the AAL atlas and employed the BrainGNN model for classification. The model effectively distinguished between tasks, achieving an accuracy of 91.67%. Significant differences were observed in brain regions such as the Temporal Pole and Supplementary Motor Area. This research contributes to a deeper understanding of the brain's functional connectivity and the neural mechanisms that underpin word and object recognition.

Original languageEnglish
Title of host publicationICSP 2024 - 2024 IEEE 17th International Conference on Signal Processing, Proceedings
EditorsYuan Baozong, Ruan Qiuqi, Wei Shikui, An Gaoyun
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages692-696
Number of pages5
ISBN (Electronic)9798350387384
DOIs
StatePublished - 2024
Externally publishedYes
Event17th IEEE International Conference on Signal Processing, ICSP 2024 - Suzhou, China
Duration: 28 Oct 202431 Oct 2024

Publication series

NameInternational Conference on Signal Processing Proceedings, ICSP
ISSN (Print)2164-5221
ISSN (Electronic)2164-523X

Conference

Conference17th IEEE International Conference on Signal Processing, ICSP 2024
Country/TerritoryChina
CitySuzhou
Period28/10/2431/10/24

Keywords

  • Functional Connectivity
  • Graph Neural Network
  • Object Recognition
  • Word Recognition
  • fMRI

Fingerprint

Dive into the research topics of 'Graph Network Modeling of Brain Connectivity: An Exploration of Word and Object Recognition Tasks'. Together they form a unique fingerprint.

Cite this