Abstract
The paper deals with functional brain-connectivity
analysis between pairs of brain lobes for a given learning task by
utilizing the fuzzy implication relations between the extracted
signal features acquired from selected channels of the lobes. The
Dienes-Rescher type fuzzy implication relation is chosen for its
closest similarity with propositional implication, resembling
implication in the true sense of its logical semantics. The DiensRecher type implication has successfully been employed to check
similarity in functional brain-connectivity for healthy (normal)
children (below 2 years) in a learning task of fruits and animals.
It is noted that children with Dyslexia disease exhibit different
brain-connectivity patterns with respect to those of
healthy subjects. This very finding opens up a new vista of
research to recognize dyslexia patients from their healthy
counterpart. Additionally, the k-means clustering algorithm is
employed to cluster children suffering from Dyslexia into groups,
based on their similarity in possible functional brainconnectivity. Such similarities of Dyslexia patients indicate
commonality in wrong terminations of neural pathways, which is
a well-known phenomenon in Dyslexia. The performance analysis
undertaken reveals that the proposed fuzzy relational approach
outperforms classical Granger causality, convergence crossmapping technique, probabilistic relative correlation adjacency
matrix and transfer entropy approaches with respect to 2
metrics: modularity and average efficiency.
analysis between pairs of brain lobes for a given learning task by
utilizing the fuzzy implication relations between the extracted
signal features acquired from selected channels of the lobes. The
Dienes-Rescher type fuzzy implication relation is chosen for its
closest similarity with propositional implication, resembling
implication in the true sense of its logical semantics. The DiensRecher type implication has successfully been employed to check
similarity in functional brain-connectivity for healthy (normal)
children (below 2 years) in a learning task of fruits and animals.
It is noted that children with Dyslexia disease exhibit different
brain-connectivity patterns with respect to those of
healthy subjects. This very finding opens up a new vista of
research to recognize dyslexia patients from their healthy
counterpart. Additionally, the k-means clustering algorithm is
employed to cluster children suffering from Dyslexia into groups,
based on their similarity in possible functional brainconnectivity. Such similarities of Dyslexia patients indicate
commonality in wrong terminations of neural pathways, which is
a well-known phenomenon in Dyslexia. The performance analysis
undertaken reveals that the proposed fuzzy relational approach
outperforms classical Granger causality, convergence crossmapping technique, probabilistic relative correlation adjacency
matrix and transfer entropy approaches with respect to 2
metrics: modularity and average efficiency.
| Original language | English |
|---|---|
| Title of host publication | 2022 International Joint Conference on Neural Networks (IJCNN) |
| Publisher | IEEE |
| Pages | 1-8 |
| Number of pages | 8 |
| ISBN (Electronic) | 978-1-7281-8671-9 |
| ISBN (Print) | 978-1-6654-9526-4 |
| DOIs | |
| Publication status | Published (in print/issue) - 18 Jul 2022 |
Publication series
| Name | 2022 International Joint Conference on Neural Networks (IJCNN) |
|---|---|
| Publisher | IEEE Control Society |
Bibliographical note
Publisher Copyright:© 2022 IEEE.
Keywords
- functional brain connectivity
- fuzzy relational approach
- Dyslexia patients
- f-NIRS
