Related Experiment Video
Updated: Jun 18, 2025
12:09
Network Analysis of the Default Mode Network Using Functional Connectivity MRI in Temporal Lobe Epilepsy
Published on: August 5, 2014
18.0K
Early MS Identification Using Non-linear Functional Connectivity and Graph-theoretic Measures of Cognitive Task-fMRI
Farzad Azarmi1, Ahmad Shalbaf1, Seyedeh Naghmeh Miri Ashtiani2
1Department of Biomedical Engineering and Medical Physics, School of Medicine, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Basic and Clinical Neuroscience
|July 29, 2024
Summary
This study reveals that non-linear brain network measures, like Kernel Mutual Information (KMI), are superior to linear measures for identifying biomarkers in multiple sclerosis (MS) patients using functional MRI. These findings enhance the understanding of MS-related cognitive deficiencies.
Area of Science:
- Neuroimaging
- Cognitive Neuroscience
- Biomarker Discovery
Background:
- Multiple Sclerosis (MS) impairs central nervous system function, affecting brain networks and cognitive abilities.
- Assessing cognitive deficiencies in MS patients using functional neuroimaging remains a challenge.
- Understanding brain network alterations is crucial for diagnosing and managing MS.
Purpose of the Study:
- To identify brain network differences between relapsing-remitting MS (RRMS) patients and healthy controls using cognitive task-based fMRI.
- To evaluate the efficacy of non-linear connectivity measures as biomarkers for cognitive impairment in MS.
- To compare the performance of non-linear versus linear connectivity measures in classifying MS patients.
Main Methods:
- Task-based functional magnetic resonance imaging (fMRI) data were collected during a cognitive task (paced auditory serial addition test).
- A non-linear brain network was constructed using Kernel Mutual Information (KMI) based on the automated anatomical labeling (AAL) atlas.
- Machine learning algorithms, including decision tree-based techniques and support vector machines, were employed for classification, with feature selection using Wilcoxon rank-sum test and Fisher score.
Main Results:
- Non-linear connectivity measures, specifically KMI, demonstrated superior classification performance compared to linear measures like the correlation coefficient.
- Significant differences in brain network performance were observed in regions including the hippocampus, parahippocampus, cuneus, pallidum, and cerebellum between MS patients and controls.
- The Wilcoxon rank-sum test indicated promising results for clinical applications.
Conclusions:
- Non-linear connectivity measures, such as KMI, are more effective than linear measures for detecting MS-related brain network alterations and identifying biomarkers.
- This approach offers a promising avenue for assessing cognitive deficiencies in MS patients.
- Further research is warranted to validate these findings in larger MS patient cohorts.
Keywords:
Cognitive task-fMRIComputational neuroscienceKernel mutual informationMachine learning systemNetwork measuresNon-linear connectivity
