CLASSIFFICATION OF MILD COGNITIVE IMPAIRMENT BASED ON DYNAMIC FUNCTIONAL CONNECTIVITY USING SPATIO-TEMPORAL...

Area of Science:

• Neuroimaging
• Artificial Intelligence
• Computational Neuroscience

Background:

• Dynamic functional connectivity (dFC) from resting-state fMRI (rs-fMRI) captures brain activity dynamics.
• Sequential information in dFC is underexplored for brain disease identification.
• Alzheimer's disease (AD) and its prodromal stage, Mild Cognitive Impairment (MCI), require early detection methods.

Purpose of the Study:

• To propose a novel framework for jointly learning spatial and temporal information from dFC.
• To leverage transformer architecture for enhanced analysis of dynamic brain connectivity.
• To improve the early prediction of MCI, a precursor to AD.

Main Methods:

• Constructing dFC networks from rs-fMRI data using a sliding window approach.
• Employing a transformer-based framework with temporal and spatial blocks to capture spatio-temporal dependencies.
• Utilizing contrastive learning to enhance feature representation robustness and reduce reliance on labeled data.

Main Results:

• The proposed method demonstrated superior performance in predicting MCI.
• Experimental validation on 345 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset.
• The framework effectively integrates spatial and temporal features for improved diagnostic accuracy.

Conclusions:

• The novel transformer-based framework significantly enhances the analysis of dFC for brain disease prediction.
• This approach shows great potential for the early identification of Mild Cognitive Impairment and Alzheimer's disease.
• The study highlights the importance of leveraging sequential and spatial information in dFC for neurodegenerative disease research.