Alzheimer disease stages identification based on correlation transfer function system using resting-state functional...

Area of Science:

• Neuroimaging
• Biomarker Discovery
• Machine Learning

Background:

• Alzheimer's disease (AD) significantly impacts quality of life, causing memory loss and cognitive decline.
• Resting-state functional magnetic resonance imaging (rs-fMRI) is a key tool for analyzing brain regions in AD.
• Identifying reliable biomarkers for early AD detection remains a critical challenge.

Purpose of the Study:

• To evaluate the efficacy of correlated transfer function (CorrTF) as a novel biomarker for AD detection using rs-fMRI.
• To distinguish between different stages of Alzheimer's disease and mild cognitive impairment (MCI).
• To explore brain regions exhibiting significant changes in CorrTF features across AD stages.

Main Methods:

• rs-fMRI data preprocessing to reduce noise and retain essential information.
• Brain parcellation into 116 regions using the Automated Anatomical Labeling (AAL) atlas.
• Extraction of CorrTF features from regional time series and classification using hierarchical and flat Support Vector Machine (SVM) schemes.

Main Results:

• The proposed framework achieved high classification accuracies: 98.2% for hierarchical SVM and 95.5% for flat SVM.
• Ten-fold cross-validation was employed to ensure robust performance evaluation.
• Significant changes in CorrTF connection strengths were observed across different AD stages.

Conclusions:

• CorrTF is a promising and effective biomarker for the early identification of Alzheimer's disease.
• Analysis of CorrTF feature strengths aids in identifying affected brain regions and their associations during AD progression.
• The study highlights the potential of rs-fMRI combined with CorrTF for improved AD diagnosis and understanding.