الفهرس 
Title : Classification Of Alzheimer Disease Based On Resting-State Functional Magnetic Resonance Imaging Using Correlation Transfer Function
SummaryOnly 14 pages are availabe for public view
 |  | from | 92 |  |  |
| | | from | 92 | | |
Abstract
Alzheimer’s disease (AD) is a type of neurodegenerative disease, considered a significant health problem due to the increase of AD patients worldwide. According to the Centers for Disease Control and Prevention (CDC) reports, AD and other dementias patients in 2020 were approximately 16% more than expected. The diagnosis of AD is currently based on the medical history of individuals and their families, physical and neurological examinations, blood tests, and brain imaging. Among the brain imaging techniques, resting state functional magnetic resonance imaging (rs-fMRI) has been widely used to assess AD progression. FMRI detects changes in Blood Oxygenation Level-Dependent (BOLD) signals to determine brain activity. Generally, it has been suggested that functional changes likely precede structural alterations. Thus, as revealed by rs-fMRI, low-frequency fluctuation of the BOLD signals provides a viable approach to assess the network functional integrity of structurally segregated brain areas. This study investigates the effectiveness of using correlated transfer function (CorrTF) as a new biomarker to extract the essential features from rs-fMRI to distinguish between different AD stages. Furthermore, for better investigation and understanding of the interregional communication between different brain areas, we study the disease in cross-sectional and longitudinal manners separately.
In a cross-sectional framework, the CorrTF features and support vector machine (SVM) have been employed to distinguish between different AD stages. Additionally, we explored the regions, showing significant changes based on the CorrTF extracted features’ strength among different AD stages. First, the process was initialized by applying the preprocessing on rs-fMRI data samples to reduce noise and retain the essential information. Then, the brain was divided into 116 regions using the Automated Anatomical Labeling atlas (AAL), where the intensity time series is calculated, and the CorrTF connections are extracted for each region. Finally, the proposed framework employed an SVM classifier in two different methodologies, hierarchical and flat multi-classification schemes, to differentiate between AD stages for early detection purposes. The proposed schemes achieved an average accuracy of 98.2% and 95.5% for hierarchical and flat multi-classification tasks, respectively, calculated using ten folds cross-validation. Therefore, CorrTF is considered a promising technique for extracting useful features to be employed as biomarkers for AD identification at an early stage. Moreover, the significant changes in the extracted CorrTF connections strengths among AD stages help us define and explore the affected brain regions and their latent associations during the progression of AD.
Neural plasticity is the ability of the nervous system to modify itself functionally and structurally due to specific input. However, longitudinal changes in functional connectivity of the brain are unknown in Alzheimer’s disease (AD). In the longitudinal framework, the CorrTF features and statistical t-test have been used to discover the significant connections between brain regions for different AD stages over time. After extracting the CorrTF features, as done in the cross-sectional framework, both the standard t-test and the analysis of variance (ANONA) were employed to explore the significant CorrTF connections among each AD stage’s visits. Our results highlight each AD stage significant connections. Furthermore, we observed that the number of significant connections decreases with the disease severity in the following six network pairs: SMC-Cereb, VC-DMN, EAN-Cereb, DMN-Cereb, SN-Cereb, and Cereb-Cereb. Our results suggest that the Cerebellum regions are highly affected by AD progression. Additionally, the hippocampus connectivity increased in LMCI compared to EMCI subjects while missed in the AD final stage.