Brain signatures based on structural MRI: Classification for MCI, PMCI, and AD.

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Medical Image Analysis

Background:

• Structural MRI (sMRI) detects cerebral atrophy, crucial for understanding neurodegenerative diseases like Alzheimer's Disease (AD).
• Developing brain networks from sMRI data offers a novel network neuroscience perspective, yet remains understudied.
• Identifying brain network alterations is key to understanding disease progression and developing diagnostic tools.

Purpose of the Study:

• To propose a framework for constructing brain networks from sMRI data.
• To extract disease-specific brain signatures and identify critical regions of interest (ROIs).
• To validate the framework's ability to detect neurodegenerative patterns in mild cognitive impairment (MCI), progressive MCI (PMCI), and AD.

Main Methods:

• Constructing brain networks where nodes represent brain atlas regions and edge weights are derived from Sorensen distance between gray matter probability maps.
• Defining brain signatures based on network changes observed between disease and control subjects.
• Validating the methodology using reference and examination cohorts encompassing control, MCI, PMCI, and AD subjects.

Main Results:

• The proposed framework successfully extracts distinct brain signatures associated with MCI, PMCI, and AD.
• Critical ROIs linked to these neurodegenerative conditions were identified.
• The methodology demonstrated efficacy in differentiating disease states based on network patterns.

Conclusions:

• The developed framework effectively constructs brain networks from sMRI data and extracts relevant disease signatures.
• This approach holds significant potential for brain mapping, understanding brain communication, and developing network-based diagnostic applications.
• The findings contribute to advancing neuroscientific insights into neurodegenerative diseases through network analysis.