Predicting Alzheimer's progression in MCI: a DTI-based white matter network model

Area of Science:

• Neuroimaging
• Neuroscience
• Biomarkers

Background:

• Mild cognitive impairment (MCI) is a precursor to Alzheimer's disease (AD).
• Early identification of MCI patients at high risk of AD progression is crucial for timely intervention.
• Diffusion tensor imaging (DTI) offers insights into white matter integrity, a potential indicator of neurodegeneration.

Purpose of the Study:

• To identify white matter network attributes from DTI that predict progression from MCI to AD.
• To construct a comprehensive predictive model for identifying MCI patients at high risk of AD.
• To develop an adjunct biomarker for early AD detection.

Main Methods:

• 121 MCI patients from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were analyzed.
• Brain networks were constructed from white matter tracts, and network features were extracted and downscaled.
• A comprehensive model integrating white matter network markers with clinical features (APOE4, ADAS scores) was developed and evaluated.

Main Results:

• The comprehensive model demonstrated high diagnostic efficacy (0.924 training, 0.919 testing).
• The model achieved high sensitivity (0.864 training, 0.900 testing) and specificity (0.871 training, 0.815 testing).
• The combined model showed significantly better diagnostic efficacy than APOE4 and ADAS scores alone.

Conclusions:

• A model integrating white matter network markers can effectively identify MCI patients at high risk of progressing to AD.
• This approach provides a valuable adjunct biomarker for early AD detection.
• The findings support the use of DTI-based network analysis in predicting AD conversion.