Predicting the apolipoprotein E ε4 allele carrier status based on gray matter volumes and cognitive function

Area of Science:

• Neuroimaging
• Machine Learning
• Genetics

Background:

• Apolipoprotein E (ApoE) ε4 carriers have an increased risk of Alzheimer's disease (AD).
• Individuals with the ApoE ε4 allele may exhibit brain atrophy and cognitive decline prior to clinical diagnosis.

Purpose of the Study:

• To develop machine learning (ML) models for predicting ApoE ε4 status.
• Utilize gray matter volume (GMV) from MRI scans and demographic data for prediction.

Main Methods:

• Recruited 74 participants (AD, MCI, cognitively normal) with known ApoE genotype.
• Acquired 3D T1-weighted (T1W) and 3D double inversion recovery (DIR) MRI scans.
• Extracted GMV from AD-related regions, incorporating age and MMSE scores as features for ML models.

Main Results:

• A cubic support vector machine (SVM3) model achieved an AUC of 0.88.
• The best model integrated age, MMSE scores, and DIR GMVs from the amygdala, hippocampus, and precuneus.
• This model surpassed predictions based solely on T1W GMV or demographic data.

Conclusions:

• Brain atrophy (DIR GMV) and cognitive decline are valuable biomarkers for predicting ApoE ε4 status.
• These findings aid in identifying individuals at higher risk of Alzheimer's disease progression.