Enhancing Brain Age Prediction and Neurodegeneration Detection with Contrastive Learning on Regional Biomechanical...

Area of Science:

• Neuroimaging
• Biophysics
• Gerontology

Background:

• Brain aging impacts structure and function, but biomechanical properties are understudied.
• Magnetic Resonance Elastography (MRE) measures tissue stiffness and damping, correlating with age and disease.
• Conventional structural MRI lacks sensitivity to early neurodegenerative changes.

Purpose of the Study:

• To investigate the potential of MRE in characterizing age-related biomechanical changes in the brain.
• To assess MRE's sensitivity in differentiating Alzheimer's disease (AD) and mild cognitive impairment (MCI) from healthy aging.
• To identify specific brain regions and biomechanical parameters as biomarkers for neurodegeneration.

Main Methods:

• Utilized a self-supervised contrastive regression framework with MRE data.
• Mapped brain tissue stiffness and damping ratio across participants.
• Performed regional analysis to identify key biomarkers of aging and disease.

Main Results:

• MRE demonstrated higher sensitivity than structural MRI for detecting age-related changes.
• Brain stiffness correlated with Alzheimer's disease (AD), while damping ratio detected mild cognitive impairment (MCI).
• Deep brain structures, including the caudate and thalamus, were identified as key aging biomarkers. Significant alterations were observed in the thalamus (AD) and hippocampus (MCI).

Conclusions:

• MRE is a sensitive biomarker for early neurodegenerative changes, surpassing conventional MRI.
• MRE can aid in early dementia risk detection and intervention by identifying specific biomechanical alterations.
• Biomechanical changes are detectable even in cognitively healthy individuals, suggesting preclinical neurodegeneration.