Blood Markers Improve Brain Age Predictions for Alzheimer's Risk Assessment
New AI model using blood tests better predicts brain aging and amyloid buildup linked to dementia risk.
Summary
Researchers developed an improved AI model that estimates brain age more accurately by incorporating blood test markers alongside brain scans. The enhanced model showed stronger associations with amyloid-beta accumulation and cognitive decline, key hallmarks of Alzheimer's disease. Traditional brain age models relied only on MRI scans and medical history, potentially missing metabolic factors that influence brain aging. The new approach better distinguished between people with and without amyloid buildup in their brains, suggesting it could be a more reliable tool for assessing dementia risk and brain health.
Detailed Summary
Brain age estimation using AI has emerged as a promising biomarker for neurodegeneration, but previous studies showed inconsistent results when linking brain age gaps to dementia markers. This inconsistency may stem from overlooking metabolic and physiological factors that influence brain aging.
Japanese researchers studied 680 participants to develop two brain age prediction models. The first used conventional criteria based on medical history and MRI findings. The second incorporated blood test parameters measuring metabolic and physiological health. They then tested both models on 38 participants who underwent brain scans and amyloid PET imaging.
The enhanced model incorporating blood markers showed significantly stronger associations with amyloid-beta accumulation and cognitive function scores. Participants with amyloid buildup had higher brain age gaps when assessed with the blood-enhanced model, suggesting accelerated brain aging in those at risk for Alzheimer's disease.
These findings suggest that metabolic health plays a crucial role in brain aging assessment. The improved model could potentially serve as a more accurate tool for early dementia risk detection and monitoring brain health optimization interventions. This approach aligns with growing evidence that metabolic factors significantly influence neurodegeneration.
However, the study's small validation sample of 38 participants limits generalizability. Larger studies are needed to confirm these exploratory findings and establish clinical utility for this enhanced brain age assessment approach.
Key Findings
- AI brain age models incorporating blood markers better predicted amyloid buildup than conventional models
- Enhanced models showed stronger associations with cognitive function decline
- Metabolic health parameters significantly improved brain aging assessment accuracy
- Blood-enhanced models better distinguished amyloid-positive from amyloid-negative participants
Methodology
Researchers developed two machine learning models using MRI data from 680 participants, with one incorporating blood test parameters. Models were validated on 38 participants with amyloid PET imaging using correlation analysis and group comparisons.
Study Limitations
The validation dataset was very small with only 38 participants, limiting generalizability. Results are exploratory and require validation in larger, diverse populations before clinical implementation.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.