WPI Researchers Predict Alzheimer's Disease From Brain Anatomy
Worcester Polytechnic Institute researchers published in Neuroscience (2025) used machine learning to analyze 815 MRI scans from the Alzheimer's Disease Neuroimaging Initiative and predicted Alzheimer's disease with 92.87% accuracy. The study found that volume loss in the hippocampus, amygdala, and entorhinal cortex varies by age and sex, highlighting right hippocampus changes as an early marker and suggesting sex-specific biomarkers for earlier diagnosis and targeted assessment.
Key Points
- 1Achieved 92.87% accuracy using machine learning on 815 ADNI MRI scans across 95 brain regions
- 2Identified hippocampus, amygdala, and entorhinal cortex volume loss as top predictors across ages and sexes
- 3Suggests right hippocampus as early diagnostic marker and reveals sex-specific neuroanatomical biomarker patterns
Scoring Rationale
Peer-reviewed study demonstrating high accuracy and sex-specific biomarkers, but limited age range may limit generalizability.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
