A cheap, 20-minute MRI scan could soon replace the painful $5,000 tests currently used to find Alzheimer’s.
April 16, 2026
Original Paper
Cross-Modal Knowledge Distillation for PET-Free Amyloid-Beta Detection from MRI
arXiv · 2604.12574
The Takeaway
Right now, confirming Alzheimer’s usually requires an invasive spinal tap or an expensive PET scan that uses radioactive tracers. This new breakthrough uses AI to "distill" the knowledge from those expensive scans and apply it to standard, widely available MRI images. It effectively teaches a basic MRI machine to "see" the amyloid-beta plaques it would normally miss. If this goes mainstream, Alzheimer’s screening could move from a specialist's last resort to a routine check-up. It’s "practical magic" that could catch the disease a decade before symptoms appear, making early treatment actually possible for millions.
From the abstract
Detecting amyloid-$\beta$ (A$\beta$) positivity is crucial for early diagnosis of Alzheimer's disease but typically requires PET imaging, which is costly, invasive, and not widely accessible, limiting its use for population-level screening. We address this gap by proposing a PET-guided knowledge distillation framework that enables A$\beta$ prediction from MRI alone, without requiring non-imaging clinical covariates or PET at inference. Our approach employs a BiomedCLIP-based teacher model that l