Cross-Modal Knowledge Distillation for PET-Free Amyloid-Beta Detection from MRI
This work addresses the need for accessible, non-invasive Alzheimer's screening by enabling PET-free amyloid-beta detection from MRI, though performance is moderate and incremental over existing MRI-based methods.
The authors propose a PET-guided knowledge distillation framework that enables amyloid-beta positivity detection from MRI alone, achieving AUCs of 0.74 on OASIS-3 and 0.68 on ADNI, eliminating the need for PET imaging and clinical covariates at inference.
Detecting amyloid-$β$ (A$β$) 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$β$ prediction from MRI alone, without requiring non-imaging clinical covariates or PET at inference. Our approach employs a BiomedCLIP-based teacher model that learns PET-MRI alignment via cross-modal attention and triplet contrastive learning with PET-informed (Centiloid-aware) online negative sampling. An MRI-only student then mimics the teacher via feature-level and logit-level distillation. Evaluated across four MRI contrasts (T1w, T2w, FLAIR, T2*) and two independent datasets, our approach demonstrates effective knowledge transfer (best AUC: 0.74 on OASIS-3, 0.68 on ADNI) while maintaining interpretability and eliminating the need for clinical variables. Saliency analysis confirms that predictions focus on anatomically relevant cortical regions, supporting the clinical viability of PET-free A$β$ screening. Code is available at https://github.com/FrancescoChiumento/pet-guided-mri-amyloid-detection.