Cross-Modal Knowledge Distillation for PET-Free Amyloid-Beta Detection from MRI

TL;DR

This paper introduces a PET-guided knowledge distillation framework enabling amyloid-beta detection from MRI alone, eliminating the need for PET imaging and clinical covariates, with promising results across multiple datasets.

Contribution

It proposes a novel cross-modal knowledge distillation method using a BiomedCLIP-based teacher to predict amyloid-beta from MRI without PET or clinical data.

Findings

Achieved AUC up to 0.74 on OASIS-3 and 0.68 on ADNI datasets.

Effectively transferred knowledge from PET to MRI modality.

Saliency analysis shows focus on relevant cortical regions.

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 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$\beta$ screening. Code is available at https://github.com/FrancescoChiumento/pet-guided-mri-amyloid-detection.