Parameter-efficient Prompt Tuning of Vision Foundation Model With Adaptive Focal Loss for Interpretable MCI Screening

TL;DR - A parameter-efficient framework adapts a frozen DINOv2-Small vision model with learnable prompt tokens to screen for Mild Cognitive Impairment from neuropsychological drawing tests, building spatial explainability directly into the architecture rather than bolting it on afterward.

• Uses a frozen DINOv2-Small backbone with just three modality-specific learnable prompt tokens (~1.19M trainable params); each token queries image patch tokens via a shared cross-attention layer, whose attention maps serve as intrinsic spatial explanations.
• Task-conditioned embeddings are fused through an attention module to quantify per-subject modality-level importance.
• A MoCA-adapted focal loss integrates continuous cognitive scores into the training target for loss modulation and adaptive sample weighting, addressing boundary ambiguity and class imbalance (framed as a generalization of soft-label approaches).
• Under stratified 5-fold cross-validation: MCI-class F1 of 0.641 and AUC of 0.795, beating the heavier ResViT baseline by 0.110 in MCI-class F1.