Anatomy-Privileged Distillation with Token Routing for MRI-Based Prediction of Perineural Invasion

arXiv:2607.11987v1 Announce Type: new Abstract: Perineural invasion (PNI) is associated with poor postoperative outcomes in intrahepatic cholangiocarcinoma, but it is confirmed by surgical pathology. Existing preoperative imaging models often rely on radiologist-defined variables, contrast-enhanced imaging, or manual annotations. We propose an anatomy-privileged teacher--student framework for patient-level PNI prediction from T2-weighted MRI. During training, the teacher uses MRI with tumor and liver masks to learn dense token routing, and the student distills this guidance to retain and aggregate informative tokens under a fixed budget. Anatomical supervision is restricted to training, and the deployed model does not require masks at inference. In 155 patients, the proposed method achieved the highest mean AUROC of 0.750 among matched MRI-only baselines evaluated under the same protocol, with 1.43 GFLOPs and 8.02 ms per case on a Jetson Orin Nano Super Developer Kit.

Related

• Beyond Visual Cues: Semantic-Driven Token Filtering and Expert Routing for Anytime Person ReID
• Structural Assessment for Understanding and Guiding Dataset Distillation in Discrete Token Space
• EMPURPLE: A Free Lunch for Diffusion Distillation based on the Information Bottleneck
• ActDistill: General Action-Guided Self-Derived Distillation for Efficient Vision-Language-Action Models

Source: arXiv cs.CV | 2026-07-15