Anatomy-Privileged Distillation with Token Routing for MRI-Based Prediction of Perineural Invasion
Quick Answer
This study introduces an anatomy-privileged teacher-student framework for predicting perineural invasion (PNI) in intrahepatic cholangiocarcinoma using T2-weighted MRI.
Quick Take
This study introduces an anatomy-privileged teacher-student framework for predicting perineural invasion (PNI) in intrahepatic cholangiocarcinoma using T2-weighted MRI. The model achieved a mean AUROC of 0.750 in 155 patients, outperforming existing MRI-only baselines with efficient processing of 1.43 GFLOPs and 8.02 ms per case on a Jetson Orin Nano Super Developer Kit.
Key Points
- Proposed a novel framework for PNI prediction using T2-weighted MRI.
- Achieved a mean AUROC of 0.750 in a cohort of 155 patients.
- Outperformed existing MRI-only models evaluated under the same conditions.
- Processing efficiency measured at 1.43 GFLOPs and 8.02 ms per case.
- Anatomical supervision is limited to the training phase, enhancing model generalization.
Paper Resources
📖 Reader Mode
~2 min readAuthors:Hyunsu Go, Youngung Han, Kyeonghun Kim, Junga Kim, Dohyun Kweon, Jinyong Jun, Sungha Park, Anna Jung, Induk Um, Yului Jeong, Suah Park, Jina Jeong, Pa Hong, Woo Kyoung Jeong, Won Jae Lee, Ken Ying-Kai Liao, Hyuk-Jae Lee, Nam-Joon Kim
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.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2607.11987 [cs.CV] |
| (or arXiv:2607.11987v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.11987 arXiv-issued DOI via DataCite |
Submission history
From: Dohyun Kweon [view email]
[v1]
Mon, 13 Jul 2026 14:18:22 UTC (1,407 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.