cs.CVNov 6, 2026

REFORM-3D: A Representation-Centric Evaluation Framework for 3D Medical Vision Foundation Models

Authors: Maregu Assefa, Muzammal Naseer, Divya Velayudhan, Gedamu Alemu Kumie, Naoufel Werghi

Organizations: Department of Computer Science, Khalifa University, Abu Dhabi, UAE · University of Western Australia

Abstract

Evaluation of 3D medical foundation models is typically dominated by downstream task scores, which entangle representation quality with adaptation choices. This is especially critical in volumetric imaging, where spacing and anisotropy can substantially alter image statistics while the anatomy itself remains unchanged. We therefore introduce REFORM-3D, a Representation-centric Evaluation FramewORk for benchmarking frozen 3D medical vision foundation Models across three questions: Acquisition Robustness, asking whether features remain stable when spacing varies while anatomy is fixed; Cross-Modal Anatomical Alignment, asking whether labeled anatomical structures remain recoverable across modalities under balanced bidirectional retrieval; and Anatomical Transfer, asking whether organ identity remains recoverable from limited labeled support under case-disjoint evaluation. These are also pretraining questions, because what a 3D encoder learns depends on how volumes are geometrically presented during self-supervised pretraining. We further introduce Med3DINO encoder family, pretrained on 62.662.6K unlabeled MR, CT, PET, CTA, and CBCT volumes under three explicit geometric assumptions: isotropic normalization, voxel-relative sampling, and spacing-aware resampling. Across Med3DINO and heterogeneous external encoders, REFORM-3D reveals module-specific strengths rather than a single model ordering. Within the controlled Med3DINO comparison, voxel-relative sampling yields lower embedding drift, whereas spacing-aware resampling favors anatomical transfer; semantic-transfer advantages depend on the dataset and feature type. Project link: REFORM-3D.

Figures & tables

Appendix figures & tables17 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Benchmarking Self-Supervised 3D Medical Vision Foundation Models and Transferability Estimation

    Oct 13, 2026Jiaqi Tang, Yu Wang, Yang Liu +7Medical Imaging Foundation ModelsMedical Image Benchmarks

  2. Do 3D Medical Foundation Models See Through MRI Artifacts? A Controlled Study of Representation Robustness

    Aug 6, 2026Julia Anna Mielcarz, Daniel Klaaby, Mostafa Mehdipour Ghazi3D Foundation ModelsMedical Imaging Foundation Models

  3. Topology-Driven Transferability Estimation for 3D Medical Vision Foundation Models

    Jul 5, 2026Jiaqi Tang, Shaoyang Zhang, Fandong Zhang +3Medical Imaging Foundation Models3D Medical Imaging