AI Paper Atlas

Research map / 3D vision / 3D generation

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

Maregu Assefa, Muhammad Muzammal Naseer, Divya Velayudhan, Gedamu Kumie, Naoufel Werghi

NeurIPS 2026 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes REFORM-3D, a representation-centric evaluation framework that assesses 3D medical vision foundation models separately from downstream adaptation choices, motivated by spacing and anisotropy effects on image statistics.

Problem

Evaluation of 3D medical foundation models relies on downstream task scores that entangle representation quality with adaptation choices, and spacing/anisotropy can change image statistics without changing anatomy.

Approach

Shifts evaluation toward intrinsic representation quality of 3D medical vision foundation models, disentangling it from downstream fine-tuning or adaptation decisions, with explicit attention to spacing and anisotropy effects.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use voxel-relative sampling instead of isotropic normalization in pretraining (ablation)Med3DINO (CLS) · Controlled same-scan spacing, AbdomenAtlascosine drift: 19.00 -> 10.90Improved
use spacing-aware resampling instead of isotropic normalization in pretraining (ablation)Med3DINO (CLS) · Controlled same-scan spacing, AbdomenAtlassemantic transfer: 73.60 -> 81.00Improved
use spacing-aware resampling instead of voxel-relative sampling (ablation)Med3DINO (CLS) · Controlled same-scan spacing, ImageCASsemantic transfer: 54.60 (REL) vs 54.20 (SA)No clear effect
use spacing-aware resampling instead of voxel-relative sampling (ablation)Med3DINO (CLS) · Controlled same-scan spacing, AbdomenAtlascosine drift: 10.90 -> 18.40Hurt
use spacing-aware resampling instead of isotropic normalization (ablation)Med3DINO (GAP) · CT-MR retrieval, TotalSegmentatormAP: 16.20 -> 27.20Improved
use spacing-aware resampling instead of isotropic normalization (ablation)Med3DINO (GAP) · CT-MR retrieval, AMOS22mAP: 21.60 -> 44.80Improved
use GAP features instead of CLS features for SA retrieval (ablation)Med3DINO-SA · CT-MR retrieval, TotalSegmentatormAP: 19.90 -> 27.20Improved
use spacing-aware resampling instead of isotropic normalization (ablation)Med3DINO (GAP) · Anatomical transfer, TotalSegmentator CT-MRprobe balanced accuracy: 44.30 -> 58.80Improved
use spacing-aware resampling instead of voxel-relative sampling (ablation)Med3DINO (GAP) · Anatomical transfer, MMWHScentroid balanced accuracy: 75.10 (REL) vs 74.50 (SA)No clear effect
use VoCo features rather than Med3DINO-SA under spacing changeVoCo · Controlled same-scan spacing, AbdomenAtlasdrift vs CKA: drift 0.80, CKA 86.70, transfer 74.60Mixed
switch from Med3DINO-ISO to Med3DINO-SA pretraining (ablation)Med3DINO segmentation · BTCVDice: 64.71 -> 75.69Improved
use Med3DINO-SA instead of 3DINOsegmentation · BTCVDice: 82.10 -> 75.69Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy