AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Large-Scale 3D Medical Image Pre-Training With Geometric Context Priors

Linshan Wu, Jiaxin Zhuang, Hao Chen

TPAMI 2025 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised neuroimaging representations · 38 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add intra-volume contrast pre-training (ablation)Swin-B · TotalSegmentatorDSC (%): 80.97 -> 81.38Improved
add inter-volume contrast on top of intra-volume contrast (ablation)VoCo Swin-B · TotalSegmentatorDSC (%): 81.38 -> 82.07Improved
add inter-volume contrast on top of intra-volume contrast (ablation)VoCo Swin-B · CC-CCIIaccuracy (%): 92.85 -> 93.64Improved
add semi-supervised learning to self-supervised contrast (omni-supervised) (ablation)VoCo Swin-B · TotalSegmentatorDSC (%): 82.07 -> 84.84Improved
add semi-supervised learning to self-supervised contrast (omni-supervised) (ablation)VoCo Swin-B · CC-CCIIaccuracy (%): 93.64 -> 93.80Improved
add self-supervised contrast to pure semi-supervised pre-training (ablation)VoCo Swin-B · CC-CCIIaccuracy (%): 91.98 -> 93.80Improved
add VoCo pre-training over scratchSwin-B/nnUNet · 7 segmentation datasetsDSC (%): +3.12%Improved
add VoCo pre-training over scratchSwin-B/nnUNet · 24 downstream segmentation tasksDSC (%): +4.42%Improved
add VoCo pre-training over scratchnnUNet/Swin-B · MSD 10-TaskDSC (%): +2.98%Improved
scale pre-training data from 10K to 160K (ablation)VoCo · TotalSegmentatorDSC (%)No clear effect
use larger model (VoCo-H over smaller) on small registration dataset (ablation)VoCo · OASISDSC (%)Hurt
add VoCo pre-training on PET AutoPETnnUNet · AutoPETDSC (%): 35.84 -> 33.02Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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