It incurs high annotation costs but also leads to models that often suffer from poor generaliza- tion to new data sources

INTRODUCTION Medical image segmentation has traditionally relied on supervised learning paradigms, which, despite their efficacy, come with notable drawbacks

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

browse 1 citing papers

representative citing papers

Medical Semantic Segmentation with Diffusion Pretrain

cs.CV · 2025-01-31 · conditional · novelty 5.0

Anatomically guided diffusion pretraining on 3D CT reaches 67.8 Dice on BTCV organ segmentation, beating restorative baselines but not the contrastive vox2vec.

citing papers explorer

Showing 1 of 1 citing paper.

Medical Semantic Segmentation with Diffusion Pretrain cs.CV · 2025-01-31 · conditional · none · ref 1
Anatomically guided diffusion pretraining on 3D CT reaches 67.8 Dice on BTCV organ segmentation, beating restorative baselines but not the contrastive vox2vec.

It incurs high annotation costs but also leads to models that often suffer from poor generaliza- tion to new data sources

fields

years

verdicts

representative citing papers

citing papers explorer