REVIEW 11 cited by
STU-Net: Scalable and Transferable Medical Image Segmentation Models Empowered by Large-Scale Supervised Pre-training
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Large-scale models pre-trained on large-scale datasets have profoundly advanced the development of deep learning. However, the state-of-the-art models for medical image segmentation are still small-scale, with their parameters only in the tens of millions. Further scaling them up to higher orders of magnitude is rarely explored. An overarching goal of exploring large-scale models is to train them on large-scale medical segmentation datasets for better transfer capacities. In this work, we design a series of Scalable and Transferable U-Net (STU-Net) models, with parameter sizes ranging from 14 million to 1.4 billion. Notably, the 1.4B STU-Net is the largest medical image segmentation model to date. Our STU-Net is based on nnU-Net framework due to its popularity and impressive performance. We first refine the default convolutional blocks in nnU-Net to make them scalable. Then, we empirically evaluate different scaling combinations of network depth and width, discovering that it is optimal to scale model depth and width together. We train our scalable STU-Net models on a large-scale TotalSegmentator dataset and find that increasing model size brings a stronger performance gain. This observation reveals that a large model is promising in medical image segmentation. Furthermore, we evaluate the transferability of our model on 14 downstream datasets for direct inference and 3 datasets for further fine-tuning, covering various modalities and segmentation targets. We observe good performance of our pre-trained model in both direct inference and fine-tuning. The code and pre-trained models are available at https://github.com/Ziyan-Huang/STU-Net.
Forward citations
Cited by 11 Pith papers
-
BATS: Resource-Efficient Volumetric Segmentation with Boundary-Aware Mixed-Resolution Tokens
Boundary-aware mixed-resolution tokens cut volumetric segmentation peak memory by >53% versus MedNeXt-L while averaging within 0.37 Dice across five nnU-Net Revisited datasets.
-
HERMES: A Hybrid Ensemble for Head-and-Neck Tumor Segmentation, TN Staging, and Recurrence-Free Survival on PET/CT
On HECKTOR 2026, mask-derived geometry features for nodal staging gave out-of-fold balanced accuracy 0.720 vs 0.691 for radiomics (paired CI includes zero); on ground-truth masks the gap was 0.897 vs 0.837.
-
Unified Start, Personalized End: Progressive Pruning for Efficient 3D Medical Image Segmentation
PSP-Seg prunes redundant modules during training to make 3D segmentation networks much smaller and faster without losing accuracy.
-
MedVista3D: Vision-Language Modeling for Reducing Diagnostic Errors in 3D CT Disease Detection, Understanding and Reporting
A 3D CT vision-language pretraining framework with global and organ-level contrastive alignment plus a text retrieval bank achieves state-of-the-art zero-shot disease classification, report retrieval, and medical visu...
-
Rethink Domain Generalization in Heterogeneous Sequence MRI Segmentation
A semi-supervised pretraining method improves cross-sequence pancreas segmentation Dice from 43.55% to 70.39% (NU) and from 35.62% to 66.61% (IH) on the new PancreasDG benchmark.
-
CADS: A Comprehensive Anatomical Dataset and Segmentation for Whole-Body Anatomy in Computed Tomography
A new public dataset of 22,022 CT volumes labeled for 167 structures, and a nnU-Net model trained on it, outperform TotalSegmentator on most shared structures and expand coverage.
-
Same Branches, Different Trees: A Bifurcation Connectedness Metric for Coronary Artery Segmentation and FFR-CT Decision Agreement
Bifurcation Connectedness Score (BCS) measures junction-level vessel connectivity that Dice misses, tracks geometric FFR-CT decision agreement in severe disease, and shows branch recovery and tree connectedness are se...
-
DpDNet: An Dual-Prompt-Driven Network for Universal PET-CT Segmentation
DpDNet, a dual-prompt network, achieves top average DSC (74.87%) and IoU (62.56%) on a four-cancer-type whole-body PET-CT segmentation benchmark, and its extracted biomarkers stratify breast cancer survival.
-
A Large Convolutional Neural Network for Clinical Target and Multi-organ Segmentation in Gynecologic Brachytherapy with Multi-stage Learning
A three-stage pretrained CNN outperforms nnU-Net and Swin-UNETR for CTV and organ-at-risk segmentation in gynecologic brachytherapy CT, with the largest gains attributed to self-supervised pretraining.
-
Good Enough? An Investigation on the Impact of Label Quality in Large-Scale Medical Datasets
Label quality matters little when pre-training medical segmentation models, but still matters for in-domain deployment; only large quality gaps affect transfer results.
-
BraTS orchestrator : Democratizing and Disseminating state-of-the-art brain tumor image analysis
BraTS orchestrator is a new open-source package that provides uniform, tutorial-based access to winning BraTS segmentation and synthesis algorithms for brain tumor MRI.
Discussion (0). Sign in to comment.