Efficient few-shot medical image segmentation via self-supervised variational autoencoder

Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registrati...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Medical image analysis Ročník 104; s. 103637
Hlavní autoři: Zhou, Yanjie, Zhou, Feng, Xi, Fengjun, Liu, Yong, Peng, Yun, Carlson, David E., Tu, Liyun
Médium: Journal Article
Jazyk:angličtina
Vydáno: Netherlands Elsevier B.V 01.08.2025
Témata:
Algorithms
Autoencoder
Few-shot learning
Humans
Image Interpretation, Computer-Assisted - methods
Image Processing, Computer-Assisted - methods
Image reconstruction
Magnetic Resonance Imaging
Medical image segmentation
Supervised Machine Learning
Variational autoencoder
Variational autoencoder
Few-shot learning
Image reconstruction
Medical image segmentation
ISSN:1361-8415, 1361-8423, 1361-8423
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Few-shot medical image segmentation typically uses a joint model for registration and segmentation. The registration model aligns a labeled atlas with unlabeled images to form initial masks, which are then refined by the segmentation model. However, inevitable spatial misalignments during registration can lead to inaccuracies and diminished segmentation quality. To address this, we developed EFS-MedSeg, an end-to-end model using two labeled atlases and few unlabeled images, enhanced by data augmentation and self-supervised learning. Initially, EFS-MedSeg applies a 3D random regional switch strategy to augment atlases, thereby enhancing supervision in segmentation tasks. This not only introduces variability to the training data but also enhances the model’s ability to generalize and prevents overfitting, resulting in natural and smooth label boundaries. Following this, we use a variational autoencoder for a weighted reconstruction task, focusing the model’s attention on areas with lower Dice scores to ensure accurate segmentation that conforms to the atlas image’s shape and structural appearance. Moreover, we introduce a self-contrastive module aimed at improving feature extraction, guided by anatomical structure priors, thus enhancing the model’s convergence and segmentation accuracy. Results on multi-modal medical image datasets show that EFS-MedSeg achieves performance comparable to fully-supervised methods. Moreover, it consistently surpasses the second-best method in Dice score by 1.4%, 9.1%, and 1.1% on the OASIS, BCV, and BCH datasets, respectively, highlighting its robustness and adaptability across diverse datasets. The source code will be made publicly available at: https://github.com/NoviceFodder/EFS-MedSeg. •Registration-free model enhances few-shot segmentation with 3D random regional switch.•Self-supervised framework refines low-Dice areas for precise segmentation.•Adaptive attention mechanism balances tissue volume discrepancies and boosts accuracy.•Self-contrastive module leverages anatomical priors for better features and outcomes.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1361-8415
1361-8423
1361-8423
DOI:10.1016/j.media.2025.103637