English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The application of deep learning to medical image segmentation has beenhampered due to the lack of abundant pixel-level annotated data. Few-shotSemantic Segmentation (FSS) is a promising strategy for breaking the deadlock.However, a high-performing FSS model still requires sufficient pixel-levelannotated classes for training to avoid overfitting, which leads to itsperformance bottleneck in medical image segmentation due to the unmet need forannotations. Thus, semi-supervised FSS for medical images is accordinglyproposed to utilize unlabeled data for further performance improvement.Nevertheless, existing semi-supervised FSS methods has two obvious defects: (1)neglecting the relationship between the labeled and unlabeled data; (2) usingunlabeled data directly for end-to-end training leads to degeneratedrepresentation learning. To address these problems, we propose a novelsemi-supervised FSS framework for medical image segmentation. The proposedframework employs Poisson learning for modeling data relationship andpropagating supervision signals, and Spatial Consistency Calibration forencouraging the model to learn more coherent representations. In this process,unlabeled samples do not involve in end-to-end training, but providesupervisory information for query image segmentation through graph-basedlearning. We conduct extensive experiments on three medical image segmentationdatasets (i.e. ISIC skin lesion segmentation, abdominal organs segmentation forMRI and abdominal organs segmentation for CT) to demonstrate thestate-of-the-art performance and broad applicability of the proposed framework.

PoissonSeg: Semi-Supervised Few-Shot Medical Image Segmentation via Poisson Learning