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

Integrating high-level semantically correlated contents and low-levelanatomical features is of central importance in medical image segmentation.Towards this end, recent deep learning-based medical segmentation methods haveshown great promise in better modeling such information. However, convolutionoperators for medical segmentation typically operate on regular grids, whichinherently blur the high-frequency regions, i.e., boundary regions. In thiswork, we propose MORSE, a generic implicit neural rendering framework designedat an anatomical level to assist learning in medical image segmentation. Ourmethod is motivated by the fact that implicit neural representation has beenshown to be more effective in fitting complex signals and solving computergraphics problems than discrete grid-based representation. The core of ourapproach is to formulate medical image segmentation as a rendering problem inan end-to-end manner. Specifically, we continuously align the coarsesegmentation prediction with the ambiguous coordinate-based pointrepresentations and aggregate these features to adaptively refine the boundaryregion. To parallelly optimize multi-scale pixel-level features, we leveragethe idea from Mixture-of-Expert (MoE) to design and train our MORSE with astochastic gating mechanism. Our experiments demonstrate that MORSE can workwell with different medical segmentation backbones, consistently achievingcompetitive performance improvements in both 2D and 3D supervised medicalsegmentation methods. We also theoretically analyze the superiority of MORSE.

Implicit Anatomical Rendering for Medical Image Segmentation with Stochastic Experts