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

Learning with multiple modalities is crucial for automated brain tumorsegmentation from magnetic resonance imaging data. Explicitly optimizing thecommon information shared among all modalities (e.g., by maximizing the totalcorrelation) has been shown to achieve better feature representations and thusenhance the segmentation performance. However, existing approaches areoblivious to partial common information shared by subsets of the modalities. Inthis paper, we show that identifying such partial common information cansignificantly boost the discriminative power of image segmentation models. Inparticular, we introduce a novel concept of partial common information mask(PCI-mask) to provide a fine-grained characterization of what partial commoninformation is shared by which subsets of the modalities. By solving a maskedcorrelation maximization and simultaneously learning an optimal PCI-mask, weidentify the latent microstructure of partial common information and leverageit in a self-attention module to selectively weight different featurerepresentations in multi-modal data. We implement our proposed framework on thestandard U-Net. Our experimental results on the Multi-modal Brain TumorSegmentation Challenge (BraTS) datasets outperform those of state-of-the-artsegmentation baselines, with validation Dice similarity coefficients of 0.920,0.897, 0.837 for the whole tumor, tumor core, and enhancing tumor onBraTS-2020.

Exploiting Partial Common Information Microstructure for Multi-Modal Brain Tumor Segmentation