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/

Zendy Plus

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

Zendy Tools

Zendy Open

Multiple stains are usually used to highlight biological substances inbiomedical image analysis. To decompose multiple stains for co-localizationquantification, blind source separation is usually performed. Prior model-basedstain separation methods usually rely on stains' spatial distributions over animage and may fail to solve the co-localization problem. With the advantage ofmachine learning, deep generative models are used for this purpose. Since priorknowledge of imaging models is ignored in purely data-driven solutions, thesemethods may be sub-optimal. In this study, a novel learning-based blind sourceseparation framework is proposed, where the physical model of biomedicalimaging is incorporated to regularize the learning process. The introducedmodel-relevant adversarial loss couples all generators in the framework andlimits the capacities of the generative models. Further more, a trainingalgorithm is innovated for the proposed framework to avoid inter-generatorconfusion during learning. This paper particularly takes fluorescence unmixingin fluorescence microscopy images as an application example of the proposedframework. Qualitative and quantitative experimentation on a publicfluorescence microscopy image set demonstrates the superiority of the proposedmethod over both prior model-based approaches and learning-based methods.

Blind stain separation using model-aware generative learning and its applications on fluorescence microscopy images