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

It has been advocated that medical imaging systems and reconstructionalgorithms should be assessed and optimized by use of objective measures ofimage quality that quantify the performance of an observer at specificdiagnostic tasks. One important source of variability that can significantlylimit observer performance is variation in the objects to-be-imaged. Thissource of variability can be described by stochastic object models (SOMs). ASOM is a generative model that can be employed to establish an ensemble ofto-be-imaged objects with prescribed statistical properties. In order toaccurately model variations in anatomical structures and object textures, it isdesirable to establish SOMs from experimental imaging measurements acquired byuse of a well-characterized imaging system. Deep generative neural networks,such as generative adversarial networks (GANs) hold great potential for thistask. However, conventional GANs are typically trained by use of reconstructedimages that are influenced by the effects of measurement noise and thereconstruction process. To circumvent this, an AmbientGAN has been proposedthat augments a GAN with a measurement operator. However, the originalAmbientGAN could not immediately benefit from modern training procedures, suchas progressive growing, which limited its ability to be applied torealistically sized medical image data. To circumvent this, in this work, a newProgressive Growing AmbientGAN (ProAmGAN) strategy is developed forestablishing SOMs from medical imaging measurements. Stylized numerical studiescorresponding to common medical imaging modalities are conducted to demonstrateand validate the proposed method for establishing SOMs.

Learning stochastic object models from medical imaging measurements using Progressively-Growing AmbientGANs