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

We present a deep learning framework for wide-field, content-aware estimationof absorption and scattering coefficients of tissues, called GenerativeAdversarial Network Prediction of Optical Properties (GANPOP). Spatialfrequency domain imaging is used to obtain ground-truth optical properties fromin vivo human hands, freshly resected human esophagectomy samples andhomogeneous tissue phantoms. Images of objects with either flat-field orstructured illumination are paired with registered optical property maps andare used to train conditional generative adversarial networks that estimateoptical properties from a single input image. We benchmark this approach bycomparing GANPOP to a single-snapshot optical property (SSOP) technique, usinga normalized mean absolute error (NMAE) metric. In human gastrointestinalspecimens, GANPOP estimates both reduced scattering and absorption coefficientsat 660 nm from a single 0.2/mm spatial frequency illumination image with 58%higher accuracy than SSOP. When applied to both in vivo and ex vivo swinetissues, a GANPOP model trained solely on human specimens and phantomsestimates optical properties with approximately 43% improvement over SSOP,indicating adaptability to sample variety. Moreover, we demonstrate that GANPOPestimates optical properties from flat-field illumination images with similarerror to SSOP, which requires structured-illumination. Given a training setthat appropriately spans the target domain, GANPOP has the potential to enablerapid and accurate wide-field measurements of optical properties, even fromconventional imaging systems with flat-field illumination.

GANPOP: Generative Adversarial Network Prediction of Optical Properties from Single Snapshot Wide-field Images