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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We demonstrate training of a Generative Adversarial Network (GAN) forprediction of optical property maps (scattering and absorption) using spatialfrequency domain imaging (SFDI) image data sets generated synthetically withfree open-source 3D modelling and rendering software, Blender. The flexibilityof Blender is exploited to simulate 3 models with real-life relevance toclinical SFDI of diseased tissue: flat samples, flat samples with spheroidaltumours and cylindrical samples with spheroidal tumours representing imaginginside a tubular organ e.g. the gastro-intestinal tract. In all 3 scenarios weshow the GAN provides accurate reconstruction of optical properties from singleSFDI images with mean normalised error ranging from 1-1.2% for absorption and0.7-1.2% for scattering, resulting in visually improved contrast for tumourspheroid structures. This compares favourably with 25% absorption error and 10%scattering error achieved using GANs on experimental SFDI data. However, someof this improvement is due to lower noise and availability of perfect groundtruths so we therefore cross-validate our synthetically-trained GAN with a GANtrained on experimental data and observe visually accurate results with errorof <40% for absorption and <25% for scattering, due largely to the presence ofspatial frequency mismatch artefacts. Our synthetically trained GAN istherefore highly relevant to real experimental samples, but providessignificant added benefits of large training datasets, perfect ground-truthsand the ability to test realistic imaging geometries, e.g. inside cylinders,for which no conventional single-shot demodulation algorithms exist. In futurewe expect that application of techniques such as domain adaptation or trainingon hybrid real-synthetic datasets will create a powerful tool for fast,accurate production of optical property maps from real clinical imagingsystems.

Training Generative Adversarial Networks for Optical Property Mapping using Synthetic Image Data