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

Deep learning has been proposed for the assessment and classification ofmedical images. However, many medical image databases with appropriatelylabeled and annotated images are small and imbalanced, and thus unsuitable totrain and validate such models. The option is to generate synthetic images andone successful technique has been patented which limits its use for others. Wehave developed a free-access, alternate method for generating synthetichigh-resolution images using Generative Adversarial Networks (GAN) for dataaugmentation and showed their effectiveness using eye-fundus images forAge-Related Macular Degeneration (AMD) identification. Ten different GANarchitectures were compared to generate synthetic eye-fundus images with andwithout AMD. Data from three public databases were evaluated using theFr\'echet Inception Distance (FID), two clinical experts and deep-learningclassification. The results show that StyleGAN2 reached the lowest FID(166.17), and clinicians could not accurately differentiate between real andsynthetic images. ResNet-18 architecture obtained the best performance with 85%accuracy and outperformed the two experts in detecting AMD fundus images, whoseaverage accuracy was 77.5%. These results are similar to a recently patentedmethod, and will provide an alternative to generating high-quality syntheticmedical images. Free access has been provided to the entire method tofacilitate the further development of this field.

Which Generative Adversarial Network Yields High-Quality Synthetic Medical Images: Investigation Using AMD Image Datasets