Perceived Realism of High-Resolution Generative Adversarial Network-derived Synthetic Mammograms.
Dimitrios KorkinofHugh HarveyAndreas HeindlEdith KarpatiGareth WilliamsTobias RijkenPeter KecskemethyBen GlockerPublished in: Radiology. Artificial intelligence (2020)
On the basis of these findings, it appears possible to generate realistic synthetic full-field digital mammograms by using a progressive GAN architecture up to a resolution of 1280 × 1024 pixels.Supplemental material is available for this article.© RSNA, 2020.