Deep learning detects genetic alterations in cancer histology generated by adversarial networks

• Published in: The Journal of pathology Vol. 254; no. 1; pp. 70 - 79
• Main Authors: Krause, Jeremias, Grabsch, Heike I, Kloor, Matthias, Jendrusch, Michael, Echle, Amelie, Buelow, Roman David, Boor, Peter, Luedde, Tom, Brinker, Titus J, Trautwein, Christian, Pearson, Alexander T, Quirke, Philip, Jenniskens, Josien, Offermans, Kelly, Brandt, Piet A, Kather, Jakob Nikolas
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01.05.2021; Wiley Subscription Services, Inc
• Subjects: artificial intelligence; Cancer; Colorectal cancer; Colorectal carcinoma; Colorectal Neoplasms - genetics; Deep Learning; digital pathology; generative adversarial network; generative model; Histology; Humans; Image Interpretation, Computer-Assisted - methods; machine learning; Microsatellite Instability; generative adversarial network; deep learning; microsatellite instability; colorectal cancer; digital pathology; generative model; machine learning; artificial intelligence
• ISSN: 0022-3417; 1096-9896
• DOI: 10.1002/path.5638

Deep learning can detect microsatellite instability (MSI) from routine histology images in colorectal cancer (CRC). However, ethical and legal barriers impede sharing of images and genetic data, hampering development of new algorithms for detection of MSI and other biomarkers. We hypothesized that histology images synthesized by conditional generative adversarial networks (CGANs) retain information about genetic alterations. To test this, we developed a ‘histology CGAN’ which was trained on 256 patients (training cohort 1) and 1457 patients (training cohort 2). The CGAN synthesized 10 000 synthetic MSI and non‐MSI images which contained a range of tissue types and were deemed realistic by trained observers in a blinded study. Subsequently, we trained a deep learning detector of MSI on real or synthetic images and evaluated the performance of MSI detection in a held‐out set of 142 patients. When trained on real images from training cohort 1, this system achieved an area under the receiver operating curve (AUROC) of 0.742 [0.681, 0.854]. Training on the larger cohort 2 only marginally improved the AUROC to 0.757 [0.707, 0.869]. Training on purely synthetic data resulted in an AUROC of 0.743 [0.658, 0.801]. Training on both real and synthetic data further increased AUROC to 0.777 [0.715, 0.821]. We conclude that synthetic histology images retain information reflecting underlying genetic alterations in colorectal cancer. Using synthetic instead of real images to train deep learning systems yields non‐inferior classifiers. This approach can be used to create large shareable data sets or to augment small data sets with rare molecular features. © 2021 The Authors. The Journal of Pathology published by John Wiley & Sons, Ltd. on behalf of The Pathological Society of Great Britain and Ireland.