Deep learning detects genetic alterations in cancer histology generated by adversarial networks.
artificial intelligence
colorectal cancer
deep learning
digital pathology
generative adversarial network
generative model
machine learning
microsatellite instability
Journal
The Journal of pathology
ISSN: 1096-9896
Titre abrégé: J Pathol
Pays: England
ID NLM: 0204634
Informations de publication
Date de publication:
05 2021
05 2021
Historique:
revised:
10
01
2021
received:
30
11
2020
accepted:
05
02
2021
pubmed:
11
2
2021
medline:
1
1
2022
entrez:
10
2
2021
Statut:
ppublish
Résumé
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.
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
70-79Informations de copyright
© 2021 The Authors. The Journal of Pathology published by John Wiley & Sons, Ltd. on behalf of The Pathological Society of Great Britain and Ireland.
Références
Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018; 68: 394-424.
Kather JN, Calderaro J. Development of AI-based pathology biomarkers in gastrointestinal and liver cancer. Nat Rev Gastroenterol Hepatol 2020; 17: 591-592.
Kather JN, Halama N, Jaeger D. Genomics and emerging biomarkers for immunotherapy of colorectal cancer. Semin Cancer Biol 2018; 52: 189-197.
Stjepanovic N, Moreira L, Carneiro F, et al. Hereditary gastrointestinal cancers: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up†. Ann Oncol 2019; 30: 1558-1571.
Molecular testing strategies for Lynch syndrome in people with colorectal cancer (section 4, Evidence) | Guidance. NICE. [Accessed 30 April 2020]. Available from: https://www.nice.org.uk/guidance/dg27/chapter/4-Evidence.
Snowsill T, Coelho H, Huxley N, et al. Molecular testing for Lynch syndrome in people with colorectal cancer: systematic reviews and economic evaluation. Health Technol Assess 2017; 21: 1-238.
Fu Y, Jung AW, Torné RV, et al. Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis. Nat Cancer 2020; 1: 800-810.
Kather JN, Heij LR, Grabsch HI, et al. Pan-cancer image-based detection of clinically actionable genetic alterations. Nat Cancer 2020; 1: 789-799.
Coudray N, Ocampo PS, Sakellaropoulos T, et al. Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning. Nat Med 2018; 24: 1559-1567.
Kather JN, Pearson AT, Halama N, et al. Deep learning can predict microsatellite instability directly from histology in gastrointestinal cancer. Nat Med 2019; 25: 1054-1056.
Echle A, Grabsch HI, Quirke P, et al. Clinical-grade detection of microsatellite instability in colorectal tumors by deep learning. Gastroenterology 2020; 159: 1406-1416.
Schmauch B, Romagnoni A, Pronier E, et al. A deep learning model to predict RNA-Seq expression of tumours from whole slide images. Nat Commun 2020; 11: 3877.
Echle A, Rindtorff NT, Brinker TJ, et al. Deep learning in cancer pathology: a new generation of clinical biomarkers. Br J Cancer 2021; 124: 686-696.
Campanella G, Hanna MG, Geneslaw L, et al. Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med 2019; 25: 1301-1309.
Calderaro J, Kather JN. Artificial intelligence-based pathology for gastrointestinal and hepatobiliary cancers. Gut 2020. https://doi.org/10.1136/gutjnl-2020-322880 [Epub ahead of print].
The Cancer Genome Atlas Network. Comprehensive molecular characterization of human colon and rectal cancer. Nature 2012; 487: 330-337.
Levine AB, Peng J, Farnell D, et al. Synthesis of diagnostic quality cancer pathology images by generative adversarial networks. J Pathol 2020; 252: 178-188.
Murali LK, Lutnick B, Ginley B, et al. Generative modeling for renal microanatomy. Proc SPIE Int Soc Opt Eng 2020; 11320: 113200F.
Gadermayr M, Gupta L, Appel V, et al. Generative adversarial networks for facilitating stain-independent supervised and unsupervised segmentation: a study on kidney histology. IEEE Trans Med Imaging 2019; 38: 2293-2302.
Safarpoor A, Kalra S, Tizhoosh HR. Generative models in pathology: synthesis of diagnostic quality pathology images. J Pathol 2020; 253: 131-132.
Han T, Nebelung S, Haarburger C, et al. Breaking medical data sharing boundaries by using synthesized radiographs. Sci Adv 2020; 6: eabb7973.
van den Brandt PA. Maastricht Pathology 2018. 11th Joint Meeting of the British Division of the International Academy of Pathology and the Pathological Society of Great Britain & Ireland, 19-22 June 2018. J Pathol 2018; 246(suppl 1): S1-S46.
van den Brandt PA, Goldbohm RA, van't Veer P, et al. A large-scale prospective cohort study on diet and cancer in The Netherlands. J Clin Epidemiol 1990; 43: 285-295.
Macenko M, Niethammer M, Marron JS, et al. A method for normalizing histology slides for quantitative analysis. Proc 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2009; 1107-1110; DOI: https://doi.org/10.1109/ISBI.2009.5193250.
Goodfellow IJ, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks. v1. arXiv.org 2014. Available from: https://arxiv.org/abs/1406.2661. [Accessed 4 February 2021]. Not peer reviewed.
Zhang X, Zhou X, Lin M, et al. ShuffleNet: an extremely efficient convolutional neural network for mobile devices. v2. arXiv.org 2017. Available from: https://arxiv.org/abs/1707.01083. [Accessed 4 February 2021]. Not peer reviewed.
Tellez D, Litjens G, Bándi P, et al. Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology. Med Image Anal 2019; 58: 101544.
van der Maaten L, Hinton G. Visualizing data using t-SNE. J Mach Learn Res 2008; 9: 2579-2605.
Coudray N, Tsirigos A. Deep learning links histology, molecular signatures and prognosis in cancer. Nat Cancer 2020; 1: 755-757.
Deshpande S, Minhas F, Graham S, et al. SAFRON: stitching across the frontier for generating colorectal cancer histology images. v1. arXiv.org 2020. Available from: http://arxiv.org/abs/2008.04526. [Accessed 4 February 2021]. Not peer reviewed.