Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning

被引:1711
作者
Coudray, Nicolas [1 ,2 ]
Ocampo, Paolo Santiago [3 ]
Sakellaropoulos, Theodore [4 ]
Narula, Navneet [3 ]
Snuderl, Matija [3 ]
Fenyo, David [5 ,6 ]
Moreira, Andre L. [3 ,7 ]
Razavian, Narges [8 ,9 ]
Tsirigos, Aristotelis [1 ,3 ]
机构
[1] NYU, Sch Med, Appl Bioinformat Labs, New York, NY 10003 USA
[2] NYU, Sch Med, Dept Cell Biol, Skirball Inst, New York, NY 10016 USA
[3] NYU, Sch Med, Dept Pathol, New York, NY 10003 USA
[4] Natl Tech Univ Athens, Sch Mech Engn, Zografos, Greece
[5] NYU, Sch Med, Inst Syst Genet, New York, NY USA
[6] NYU, Sch Med, Dept Biochem & Mol Pharmacol, New York, NY USA
[7] NYU, Ctr Biospecimen Res & Dev, New York, NY USA
[8] NYU, Sch Med, Dept Populat Hlth, New York, NY 10003 USA
[9] NYU, Sch Med, Ctr Healthcare Innovat & Delivery Sci, New York, NY 10003 USA
关键词
TARGETED THERAPY; NEURAL-NETWORKS; EGFR MUTATION; LKB1; ADENOCARCINOMAS; INACTIVATION; SOCIETY; PROMISE; KRAS;
D O I
10.1038/s41591-018-0177-5
中图分类号
Q5 [生物化学]; Q7 [分子生物学];
学科分类号
071010 ; 081704 ;
摘要
Visual inspection of histopathology slides is one of the main methods used by pathologists to assess the stage, type and subtype of lung tumors. Adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC) are the most prevalent subtypes of lung cancer, and their distinction requires visual inspection by an experienced pathologist. In this study, we trained a deep convolutional neural network (inception v3) on whole-slide images obtained from The Cancer Genome Atlas to accurately and automatically classify them into LUAD, LUSC or normal lung tissue. The performance of our method is comparable to that of pathologists, with an average area under the curve (AUC) of 0.97. Our model was validated on independent datasets of frozen tissues, formalin-fixed paraffin-embedded tissues and biopsies. Furthermore, we trained the network to predict the ten most commonly mutated genes in LUAD. We found that six of them-STK11, EGFR, FAT1, SETBP1, KRAS and TP53-can be predicted from pathology images, with AUCs from 0.733 to 0.856 as measured on a held-out population. These findings suggest that deep-learning models can assist pathologists in the detection of cancer subtype or gene mutations. Our approach can be applied to any cancer type, and the code is available at https://github.com/ncoudray/DeepPATH.
引用
收藏
页码:1559 / +
页数:11
相关论文
共 67 条
  • [11] Automated segmentation of epithelial tissue in prostatectomy slides using deep learning
    Bulten, Wouter
    Hulsbergen-van de Kaa, Christina A.
    van der Laak, Jeroen
    Litjens, Geert J. S.
    [J]. MEDICAL IMAGING 2018: DIGITAL PATHOLOGY, 2018, 10581
  • [12] Targeted therapy for non-small cell lung cancer: current standards and the promise of the future
    Chan, Bryan A.
    Hughes, Brett G. M.
    [J]. TRANSLATIONAL LUNG CANCER RESEARCH, 2015, 4 (01) : 36 - 54
  • [13] Cheng JZ, 2016, SCI REP-UK, V6, DOI [10.1038/srep24454, 10.1038/srep25671]
  • [14] IDH2 Mutations Define a Unique Subtype of Breast Cancer with Altered Nuclear Polarity
    Chiang, Sarah
    Weigelt, Britta
    Wen, Huei-Chi
    Pareja, Fresia
    Raghavendra, Ashwini
    Martelotto, Luciano G.
    Burke, Kathleen A.
    Basili, Thais
    Li, Anqi
    Geyer, Felipe C.
    Piscuoglio, Salvatore
    Ng, Charlotte K. Y.
    Jungbluth, Achim A.
    Balss, Joerg
    Pusch, Stefan
    Baker, Gabrielle M.
    Cole, Kimberly S.
    von Deimling, Andreas
    Batten, Julie M.
    Marotti, Jonathan D.
    Soh, Hwei-Choo
    McCalip, Benjamin L.
    Serrano, Jonathan
    Lim, Raymond S.
    Siziopikou, Kalliopi P.
    Lu, Song
    Liu, Xiaolong
    Hammour, Tarek
    Brogi, Edi
    Snuderl, Matija
    Iafrate, A. John
    Reis-Filho, Jorge S.
    Schnitt, Stuart J.
    [J]. CANCER RESEARCH, 2016, 76 (24) : 7118 - 7129
  • [15] A COEFFICIENT OF AGREEMENT FOR NOMINAL SCALES
    COHEN, J
    [J]. EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT, 1960, 20 (01) : 37 - 46
  • [16] Comprehensive molecular profiling of lung adenocarcinoma
    Collisson, Eric A.
    Campbell, Joshua D.
    Brooks, Angela N.
    Berger, Alice H.
    Lee, William
    Chmielecki, Juliann
    Beer, David G.
    Cope, Leslie
    Creighton, Chad J.
    Danilova, Ludmila
    Ding, Li
    Getz, Gad
    Hammerman, Peter S.
    Hayes, D. Neil
    Hernandez, Bryan
    Herman, James G.
    Heymach, John V.
    Jurisica, Igor
    Kucherlapati, Raju
    Kwiatkowski, David
    Ladanyi, Marc
    Robertson, Gordon
    Schultz, Nikolaus
    Shen, Ronglai
    Sinha, Rileen
    Sougnez, Carrie
    Tsao, Ming-Sound
    Travis, William D.
    Weinstein, John N.
    Wigle, Dennis A.
    Wilkerson, Matthew D.
    Chu, Andy
    Cherniack, Andrew D.
    Hadjipanayis, Angela
    Rosenberg, Mara
    Weisenberger, Daniel J.
    Laird, Peter W.
    Radenbaugh, Amie
    Ma, Singer
    Stuart, Joshua M.
    Byers, Lauren Averett
    Baylin, Stephen B.
    Govindan, Ramaswamy
    Meyerson, Matthew
    Rosenberg, Mara
    Gabriel, Stacey B.
    Cibulskis, Kristian
    Sougnez, Carrie
    Kim, Jaegil
    Stewart, Chip
    [J]. NATURE, 2014, 511 (7511) : 543 - 550
  • [17] Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent
    Cruz-Roa, Angel
    Gilmore, Hannah
    Basavanhally, Ajay
    Feldman, Michael
    Ganesan, Shridar
    Shih, Natalie N. C.
    Tomaszewski, John
    Gonzalez, Fabio A.
    Madabhushi, Anant
    [J]. SCIENTIFIC REPORTS, 2017, 7
  • [18] Automatic segmentation of histopathological slides of renal tissue using deep learning
    de Bel, Thomas
    Hermsen, Meyke
    Smeets, Bart
    Hilbrands, Luuk
    van der Laak, Jeroen
    Litjens, Geert
    [J]. MEDICAL IMAGING 2018: DIGITAL PATHOLOGY, 2018, 10581
  • [19] Mutation incidence and coincidence in non small-cell lung cancer: meta-analyses by ethnicity and histology (mutMap)
    Dearden, S.
    Stevens, J.
    Wu, Y. -L.
    Blowers, D.
    [J]. ANNALS OF ONCOLOGY, 2013, 24 (09) : 2371 - 2376
  • [20] Standardization of Epidermal Growth Factor Receptor (EGFR) Measurement by Quantitative Immunofluorescence and Impact on Antibody-Based Mutation Detection in Non Small Cell Lung Cancer
    Dimou, Anastasios
    Agarwal, Seema
    Anagnostou, Valsamo
    Viray, Hollis
    Christensen, Stephen
    Rothberg, Bonnie Gould
    Zolota, Vassiliki
    Syrigos, Konstantinos
    Rimm, David L.
    [J]. AMERICAN JOURNAL OF PATHOLOGY, 2011, 179 (02) : 580 - 589