{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,7]],"date-time":"2026-01-07T23:03:13Z","timestamp":1767826993645,"version":"3.49.0"},"reference-count":39,"publisher":"Oxford University Press (OUP)","issue":"13","license":[{"start":{"date-parts":[[2017,1,31]],"date-time":"2017-01-31T00:00:00Z","timestamp":1485820800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/about_us\/legal\/notices"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2017,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Motivation<\/jats:title><jats:p>Whole genome expression profiling of large cohorts of different types of cancer led to the identification of distinct molecular subcategories (subtypes) that may partially explain the observed inter-tumoral heterogeneity. This is also the case of colorectal cancer (CRC) where several such categorizations have been proposed. Despite recent developments, the problem of subtype definition and recognition remains open, one of the causes being the intrinsic heterogeneity of each tumor, which is difficult to estimate from gene expression profiles. However, one of the observations of these studies indicates that there may be links between the dominant tumor morphology characteristics and the molecular subtypes. Benefiting from a large collection of CRC samples, comprising both gene expression and histopathology images, we investigated the possibility of building image-based classifiers able to predict the molecular subtypes. We employed deep convolutional neural networks for extracting local descriptors which were then used for constructing a dictionary-based representation of each tumor sample. A set of support vector machine classifiers were trained to solve different binary decision problems, their combined outputs being used to predict one of the five molecular subtypes.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>A hierarchical decomposition of the multi-class problem was obtained with an overall accuracy of 0.84 (95%CI=0.79\u20130.88). The predictions from the image-based classifier showed significant prognostic value similar to their molecular counterparts.<\/jats:p><\/jats:sec><jats:sec><jats:title>Availability and Implementation<\/jats:title><jats:p>Source code used for the image analysis is freely available from https:\/\/github.com\/higex\/qpath.<\/jats:p><\/jats:sec><jats:sec><jats:title>Supplementary information<\/jats:title><jats:p>Supplementary data are available at Bioinformatics online.<\/jats:p><\/jats:sec>","DOI":"10.1093\/bioinformatics\/btx027","type":"journal-article","created":{"date-parts":[[2017,1,30]],"date-time":"2017-01-30T13:28:22Z","timestamp":1485782902000},"page":"2002-2009","source":"Crossref","is-referenced-by-count":26,"title":["Image-based surrogate biomarkers for molecular subtypes of colorectal cancer"],"prefix":"10.1093","volume":"33","author":[{"given":"Vlad","family":"Popovici","sequence":"first","affiliation":[{"name":"Faculty of Medicine, Institute of Biostatistics and Analyses, Masaryk University, Brno, Czech Republic"}]},{"given":"Eva","family":"Budinsk\u00e1","sequence":"additional","affiliation":[{"name":"Faculty of Science, Research Centre for Toxic Compounds in the Environment, Masaryk University, Brno, Czech Republic"}]},{"given":"Ladislav","family":"Du\u0161ek","sequence":"additional","affiliation":[{"name":"Faculty of Medicine, Institute of Biostatistics and Analyses, Masaryk University, Brno, Czech Republic"}]},{"given":"Michal","family":"Kozubek","sequence":"additional","affiliation":[{"name":"Faculty of Informatics, Masaryk University, Brno, Czech Republic"}]},{"given":"Fred","family":"Bosman","sequence":"additional","affiliation":[{"name":"University Institute of Pathology, University of Lausanne, Switzerland"}]}],"member":"286","published-online":{"date-parts":[[2017,1,31]]},"reference":[{"key":"2023020206012311000_btx027-B1","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1148\/radiol.14121031","article-title":"Computerized image analysis for identifying triple-negative breast cancers and differentiating them from other molecular subtypes of breast cancer on dynamic contrast-enhanced MR images: a feasibility study","volume":"272","author":"Agner","year":"2014","journal-title":"Radiology"},{"key":"2023020206012311000_btx027-B2","first-page":"119","article-title":"Approximate is better than \u201cExact\u201d for interval estimation of binomial proportions","volume":"52","author":"Agresti","year":"1998","journal-title":"Am. Stat"},{"key":"2023020206012311000_btx027-B3","first-page":"1578","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition","author":"Arandjelovic","year":"2013"},{"key":"2023020206012311000_btx027-B4","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1002\/path.4212","article-title":"Gene expression patterns unveil a new level of molecular heterogeneity in colorectal cancer","volume":"231","author":"Budinsk\u00e1","year":"2013","journal-title":"J. Pathol"},{"key":"2023020206012311000_btx027-B5","first-page":"1168","volume-title":"International Symposium on Biomedical Imaging","author":"Budinsk\u00e1","year":"2016"},{"key":"2023020206012311000_btx027-B6","doi-asserted-by":"crossref","first-page":"484.","DOI":"10.1186\/1471-2105-12-484","article-title":"Morphometic analysis of TCGA glioblastoma multiforme","volume":"12","author":"Chang","year":"2011","journal-title":"BMC Bioinform"},{"key":"2023020206012311000_btx027-B7","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1007\/s11263-014-0790-9","article-title":"Stacked predictive sparse decomposition for classification of histology sections","volume":"113","author":"Chang","year":"2015","journal-title":"Int. J. Comput. Vis"},{"key":"2023020206012311000_btx027-B8","volume-title":"British Machine Vision Conference","author":"Chatfield","year":"2014"},{"key":"2023020206012311000_btx027-B9","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1007\/s11263-015-0872-3","article-title":"Deep filter banks for texture recognition, description, and segmentation","volume":"118","author":"Cimpoi","year":"2016","journal-title":"Int. J. Comput. Vis"},{"key":"2023020206012311000_btx027-B10","doi-asserted-by":"crossref","first-page":"317","DOI":"10.1136\/amiajnl-2011-000700","article-title":"Integrated morphologic analysis for the identification and characterization of disease subtypes","volume":"19","author":"Cooper","year":"2012","journal-title":"J. Am. Med. Inf. Assoc"},{"key":"2023020206012311000_btx027-B11","first-page":"59","volume-title":"Workshop on Statistical Learning in Computer Vision, ECCV","author":"Csurka","year":"2004"},{"key":"2023020206012311000_btx027-B12","doi-asserted-by":"crossref","first-page":"614","DOI":"10.1038\/nm.3174","article-title":"Poor-prognosis colon cancer is defined by a molecularly distinct subtype and develops from serrated precursor lesions","volume":"19","author":"De Sousa","year":"2013","journal-title":"Nat. Med"},{"key":"2023020206012311000_btx027-B13","doi-asserted-by":"crossref","first-page":"vi23","DOI":"10.1093\/annonc\/mds191","article-title":"Imaging of triple-negative breast cancer","volume":"23","author":"Dogan","year":"2012","journal-title":"Ann. Oncol.: Off. J. Eur. Soc. Med. Oncol.\/ESMO"},{"key":"2023020206012311000_btx027-B14","doi-asserted-by":"crossref","first-page":"4095","DOI":"10.1158\/1078-0432.CCR-16-0032","article-title":"Challenging the cancer molecular stratification dogma: intratumoral heterogeneity undermines consensus molecular subtypes and potential diagnostic value in colorectal cancer","volume":"22","author":"Dunne","year":"2016","journal-title":"Clin. Cancer Res"},{"key":"2023020206012311000_btx027-B15","doi-asserted-by":"crossref","first-page":"1350","DOI":"10.1038\/nm.3967","article-title":"The consensus molecular subtypes of colorectal cancer","volume":"21","author":"Guinney","year":"2015","journal-title":"Nat. Med"},{"key":"2023020206012311000_btx027-B16","first-page":"3304","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"J\u00e9gou","year":"2010"},{"key":"2023020206012311000_btx027-B17","doi-asserted-by":"crossref","first-page":"1704","DOI":"10.1109\/TPAMI.2011.235","article-title":"Aggregating local image descriptors into compact codes","volume":"34","author":"J\u00e9gou","year":"2012","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell"},{"key":"2023020206012311000_btx027-B18","first-page":"1397","volume-title":"IEEE International Symposium on Biomedical Imaging","author":"Kawahara","year":"2016"},{"key":"2023020206012311000_btx027-B19","first-page":"1106","volume-title":"NIPS","author":"Krizhevsky","year":"2012"},{"key":"2023020206012311000_btx027-B20","doi-asserted-by":"crossref","first-page":"16317\u201316317.","DOI":"10.1038\/srep16317","article-title":"Quantitative histology analysis of the ovarian tumour microenvironment","volume":"5","author":"Lan","year":"2015","journal-title":"Sci. Rep"},{"key":"2023020206012311000_btx027-B21","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1162\/neco.1989.1.4.541","article-title":"Backpropagation applied to handwritten zip code recognition","volume":"1","author":"LeCun","year":"1989","journal-title":"Neural Comput"},{"key":"2023020206012311000_btx027-B22","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"2023020206012311000_btx027-B23","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbw044","article-title":"Embracing an integromic approach to tissue biomarker research in cancer: Perspectives and lessons learned","author":"Li","year":"2016","journal-title":"Brief. Bioinform."},{"key":"2023020206012311000_btx027-B24","first-page":"1150","volume-title":"Proceedings of the IEEE International Conference on Computer Vision","author":"Lowe","year":"1999"},{"key":"2023020206012311000_btx027-B25","doi-asserted-by":"crossref","first-page":"e1001453.","DOI":"10.1371\/journal.pmed.1001453","article-title":"Gene expression classification of colon cancer into molecular subtypes: characterization, validation, and prognostic value","volume":"10","author":"Marisa","year":"2013","journal-title":"PLoS Med"},{"key":"2023020206012311000_btx027-B26","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1016\/0031-3203(95)00067-4","article-title":"A comparative study of texture measures with classification based on featured distributions","volume":"29","author":"Ojala","year":"1996","journal-title":"Pattern Recognit"},{"key":"2023020206012311000_btx027-B27","doi-asserted-by":"crossref","first-page":"747","DOI":"10.1038\/35021093","article-title":"Molecular portraits of human breast tumours","volume":"406","author":"Perou","year":"2000","journal-title":"Nature"},{"key":"2023020206012311000_btx027-B28","first-page":"1","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Perronnin","year":"2007"},{"key":"2023020206012311000_btx027-B29","doi-asserted-by":"crossref","first-page":"552","DOI":"10.1002\/ijc.28387","article-title":"Colorectal cancer intrinsic subtypes predict chemotherapy benefit, deficient mismatch repair and epithelial-to-mesenchymal transition","volume":"134","author":"Roepman","year":"2013","journal-title":"Int. J. Cancer"},{"key":"2023020206012311000_btx027-B30","first-page":"234","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Ronneberger","year":"2015"},{"key":"2023020206012311000_btx027-B31","doi-asserted-by":"crossref","first-page":"619","DOI":"10.1038\/nm.3175","article-title":"A colorectal cancer classification system that associates cellular phenotype and responses to therapy","volume":"19","author":"Sadanandam","year":"2013","journal-title":"Nat. Med"},{"key":"2023020206012311000_btx027-B32","doi-asserted-by":"crossref","first-page":"27.","DOI":"10.4103\/2153-3539.119005","article-title":"OpenSlide: a vendor-neutral software foundation for digital pathology","volume":"4","author":"Satyanarayanan","year":"2013","journal-title":"J. Pathol. Inform"},{"key":"2023020206012311000_btx027-B33","doi-asserted-by":"crossref","first-page":"1470","DOI":"10.1109\/ICCV.2003.1238663","volume-title":"Proceedings of the IEEE International Conference on Computer Vision","author":"Sivic","year":"2003"},{"key":"2023020206012311000_btx027-B34","doi-asserted-by":"crossref","first-page":"318","DOI":"10.1038\/modpathol.2016.34","article-title":"Digital image analysis outperforms manual biomarker assessment in breast cancer","volume":"29","author":"St\u00e5lhammar","year":"2016","journal-title":"Modern Pathol"},{"key":"2023020206012311000_btx027-B35","first-page":"1","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition","author":"Szegedy","year":"2015"},{"key":"2023020206012311000_btx027-B36","doi-asserted-by":"crossref","first-page":"3117","DOI":"10.1200\/JCO.2008.21.6663","article-title":"Randomized phase III trial comparing biweekly infusional fluorouracil\/leucovorin alone or with irinotecan in the adjuvant treatment of stage III colon cancer: PETACC-3","volume":"27","author":"Van Cutsem","year":"2009","journal-title":"J. Clin. Oncol"},{"key":"2023020206012311000_btx027-B37","first-page":"286","volume-title":"2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI","author":"van Ginneken","year":"2015"},{"key":"2023020206012311000_btx027-B38","first-page":"1","volume-title":"ACM International Conference on Multimedia","author":"Vedaldi","year":"2015"},{"key":"2023020206012311000_btx027-B39","doi-asserted-by":"crossref","first-page":"192","DOI":"10.1016\/j.molonc.2010.04.004","article-title":"Histological types of breast cancer: how special are they?","volume":"4","author":"Weigelt","year":"2010","journal-title":"Molecular Oncology"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/33\/13\/2002\/49040387\/bioinformatics_33_13_2002.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/33\/13\/2002\/49040387\/bioinformatics_33_13_2002.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,14]],"date-time":"2025-06-14T21:40:27Z","timestamp":1749937227000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/33\/13\/2002\/2964790"}},"subtitle":[],"editor":[{"given":"Robert","family":"Murphy","sequence":"additional","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2017,1,31]]},"references-count":39,"journal-issue":{"issue":"13","published-print":{"date-parts":[[2017,7,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btx027","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2017,7,1]]},"published":{"date-parts":[[2017,1,31]]}}}