{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,27]],"date-time":"2026-04-27T16:48:46Z","timestamp":1777308526934,"version":"3.51.4"},"reference-count":16,"publisher":"World Scientific Pub Co Pte Ltd","issue":"02","funder":[{"DOI":"10.13039\/100010663","name":"H2020 European Research Council","doi-asserted-by":"publisher","award":["H2020-ICT-24-2015-GA:688592"],"award-info":[{"award-number":["H2020-ICT-24-2015-GA:688592"]}],"id":[{"id":"10.13039\/100010663","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Med. Robot. Res."],"published-print":{"date-parts":[[2018,6]]},"abstract":"<jats:p> Early diagnosis is essential for the successful treatment of bowel cancers including colorectal cancer (CRC), and capsule endoscopic imaging with robotic actuation can be a valuable diagnostic tool when combined with automated image analysis. We present a deep learning rooted detection and segmentation framework for recognizing lesions in colonoscopy and capsule endoscopy images. We restructure established convolution architectures, such as VGG and ResNets, by converting them into fully-connected convolution networks (FCNs), fine-tune them and study their capabilities for polyp segmentation and detection. We additionally use shape-from-shading (SfS) to recover depth and provide a richer representation of the tissue\u2019s structure in colonoscopy images. Depth is incorporated into our network models as an additional input channel to the RGB information and we demonstrate that the resulting network yields improved performance. Our networks are tested on publicly available datasets and the most accurate segmentation model achieved a mean segmentation interception over union (IU) of 47.78% and 56.95% on the ETIS-Larib and CVC-Colon datasets, respectively. For polyp detection, the top performing models we propose surpass the current state-of-the-art with detection recalls superior to 90% for all datasets tested. To our knowledge, we present the first work to use FCNs for polyp segmentation in addition to proposing a novel combination of SfS and RGB that boosts performance. <\/jats:p>","DOI":"10.1142\/s2424905x18400020","type":"journal-article","created":{"date-parts":[[2018,2,22]],"date-time":"2018-02-22T22:15:21Z","timestamp":1519337721000},"page":"1840002","source":"Crossref","is-referenced-by-count":67,"title":["Towards a Computed-Aided Diagnosis System in Colonoscopy: Automatic Polyp Segmentation Using Convolution Neural Networks"],"prefix":"10.1142","volume":"03","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1301-2700","authenticated-orcid":false,"given":"Patrick","family":"Brandao","sequence":"first","affiliation":[{"name":"Centre for Medical Image Computing, University College London, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Odysseas","family":"Zisimopoulos","sequence":"additional","affiliation":[{"name":"Centre for Medical Image Computing, University College London, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Evangelos","family":"Mazomenos","sequence":"additional","affiliation":[{"name":"Centre for Medical Image Computing, University College London, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gastone","family":"Ciuti","sequence":"additional","affiliation":[{"name":"The BioRobotics Institute, Scuola Superiore Sant\u2019Anna, Pisa, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jorge","family":"Bernal","sequence":"additional","affiliation":[{"name":"Department of Computer Science Universitat Autnoma de Barcelona, Barcelona, Spain"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Marco","family":"Visentini-Scarzanella","sequence":"additional","affiliation":[{"name":"Multimedia Laboratory, Corporate Research and Development Center, Toshiba Kawasaki, Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Arianna","family":"Menciassi","sequence":"additional","affiliation":[{"name":"The BioRobotics Institute, Scuola Superiore Sant\u2019Anna, Pisa, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paolo","family":"Dario","sequence":"additional","affiliation":[{"name":"The BioRobotics Institute, Scuola Superiore Sant\u2019Anna, Pisa, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anastasios","family":"Koulaouzidis","sequence":"additional","affiliation":[{"name":"UEndoscopy Unit, The Royal Infirmary of Edinburgh, Edinburgh, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alberto","family":"Arezzo","sequence":"additional","affiliation":[{"name":"Department of Surgical Sciences, University of Turin, Turin, Italy"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David J","family":"Hawkes","sequence":"additional","affiliation":[{"name":"Centre for Medical Image Computing, University College London, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Danail","family":"Stoyanov","sequence":"additional","affiliation":[{"name":"Centre for Medical Image Computing, University College London, London, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"219","published-online":{"date-parts":[[2018,3,13]]},"reference":[{"issue":"1","key":"S2424905X18400020BIB001","first-page":"1","volume":"11","author":"Ciuti G.","year":"2016","journal-title":"J. Microbio Robot."},{"key":"S2424905X18400020BIB002","doi-asserted-by":"publisher","DOI":"10.1055\/s-0031-1291666"},{"key":"S2424905X18400020BIB003","doi-asserted-by":"publisher","DOI":"10.1109\/TITB.2003.813794"},{"issue":"25","key":"S2424905X18400020BIB004","first-page":"346","volume":"22","author":"Ameling S.","year":"2009","journal-title":"Bildverarbeitung die Medizin"},{"key":"S2424905X18400020BIB005","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2015.2487997"},{"key":"S2424905X18400020BIB006","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2015.02.007"},{"key":"S2424905X18400020BIB007","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"issue":"1","key":"S2424905X18400020BIB008","first-page":"98136","volume":"111","author":"Everingham M.","year":"2014","journal-title":"Int. J. Comput. Vis."},{"key":"S2424905X18400020BIB010","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"S2424905X18400020BIB020","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2012.03.002"},{"key":"S2424905X18400020BIB023","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2535302"},{"key":"S2424905X18400020BIB025","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2012.03.002"},{"key":"S2424905X18400020BIB031","doi-asserted-by":"publisher","DOI":"10.1007\/s00138-005-0006-y"},{"key":"S2424905X18400020BIB034","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2009.03.025"},{"key":"S2424905X18400020BIB036","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2003.11.007"},{"key":"S2424905X18400020BIB038","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2017.2664042"}],"container-title":["Journal of Medical Robotics Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.worldscientific.com\/doi\/pdf\/10.1142\/S2424905X18400020","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2019,9,21]],"date-time":"2019-09-21T01:37:08Z","timestamp":1569029828000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.worldscientific.com\/doi\/abs\/10.1142\/S2424905X18400020"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,3,13]]},"references-count":16,"journal-issue":{"issue":"02","published-online":{"date-parts":[[2018,3,13]]},"published-print":{"date-parts":[[2018,6]]}},"alternative-id":["10.1142\/S2424905X18400020"],"URL":"https:\/\/doi.org\/10.1142\/s2424905x18400020","relation":{},"ISSN":["2424-905X","2424-9068"],"issn-type":[{"value":"2424-905X","type":"print"},{"value":"2424-9068","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,3,13]]}}}