{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T06:48:34Z","timestamp":1789541314643,"version":"build-2803163510"},"reference-count":34,"publisher":"BMJ","issue":"12","license":[{"start":{"date-parts":[[2022,2,2]],"date-time":"2022-02-02T00:00:00Z","timestamp":1643760000000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000360","name":"Scottish Funding Council","doi-asserted-by":"publisher","award":["Cancer Innovation Challenge 2018"],"award-info":[{"award-number":["Cancer Innovation Challenge 2018"]}],"id":[{"id":"10.13039\/501100000360","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000351","name":"British Lung Foundation","doi-asserted-by":"publisher","award":["MPG16-7"],"award-info":[{"award-number":["MPG16-7"]}],"id":[{"id":"10.13039\/501100000351","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000589","name":"Chief Scientist Office","doi-asserted-by":"publisher","award":["ETM\/285"],"award-info":[{"award-number":["ETM\/285"]}],"id":[{"id":"10.13039\/501100000589","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["Thorax"],"published-print":{"date-parts":[[2022,12]]},"abstract":"<jats:sec><jats:title>Background<\/jats:title><jats:p>In malignant pleural mesothelioma (MPM), complex tumour morphology results in inconsistent radiological response assessment. Promising volumetric methods require automation to be practical. We developed a fully automated Convolutional Neural Network (CNN) for this purpose, performed blinded validation and compared CNN and human response classification and survival prediction in patients treated with chemotherapy.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>In a multicentre retrospective cohort study; 183 CT datasets were split into training and internal validation (123 datasets (80 fully annotated); 108 patients; 1 centre) and external validation (60 datasets (all fully annotated); 30 patients; 3 centres). Detailed manual annotations were used to train the CNN, which used two-dimensional U-Net architecture. CNN performance was evaluated using correlation, Bland-Altman and Dice agreement. Volumetric response\/progression were defined as \u226430%\/\u226520% change and compared with modified Response Evaluation Criteria In Solid Tumours (mRECIST) by Cohen\u2019s kappa. Survival was assessed using Kaplan-Meier methodology.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Human and artificial intelligence (AI) volumes were strongly correlated (validation set r=0.851, p&lt;0.0001). Agreement was strong (validation set mean bias +31 cm<jats:sup>3<\/jats:sup>(p=0.182), 95% limits 345 to +407 cm<jats:sup>3<\/jats:sup>). Infrequent AI segmentation errors (4\/60 validation cases) were associated with fissural tumour, contralateral pleural thickening and adjacent atelectasis. Human and AI volumetric responses agreed in 20\/30 (67%) validation cases \u03ba=0.439 (0.178 to 0.700). AI and mRECIST agreed in 16\/30 (55%) validation cases \u03ba=0.284 (0.026 to 0.543). Higher baseline tumour volume was associated with shorter survival.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>We have developed and validated the first fully automated CNN for volumetric MPM segmentation. CNN performance may be further improved by enriching future training sets with morphologically challenging features. Volumetric response thresholds require further calibration in future studies.<\/jats:p><\/jats:sec>","DOI":"10.1136\/thoraxjnl-2021-217808","type":"journal-article","created":{"date-parts":[[2022,2,2]],"date-time":"2022-02-02T18:28:56Z","timestamp":1643826536000},"page":"1251-1259","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":43,"title":["Fully automated volumetric measurement of malignant pleural mesothelioma by deep learning AI: validation and comparison with modified RECIST response criteria"],"prefix":"10.1136","volume":"77","author":[{"given":"Andrew C","family":"Kidd","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Owen","family":"Anderson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gordon W","family":"Cowell","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexander J","family":"Weir","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeremy P","family":"Voisey","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matthew","family":"Evison","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Selina","family":"Tsim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Keith A","family":"Goatman","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kevin G","family":"Blyth","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2022,2,2]]},"reference":[{"key":"2023100202050733000_77.12.1251.1","first-page":"2636","article-title":"Phase III study of pemetrexed in combination with cisplatin versus cisplatin alone in patients with malignant pleural mesothelioma","volume":"21","author":"Vogelzang","year":"2003","journal-title":"Journal of Oncol"},{"key":"2023100202050733000_77.12.1251.2","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(20)32714-8"},{"key":"2023100202050733000_77.12.1251.3","doi-asserted-by":"crossref","unstructured":"Fennell DA , Kirkpatrick E , Cozens K , et al . Confirm: a double-blind, placebo-controlled phase III clinical trial investigating the effect of nivolumab in patients with relapsed mesothelioma: study protocol for a randomised controlled trial. Trials 2018;19:233.doi:10.1186\/s13063-018-2602-y","DOI":"10.1186\/s13063-018-2602-y"},{"key":"2023100202050733000_77.12.1251.4","doi-asserted-by":"crossref","unstructured":"van KRJ , JGJV A , de BH . Inadequacy of the RECIST criteria for response evaluation in patients with malignant pleural mesothelioma. Poster Abstracts of the 13th Annual British Thoracic Oncology Group Conference 2015; 28-30 January 2015, Dublin, Ireland, 2004:63\u20139.","DOI":"10.1016\/S0169-5002(03)00292-7"},{"key":"2023100202050733000_77.12.1251.5","doi-asserted-by":"publisher","DOI":"10.1093\/annonc\/mdh059"},{"key":"2023100202050733000_77.12.1251.6","doi-asserted-by":"publisher","DOI":"10.1118\/1.1688211"},{"key":"2023100202050733000_77.12.1251.7","doi-asserted-by":"publisher","DOI":"10.1097\/JTO.0000000000000211"},{"key":"2023100202050733000_77.12.1251.8","doi-asserted-by":"publisher","DOI":"10.1593\/tlo.11232"},{"key":"2023100202050733000_77.12.1251.9","doi-asserted-by":"crossref","first-page":"155","DOI":"10.1102\/1470-7330.2011.0022","article-title":"Volume and attenuation computed tomography measurements for interim evaluation of Hodgkin and follicular lymphoma as an additional surrogate parameter for more confident response monitoring: a pilot study","volume":"11","author":"Spira","year":"2011","journal-title":"Cancer Imaging"},{"key":"2023100202050733000_77.12.1251.10","doi-asserted-by":"crossref","first-page":"737","DOI":"10.1111\/j.1445-2197.2012.06238.x","article-title":"Different approaches to volume assessment of lymph nodes in computer tomography scans of head and neck squamous cell carcinoma in comparison with a real gold standard","volume":"82","author":"Mueller","year":"2012","journal-title":"ANZ J Surg"},{"key":"2023100202050733000_77.12.1251.11","doi-asserted-by":"publisher","DOI":"10.1111\/j.1365-2125.2012.04179.x"},{"key":"2023100202050733000_77.12.1251.12","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1016\/j.lungcan.2020.09.025","article-title":"A comparison between MRI and CT in the assessment of primary tumour volume in mesothelioma","volume":"150","author":"Tsim","year":"2020","journal-title":"Lung Cancer"},{"key":"2023100202050733000_77.12.1251.13","doi-asserted-by":"crossref","first-page":"1059","DOI":"10.1016\/j.athoracsur.2016.06.069","article-title":"A multicenter study of volumetric computed tomography for staging malignant pleural mesothelioma","volume":"102","author":"Rusch","year":"2016","journal-title":"Ann Thorac Surg"},{"key":"2023100202050733000_77.12.1251.14","doi-asserted-by":"crossref","first-page":"1335","DOI":"10.1016\/j.jtho.2016.04.027","article-title":"North American multicenter volumetric CT study for clinical staging of malignant pleural mesothelioma: feasibility and logistics of setting up a quantitative imaging study","volume":"11","author":"Gill","year":"2016","journal-title":"J Thorac Oncol"},{"key":"2023100202050733000_77.12.1251.15","doi-asserted-by":"crossref","first-page":"84","DOI":"10.1145\/3065386","article-title":"ImageNet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Commun ACM"},{"key":"2023100202050733000_77.12.1251.16","doi-asserted-by":"publisher","DOI":"10.1136\/bmjqs-2015-004411"},{"key":"2023100202050733000_77.12.1251.17","doi-asserted-by":"crossref","first-page":"1705","DOI":"10.1016\/j.jtho.2021.05.018","article-title":"Serum proteomics and plasma fibulin-3 in differentiation of mesothelioma from asbestos-exposed controls and patients with other pleural diseases","volume":"16","author":"Tsim","year":"2021","journal-title":"J Thorac Oncol"},{"key":"2023100202050733000_77.12.1251.18","doi-asserted-by":"crossref","first-page":"S26","DOI":"10.1016\/S0169-5002(18)30090-4","article-title":"An update regarding the prediction of resistance to chemotherapy using somatic copy number variation in mesothelioma (PriSM) study","volume":"115","author":"Blyth","year":"2018","journal-title":"Lung Cancer"},{"key":"2023100202050733000_77.12.1251.19","doi-asserted-by":"publisher","DOI":"10.1016\/S1010-7940(03)00609-2"},{"key":"2023100202050733000_77.12.1251.20","doi-asserted-by":"publisher","DOI":"10.1097\/RLI.0b013e3181817b3d"},{"key":"2023100202050733000_77.12.1251.21","doi-asserted-by":"crossref","unstructured":"Ronneberger O , Fischer P . Brox T. U-Net: Convolutional networks for biomedical image segmentation. Arxiv 2015.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"2023100202050733000_77.12.1251.22","doi-asserted-by":"crossref","unstructured":"Anderson O , Kidd A , Goatman K . Fully automated volumetric measurement of malignant pleural mesothelioma from computed tomography images by deep learning: preliminary results of an internal validation 2020:64\u201373.","DOI":"10.5220\/0008976100640073"},{"key":"2023100202050733000_77.12.1251.23","doi-asserted-by":"crossref","unstructured":"Oxnard GR , Armato SG , Kindler HL . Modeling of mesothelioma growth demonstrates weaknesses of current response criteria. Poster Abstracts of the 13th Annual British Thoracic Oncology Group Conference 2015; 28-30 January 2015, Dublin, Ireland, 2006:141\u20138.","DOI":"10.1016\/j.lungcan.2005.12.013"},{"key":"2023100202050733000_77.12.1251.24","doi-asserted-by":"publisher","DOI":"10.1016\/S0022-5223(98)70274-0"},{"key":"2023100202050733000_77.12.1251.25","doi-asserted-by":"publisher","DOI":"10.1097\/JTO.0b013e31817c73d6"},{"key":"2023100202050733000_77.12.1251.26","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejca.2012.05.018"},{"key":"2023100202050733000_77.12.1251.27","doi-asserted-by":"crossref","first-page":"1405","DOI":"10.1016\/S0140-6736(15)01238-6","article-title":"Bevacizumab for newly diagnosed pleural mesothelioma in the mesothelioma Avastin cisplatin pemetrexed study (maps): a randomised, controlled, open-label, phase 3 trial","volume":"387","author":"Zalcman","year":"2016","journal-title":"The Lancet"},{"key":"2023100202050733000_77.12.1251.28","doi-asserted-by":"publisher","DOI":"10.1183\/09031936.00146110"},{"key":"2023100202050733000_77.12.1251.29","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2009.02.002"},{"key":"2023100202050733000_77.12.1251.30","doi-asserted-by":"crossref","first-page":"529","DOI":"10.1007\/s11548-016-1511-3","article-title":"Computer-Aided volumetric assessment of malignant pleural mesothelioma on CT using a random walk-based method","volume":"12","author":"Chen","year":"2017","journal-title":"Int J Comput Assist Radiol Surg"},{"key":"2023100202050733000_77.12.1251.31","doi-asserted-by":"publisher","DOI":"10.1117\/1.jmi.5.1.015002"},{"key":"2023100202050733000_77.12.1251.32","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1016\/j.compmedimag.2017.05.006","article-title":"Malignant pleural mesothelioma segmentation for photodynamic therapy planning","volume":"65","author":"Brahim","year":"2018","journal-title":"Comput Med Imaging Graph"},{"key":"2023100202050733000_77.12.1251.33","doi-asserted-by":"publisher","DOI":"10.1118\/1.3525836"},{"key":"2023100202050733000_77.12.1251.34","doi-asserted-by":"publisher","DOI":"10.1117\/1.jmi.7.5.054501"}],"container-title":["Thorax"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/thoraxjnl-2021-217808","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,2]],"date-time":"2023-10-02T09:08:13Z","timestamp":1696237693000},"score":1,"resource":{"primary":{"URL":"https:\/\/thorax.bmj.com\/lookup\/doi\/10.1136\/thoraxjnl-2021-217808"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,2,2]]},"references-count":34,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2022,11,15]]},"published-print":{"date-parts":[[2022,12]]}},"alternative-id":["10.1136\/thoraxjnl-2021-217808"],"URL":"https:\/\/doi.org\/10.1136\/thoraxjnl-2021-217808","relation":{},"ISSN":["0040-6376","1468-3296"],"issn-type":[{"value":"0040-6376","type":"print"},{"value":"1468-3296","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,2,2]]}}}