{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T15:21:21Z","timestamp":1781536881600,"version":"3.54.5"},"reference-count":26,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2021,10,15]],"date-time":"2021-10-15T00:00:00Z","timestamp":1634256000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:p><jats:bold>Purpose:<\/jats:bold> Portable chest radiographs are diagnostically indispensable in intensive care units (ICU). This study aimed to determine if the proposed machine learning technique increased in accuracy as the number of radiograph readings increased and if it was accurate in a clinical setting.<\/jats:p><jats:p><jats:bold>Methods:<\/jats:bold> Two independent data sets of portable chest radiographs (<jats:italic>n<\/jats:italic> = 380, a single Japanese hospital; <jats:italic>n<\/jats:italic> = 1,720, The National Institution of Health [NIH] ChestX-ray8 dataset) were analyzed. Each data set was divided training data and study data. Images were classified as atelectasis, pleural effusion, pneumonia, or no emergency. DenseNet-121, as a pre-trained deep convolutional neural network was used and ensemble learning was performed on the best-performing algorithms. Diagnostic accuracy and processing time were compared to those of ICU physicians.<\/jats:p><jats:p><jats:bold>Results:<\/jats:bold> In the single Japanese hospital data, the area under the curve (AUC) of diagnostic accuracy was 0.768. The area under the curve (AUC) of diagnostic accuracy significantly improved as the number of radiograph readings increased from 25 to 100% in the NIH data set. The AUC was higher than 0.9 for all categories toward the end of training with a large sample size. The time to complete 53 radiographs by machine learning was 70 times faster than the time taken by ICU physicians (9.66 s vs. 12 min). The diagnostic accuracy was higher by machine learning than by ICU physicians in most categories (atelectasis, AUC 0.744 vs. 0.555, <jats:italic>P<\/jats:italic> &amp;lt; 0.05; pleural effusion, 0.856 vs. 0.706, <jats:italic>P<\/jats:italic> &amp;lt; 0.01; pneumonia, 0.720 vs. 0.744, <jats:italic>P<\/jats:italic> = 0.88; no emergency, 0.751 vs. 0.698, <jats:italic>P<\/jats:italic> = 0.47).<\/jats:p><jats:p><jats:bold>Conclusions:<\/jats:bold> We developed an automatic detection system for portable chest radiographs in ICU setting; its performance was superior and quite faster than ICU physicians.<\/jats:p>","DOI":"10.3389\/fmed.2021.676277","type":"journal-article","created":{"date-parts":[[2021,10,16]],"date-time":"2021-10-16T00:08:11Z","timestamp":1634342891000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Superiority of Supervised Machine Learning on Reading Chest X-Rays in Intensive Care Units"],"prefix":"10.3389","volume":"8","author":[{"given":"Kumiko","family":"Tanaka","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Taka-aki","family":"Nakada","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nozomi","family":"Takahashi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takahiro","family":"Dozono","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuichiro","family":"Yoshimura","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hajime","family":"Yokota","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Takuro","family":"Horikoshi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Toshiya","family":"Nakaguchi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Koichiro","family":"Shinozaki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2021,10,15]]},"reference":[{"key":"B1","first-page":"469","article-title":"Chest radiography in intensive care units","volume":"129","author":"Savoca","year":"1978","journal-title":"West J Med."},{"key":"B2","doi-asserted-by":"publisher","first-page":"801","DOI":"10.1097\/00003246-199705000-00015","article-title":"Routine portable chest radiographs in the medical intensive care unit: effects and costs","volume":"25","author":"Brainsky","year":"1997","journal-title":"Crit Care Med"},{"key":"B3","doi-asserted-by":"publisher","first-page":"427","DOI":"10.4187\/respcare.01712","article-title":"Bedside chest radiography","volume":"57","author":"Eisenhuber","year":"2012","journal-title":"Respir Care."},{"key":"B4","doi-asserted-by":"publisher","first-page":"388","DOI":"10.1164\/rccm.200405-644st","article-title":"Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia","volume":"171","year":"2005","journal-title":"Am J Respir Crit Care Med"},{"key":"B5","doi-asserted-by":"publisher","first-page":"867","DOI":"10.1164\/ajrccm.165.7.2105078","article-title":"Ventilator-associated pneumonia","volume":"165","author":"Chastre","year":"2002","journal-title":"Am J Respir Crit Care Med."},{"key":"B6","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/nature21056","article-title":"Dermatologist-level classification of skin cancer with deep neural networks","volume":"542","author":"Esteva","year":"2017","journal-title":"Nature."},{"key":"B7","first-page":"1035","article-title":"Respiratory Care, Principles & Practice","author":"Simmons","year":"2002"},{"key":"B8","doi-asserted-by":"publisher","first-page":"637","DOI":"10.1128\/cmr.00051-05","article-title":"Ventilator-associated pneumonia: diagnosis, treatment, and prevention","volume":"19","author":"Koenig","year":"2006","journal-title":"Clin Microbiol Rev."},{"key":"B9","doi-asserted-by":"publisher","first-page":"208","DOI":"10.1186\/cc13775","article-title":"Ventilator-associated pneumonia in the ICU","volume":"18","author":"Kalanuria","year":"2014","journal-title":"Crit Care."},{"key":"B10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40537-014-0007-7","article-title":"Deep learning applications and challenges in big data analytics","volume":"2","author":"Najafabadi","year":"2015","journal-title":"J Big Data."},{"key":"B11","doi-asserted-by":"publisher","first-page":"133","DOI":"10.1007\/978-3-319-44550-2_5","article-title":"Deep learning techniques in big data analytics","author":"Najafabadi","year":"2016","journal-title":"Big Data Technol Appl"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1707","DOI":"10.1109\/iccv.2017.97","article-title":"Revisiting unreasonable effectiveness of data in deep learning era","author":"Sun","year":"2017","journal-title":"Arxivorg"},{"key":"B13","doi-asserted-by":"publisher","first-page":"1637","DOI":"10.1007\/s11548-016-1359-6","article-title":"Edge map analysis in chest X-rays for automatic pulmonary abnormality screening","volume":"11","author":"Santosh","year":"2016","journal-title":"Int J Comput Assist Radiol Surg."},{"key":"B14","doi-asserted-by":"publisher","first-page":"1168","DOI":"10.1109\/tmi.2017.2775636","article-title":"Automated chest X-ray screening: can lung region symmetry help detect pulmonary abnormalities?","volume":"37","author":"Santosh","year":"2018","journal-title":"IEEE Trans Medical Imaging."},{"key":"B15","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1186\/s12938-018-0544-y","article-title":"Computer-aided detection in chest radiography based on artificial intelligence: a survey","volume":"17","author":"Qin","year":"2018","journal-title":"Biomed Eng Online."},{"key":"B16","doi-asserted-by":"publisher","first-page":"6536","DOI":"10.3934\/mbe.2019326","article-title":"Convolutional neural networks for computer-aided detection or diagnosis in medical image analysis: an overview","volume":"16","author":"Gao","year":"2019","journal-title":"Math Biosci Eng."},{"key":"B17","doi-asserted-by":"publisher","first-page":"20190580","DOI":"10.1259\/bjr.20190580","article-title":"CAD. and AI for breast cancer-recent development and challenges","volume":"93","author":"Chan","year":"2020","journal-title":"Br J Radiol."},{"key":"B18","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1007\/s10916-018-1010-x","article-title":"Framework of computer aided diagnosis systems for cancer classification based on medical images","volume":"42","author":"El Houby","year":"2018","journal-title":"J Med Syst."},{"key":"B19","doi-asserted-by":"publisher","first-page":"145","DOI":"10.3348\/kjr.2011.12.2.145","article-title":"A computer-aided diagnosis for evaluating lung nodules on chest CT: the current status and perspective","volume":"12","author":"Goo","year":"2011","journal-title":"Korean J Radiol."},{"key":"B20","doi-asserted-by":"publisher","first-page":"2402","DOI":"10.1001\/jama.2016.17216","article-title":"Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs","volume":"316","author":"Gulshan","year":"2016","journal-title":"JAMA."},{"key":"B21","doi-asserted-by":"publisher","first-page":"158","DOI":"10.1080\/00365521.2019.1577486","article-title":"Application of convolutional neural networks for evaluating helicobacter pylori infection status on the basis of endoscopic images","volume":"54","author":"Shichijo","year":"2019","journal-title":"Scand J Gastroenterol."},{"key":"B22","doi-asserted-by":"publisher","first-page":"2097","DOI":"10.1109\/CVPR.2017.369","article-title":"Chestx-ray8: hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases","author":"Wang","year":"2017","journal-title":"IEEE CVPR"},{"key":"B23","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1007\/s10916-020-01562-1","article-title":"AI-driven tools for coronavirus outbreak: need of active learning and cross-population train\/test models on multitudinal\/multimodal data","volume":"44","author":"Santosh","year":"2020","journal-title":"J Medical Syst."},{"key":"B24","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1007\/s10916-021-01747-2","article-title":"COVID-19 imaging tools: how big data is big?","volume":"45","author":"Santosh","year":"2021","journal-title":"J Medical Syst."},{"key":"B25","doi-asserted-by":"publisher","first-page":"915","DOI":"10.1007\/s13246-020-00888-x","article-title":"Truncated inception net: COVID-19 outbreak screening using chest X-rays","volume":"43","author":"Das","year":"2020","journal-title":"Phys Eng Sci Med."},{"key":"B26","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/978-981-15-9682-7","article-title":"COVID-19: prediction, decision-making, and its impacts, book series in lecture notes on data engineering and communications technologies, springer","author":"Santosh","year":"2020","journal-title":"Nature"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2021.676277\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,10,16]],"date-time":"2021-10-16T00:08:16Z","timestamp":1634342896000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2021.676277\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,10,15]]},"references-count":26,"alternative-id":["10.3389\/fmed.2021.676277"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2021.676277","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,10,15]]},"article-number":"676277"}}