{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T13:59:51Z","timestamp":1790171991175,"version":"4.1.0"},"reference-count":22,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T00:00:00Z","timestamp":1782950400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T00:00:00Z","timestamp":1790121600000},"content-version":"vor","delay-in-days":83,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Emerg Med"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Pneumothorax is a finding that is frequently missed in the emergency department, particularly in cases with subtle radiographic features, and can rapidly progress to tension pneumothorax when overlooked on the initial chest radiograph. Although artificial intelligence (AI)\u2013assisted chest radiograph interpretation has been shown to improve diagnostic accuracy, its effect on the actual decision behaviour of emergency physicians \u2014 particularly on chest computed tomography (CT) ordering \u2014 remains insufficiently studied. This study evaluated the impact of AI-assisted chest radiograph interpretation on emergency physicians\u2019 pneumothorax diagnostic decisions and chest CT ordering behaviour, with a particular focus on the capacity of the AI alert to function as a second reader and capture diagnoses initially missed by the physician.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This was a nationwide, online, cross-sectional simulation study conducted between 1 and 15 April 2026. Emergency physicians practising in T\u00fcrkiye were asked to interpret 20 standardised posteroanterior chest radiographs presented in a randomised order (10 pneumothorax-positive [5 overt and 5 subtle] and 10 pneumothorax-negative cases, all confirmed by chest CT). For each case, participants first indicated whether they suspected pneumothorax on the unannotated radiograph, and then \u2014 after being shown the same image annotated by a CE-certified, deep learning\u2013based chest radiograph interpretation system \u2014 indicated whether they would order a chest CT. The primary endpoint was the paired pre-AI versus post-AI change in simulated CT-ordering intent, derived from a predefined survey coding scheme and analysed using the McNemar exact test. As the pre- and post-AI items do not capture CT-ordering intent identically, this endpoint reflects simulated imaging intent rather than observed CT-ordering behaviour.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      A total of 231 physicians (51.5% general practitioners, 25.5% emergency medicine residents, 22.9% emergency medicine specialists) completed the survey, yielding 4,620 physician\u2013case response pairs. Simulated CT-ordering intent rose from 8.1% before AI exposure to 26.7% after AI exposure (absolute difference\u2009+\u200918.6% points;\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001). The increase was concentrated in pneumothorax-positive cases (+\u200931.6 points;\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001) and remained limited in pneumothorax-negative cases (+\u20095.7 points;\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001). In subtle pneumothorax cases, the integrated diagnostic capture rate rose from 54.8% to 85.0% (+\u200930.2 points); among the 339 cases in which physicians had initially answered \u201cno pneumothorax,\u201d 211 (62.2%) were converted into a CT order after the AI alert. The effect was statistically significant across all professional titles and seniority strata (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001 for all). The mean perceived usability score of the system was 8.1 out of 10 (interquartile range 7\u201310).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>In this simulation, AI-assisted chest radiograph interpretation did not reduce simulated CT-ordering intent; rather, the increase in intent was concentrated in true positive cases, with only a limited effect on negative cases. These findings are hypothesis-generating and require prospective confirmation with real CT-ordering data. The principal clinical value of the system lies in capturing pneumothorax diagnoses initially missed by emergency physicians, supporting its positioning as a second reader rather than as a CT-reduction tool.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Trial registration<\/jats:title>\n                    <jats:p>Not applicable.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s12873-026-01673-x","type":"journal-article","created":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T12:45:50Z","timestamp":1782996350000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Artificial intelligence as a second reader for pneumothorax detection: redirecting rather than reducing chest CT use"],"prefix":"10.1186","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6946-0860","authenticated-orcid":false,"given":"Muhammed Fatih","family":"C\u0131r\u0131l","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-1476-2147","authenticated-orcid":false,"given":"Ferhat","family":"Y\u0131ld\u0131z","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-7197-3831","authenticated-orcid":false,"given":"Mustafa","family":"Akarca","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,7,2]]},"reference":[{"issue":"1","key":"1673_CR1","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1186\/1755-7682-4-30","volume":"4","author":"HR Omar","year":"2011","unstructured":"Omar HR, Mangar D, Khetarpal S, Shapiro DH, Kolla J, Rashad R, et al. Anteroposterior chest radiograph vs. chest CT scan in early detection of pneumothorax in trauma patients. Int Arch Med. 2011;4(1):30. https:\/\/doi.org\/10.1186\/1755-7682-4-30.","journal-title":"Int Arch Med"},{"issue":"6","key":"1673_CR2","doi-asserted-by":"publisher","first-page":"1068","DOI":"10.1097\/SLA.0000000000001073","volume":"261","author":"DJ Roberts","year":"2015","unstructured":"Roberts DJ, Leigh-Smith S, Faris PD, Blackmore C, Ball CG, Robertson HL, et al. Clinical presentation of patients with tension pneumothorax: a systematic review. Ann Surg. 2015;261(6):1068\u201378. https:\/\/doi.org\/10.1097\/SLA.0000000000001073.","journal-title":"Ann Surg"},{"issue":"10","key":"1673_CR3","doi-asserted-by":"publisher","first-page":"1868","DOI":"10.3390\/diagnostics11101868","volume":"11","author":"J Rudolph","year":"2021","unstructured":"Rudolph J, Fink N, Dinkel J, Koliogiannis V, Schwarze V, Goller S, et al. Interpretation of thoracic radiography shows large discrepancies depending on the qualification of the physician\u2014quantitative evaluation of interobserver agreement in a representative emergency department scenario. Diagnostics (Basel). 2021;11(10):1868. https:\/\/doi.org\/10.3390\/diagnostics11101868.","journal-title":"Diagnostics (Basel)"},{"issue":"10","key":"1673_CR4","doi-asserted-by":"publisher","first-page":"602","DOI":"10.1136\/emermed-2023-213620","volume":"41","author":"A Novak","year":"2024","unstructured":"Novak A, Ather S, Gill A, Aylward P, Maskell G, Cowell GW, et al. Evaluation of the impact of artificial intelligence-assisted image interpretation on the diagnostic performance of clinicians in identifying pneumothoraces on plain chest X-ray: a multi-case multi-reader study. Emerg Med J. 2024;41(10):602\u20139. https:\/\/doi.org\/10.1136\/emermed-2023-213620.","journal-title":"Emerg Med J"},{"issue":"1","key":"1673_CR5","doi-asserted-by":"publisher","first-page":"311","DOI":"10.1186\/s12911-021-01679-4","volume":"21","author":"JH Kim","year":"2021","unstructured":"Kim JH, Han SG, Cho A, Shin HJ, Baek SE. Effect of deep learning-based assistive technology use on chest radiograph interpretation by emergency department physicians: a prospective interventional simulation-based study. BMC Med Inf Decis Mak. 2021;21(1):311. https:\/\/doi.org\/10.1186\/s12911-021-01679-4.","journal-title":"BMC Med Inf Decis Mak"},{"issue":"1","key":"1673_CR6","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1016\/j.chest.2024.01.039","volume":"166","author":"J Rudolph","year":"2024","unstructured":"Rudolph J, Huemmer C, Preuhs A, Buizza G, Hoppe BF, Dinkel J, et al. Nonradiology health care professionals significantly benefit from AI assistance in emergency-related chest radiography interpretation. Chest. 2024;166(1):157\u201370. https:\/\/doi.org\/10.1016\/j.chest.2024.01.039.","journal-title":"Chest"},{"issue":"20","key":"1673_CR7","doi-asserted-by":"publisher","first-page":"2639","DOI":"10.3390\/diagnostics15202639","volume":"15","author":"CW Ho","year":"2025","unstructured":"Ho CW, Wu YL, Chen YC, Ju YJ, Wu MT. Impact of AI assistance in pneumothorax detection on chest radiographs among readers of varying experience. Diagnostics (Basel). 2025;15(20):2639. https:\/\/doi.org\/10.3390\/diagnostics15202639.","journal-title":"Diagnostics (Basel)"},{"key":"1673_CR8","doi-asserted-by":"publisher","first-page":"4099","DOI":"10.2147\/JMDH.S535405","volume":"18","author":"\u0130 Dal","year":"2025","unstructured":"Dal \u0130, Kaya HB. Multidisciplinary evaluation of an AI-based pneumothorax detection model: clinical comparison with physicians in edge and cloud environments. J Multidiscip Healthc. 2025;18:4099\u2013111. https:\/\/doi.org\/10.2147\/JMDH.S535405.","journal-title":"J Multidiscip Healthc"},{"issue":"3","key":"1673_CR9","doi-asserted-by":"publisher","first-page":"e231236","DOI":"10.1148\/radiol.231236","volume":"308","author":"L Lind Plesner","year":"2023","unstructured":"Lind Plesner L, M\u00fcller FC, Brejneb\u00f8l MW, Laustrup LC, Rasmussen F, Nielsen OW, et al. Commercially available chest radiograph AI tools for detecting airspace disease, pneumothorax, and pleural effusion. Radiology. 2023;308(3):e231236. https:\/\/doi.org\/10.1148\/radiol.231236.","journal-title":"Radiology"},{"key":"1673_CR10","doi-asserted-by":"publisher","first-page":"94","DOI":"10.1186\/s13049-026-01616-2","volume":"34","author":"E G\u00fcnsoy","year":"2026","unstructured":"G\u00fcnsoy E, Aykut A, Y\u0131ld\u0131r\u0131m C, \u00d6nc\u00fcl MV, T\u00fcrko\u011flu S. Chest X-ray evaluation using GPT for tube thoracostomy or conservative care in non-tension spontaneous pneumothorax. Scand J Trauma Resusc Emerg Med. 2026;34:94. https:\/\/doi.org\/10.1186\/s13049-026-01616-2.","journal-title":"Scand J Trauma Resusc Emerg Med"},{"issue":"10","key":"1673_CR11","doi-asserted-by":"publisher","first-page":"e2336100","DOI":"10.1001\/jamanetworkopen.2023.36100","volume":"6","author":"J Huang","year":"2023","unstructured":"Huang J, Neill L, Wittbrodt M, Melnick D, Klug M, Thompson M, et al. Generative artificial intelligence for chest radiograph interpretation in the emergency department. JAMA Netw Open. 2023;6(10):e2336100. https:\/\/doi.org\/10.1001\/jamanetworkopen.2023.36100.","journal-title":"JAMA Netw Open"},{"issue":"7","key":"1673_CR12","doi-asserted-by":"publisher","first-page":"839","DOI":"10.1111\/acem.13195","volume":"24","author":"K Bookman","year":"2017","unstructured":"Bookman K, West D, Ginde A, Wiler J, McIntyre R, Hammes A, et al. Embedded clinical decision support in electronic health record decreases use of high-cost imaging in the emergency department: EmbED study. Acad Emerg Med. 2017;24(7):839\u201345. https:\/\/doi.org\/10.1111\/acem.13195.","journal-title":"Acad Emerg Med"},{"issue":"5","key":"1673_CR13","doi-asserted-by":"publisher","first-page":"631","DOI":"10.1007\/s43678-021-00170-3","volume":"23","author":"JE Andruchow","year":"2021","unstructured":"Andruchow JE, Grigat D, McRae AD, Innes G, Vatanpour S, Wang D, et al. Decision support for computed tomography in the emergency department: a multicenter cluster-randomized controlled trial. CJEM. 2021;23(5):631\u201340. https:\/\/doi.org\/10.1007\/s43678-021-00170-3.","journal-title":"CJEM"},{"issue":"3","key":"1673_CR14","doi-asserted-by":"publisher","first-page":"e211276","DOI":"10.1001\/jamanetworkopen.2021.1276","volume":"4","author":"B Vasey","year":"2021","unstructured":"Vasey B, Ursprung S, Beddoe B, Taylor EH, Marlow N, Bilbro N, et al. Association of clinician diagnostic performance with machine learning-based decision support systems: a systematic review. JAMA Netw Open. 2021;4(3):e211276. https:\/\/doi.org\/10.1001\/jamanetworkopen.2021.1276.","journal-title":"JAMA Netw Open"},{"issue":"12","key":"1673_CR15","doi-asserted-by":"publisher","first-page":"774","DOI":"10.1136\/emermed-2024-214781","volume":"42","author":"D Lyell","year":"2025","unstructured":"Lyell D, Dinh M, Gillett M, Abraham N, Symes ER, Susanto AP, et al. Evaluating the impact of AI assistance on decision-making in emergency doctors interpreting chest X-rays: a multi-reader multi-case study. Emerg Med J. 2025;42(12):774\u201382. https:\/\/doi.org\/10.1136\/emermed-2024-214781.","journal-title":"Emerg Med J"},{"issue":"12","key":"1673_CR16","doi-asserted-by":"publisher","first-page":"2050","DOI":"10.1093\/jamia\/ocad180","volume":"30","author":"AP Susanto","year":"2023","unstructured":"Susanto AP, Lyell D, Widyantoro B, Berkovsky S, Magrabi F. Effects of machine learning-based clinical decision support systems on decision-making, care delivery, and patient outcomes: a scoping review. J Am Med Inf Assoc. 2023;30(12):2050\u201363. https:\/\/doi.org\/10.1093\/jamia\/ocad180.","journal-title":"J Am Med Inf Assoc"},{"issue":"5","key":"1673_CR17","doi-asserted-by":"publisher","first-page":"3469","DOI":"10.1007\/s00330-021-08397-5","volume":"32","author":"KN Jin","year":"2022","unstructured":"Jin KN, Kim EY, Kim YJ, Lee GP, Kim H, Oh S, et al. Diagnostic effect of artificial intelligence solution for referable thoracic abnormalities on chest radiography: a multicenter respiratory outpatient diagnostic cohort study. Eur Radiol. 2022;32(5):3469\u201379. https:\/\/doi.org\/10.1007\/s00330-021-08397-5.","journal-title":"Eur Radiol"},{"issue":"4","key":"1673_CR18","doi-asserted-by":"publisher","first-page":"e222176","DOI":"10.1148\/radiol.222176","volume":"307","author":"T Dratsch","year":"2023","unstructured":"Dratsch T, Chen X, Rezazade Mehrizi M, Kloeckner R, M\u00e4hringer-Kunz A, P\u00fcsken M, et al. Automation bias in mammography: the impact of artificial intelligence BI-RADS suggestions on reader performance. Radiology. 2023;307(4):e222176. https:\/\/doi.org\/10.1148\/radiol.222176.","journal-title":"Radiology"},{"issue":"3","key":"1673_CR19","doi-asserted-by":"publisher","first-page":"381","DOI":"10.1177\/0018720810376055","volume":"52","author":"R Parasuraman","year":"2010","unstructured":"Parasuraman R, Manzey DH. Complacency and bias in human use of automation: an attentional integration. Hum Factors. 2010;52(3):381\u2013410. https:\/\/doi.org\/10.1177\/0018720810376055.","journal-title":"Hum Factors"},{"issue":"1","key":"1673_CR20","doi-asserted-by":"publisher","first-page":"121","DOI":"10.1136\/amiajnl-2011-000089","volume":"19","author":"K Goddard","year":"2012","unstructured":"Goddard K, Roudsari A, Wyatt JC. Automation bias: a systematic review of frequency, effect mediators, and mitigators. J Am Med Inf Assoc. 2012;19(1):121\u20137. https:\/\/doi.org\/10.1136\/amiajnl-2011-000089.","journal-title":"J Am Med Inf Assoc"},{"issue":"1","key":"1673_CR21","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1038\/s41746-021-00385-9","volume":"4","author":"S Gaube","year":"2021","unstructured":"Gaube S, Suresh H, Raue M, Merritt A, Berkowitz SJ, Lermer E, et al. Do as AI say: susceptibility in deployment of clinical decision-aids. NPJ Digit Med. 2021;4(1):31. https:\/\/doi.org\/10.1038\/s41746-021-00385-9.","journal-title":"NPJ Digit Med"},{"issue":"22","key":"1673_CR22","doi-asserted-by":"publisher","first-page":"2592","DOI":"10.3390\/diagnostics14222592","volume":"14","author":"J L\u00f3pez Alcolea","year":"2024","unstructured":"L\u00f3pez Alcolea J, Fern\u00e1ndez Alfonso A, Cano Alonso R, \u00c1lvarez V\u00e1zquez A, D\u00edaz Moreno A, Garc\u00eda Castellanos D, et al. Diagnostic performance of artificial intelligence in chest radiographs referred from the emergency department. Diagnostics (Basel). 2024;14(22):2592. https:\/\/doi.org\/10.3390\/diagnostics14222592.","journal-title":"Diagnostics (Basel)"}],"container-title":["BMC Emergency Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12873-026-01673-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12873-026-01673-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12873-026-01673-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T13:03:49Z","timestamp":1790168629000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s12873-026-01673-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,2]]},"references-count":22,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["1673"],"URL":"https:\/\/doi.org\/10.1186\/s12873-026-01673-x","relation":{},"ISSN":["1471-227X"],"issn-type":[{"value":"1471-227X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,2]]},"assertion":[{"value":"10 May 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 July 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The study protocol was approved by the Scientific Research Ethics Committee of Mardin Training and Research Hospital (decision date: 27 March 2026; protocol number: 2026\/03\u2009\u2212\u200901). The study was conducted in accordance with the current version of the Declaration of Helsinki. Electronic consent was obtained from all physician participants by reading the informational text presented on the introductory page of the online survey and proceeding with the survey. Additional patient consent for the chest radiographs was not deemed necessary, as the images were retrieved anonymously from a retrospective archive and all patient-identifying information had been masked.","order":1,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable.","order":2,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"During the preparation of this manuscript, the authors used Claude (Anthropic, San Francisco, CA, USA), a large language model, to assist with manuscript layout, table visualisation, and reference formatting. All scientific content, data interpretation, statistical analyses, and conclusions are the authors\u2019 own work. The authors reviewed and verified all outputs and assume full responsibility for the accuracy and integrity of the final manuscript.","order":3,"name":"Ethics","label":"Use of AI tools in manuscript preparation","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"256"}}