{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T12:36:19Z","timestamp":1768307779995,"version":"3.49.0"},"reference-count":40,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T00:00:00Z","timestamp":1768262400000},"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. Artif. Intell."],"abstract":"\n Introduction<\/jats:title>\n Pulmonary hypertension (PH) has an incidence of approximately 6 cases per million adults, with a global prevalence ranging from 49 to 55 cases per million adults. Recent advancements in artificial intelligence (AI) have demonstrated promising improvements in the diagnostic accuracy of imaging for PH, achieving an area under the curve (AUC) of 0.94, compared to seasoned professionals.<\/jats:p>\n <\/jats:sec>\n \n Research objective<\/jats:title>\n To systematically synthesize available evidence on the comparative accuracy of AI versus manual interpretation in detecting PH across various chest imaging modalities, i.e., chest X-ray, echocardiography, CT scan and cardiac MRI.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n Following PRISMA guidelines, a comprehensive search was conducted across five databases\u2014PubMed, Embase, ScienceDirect, Scopus, and the Cochrane Library\u2014from inception through March 2025. Statistical analysis was performed using R (version 2024.12.1\u202f+\u202f563) with 2\u202f\u00d7\u202f2 contingency data. Sensitivity, specificity, and diagnostic odds ratio (DOR) were pooled using a bivariate random-effects model (reitsma() from the mada package), while the AUC were meta-analyzed using logit-transformed values via the metagen() function from the meta package.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n This meta-analysis of 12 studies, encompassing 7,459 patients, demonstrated a statistically significant improvement in diagnostic accuracy of PH with AI integration, evidenced by a logit mean difference in AUC of 0.43 (95% CI: 0.23\u20130.64;\n p<\/jats:italic>\n \u202f&lt;\u202f0.0001) and low heterogeneity (\n I<\/jats:italic>\n 2<\/jats:sup>\n \u202f=\u202f21.0%,\n \u03c4<\/jats:italic>\n 2<\/jats:sup>\n \u202f&lt;\u202f0.0001,\n p<\/jats:italic>\n \u202f=\u202f0.2090), which was consolidated by pooled AUC of 0.934 on bivariate model. Pooled sensitivity and specificity for AI models were 0.83 (95% CI: 0.73\u20130.90) and 0.91 (95% CI: 0.86\u20130.95), respectively, with substantial heterogeneity for sensitivity (\n I<\/jats:italic>\n 2<\/jats:sup>\n \u202f=\u202f83.8%,\n \u03c4<\/jats:italic>\n 2<\/jats:sup>\n \u202f=\u202f0.4934,\n p<\/jats:italic>\n \u202f&lt;\u202f0.0001) and moderate for specificity (\n I<\/jats:italic>\n 2<\/jats:sup>\n \u202f=\u202f41.5%,\n \u03c4<\/jats:italic>\n 2<\/jats:sup>\n \u202f=\u202f0.1015,\n p<\/jats:italic>\n \u202f=\u202f0.1146); the diagnostic odds ratio was 54.26 (95% CI: 22.50\u2013130.87) with substantial heterogeneity (\n I<\/jats:italic>\n 2<\/jats:sup>\n \u202f=\u202f70.7%,\n \u03c4<\/jats:italic>\n 2<\/jats:sup>\n \u202f=\u202f0.8451,\n p<\/jats:italic>\n \u202f=\u202f0.0023). Sensitivity analysis showed stable estimates and did not reduce heterogeneity across outcomes.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusion<\/jats:title>\n AI-integrated imaging significantly enhances diagnostic accuracy for pulmonary hypertension, with higher sensitivity (0.83) and specificity (0.91) compared to manual interpretation across chest imaging modalities. However, further high-quality trials with externally validated cohorts may be needed to confirm these findings and reduce variability among AI models across diverse clinical settings.<\/jats:p>\n <\/jats:sec>","DOI":"10.3389\/frai.2025.1709489","type":"journal-article","created":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T06:32:23Z","timestamp":1768285943000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparative accuracy of artificial intelligence versus manual interpretation in detecting pulmonary hypertension across chest imaging modalities: a diagnostic test accuracy meta-analysis"],"prefix":"10.3389","volume":"8","author":[{"given":"Faizan","family":"Ahmed","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Faseeh","family":"Haider","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ramsha","family":"Ali","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Muhammad","family":"Arham","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yusra","family":"Junaid","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Allah","family":"Dad","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kinza","family":"Bakht","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Maryam","family":"Abbasi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bareera Tanveer","family":"Malik","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Abdul","family":"Mateen","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Najam","family":"Gohar","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rubiya","family":"Ali","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yasar","family":"Sattar","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mushood","family":"Ahmed","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mohamed","family":"Bakr","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Swapnil","family":"Patel","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jesus","family":"Almendral","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fawaz","family":"Alenezi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,1,13]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1038\/s41746-021-00438-z","article-title":"Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis","volume":"4","author":"Aggarwal","year":"2021","journal-title":"npj Digital Med"},{"key":"ref2","volume-title":"ESC\/ERS guidelines for the diagnosis and treatment of pulmonary hypertension: Ten points to remember","year":"2022"},{"key":"ref3","volume-title":"ACR appropriateness criteria: Suspected pulmonary hypertension","year":"2024"},{"key":"ref4","volume-title":"Pulmonary hypertension: Diagnosis and treatment. ATS patient information series","year":"2023"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"25151","DOI":"10.1038\/s41598-024-76608-2","article-title":"Deep learning improves physician accuracy in the comprehensive detection of abnormalities on chest X-rays","volume":"14","author":"Anderson","year":"2024","journal-title":"Sci. Rep."},{"key":"ref6","doi-asserted-by":"publisher","first-page":"8263","DOI":"10.1007\/s00330-023-09747-1","article-title":"Can incorrect artificial intelligence (AI) results impact radiologists, and if so, what can we do about it? A multi-reader pilot study of lung cancer detection with chest radiography","volume":"33","author":"Bernstein","year":"2023","journal-title":"Eur. Radiol."},{"key":"ref7","first-page":"3","article-title":"Discrepancy and error in radiology: concepts, causes and consequences","volume":"81","author":"Brady","year":"2012","journal-title":"Ulster Med. J."},{"key":"ref8","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1007\/s13755-021-00138-8","article-title":"Studying human-AI collaboration protocols: the case of the Kasparov's law in radiological double reading","volume":"9","author":"Cabitza","year":"2021","journal-title":"Health Inf. Sci. Syst."},{"key":"ref9","doi-asserted-by":"publisher","first-page":"e024969","DOI":"10.1161\/JAHA.121.024969","article-title":"Mortality in pulmonary arterial hypertension in the modern era: early insights from the Pulmonary Hypertension Association registry","volume":"11","author":"Chang","year":"2022","journal-title":"J. Am. Heart Assoc."},{"key":"ref10","doi-asserted-by":"publisher","first-page":"e500","DOI":"10.1016\/j.crad.2022.03.009","article-title":"Diagnostic accuracy of an automated artificial intelligence derived right ventricular to left ventricular diameter ratio tool on CT pulmonary angiography to predict pulmonary hypertension at right heart catheterisation","volume":"77","author":"Charters","year":"2022","journal-title":"Clin. Radiol."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"111277","DOI":"10.1016\/j.ejrad.2023.111277","article-title":"Artificial intelligence-assisted diagnosis of congenital heart disease and associated pulmonary arterial hypertension from chest radiographs: a multi-reader multi-case study","volume":"171","author":"Han","year":"2024","journal-title":"Eur. J. Radiol."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"1848","DOI":"10.2967\/jnumed.123.266080","article-title":"Ethical considerations for artificial intelligence in medical imaging: data collection, development, and evaluation","volume":"64","author":"Herington","year":"2023","journal-title":"J. Nucl. Med."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"101","DOI":"10.1186\/s12890-024-02891-4","article-title":"Artificial intelligence-based model for predicting pulmonary arterial hypertension on chest x-ray images","volume":"24","author":"Imai","year":"2024","journal-title":"BMC Pulm. Med."},{"key":"ref14","doi-asserted-by":"publisher","first-page":"100438","DOI":"10.1016\/j.ejro.2022.100438","article-title":"Artificial intelligence model on chest imaging to diagnose COVID-19 and other pneumonias: a systematic review and meta-analysis","volume":"9","author":"Jia","year":"2022","journal-title":"Eur J Radiol Open."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"103962","DOI":"10.1016\/j.compbiomed.2020.103962","article-title":"A lung graph model for the radiological assessment of chronic thromboembolic pulmonary hypertension in CT","volume":"125","author":"Jimenez-Del-Toro","year":"2020","journal-title":"Comput. Biol. Med."},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.3390\/healthcare10122493","article-title":"Economics of artificial intelligence in healthcare: diagnosis vs. treatment","volume":"10","author":"Khanna","year":"2022","journal-title":"Healthcare"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1016\/j.jjcc.2023.08.004","article-title":"Association of delayed diagnosis of pulmonary arterial hypertension with its prognosis","volume":"83","author":"Kubota","year":"2024","journal-title":"J. Cardiol."},{"key":"ref18","doi-asserted-by":"publisher","first-page":"19311","DOI":"10.1038\/s41598-020-76359-w","article-title":"Deep learning to predict elevated pulmonary artery pressure in patients with suspected pulmonary hypertension using standard chest X ray","volume":"10","author":"Kusunose","year":"2020","journal-title":"Sci. Rep."},{"key":"ref19","doi-asserted-by":"publisher","first-page":"891703","DOI":"10.3389\/fcvm.2022.891703","article-title":"Deep learning for detection of exercise-induced pulmonary hypertension using chest X-ray images","volume":"9","author":"Kusunose","year":"2022","journal-title":"Front. Cardiovasc. Med."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"413","DOI":"10.1016\/j.jacr.2020.09.060","article-title":"Regulatory frameworks for development and evaluation of artificial intelligence-based diagnostic imaging algorithms: summary and recommendations","volume":"18","author":"Larson","year":"2021","journal-title":"J. Am. Coll. Radiol."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"e0224453","DOI":"10.1371\/journal.pone.0224453","article-title":"A machine learning approach for the prediction of pulmonary hypertension","volume":"14","author":"Leha","year":"2019","journal-title":"PLoS One"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"1330685","DOI":"10.3389\/fcvm.2024.1330685","article-title":"Explanatory deep learning to predict elevated pulmonary artery pressure in children with ventricular septal defects using standard chest x-rays: a novel approach","volume":"11","author":"Li","year":"2024","journal-title":"Front. Cardiovasc. Med."},{"key":"ref23","doi-asserted-by":"publisher","first-page":"e12272","DOI":"10.1002\/pul2.12272","article-title":"Automatic echocardiographic evaluation of the probability of pulmonary hypertension using machine learning","volume":"13","author":"Liao","year":"2023","journal-title":"Pulm Circ."},{"key":"ref24","volume-title":"Pulmonary hypertension","author":"Manek","year":"2025"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"e029024","DOI":"10.1161\/JAHA.122.029024","article-title":"Revised definition of pulmonary hypertension and approach to management: a clinical primer","volume":"12","author":"Maron","year":"2023","journal-title":"J. Am. Heart Assoc."},{"key":"ref26","doi-asserted-by":"publisher","first-page":"388","DOI":"10.1001\/jama.2017.19163","article-title":"Preferred reporting items for a systematic review and meta-analysis of diagnostic test accuracy studies: the PRISMA-DTA statement","volume":"319","author":"McInnes","year":"2018","journal-title":"JAMA"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"100591","DOI":"10.1016\/j.imu.2021.100591","article-title":"The diagnostic accuracy of artificial intelligence-assisted CT imaging in COVID-19 disease: a systematic review and meta-analysis","volume":"24","author":"Moezzi","year":"2021","journal-title":"Inform Med Unlocked"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"324","DOI":"10.1016\/j.rxeng.2022.04.001","article-title":"Barriers to artificial intelligence implementation in radiology practice: what the radiologist needs to know","volume":"64","author":"Nair","year":"2022","journal-title":"Radiologia"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13172760","article-title":"Redefining radiology: a review of artificial intelligence integration in medical imaging","volume":"13","author":"Najjar","year":"2023","journal-title":"Diagnostics"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"e086061","DOI":"10.1136\/bmjopen-2024-086061","article-title":"Evaluating the impact of artificial intelligence-assisted image analysis on the diagnostic accuracy of front-line clinicians in detecting fractures on plain X-rays (FRACT-AI): protocol for a prospective observational study","volume":"14","author":"Novak","year":"2024","journal-title":"BMJ Open"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1016\/j.jjcc.2023.08.004","article-title":"Prognostic value of right ventricular\u2013pulmonary artery coupling in patients with pulmonary hypertension","volume":"83","author":"Ono","year":"2024","journal-title":"J. Cardiol."},{"key":"ref32","volume-title":"ESC\/ERS guidelines for the diagnosis and treatment of pulmonary hypertension","year":"2022"},{"key":"ref33","first-page":"223","article-title":"CT features of pulmonary hypertension: a guide for the radiologist","volume":"70","author":"Rajaram","year":"2015","journal-title":"Clin. Radiol."},{"key":"ref34","doi-asserted-by":"publisher","first-page":"982","DOI":"10.1016\/j.jclinepi.2005.02.022","article-title":"Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews","volume":"58","author":"Reitsma","year":"2005","journal-title":"J. Clin. Epidemiol."},{"key":"ref35","doi-asserted-by":"publisher","first-page":"859","DOI":"10.21037\/cdt-20-295","article-title":"The role of imaging in pulmonary hypertension","volume":"11","author":"Sharma","year":"2021","journal-title":"Cardiovasc Diagn Therapy"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"e2019007","DOI":"10.4178\/epih.e2019007","article-title":"Diagnostic test accuracy: application and practice using R software","volume":"41","author":"Shim","year":"2019","journal-title":"Epidemiol Health."},{"key":"ref37","doi-asserted-by":"publisher","first-page":"1066","DOI":"10.1536\/ihj.24-111","article-title":"Deep learning to detect pulmonary hypertension from the chest X-ray images of patients with systemic sclerosis","volume":"65","author":"Shimbo","year":"2024","journal-title":"Int. Heart J."},{"key":"ref38","doi-asserted-by":"publisher","first-page":"236","DOI":"10.1093\/ehjci\/jeaa001","article-title":"A machine learning cardiac magnetic resonance approach to extract disease features and automate pulmonary arterial hypertension diagnosis","volume":"22","author":"Swift","year":"2021","journal-title":"Eur. Heart J. Cardiovasc. Imaging"},{"key":"ref39","doi-asserted-by":"publisher","first-page":"1623","DOI":"10.1161\/CIRCULATIONAHA.118.034338","article-title":"Fully automated echocardiogram interpretation in clinical practice","volume":"138","author":"Zhang","year":"2018","journal-title":"Circulation"},{"key":"ref40","doi-asserted-by":"publisher","first-page":"e0236378","DOI":"10.1371\/journal.pone.0236378","article-title":"A promising approach for screening pulmonary hypertension based on frontal chest radiographs using deep learning: a retrospective study","volume":"15","author":"Zou","year":"2020","journal-title":"PLoS One"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1709489\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T06:32:29Z","timestamp":1768285949000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1709489\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,13]]},"references-count":40,"alternative-id":["10.3389\/frai.2025.1709489"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1709489","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,13]]},"article-number":"1709489"}}