{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:43:51Z","timestamp":1785545031558,"version":"3.56.0"},"reference-count":30,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Form Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n SARS-CoV-2 remains a common cause of acute respiratory illness; however, symptom-based triage poorly discriminates it from other febrile conditions. A recently developed artificial intelligence (AI)\u2013powered pharyngeal camera acquires pharyngeal images and clinical data to assist in influenza diagnosis; leveraging this workflow, we evaluated an adjunct AI algorithm (COVID-19-AI) that reports high, medium, or low suspicion to guide whether SARS-CoV-2 testing should subsequently be performed.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to report diagnostic accuracy outcomes and clinical utility of the COVID-19-AI as a triage support tool.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We conducted a performance evaluation using a prospectively collected multicenter dataset from 26 Japanese institutions between December 2023 and March 2024. Patients with suspected influenza or COVID-19 were eligible. The COVID-19-AI algorithm, a stacked ensemble of a Swin Transformer and boosting models, was developed using pharyngeal images combined with routine clinical variables from 2133 patients, and it produced a 3-tier output. Classification thresholds were predefined to optimize clinical rule-out and rule-in utilities. Diagnostic performance was assessed in 696 independent patients against centralized reverse transcription polymerase chain reaction\u2013confirmed SARS-CoV-2 infection under 2 prespecified operating criteria: inclusive (high or medium=positive and low=negative) and strict (high=positive and medium or low=negative). A subanalysis stratified accuracy by time from symptom onset (12-hour bins to 72 hours).<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n Among 696 analyzed participants (all Asian), 247 (35.5%) had reverse transcription polymerase chain reaction\u2013confirmed SARS-CoV-2 infection. The COVID-19-AI categorized 12.4% (n=86), 72.8% (n=507), and 14.8% (n=103) patients as high, medium, and low suspicion, respectively. Under the inclusive criteria, sensitivity of COVID-19-AI was 93.9% (95% CI 90.4%\u201096.4%), specificity was 19.6% (95% CI 16.1%\u201023.5%), and negative predictive value was 85.4% (95% CI 77.6%\u201091.3%). Under the strict criteria, sensitivity was 24.7% (95% CI 19.6%\u201030.4%), specificity was 94.4% (95% CI 92.0%\u201096.3%), and positive predictive value was 70.9% (95% CI 60.7%\u201079.8%). Across 12-hour onset strata, sensitivity under the inclusive criteria remained \u226592.0% and specificity under the strict criteria remained \u226583.3%; no pronounced temporal trend was observed. Additionally, an integrated model using both pharyngeal images and clinical variables (area under the receiver operating characteristic curve [AUROC] 0.78) outperformed models using only clinical variables (AUROC 0.75) or images alone (AUROC 0.71); feature importance analysis further confirmed that pharyngeal image information was the most influential individual predictor, providing greater predictive value than any single clinical variable.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Embedded within AI-powered pharyngeal camera workflows, a 3-tier AI suspicion output enables complementary operating behaviors\u2014high sensitivity to rule out COVID-19 (inclusive criteria) and high specificity to support immediate infection control measures (strict criteria), although these criteria involve inherent trade-offs with low specificity and low sensitivity, respectively. Performance stability across onset times suggests robustness to symptom chronology, offering a standardized tool for clinical triage.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/87705","type":"journal-article","created":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T13:30:08Z","timestamp":1780147808000},"page":"e87705-e87705","source":"Crossref","is-referenced-by-count":0,"title":["A 3-Tier AI Model for COVID-19 Triage Using Pharyngeal Images: Algorithm Development and Validation"],"prefix":"10.2196","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6087-7865","authenticated-orcid":false,"given":"Sho","family":"Okiyama","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0794-9940","authenticated-orcid":false,"given":"Tomonori","family":"Aoki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3860-9316","authenticated-orcid":false,"given":"Memori","family":"Fukuda","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-6890-4831","authenticated-orcid":false,"given":"Yuji","family":"Ariyasu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-7598-7386","authenticated-orcid":false,"given":"Saho","family":"Kameyama","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,7,20]]},"reference":[{"issue":"2","key":"R1","doi-asserted-by":"publisher","DOI":"10.1128\/cmr.00072-23","article-title":"Testing for SARS-CoV-2: lessons learned and current use cases","volume":"37","author":"Theel","journal-title":"Clin Microbiol Rev"},{"key":"R2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2024-081165","article-title":"Drug treatments for mild or moderate Covid-19: systematic review and network meta-analysis","volume":"389","author":"Ibrahim","journal-title":"BMJ"},{"key":"R3","doi-asserted-by":"publisher","DOI":"10.7189\/jogh.13.06026","article-title":"Diagnostic accuracy of clinical signs and symptoms of COVID-19: a systematic review and meta-analysis to investigate the different estimates in a different stage of the pandemic outbreak","volume":"13","author":"Chen","journal-title":"J Glob Health"},{"issue":"1","key":"R4","doi-asserted-by":"publisher","DOI":"10.1186\/s10194-021-01367-8","article-title":"Headache, anosmia, ageusia and other neurological symptoms in COVID-19: a cross-sectional study","volume":"23","author":"Sampaio Rocha-Filho","journal-title":"J Headache Pain"},{"issue":"12","key":"R5","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0260879","article-title":"COVID-19 symptoms at time of testing and association with positivity among outpatients tested for SARS-CoV-2","volume":"16","author":"Wohl","journal-title":"PLoS One"},{"issue":"3","key":"R6","doi-asserted-by":"publisher","first-page":"428","DOI":"10.1016\/j.jhin.2020.05.028","article-title":"Symptom-based screening for COVID-19 in health care workers: the importance of fever","volume":"105","author":"Yombi","journal-title":"J Hosp Infect"},{"issue":"1","key":"R7","doi-asserted-by":"publisher","DOI":"10.1186\/s12879-021-05930-1","article-title":"Diagnostic accuracy of symptoms as a diagnostic tool for SARS-CoV 2 infection: a cross-sectional study in a cohort of 2,173 patients","volume":"21","author":"Romero-Gameros","journal-title":"BMC Infect Dis"},{"issue":"5-6","key":"R8","doi-asserted-by":"publisher","first-page":"1196","DOI":"10.1093\/milmed\/usad138","article-title":"Sensitivity of symptom-based screening for COVID-19 in active duty basic trainees","volume":"189","author":"Matthews","journal-title":"Mil Med"},{"issue":"7","key":"R9","doi-asserted-by":"publisher","first-page":"e350","DOI":"10.1093\/cid\/ciad032","article-title":"The Infectious Diseases Society of America guidelines on the diagnosis of COVID-19: antigen testing (January 2023)","volume":"78","author":"Hayden","journal-title":"Clin Infect Dis"},{"issue":"11","key":"R10","doi-asserted-by":"publisher","DOI":"10.3346\/jkms.2022.37.e88","article-title":"Complications of nasopharyngeal swabs and safe procedures for COVID-19 testing based on anatomical knowledge","volume":"37","author":"Kim","journal-title":"J Korean Med Sci"},{"issue":"7","key":"R11","doi-asserted-by":"publisher","first-page":"672","DOI":"10.1001\/jamaoto.2021.0715","article-title":"Complications of COVID-19 nasopharyngeal swab test","volume":"147","author":"Koskinen","journal-title":"JAMA Otolaryngol Head Neck Surg"},{"issue":"3","key":"R12","doi-asserted-by":"publisher","first-page":"215","DOI":"10.5152\/FNJN.2024.22289","article-title":"Factors affecting procedural pain and discomfort experienced by individuals during nasopharyngeal swabbing: a cross-sectional study","volume":"32","author":"Do\u011fan","journal-title":"Florence Nightingale J Nurs"},{"issue":"1","key":"R13","doi-asserted-by":"publisher","DOI":"10.1186\/s12879-020-05604-4","article-title":"Proposal of COVID-19 clinical risk score for the management of suspected COVID-19 cases: a case control study","volume":"20","author":"Nakakubo","journal-title":"BMC Infect Dis"},{"issue":"9","key":"R14","doi-asserted-by":"publisher","first-page":"1587","DOI":"10.1515\/cclm-2020-0593","article-title":"Rapid identification of SARS-CoV-2-infected patients at the emergency department using routine testing","volume":"58","author":"Kurstjens","journal-title":"Clin Chem Lab Med"},{"issue":"2","key":"R15","doi-asserted-by":"publisher","first-page":"221","DOI":"10.4269\/ajtmh.22-0747","article-title":"Cobblestone throat in a younger patient infected with the Omicron variant of the SARS-CoV-2 virus","volume":"109","author":"Inui","journal-title":"Am J Trop Med Hyg"},{"issue":"2","key":"R16","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1002\/jgf2.589","article-title":"Appearance of a sore throat caused by the SARS-CoV-2 Omicron variant","volume":"24","author":"Yamashita","journal-title":"J Gen Fam Med"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijid.2022.03.056","article-title":"Variations in the appearance of posterior pharyngeal wall follicles in individuals with viral upper respiratory infections according to the virus and the stage of infection: a case series","volume":"119","author":"Takahashi","journal-title":"Int J Infect Dis"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijid.2024.107244","article-title":"Diagnostic utility of pharyngeal follicular structures in COVID-19: a large-scale cross-sectional study","volume":"149","author":"Takahashi","journal-title":"Int J Infect Dis"},{"issue":"1","key":"R19","doi-asserted-by":"publisher","DOI":"10.5826\/dpc.1401a45","article-title":"Oral mucosal manifestations associated with PCR positivity in patients with COVID-19","volume":"14","author":"Metin","journal-title":"Dermatol Pract Concept"},{"issue":"12","key":"R20","doi-asserted-by":"publisher","DOI":"10.2196\/38751","article-title":"Examining the use of an artificial intelligence model to diagnose influenza: development and validation study","volume":"24","author":"Okiyama","journal-title":"J Med Internet Res"},{"issue":"6","key":"R21","doi-asserted-by":"publisher","DOI":"10.1093\/ofid\/ofae304","article-title":"Performance of and severe acute respiratory syndrome coronavirus 2 diagnostics based on symptom onset and close contact exposure: an analysis from the Test Us at Home prospective cohort study","volume":"11","author":"Herbert","journal-title":"Open Forum Infect Dis"},{"issue":"16","key":"R22","doi-asserted-by":"publisher","first-page":"365","DOI":"10.15585\/mmwr.mm7316a2","article-title":"SARS-CoV-2 viral shedding and rapid antigen test performance - Respiratory Virus Transmission Network, November 2022-May 2023","volume":"73","author":"Smith-Jeffcoat","journal-title":"MMWR Morb Mortal Wkly Rep"},{"issue":"8","key":"R23","doi-asserted-by":"publisher","DOI":"10.3390\/pathogens14080748","article-title":"AI methods tailored to influenza, RSV, HIV, and SARS-CoV-2: a focused review","volume":"14","author":"Livieratos","journal-title":"Pathogens"},{"issue":"1","key":"R24","doi-asserted-by":"publisher","first-page":"351","DOI":"10.1007\/s00405-024-09049-2","article-title":"Diagnostic accuracy of deep learning-based algorithms in laryngoscopy: a systematic review and meta-analysis","volume":"282","author":"Du","journal-title":"Eur Arch Otorhinolaryngol"},{"key":"R25","doi-asserted-by":"publisher","DOI":"10.1109\/RBME.2020.2987975","article-title":"Review of artificial intelligence techniques in imaging data acquisition, segmentation, and diagnosis for COVID-19","volume":"14","author":"Shi","journal-title":"IEEE Rev Biomed Eng"},{"issue":"2","key":"R26","doi-asserted-by":"publisher","first-page":"460","DOI":"10.1002\/lio2.754","article-title":"Automatic classification of informative laryngoscopic images using deep learning","volume":"7","author":"Yao","journal-title":"Laryngoscope Investig Otolaryngol"},{"issue":"6","key":"R27","doi-asserted-by":"publisher","first-page":"102699","DOI":"10.1016\/j.jiac.2025.102699","article-title":"Comparison of pain between nasopharyngeal swab collection for influenza PCR testing and pharyngeal image capture for AI-based influenza diagnosis","volume":"31","author":"Kido","journal-title":"J Infect Chemother"},{"issue":"2","key":"R28","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pdig.0001233","article-title":"Infectious disease diagnosis by artificial intelligence (AI): differences in patient backgrounds and symptoms between antigen test positives and novel AI-powered pharyngeal endoscopy test positives","volume":"5","author":"Mori","journal-title":"PLOS Digit Health"},{"issue":"1","key":"R29","doi-asserted-by":"publisher","DOI":"10.1186\/s12985-024-02388-w","article-title":"Analyzing factors affecting positivity in drive-through COVID-19 testing: a cross-sectional study","volume":"21","author":"Mori","journal-title":"Virol J"},{"key":"R30","doi-asserted-by":"publisher","DOI":"10.2196\/74102","article-title":"Tongue image-based diagnosis of acute respiratory tract infection using machine learning: algorithm development and validation","volume":"13","author":"Che","journal-title":"JMIR Med Inform"}],"container-title":["JMIR Formative Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e87705","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:25:09Z","timestamp":1785536709000},"score":1,"resource":{"primary":{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e87705"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,20]]},"references-count":30,"URL":"https:\/\/doi.org\/10.2196\/87705","relation":{},"ISSN":["2561-326X"],"issn-type":[{"value":"2561-326X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,20]]},"article-number":"v10i2e87705"}}