{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,10]],"date-time":"2026-10-10T13:24:31Z","timestamp":1791638671933,"version":"4.3.4"},"reference-count":25,"publisher":"Bioscientifica","issue":"1","license":[{"start":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T00:00:00Z","timestamp":1781481600000},"content-version":"vor","delay-in-days":151,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["journals.bioscientifica.com"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,2,1]]},"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Given the limitations of conventional approaches in managing indeterminate thyroid nodules, there remains an unmet need for non-invasive assistant tools to improve risk stratification. This study aimed to evaluate the clinical applicability of an artificial intelligence (AI) model for thyroid nodules with atypia of undetermined significance (AUS) cytology.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We retrospectively reviewed patients who underwent fine-needle aspiration (FNA) for thyroid nodules between January 2019 and December 2020 across five medical institutions in Korea. Nodules initially diagnosed as AUS and later confirmed as benign or malignant were included. A previously developed deep learning-based AI model, AI-Thyroid, was employed to provide binary classifications (benign or malignant) and malignancy risk estimates.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>A total of 165 thyroid nodules were analyzed. The median (interquartile range) longest diameter was 1.30 cm (0.80\u20132.10), and the malignancy rate of the cohort was 39%. In binary classification tasks, the model achieved a sensitivity of 0.91 and a negative predictive value of 0.87. The area under the curve (AUC) based on estimated malignancy risk was 0.75 (95% confidence interval (CI): 0.68\u20130.83), and the AUC derived from K-TIRADS categories 2\u20135 was 0.76 (95% CI: 0.69\u20130.83), indicating comparable diagnostic accuracy with the traditional scoring system. Subgroup analyses demonstrated that the model achieved a sensitivity of 98% in nodules smaller than 1.5 cm.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>AI-assisted ultrasound analysis offers supplementary diagnostic information for thyroid nodules with AUS cytology. Its high sensitivity and negative predictive value may assist clinicians in decision-making processes, particularly for small, low-risk thyroid nodules.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1530\/etj-25-0268","type":"journal-article","created":{"date-parts":[[2026,1,19]],"date-time":"2026-01-19T13:11:30Z","timestamp":1768828290000},"update-policy":"https:\/\/doi.org\/10.1530\/crossmarkpolicy-17","source":"Crossref","is-referenced-by-count":6,"title":["Artificial intelligence-assisted risk stratification of thyroid nodules with atypia of undetermined significance"],"prefix":"10.1530","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4581-9493","authenticated-orcid":true,"given":"Jinyoung","family":"Kim","sequence":"first","affiliation":[{"name":"1"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9074-8087","authenticated-orcid":true,"given":"Jeongmin","family":"Lee","sequence":"additional","affiliation":[{"name":"1"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9219-7135","authenticated-orcid":true,"given":"Jeonghoon","family":"Ha","sequence":"additional","affiliation":[{"name":"1"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8293-9577","authenticated-orcid":true,"given":"Ohjoon","family":"Kwon","sequence":"additional","affiliation":[{"name":"2"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4097-1077","authenticated-orcid":true,"given":"Ki-Hyun","family":"Baek","sequence":"additional","affiliation":[{"name":"1"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5267-0135","authenticated-orcid":true,"given":"Chang Myeon","family":"Song","sequence":"additional","affiliation":[{"name":"3"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4452-8053","authenticated-orcid":true,"given":"Hyeon A","family":"Lee","sequence":"additional","affiliation":[{"name":"3"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4257-3133","authenticated-orcid":true,"given":"Hyemi","family":"Kwon","sequence":"additional","affiliation":[{"name":"4"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3518-3332","authenticated-orcid":true,"given":"Inyoung","family":"Youn","sequence":"additional","affiliation":[{"name":"5"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8225-3690","authenticated-orcid":true,"given":"Mi-ri","family":"Kwon","sequence":"additional","affiliation":[{"name":"5"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0995-6482","authenticated-orcid":true,"given":"Dong-Jun","family":"Lim","sequence":"additional","affiliation":[{"name":"1"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"416","published-online":{"date-parts":[[2026,1,15]]},"reference":[{"key":"2026070714014426800_bib1","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1016\/S2213-8587(22)00101-2","article-title":"Diagnosis of thyroid nodules","volume":"10","author":"Alexander","year":"2022","journal-title":"Lancet Diabetes Endocrinol"},{"key":"2026070714014426800_bib2","doi-asserted-by":"crossref","first-page":"705","DOI":"10.1056\/NEJMoa1203208","article-title":"Preoperative diagnosis of benign thyroid nodules with indeterminate cytology","volume":"367","author":"Alexander","year":"2012","journal-title":"N Engl J Med"},{"key":"2026070714014426800_bib3","doi-asserted-by":"crossref","first-page":"1341","DOI":"10.1089\/thy.2017.0500","article-title":"The 2017 Bethesda system for reporting thyroid cytopathology","volume":"27","author":"Cibas","year":"2017","journal-title":"Thyroid"},{"key":"2026070714014426800_bib4","doi-asserted-by":"crossref","first-page":"1","DOI":"10.11106\/ijt.2023.16.1.1","article-title":"2023 Korean Thyroid Association management guidelines for patients with thyroid nodules","volume":"16","author":"Park","year":"2023","journal-title":"Int J Thyroidol"},{"key":"2026070714014426800_bib5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1089\/thy.2015.0020","article-title":"2015 American Thyroid Association Management Guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association Guidelines Task Force","volume":"26","author":"Haugen","year":"2016","journal-title":"Thyroid"},{"key":"2026070714014426800_bib6","doi-asserted-by":"crossref","first-page":"908","DOI":"10.4158\/EP-2018-0559","article-title":"Using the ATA and ACR TI-RADS sonographic classifications as adjunctive predictors of malignancy for indeterminate thyroid nodules","volume":"25","author":"Ahmadi","year":"2019","journal-title":"Endocr Pract"},{"key":"2026070714014426800_bib7","doi-asserted-by":"crossref","first-page":"570","DOI":"10.3348\/kjr.2017.18.4.570","article-title":"Deep learning in medical imaging: general overview","volume":"18","author":"Lee","year":"2017","journal-title":"Korean J Radiol"},{"key":"2026070714014426800_bib8","doi-asserted-by":"crossref","first-page":"1024","DOI":"10.1089\/thy.2017.0525","article-title":"Predicting malignancy in thyroid nodules: radiomics score versus 2017 ACR TI-RADS","volume":"28","author":"Liang","year":"2018","journal-title":"Thyroid"},{"key":"2026070714014426800_bib9","doi-asserted-by":"crossref","first-page":"695","DOI":"10.1148\/radiol.2019181343","article-title":"Management of thyroid nodules seen on US images: deep learning may match performance of radiologists","volume":"292","author":"Buda","year":"2019","journal-title":"Radiology"},{"key":"2026070714014426800_bib10","doi-asserted-by":"crossref","first-page":"527","DOI":"10.1210\/clinem\/dgad503","article-title":"Artificial intelligence model assisting thyroid nodule diagnosis and management: a multicenter diagnostic study","volume":"109","author":"Ha","year":"2024","journal-title":"J Clin Endocrinol Metab"},{"key":"2026070714014426800_bib11","first-page":"1974","article-title":"Ultrasonography diagnosis and imaging-based management of thyroid nodules: revised KSThR consensus statement","volume":"22","author":"Shin","year":"2021","journal-title":"Korean J Radiol"},{"key":"2026070714014426800_bib12","doi-asserted-by":"crossref","first-page":"1159","DOI":"10.1089\/thy.2009.0274","article-title":"The Bethesda system for reporting thyroid cytopathology","volume":"19","author":"Cibas","year":"2009","journal-title":"Thyroid"},{"key":"2026070714014426800_bib13","first-page":"1039","article-title":"The 2023 Bethesda system for reporting thyroid cytopathology","volume":"33","author":"Ali","year":"2023","journal-title":"Thyroid"},{"key":"2026070714014426800_bib14","doi-asserted-by":"crossref","first-page":"1111","DOI":"10.3803\/EnM.2021.1208","article-title":"Comparison of Korean vs American TI-RADS in malignancy risk assessment of indeterminate thyroid nodules","volume":"36","author":"Kang","year":"2021","journal-title":"Endocrinol Metab"},{"key":"2026070714014426800_bib15","doi-asserted-by":"crossref","first-page":"W854","DOI":"10.2214\/AJR.12.9901","article-title":"Outcome of thyroid nodules with AUS\/FLUS and correlation with ultrasound and BRAF(V600E)","volume":"201","author":"Jeong","year":"2013","journal-title":"AJR Am J Roentgenol"},{"key":"2026070714014426800_bib16","doi-asserted-by":"crossref","first-page":"2326","DOI":"10.1245\/s10434-014-3568-y","article-title":"AUS in thyroid FNA cytology: prediction of malignancy by US and management strategies","volume":"21","author":"Lee","year":"2014","journal-title":"Ann Surg Oncol"},{"key":"2026070714014426800_bib17","doi-asserted-by":"crossref","first-page":"e250","DOI":"10.1016\/S2589-7500(21)00041-8","article-title":"Deep learning\u2013based AI model to assist thyroid nodule diagnosis and management: a multicentre study","volume":"3","author":"Peng","year":"2021","journal-title":"Lancet Digit Health"},{"key":"2026070714014426800_bib18","doi-asserted-by":"crossref","first-page":"646","DOI":"10.1111\/cen.14612","article-title":"Risk stratification of indeterminate thyroid nodules using ultrasound and ML algorithms","volume":"96","author":"Gild","year":"2022","journal-title":"Clin Endocrinol"},{"key":"2026070714014426800_bib19","doi-asserted-by":"crossref","first-page":"343","DOI":"10.1016\/j.jclinepi.2011.09.002","article-title":"Trade-offs between accuracy measures for EHR algorithms","volume":"65","author":"Chubak","year":"2012","journal-title":"J Clin Epidemiol"},{"key":"2026070714014426800_bib20","doi-asserted-by":"crossref","first-page":"150","DOI":"10.1089\/thy.2022.0560","article-title":"Artificial intelligence for evaluation of thyroid nodules: a primer","volume":"33","author":"Tessler","year":"2023","journal-title":"Thyroid"},{"key":"2026070714014426800_bib21","doi-asserted-by":"crossref","first-page":"216","DOI":"10.3803\/EnM.2024.2058","article-title":"Deep learning for classification of thyroid nodules using multi-view ultrasound: benefits and challenges","volume":"40","author":"Kim","year":"2025","journal-title":"Endocrinol Metab"},{"key":"2026070714014426800_bib22","doi-asserted-by":"crossref","first-page":"648","DOI":"10.3390\/bioengineering11070648","article-title":"A multi-view deep learning model for thyroid nodule detection and characterization in US imaging","volume":"11","author":"Vahdati","year":"2024","journal-title":"Bioengineering"},{"key":"2026070714014426800_bib23","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1016\/j.surg.2023.06.053","article-title":"Generating a multimodal AI model to differentiate benign and malignant follicular neoplasms: proof-of-concept","volume":"175","author":"Lin","year":"2024","journal-title":"Surgery"},{"key":"2026070714014426800_bib24","doi-asserted-by":"crossref","first-page":"590","DOI":"10.1089\/thy.2024.0584","article-title":"Combining ultrasound and molecular testing in a multimodal deep learning model for indeterminate thyroid nodules","volume":"35","author":"Athreya","year":"2025","journal-title":"Thyroid"},{"key":"2026070714014426800_bib25","doi-asserted-by":"crossref","first-page":"853","DOI":"10.3803\/EnM.2024.2251","article-title":"Trends in thyroid cancer mortality rates in Korea: insights from national health database","volume":"39","author":"Kim","year":"2024","journal-title":"Endocrinol Metab"}],"container-title":["European Thyroid Journal"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/etj.bioscientifica.com\/view\/journals\/etj\/15\/1\/ETJ-25-0268.xml","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/etj.bioscientifica.com\/downloadpdf\/journals\/etj\/15\/1\/ETJ-25-0268.xml","content-type":"text\/html","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/journals.bioscientifica.com\/etj\/article-pdf\/doi\/10.1530\/ETJ-25-0268\/508622\/etj-25-0268.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/journals.bioscientifica.com\/etj\/article-pdf\/doi\/10.1530\/ETJ-25-0268\/508622\/etj-25-0268.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T18:01:49Z","timestamp":1783447309000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.bioscientifica.com\/etj\/article\/15\/1\/ETJ250268\/75665\/Artificial-intelligence-assisted-risk"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,15]]},"references-count":25,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,2,1]]}},"URL":"https:\/\/doi.org\/10.1530\/etj-25-0268","relation":{},"ISSN":["2235-0802","2235-0640"],"issn-type":[{"value":"2235-0802","type":"electronic"},{"value":"2235-0640","type":"print"}],"subject":[],"published":{"date-parts":[[2026,1,15]]},"article-number":"ETJ250268"}}