{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T09:40:29Z","timestamp":1777887629096,"version":"3.51.4"},"reference-count":24,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"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. Endocrinol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>We evaluated whether a generative adversarial network (GAN)-based super-resolution (SR) radiomics pipeline could improve the pre-fine-needle aspiration (FNA) prediction of nondiagnostic (Bethesda I) outcomes in thyroid nodules compared with normal-resolution (NR) ultrasound analysis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This single-center retrospective study included 437 patients with thyroid nodules, with 338 assigned to the development cohort and 99 to a temporally independent validation cohort. Two-dimensional B-mode ultrasound images were processed using a pretrained GAN-based SR model in inference-only mode. Handcrafted radiomic features were extracted using PyRadiomics and subjected to a multistep selection pipeline comprising the Mann-Whitney U test, Spearman correlation-based redundancy pruning, random forest-based recursive feature elimination (RFE), and least absolute shrinkage and selection operator (LASSO) regression. Four machine-learning classifiers, including random forest (RF), LightGBM, ExtraTrees, and XGBoost, were trained using five repeated stratified 80\/20 splits. Post hoc probability calibration was performed using Platt (sigmoid) scaling fitted on the training set and applied unchanged to the independent validation set. In addition, qualitative error analysis was conducted to characterize false-positive and false-negative cases and to assess the potential effect of operator experience on model performance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The SR-based models consistently outperformed the NR-based models. The SR-RF model achieved an area under the receiver operating characteristic curve (AUC) of 0.808 in training and 0.733 in internal testing, compared with 0.672 and 0.596, respectively, for the NR-RF model. In the independent validation cohort, the SR-RF model yielded an AUC of 0.7435. Platt calibration improved the Brier score from 0.1885 to 0.1702 without affecting AUC, indicating improved probability reliability. Across classifiers, RF and LightGBM showed the most balanced overall performance. Qualitative error analysis showed that false-positive cases were often associated with cystic or heterogeneous echotexture and posterior acoustic artifacts, whereas false-negative cases involved isoechoic solid nodules with subtle margins and limited region-of-interest (ROI) coverage. No systematic effect of operator experience on model performance was observed.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>GAN-based SR radiomics significantly improves the pre-FNA prediction of nondiagnostic cytology in thyroid nodules. When combined with post hoc calibration, this approach provides more reliable individualized risk estimates and may help reduce unnecessary repeat procedures.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fendo.2026.1710097","type":"journal-article","created":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:19:22Z","timestamp":1777609162000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Super-resolution ultrasound radiomics for pre-FNA prediction of nondiagnostic (Bethesda I) thyroid nodules"],"prefix":"10.3389","volume":"17","author":[{"given":"Shaozheng","family":"He","sequence":"first","affiliation":[{"name":"Department of Ultrasound, Second Affiliated Hospital of Fujian Medical University","place":["Quanzhou, Fujian, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guo-Rong","family":"Lyu","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Second Affiliated Hospital of Fujian Medical University","place":["Quanzhou, Fujian, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mingli","family":"Cai","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Jinjiang Municipal Hospital","place":["Quanzhou, Fujian, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jian","family":"Lin","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Jinjiang Municipal Hospital","place":["Quanzhou, Fujian, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kunzhang","family":"Zeng","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Jinjiang Municipal Hospital","place":["Quanzhou, Fujian, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Junfa","family":"Sheng","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Jinjiang Municipal Hospital","place":["Quanzhou, Fujian, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,5,1]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ecl.2017.04.011","article-title":"Thyroid cancer","volume":"401","author":"Chen","year":"2023","journal-title":"Lancet"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.3238\/arztebl.2016.0353","article-title":"Fine needle aspiration in the investigation of thyroid nodules","volume":"113","author":"Feldkamp","year":"2016","journal-title":"Dtsch Arztebl Int"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1016\/j.jasc.2017.09.002","article-title":"The 2017 Bethesda system for reporting thyroid cytopathology","volume":"27","author":"Cibas","year":"2017","journal-title":"Thyroid"},{"key":"B4","doi-asserted-by":"publisher","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 on Thyroid Nodules and Differentiated Thyroid Cancer","volume":"26","author":"Haugen","year":"2016","journal-title":"Thyroid"},{"key":"B5","doi-asserted-by":"publisher","first-page":"196","DOI":"10.1111\/cyt.12508","article-title":"The impact of repeat FNA in non-diagnostic and indeterminate thyroid nodules: A 5-year single-centre experience","volume":"29","author":"Saieg","year":"2018","journal-title":"Cytopathology"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2022.1022438","article-title":"Clinical value of FNA puncture feeling in the diagnosis of non-diagnostic and indeterminate thyroid nodules","volume":"13","author":"Wu","year":"2022","journal-title":"Front Endocrinol (Lausanne)"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.118.222893","article-title":"Introduction to radiomics","volume":"61","author":"Mayerhoefer","year":"2020","journal-title":"J Nucl Med"},{"key":"B8","doi-asserted-by":"publisher","first-page":"340","DOI":"10.1186\/s12885-023-10743-3","article-title":"Automated breast ultrasound (ABUS)-based radiomics nomogram: An individualized tool for predicting axillary lymph node tumor burden in patients with early breast cancer","volume":"23","author":"Chen","year":"2023","journal-title":"BMC Cancer"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.2147\/bctt.s398300","article-title":"Optimized radiomics nomogram based on automated breast ultrasound system: A potential tool for preoperative prediction of metastatic lymph node burden in breast cancer","volume":"15","author":"Li","year":"2023","journal-title":"Breast Cancer (Dove Med Press)"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1002\/nbm.4345","article-title":"Generative adversarial network-based super-resolution of diffusion-weighted imaging: Application to tumour radiomics in breast cancer","volume":"33","author":"Fan","year":"2020","journal-title":"NMR BioMed"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s00330-022-08952-8","article-title":"Deep-learning-based 3D super-resolution MRI radiomics model: Superior predictive performance in preoperative T-staging of rectal cancer","volume":"33","author":"Hou","year":"2023","journal-title":"Eur Radiol"},{"key":"B12","doi-asserted-by":"publisher","first-page":"6362","DOI":"10.1038\/s41598-022-10298-6","article-title":"Deep learning-based single image super-resolution for low-field MR brain images","volume":"12","author":"de Leeuw","year":"2022","journal-title":"Sci Rep"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2021.118869","article-title":"Development and evaluation of a high resolution 0.5mm isotropic T1-weighted template of the older adult brain","volume":"248","author":"Niaz","year":"2022","journal-title":"Neuroimage"},{"key":"B14","doi-asserted-by":"publisher","first-page":"21361","DOI":"10.1038\/s41598-021-00898-z","article-title":"Impact of GAN-based lesion-focused medical image super-resolution on the robustness of radiomic features","volume":"11","author":"de Farias","year":"2021","journal-title":"Sci Rep"},{"key":"B15","first-page":"63","volume-title":"ESRGAN: Enhanced super-resolution generative adversarial networks","author":"Wang","year":"2019"},{"key":"B16","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2017.632","article-title":"Image-to-Image Translation with Conditional Adversarial Networks","volume-title":"2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Isola","year":"2017"},{"key":"B17","article-title":"Photo-realistic single image super-resolution using a generative adversarial network: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Ledig","year":"2017"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-024-11093-9","article-title":"ESR essentials: Radiomics-practice recommendations by the European Society of Medical Imaging Informatics","volume":"35","author":"Santinha","year":"2025","journal-title":"Eur Radiol"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2023-074820","article-title":"Evaluation of clinical prediction models (part 2): How to undertake an external validation study","volume":"384","author":"Riley","year":"2024","journal-title":"BMJ"},{"key":"B20","doi-asserted-by":"publisher","first-page":"102","DOI":"10.1186\/s12958-025-01437-5","article-title":"The application of super-resolution ultrasound radiomics models in predicting the failure of conservative treatment for ectopic pregnancy","volume":"23","author":"Zhang","year":"2025","journal-title":"Reprod Biol Endocrinol"},{"key":"B21","doi-asserted-by":"publisher","first-page":"36200","DOI":"10.1038\/s41598-025-19908-5","article-title":"Prediction model for extrathyroidal extension in thyroid papillary carcinoma based on ultrasound radiomics","volume":"15","author":"Yuan","year":"2025","journal-title":"Sci Rep"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.21037\/gs-24-416","article-title":"Intranodular and perinodular ultrasound radiomics distinguishes benign and Malignant thyroid nodules: A multicenter study","volume":"13","author":"Zhu","year":"2024","journal-title":"Gland Surg"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.3390\/s24196203","article-title":"Multimodal ultrasound radiomic technology for diagnosing benign and Malignant thyroid nodules of Ti-Rads 4-5: A multicenter study","volume":"24","author":"Wang","year":"2024","journal-title":"Sensors (Basel)"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1016\/s2589-7500(24)00085-2","article-title":"Deep learning models for thyroid nodules diagnosis of fine-needle aspiration biopsy: A retrospective, prospective, multicentre study in China","volume":"6","author":"Wang","year":"2024","journal-title":"Lancet Digit Health"}],"container-title":["Frontiers in Endocrinology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2026.1710097\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:19:24Z","timestamp":1777609164000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2026.1710097\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,1]]},"references-count":24,"alternative-id":["10.3389\/fendo.2026.1710097"],"URL":"https:\/\/doi.org\/10.3389\/fendo.2026.1710097","relation":{},"ISSN":["1664-2392"],"issn-type":[{"value":"1664-2392","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,1]]},"article-number":"1710097"}}