{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T13:00:48Z","timestamp":1791464448263,"version":"4.3.4"},"reference-count":42,"publisher":"Ovid Technologies (Wolters Kluwer Health)","issue":"4","license":[{"start":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T00:00:00Z","timestamp":1772668800000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,4]]},"abstract":"<jats:sec>\n                    <jats:title>Objective:<\/jats:title>\n                    <jats:p>Accurately distinguishing between benign and malignant thyroid nodules (TNs) in the context of Hashimoto\u2019s thyroiditis (HT) is challenging. This study aimed to explore the diagnostic efficacy of artificial intelligence models constructed using radiomics and deep learning (DL) features extracted from the ultrasound images of TNs in the setting of HT. The study also aimed to quantitatively compare the diagnostic performance of these models against that of fine-needle aspiration (FNA) cytology combined with BRAFV600E gene mutation testing so as to establish a superior diagnostic paradigm for TNs in HT.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods:<\/jats:title>\n                    <jats:p>We analyzed the clinical data and preoperative ultrasound images of 1585 patients with HT admitted to eight hospitals in China between 1 January 2018, and 30 December 2023. Radiomics features were extracted from each manually annotated and standardized region of interest in the images. The DL features based on various convolutional neural network models were also extracted, including 11 pre-trained DL models based on real images and the image features extracted using a combined DL\u2013radiomics (DLR) approach. Least absolute shrinkage and selection operator regression was used to select features with nonzero coefficients. Further, eight machine learning methods were employed to construct prediction models, namely the DLR models, incorporating both DL and radiomics features. The importance of features in contributing to the model was prioritized using the SHapley Additive exPlanations (SHAP) method for interpretation. The predictive performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Two rounds of reader studies were conducted (first round: independent reading; second round: guided reading) to validate the clinical application value of the DLR model, with results compared with FNA biopsy outcomes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results:<\/jats:title>\n                    <jats:p>A total of 1561 radiomics features and 256 DL features were extracted from the original images. The DLR model, leveraging the ResNet152 neural network, could effectively differentiate between benign and malignant nodules in the context of HT. The AUC, accuracy, sensitivity, and specificity of the logistic regression (LR) model, based on these features, were 0.917 [95% confidence interval (CI): 0.838\u20130.988], 85.7%, 81.8%, and 86.6%, respectively, in the validation cohort and 0.827(95% CI: 0.777\u20130.876), 93.9%, 80.9%, and 96.3%, respectively, in the external test cohort. The SHAP summary plot illustrated how feature values influenced their impact on the model, whereas the SHAP force plot showed the integrated impact of features on individual responses. Gradient-weighted class activation mapping (Grad-CAM)-generated heatmaps from DL models visually highlighted high-risk areas in HT. The DLR model outperformed five junior-level physicians in terms of diagnostic accuracy, sensitivity, and specificity in the validation cohort. The diagnostic performance of all clinicians was significantly enhanced with the assistance of the DLR model, with no statistical difference detected compared with the FNA biopsy results.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions:<\/jats:title>\n                    <jats:p>The DLR model combined with the SHAP and Grad-CAM method can improve the diagnostic performance of radiologists in identifying benign and malignant TNs in the context of HT. The diagnostic efficacy of this visualization model is comparable to that of FNA cytology combined with gene mutation testing. The DLR model can enhance the diagnostic ability of radiologists in differentiating between benign and malignant TNs in the context of HT, thereby minimizing unnecessary biopsies. Additionally, it can aid clinicians in making personalized decisions regarding the necessity of biopsy or even surgery by providing intuitive visual explanations.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1097\/js9.0000000000004546","type":"journal-article","created":{"date-parts":[[2025,12,17]],"date-time":"2025-12-17T15:01:21Z","timestamp":1765983681000},"page":"10157-10171","update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":6,"title":["Revolutionizing thyroid nodule diagnosis in Hashimoto\u2019s thyroiditis: AI-driven radiomics and deep learning model"],"prefix":"10.1097","volume":"112","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3740-2762","authenticated-orcid":false,"given":"Fan","family":"Wu","sequence":"first","affiliation":[{"name":"Department of Oncological Surgery, Affiliated Hangzhou First People\u2019s Hospital, Westlake University School of Medicine, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ting","family":"Pan","sequence":"additional","affiliation":[{"name":"Cancer Center, Department of Pathology, Zhejiang Provincial People\u2019s Hospital (Affiliated People\u2019s Hospital), Hangzhou Medical College, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuanwei","family":"Huang","sequence":"additional","affiliation":[{"name":"The Fourth Clinical Medical College, Zhejiang Chinese Medical University, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kaiyuan","family":"Huang","sequence":"additional","affiliation":[{"name":"The Fourth Clinical Medical College, Zhejiang Chinese Medical University, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingjing","family":"Shi","sequence":"additional","affiliation":[{"name":"Department of Oncological Surgery, Affiliated Hangzhou First People\u2019s Hospital, Westlake University School of Medicine, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linlin","family":"Mao","sequence":"additional","affiliation":[{"name":"Department of Oncological Surgery, Affiliated Hangzhou First People\u2019s Hospital, Westlake University School of Medicine, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yeqin","family":"Ni","sequence":"additional","affiliation":[{"name":"Department of Oncological Surgery, Affiliated Hangzhou First People\u2019s Hospital, Westlake University School of Medicine, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dingcun","family":"Luo","sequence":"additional","affiliation":[{"name":"Department of Oncological Surgery, Affiliated Hangzhou First People\u2019s Hospital, Westlake University School of Medicine, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Oncological Surgery, Affiliated Hangzhou First People\u2019s Hospital, Westlake University School of Medicine, Hangzhou, Zhejiang, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"276","published-online":{"date-parts":[[2026,3,5]]},"reference":[{"key":"R1-20260424","doi-asserted-by":"crossref","first-page":"101367","DOI":"10.1016\/j.beem.2019.101367","article-title":"Hashimotos\u2019 thyroiditis: epidemiology, pathogenesis, clinic and therapy","volume":"33","author":"Ragusa","year":"2019","journal-title":"Best Pract Res Clin Endocrinol Metab"},{"key":"R2-20260424","doi-asserted-by":"crossref","first-page":"252\u201365","DOI":"10.1007\/s12020-012-9703-2","article-title":"The incidence and prevalence of thyroid autoimmunity","volume":"42","author":"McLeod","year":"2012","journal-title":"Endocrine"},{"key":"R3-20260424","doi-asserted-by":"crossref","first-page":"126","DOI":"10.3390\/antiox8050126","article-title":"H(2)O(2) Metabolism in normal thyroid cells and in thyroid tumorigenesis: focus on NADPH oxidases","volume":"8","author":"Szanto","year":"2019","journal-title":"Antioxidants (Basel)"},{"key":"R4-20260424","doi-asserted-by":"crossref","first-page":"791\u2013800","DOI":"10.1210\/js.2018-00427","article-title":"The impact of hashimoto thyroiditis on thyroid nodule cytology and risk of thyroid cancer","volume":"3","author":"Silva de Morais","year":"2019","journal-title":"J Endocr Soc"},{"key":"R5-20260424","doi-asserted-by":"crossref","first-page":"506\u201311","DOI":"10.1080\/00325481.2020.1739462","article-title":"The impact of concurrent Hashimoto thyroiditis on thyroid nodule cytopathology assessed by ultrasound-guided fine-needle aspiration cytology","volume":"132","author":"Hu","year":"2020","journal-title":"Postgrad Med"},{"key":"R6-20260424","doi-asserted-by":"crossref","first-page":"697\u2013704","DOI":"10.7863\/ultra.34.4.697","article-title":"Sonographic appearance of thyroid cancer in patients with Hashimoto thyroiditis","volume":"34","author":"Durfee","year":"2015","journal-title":"J Ultrasound Med"},{"key":"R7-20260424","doi-asserted-by":"crossref","first-page":"471\u201380","DOI":"10.1016\/j.ultrasmedbio.2018.10.020","article-title":"Sonographic characteristics of papillary thyroid carcinoma with coexistent Hashimoto\u2019s thyroiditis: conventional ultrasound, acoustic radiation force impulse imaging and contrast-enhanced ultrasound","volume":"45","author":"Peng","year":"2019","journal-title":"Ultrasound Med Biol"},{"key":"R8-20260424","doi-asserted-by":"crossref","first-page":"201","DOI":"10.3389\/fpls.2019.00201","article-title":"Expression of the tobacco non-symbiotic class 1 hemoglobin gene Hb1 reduces cadmium levels by modulating Cd transporter expression through decreasing nitric oxide and ROS level in arabidopsis","volume":"10","author":"Bahmani","year":"2019","journal-title":"Front Plant Sci"},{"key":"R9-20260424","doi-asserted-by":"crossref","first-page":"603\u201308","DOI":"10.1089\/thy.2007.0313","article-title":"The value of fine-needle aspiration biopsy in subcentimeter thyroid nodules","volume":"18","author":"Berker","year":"2008","journal-title":"Thyroid"},{"key":"R10-20260424","first-page":"232\u201338","article-title":"Roles of ultrasound and power Doppler ultrasound for diagnosis of Hashimoto thyroiditis in anti-thyroid marker-positive euthyroid subjects","volume":"4","author":"Ceylan","year":"2014","journal-title":"Quant Imaging Med Surg"},{"key":"R11-20260424","doi-asserted-by":"crossref","first-page":"2436","DOI":"10.3390\/cancers13102436","article-title":"Radiomics in differentiated thyroid cancer and nodules: explorations, application, and limitations","volume":"13","author":"Cao","year":"2021","journal-title":"Cancers (Basel)"},{"key":"R12-20260424","doi-asserted-by":"crossref","first-page":"1497\u2013509","DOI":"10.1109\/TMI.2022.3140797","article-title":"A local and global feature disentangled network: toward classification of benign-malignant thyroid nodules from ultrasound image","volume":"41","author":"Zhao","year":"2022","journal-title":"IEEE Trans Med Imaging"},{"key":"R13-20260424","doi-asserted-by":"crossref","first-page":"563\u201377","DOI":"10.1148\/radiol.2015151169","article-title":"Radiomics: images are more than pictures, they are data","volume":"278","author":"Gillies","year":"2016","journal-title":"Radiology"},{"key":"R14-20260424","first-page":"2157\u201364","article-title":"Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer","volume":"34","author":"Huang","year":"2016","journal-title":"J Clin Oncol"},{"key":"R15-20260424","doi-asserted-by":"crossref","first-page":"76\u201384","DOI":"10.1016\/j.ebiom.2018.07.029","article-title":"Development and validation of an MRI-based radiomics signature for the preoperative prediction of lymph node metastasis in bladder cancer","volume":"34","author":"Wu","year":"2018","journal-title":"EBioMedicine"},{"key":"R16-20260424","doi-asserted-by":"crossref","first-page":"858\u201367","DOI":"10.1089\/thy.2018.0380","article-title":"Machine learning-assisted system for thyroid nodule diagnosis","volume":"29","author":"Zhang","year":"2019","journal-title":"Thyroid"},{"key":"R17-20260424","doi-asserted-by":"crossref","first-page":"953\u201363","DOI":"10.1210\/clinem\/dgab870","article-title":"Convolutional neural network-based computer-assisted diagnosis of Hashimoto\u2019s thyroiditis on ultrasound","volume":"107","author":"Zhao","year":"2022","journal-title":"J Clin Endocrinol Metab"},{"key":"R18-20260424","doi-asserted-by":"crossref","first-page":"500\u201310","DOI":"10.1038\/s41568-018-0016-5","article-title":"Artificial intelligence in radiology","volume":"18","author":"Hosny","year":"2018","journal-title":"Nat Rev Cancer"},{"key":"R19-20260424","doi-asserted-by":"crossref","first-page":"193\u2013201","DOI":"10.1016\/S1470-2045(18)30762-9","article-title":"Diagnosis of thyroid cancer using deep convolutional neural network models applied to sonographic images: a retrospective, multicohort, diagnostic study","volume":"20","author":"Li","year":"2019","journal-title":"Lancet Oncol"},{"key":"R20-20260424","doi-asserted-by":"crossref","first-page":"941673","DOI":"10.3389\/fonc.2022.941673","article-title":"Investigating the diagnostic efficiency of a computer-aided diagnosis system for thyroid nodules in the context of Hashimoto\u2019s thyroiditis","volume":"12","author":"Gong","year":"2022","journal-title":"Front Oncol"},{"key":"R21-20260424","doi-asserted-by":"crossref","first-page":"1022305","DOI":"10.3389\/fonc.2022.1022305","article-title":"Contrast-enhanced ultrasound and shear wave elastography in the diagnosis of ACR TI-RADS 4 and 5 category thyroid nodules coexisting with Hashimoto\u2019s thyroiditis","volume":"12","author":"Wang","year":"2022","journal-title":"Front Oncol"},{"key":"R22-20260424","doi-asserted-by":"crossref","first-page":"3834\u201342","DOI":"10.1007\/s00330-020-06737-5","article-title":"Machine learning and radiomic phenotyping of lower grade gliomas: improving survival prediction","volume":"30","author":"Choi","year":"2020","journal-title":"Eur Radiol"},{"key":"R23-20260424","doi-asserted-by":"crossref","first-page":"3170","DOI":"10.1038\/s41467-019-11007-0","article-title":"Multicenter study demonstrates radiomic features derived from magnetic resonance perfusion images identify pseudoprogression in glioblastoma","volume":"10","author":"Elshafeey","year":"2019","journal-title":"Nat Commun"},{"key":"R24-20260424","first-page":"100179","article-title":"Artificial intelligence: a powerful paradigm for scientific research","volume":"2","author":"Xu","year":"2021","journal-title":"Innovation (Camb)"},{"key":"R25-20260424","doi-asserted-by":"crossref","first-page":"8761\u201377","DOI":"10.1021\/acs.jmedchem.9b01101","article-title":"Interpretation of compound activity predictions from complex machine learning models using local approximations and shapley values","volume":"63","author":"Rodriguez-Perez","year":"2020","journal-title":"J Med Chem"},{"key":"R26-20260424","doi-asserted-by":"crossref","first-page":"57","DOI":"10.3390\/cancers13010057","article-title":"Interpretable machine learning model for locoregional relapse prediction in oropharyngeal cancers","volume":"13","author":"Giraud","year":"2020","journal-title":"Cancers (Basel)"},{"key":"R27-20260424","first-page":"637\u201346","article-title":"Machine learning-based interpretation and visualization of nonlinear interactions in prostate cancer survival","volume":"4","author":"Li","year":"2020","journal-title":"JCO Clin Cancer Inform"},{"key":"R28-20260424","doi-asserted-by":"crossref","first-page":"336\u201359","DOI":"10.1007\/s11263-019-01228-7","article-title":"Grad-CAM: visual explanations from deep networks via gradient-based localization","volume":"128","author":"Selvaraju","year":"2020","journal-title":"Int J Comput Vis"},{"key":"R29-20260424","first-page":"100082","article-title":"Transparency in the reporting of Artificial INtelligence \u2013 the TITAN guideline","volume":"10","author":"Agha","year":"2025","journal-title":"Prem J Sci"},{"key":"R30-20260424","doi-asserted-by":"crossref","first-page":"e104","DOI":"10.1158\/0008-5472.CAN-17-0339","article-title":"Computational radiomics system to decode the radiographic phenotype","volume":"77","author":"van Griethuysen","year":"2017","journal-title":"Cancer Res"},{"key":"R32-20260424","first-page":"1180\u201384","article-title":"Reporting recommendations for tumor marker prognostic studies (REMARK)","volume":"97","author":"McShane","year":"2005","journal-title":"J Natl Cancer Inst"},{"key":"R33-20260424","doi-asserted-by":"crossref","first-page":"103018","DOI":"10.1016\/j.ebiom.2020.103018","article-title":"Deep learning radiomics of ultrasonography: identifying the risk of axillary non-sentinel lymph node involvement in primary breast cancer","volume":"60","author":"Guo","year":"2020","journal-title":"EBioMedicine"},{"key":"R34-20260424","first-page":"1339\u201349","article-title":"Integrating deep learning and radiomics in differentiating papillary thyroid microcarcinoma from papillary thyroid carcinoma with ultrasound images","volume":"17","author":"Yu","year":"2025","journal-title":"Cancer Manag Res"},{"key":"R35-20260424","doi-asserted-by":"crossref","first-page":"1267886","DOI":"10.3389\/fendo.2023.1267886","article-title":"Radiomics-based ultrasound models for thyroid nodule differentiation in Hashimoto\u2019s thyroiditis","volume":"14","author":"Fang","year":"2023","journal-title":"Front Endocrinol (Lausanne)"},{"key":"R36-20260424","doi-asserted-by":"crossref","first-page":"13318","DOI":"10.1038\/s41598-025-97247-1","article-title":"Interpretable model based on MRI radiomics to predict the expression of Ki-67 in breast cancer","volume":"15","author":"Zhang","year":"2025","journal-title":"Sci Rep"},{"key":"R37-20260424","doi-asserted-by":"crossref","first-page":"3638\u201346","DOI":"10.1007\/s00330-023-09450-1","article-title":"Prediction of lymph node metastasis in stage T1-2 rectal cancers with MRI-based deep learning","volume":"33","author":"Wan","year":"2023","journal-title":"Eur Radiol"},{"key":"R38-20260424","doi-asserted-by":"crossref","first-page":"69","DOI":"10.1186\/s12885-024-11838-1","article-title":"An integrated model incorporating deep learning, hand-crafted radiomics and clinical and US features to diagnose central lymph node metastasis in patients with papillary thyroid cancer","volume":"24","author":"Gao","year":"2024","journal-title":"BMC Cancer"},{"key":"R39-20260424","doi-asserted-by":"crossref","first-page":"557169","DOI":"10.3389\/fonc.2020.557169","article-title":"A clinical assessment of an ultrasound computer-aided diagnosis system in differentiating thyroid nodules with radiologists of different diagnostic experience","volume":"10","author":"Zhang","year":"2020","journal-title":"Front Oncol"},{"key":"R40-20260424","doi-asserted-by":"crossref","first-page":"169\u201374","DOI":"10.1007\/s40477-020-00453-y","article-title":"Computer-aided diagnostic system for thyroid nodule sonographic evaluation outperforms the specificity of less experienced examiners","volume":"23","author":"Fresilli","year":"2020","journal-title":"J Ultrasound"},{"key":"R41-20260424","doi-asserted-by":"crossref","first-page":"611436","DOI":"10.3389\/fonc.2021.611436","article-title":"A computer-aided diagnosis system and thyroid imaging reporting and data system for dual validation of ultrasound-guided fine-needle aspiration of indeterminate thyroid nodules","volume":"11","author":"Liang","year":"2021","journal-title":"Front Oncol"},{"key":"R42-20260424","article-title":"Thyroid cancer polygenic risk score improves classification of thyroid nodules as benign or malignant","author":"Pozdeyev","year":"2023","journal-title":"J Clin Endocrinol Metab"},{"key":"R43-20260424","doi-asserted-by":"crossref","first-page":"23\u201329","DOI":"10.14366\/usg.20068","article-title":"Applications of machine learning and deep learning to thyroid imaging: where do we stand?","volume":"40","author":"Ha","year":"2021","journal-title":"Ultrasonography"}],"container-title":["International Journal of Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.lww.com\/10.1097\/JS9.0000000000004546","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T19:00:19Z","timestamp":1777057219000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.lww.com\/10.1097\/JS9.0000000000004546"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,5]]},"references-count":42,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026]]}},"URL":"https:\/\/doi.org\/10.1097\/js9.0000000000004546","relation":{},"ISSN":["1743-9159"],"issn-type":[{"value":"1743-9159","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,5]]},"assertion":[{"value":"2025-03-31","name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-30","name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}}]}}