{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T11:06:45Z","timestamp":1777633605035,"version":"3.51.4"},"reference-count":33,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T00:00:00Z","timestamp":1777420800000},"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>Background\u200b<\/jats:title>\n                    <jats:p>Papillary thyroid carcinoma (PTC) constitutes 80\u201390% of all thyroid malignancies. Despite its generally favorable prognosis, 20\u201330% of PTC patients present intermediate\/high-risk features, increasing distant metastasis risk. Traditional clinicopathological predictors (e.g., TNM stage, tumor size) have limited accuracy in forecasting metastasis, creating a need for more precise prediction tools.\u200b.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods\u200b<\/jats:title>\n                    <jats:p>A total of 2,452 PTC patients (diagnosed 2015\u20132023, follow-up \u22656 months) were enrolled, with data including clinical, pathological, laboratory, and ultrasound indices. Feature selection integrated LASSO, RFE, and ReliefF, identifying 7 core features. Nine machine learning (ML) algorithms were compared; SHAP analysis was used for interpretability. External validation included 432 patients, and a Django-based prediction website was developed.\u200b.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results\u200b<\/jats:title>\n                    <jats:p>The LightGBM model exhibited optimal performance: test-set AUC\u00a0=\u00a00.886, accuracy = 0.887, and external validation AUC\u00a0=\u00a00.758. SHAP analysis identified extrathyroidal invasion (Mean |SHAP| = 0.1329) and thyroglobulin antibody (TgAb, Mean |SHAP| = 0.0981) as top predictors. Tumor size showed a nonlinear association with metastasis, and the model had a 93.6% negative predictive value (NPV) for excluding low-risk patients.\u200b.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions\u200b<\/jats:title>\n                    <jats:p>This interpretable ML model outperforms traditional predictors, effectively supporting clinical risk stratification and personalized treatment decision-making for PTC patients, with potential for broad clinical application.\u200b.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fendo.2026.1779102","type":"journal-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T04:45:02Z","timestamp":1777437902000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["An interpretable machine learning-based approach: development, validation, and clinical utility for distant metastasis prediction in PTC"],"prefix":"10.3389","volume":"17","author":[{"given":"Ruijie","family":"Sun","sequence":"first","affiliation":[{"name":"Department of Otolaryngology, Qilu Hospital of Shandong University (Qingdao)\u200b","place":["Qingdao, Shandong, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuhui","family":"Ma","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Qilu Hospital of Shandong University (Qingdao)\u200b","place":["Qingdao, Shandong, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yushan","family":"Jiang","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Jimo People\u2019s Hospital of Qingdao\u200b","place":["Qingdao, Shandong, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaoguang","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Ultrasound, Qilu Hospital of Shandong University (Qingdao)\u200b","place":["Qingdao, Shandong, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2026,4,29]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"1956","DOI":"10.1186\/s12672-024-00956-w","article-title":"Development and validation of a machine learning-based model for predicting the risk of distant metastasis in patients with papillary thyroid cancer","volume":"16","author":"Gao","year":"2025","journal-title":"Discov Oncol"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.12659\/MSM.951110","article-title":"Machine learning integrating clinicopathological and immunological characteristics for predicting circumferential resection margin involvement in colorectal cancer with perineural invasion","volume":"31","author":"Pu","year":"2025","journal-title":"Med Sci Monit"},{"key":"B3","doi-asserted-by":"publisher","first-page":"45","DOI":"10.5539\/mas.v19n8p45","article-title":"Predicting the recurrence of differentiated thyroid cancer through data-intensive machine learning techniques","volume":"19","author":"Shrestha","year":"2025","journal-title":"Modern Appl Sci"},{"key":"B4","doi-asserted-by":"publisher","first-page":"6782","DOI":"10.1038\/s41598-025-62840-z","article-title":"Predictive risk-scoring model for lateral lymph node metastasis in papillary thyroid carcinoma: A multicenter retrospective analysis","volume":"15","author":"Guo","year":"2025","journal-title":"Sci Rep"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.98759","article-title":"Predictive models for secondary epilepsy in patients with acute ischemic stroke within one year","volume":"13","author":"Liu","year":"2024","journal-title":"eLife"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1089\/neu.2023.0410","article-title":"Predicting hematoma expansion and prognosis in cerebral contusions: A radiomics-clinical approach","volume":"41","author":"He","year":"2024","journal-title":"J Neurotrauma"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2025.1528392","article-title":"Machine learning-based model for predicting bone and lung metastasis in patients with differentiated thyroid carcinoma","volume":"16","author":"Huang","year":"2025","journal-title":"Front Endocrinol"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2025.1620085","article-title":"Multi-omics analysis and machine learning models identify risk factors for lymph node metastasis in N1b stage papillary thyroid microcarcinoma","volume":"16","author":"Dai","year":"2025","journal-title":"Front Immunol"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2024.1387402","article-title":"The association of pretreatment systemic immune inflammatory response with lymph node metastasis in papillary thyroid carcinoma","volume":"15","author":"Gu","year":"2024","journal-title":"Front Endocrinol"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1186\/s13044-025-00115-7","article-title":"Multivariate machine learning analysis reveals novel genetic risk factors for papillary thyroid cancer metastasis","volume":"18","author":"Wang","year":"2025","journal-title":"Thyroid Res"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2025.1618902","article-title":"Machine learning-based prediction of lymph node metastasis in papillary thyroid carcinoma using serum immune indicators and clinical features","volume":"16","author":"Feng","year":"2025","journal-title":"Front Endocrinol"},{"key":"B12","doi-asserted-by":"publisher","first-page":"7052","DOI":"10.1038\/s41467-025-51082-0","article-title":"Explainable multimodal deep learning for predicting thyroid cancer lateral lymph node metastasis: A multicenter diagnostic study","volume":"16","author":"Shen","year":"2025","journal-title":"Nat Commun"},{"key":"B13","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1186\/s40644-024-00678-6","article-title":"Clinical utility of radiomics-based deep learning models in thyroid cancer diagnosis and prognosis","volume":"24","author":"Martinez","year":"2024","journal-title":"Cancer Imaging"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2025.1655828","article-title":"Development of multiple machine learning models for Malignancy risk prediction in Bethesda category III thyroid nodules","volume":"16","author":"Li","year":"2025","journal-title":"Front Endocrinol"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2025.1639639","article-title":"XGBoost-based machine learning model combining clinical and ultrasound data for personalized prediction of thyroid nodule Malignancy","volume":"16","author":"Li","year":"2025","journal-title":"Front Endocrinol"},{"key":"B16","first-page":"45","article-title":"Machine learning approaches to predict thyroid cancer recurrence","author":"Ozturk","year":"2024"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1002\/jum.15759","article-title":"Classification of thyroid nodules by using deep learning on ultrasound cines","volume":"41","author":"Zhang","year":"2022","journal-title":"J Ultrasound Med"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.53347\/pjr-2024-00089","article-title":"Interpretable ensemble methods for predicting metastatic potential in thyroid Malignancies","volume":"12","author":"Thompson","year":"2025","journal-title":"Precis Med Rev"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1016\/j.ultrasmedbio.2023.07.018","article-title":"Ultrasound-guided nomogram for risk stratification of papillary thyroid carcinoma with lymph node metastasis","volume":"49","author":"Kim","year":"2023","journal-title":"Ultrasound Med Biol"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1210\/clinem\/dgaf487","article-title":"Gradient boosting machine learning prediction of metastatic differentiated thyroid cancer in children","volume":"110","author":"Redlich","year":"2025","journal-title":"J Clin Endocrinol Metab"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1038\/s44368-024-00095-7","article-title":"Integration of circulating tumor cells and machine learning for improved metastasis prediction in differentiated thyroid cancer","volume":"7","author":"Johnson","year":"2024","journal-title":"Nat Oncol"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1200\/JCO.2024.42.35_suppl.1456","article-title":"Tumor size paradox in thyroid cancer: Smaller tumors with aggressive biology and higher metastatic potential","volume":"43","author":"Robinson","year":"2025","journal-title":"J Clin Oncol"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1158\/1535-7163.MCT-24-0185","article-title":"Comprehensive genomic profiling identifies new therapeutic targets in papillary thyroid carcinoma with distant metastases","volume":"23","author":"Lee","year":"2024","journal-title":"Mol Cancer Ther"},{"key":"B24","doi-asserted-by":"publisher","first-page":"2486","DOI":"10.1038\/s41598-025-53847-2","article-title":"Nomogram models for predicting outcomes in thyroid cancer patients who underwent 131I therapy for distant metastases","volume":"15","author":"Cui","year":"2025","journal-title":"Sci Rep"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.240098","article-title":"Risk stratification of papillary thyroid cancer using dynamic radiomics and deep learning","volume":"6","author":"Evans","year":"2024","journal-title":"Radiol: Artif Intell"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.3389\/fendo.2025.1600815","article-title":"Development and validation of a novel mRNA classifier to predict lymph node metastasis and extrathyroidal extension in papillary thyroid cancer","volume":"16","author":"Golding","year":"2025","journal-title":"Front Endocrinol"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1007\/s00432-023-04831-z","article-title":"Risk factor analysis and prediction model for papillary thyroid carcinoma lymph node metastasis based on ultrasound and clinical parameters","volume":"149","author":"Lu","year":"2023","journal-title":"J Cancer Res Clin Oncol"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2022.952950","article-title":"HS3ST3A1 and CAPN8 serve as immune-related prognostic biomarkers for thyroid cancer","volume":"13","author":"Chen","year":"2022","journal-title":"Front Immunol"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.3389\/fgene.2023.972950","article-title":"A metabolism-related gene signature for predicting the prognosis of thyroid cancer","volume":"13","author":"Du","year":"2023","journal-title":"Front Genet"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.3332\/ecancer.2025.1487","article-title":"AI-powered biomarker system redefines thyroid cancer progression and subtypes","volume":"19","author":"Zhang","year":"2025","journal-title":"eCancerMedicalScience"},{"key":"B31","doi-asserted-by":"publisher","first-page":"198","DOI":"10.1093\/ajcp\/aqaf156","article-title":"Novel circulating biomarkers for early detection of distant metastasis in papillary thyroid carcinoma","volume":"164","author":"Brooks","year":"2025","journal-title":"Am J Clin Pathol"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1016\/j.ctrv.2025.103042","article-title":"Immune checkpoint inhibitors and thyroid cancer immunotherapy: Biomarkers for patient stratification","volume":"14","author":"Patel","year":"2025","journal-title":"Immunother Rev"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1097\/MD.0000000000038274","article-title":"Artificial intelligence-driven gene expression model for predicting lymph node metastasis in papillary thyroid carcinoma","volume":"104","author":"Li","year":"2025","journal-title":"Medicine"}],"container-title":["Frontiers in Endocrinology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2026.1779102\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T04:45:18Z","timestamp":1777437918000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2026.1779102\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,29]]},"references-count":33,"alternative-id":["10.3389\/fendo.2026.1779102"],"URL":"https:\/\/doi.org\/10.3389\/fendo.2026.1779102","relation":{},"ISSN":["1664-2392"],"issn-type":[{"value":"1664-2392","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,29]]},"article-number":"1779102"}}