{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T17:47:50Z","timestamp":1792086470220,"version":"4.4.0"},"reference-count":20,"publisher":"Public Library of Science (PLoS)","issue":"10","license":[{"start":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T00:00:00Z","timestamp":1792022400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>\n                      Accurate prediction of lymph node metastasis (LNM) in non-small cell lung cancer (NSCLC) is crucial for staging and treatment planning. This study aimed to establish a standardized machine learning workflow based on routinely available clinical,\n                      <jats:bold>serological,<\/jats:bold>\n                      and PET\/CT-derived parameters to enhance preoperative LNM risk stratification.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This retrospective study included 255 patients with histologically confirmed NSCLC, all of whom underwent systematic lymph-node dissection. Patients were randomly divided, with stratification by pathological lymph-node status, into a training cohort (n\u2009=\u2009178) and an internal held-out test test cohort (n\u2009=\u200977). Candidate predictors were screened using univariable and multivariable logistic regression in the training cohort. Six machine-learning algorithms were trained and optimized using repeated nested five-fold cross-validation. After model development was completed, all six models were applied without modification to the test cohort. Test-set AUCs were compared using paired DeLong tests with Holm correction. The primary model was further evaluated for calibration, clinical utility, and feature contribution using SHAP. A conventional multivariable logistic regression model using the same four predictors was added as a simple reference model, and its discrimination and calibration were compared with those of XGBoost.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Central tumor location, elevated carcinoembryonic antigen, elevated squamous cell carcinoma antigen, and solid nodule status were retained as predictors; none of the quantitative PET metabolic parameters remained independently associated with LNM. XGBoost achieved the highest cross-validated AUC among the six candidate machine-learning algorithms in the training cohort (0.735; 95% CI, 0.653\u20130.813) and was therefore designated the primary model. In the internal held-out test cohort, XGBoost achieved an AUC of 0.747 (95% CI, 0.631\u20130.846). The conventional logistic regression reference model achieved a test AUC of 0.733 (95% CI, 0.612\u20130.841), with no significant difference from XGBoost (DeLong P\u2009=\u20090.301). Its Brier score was 0.152, compared with 0.198 for XGBoost.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>XGBoost model showed moderate discrimination and a high negative predictive value for pathological LNM in NSCLC. However, its limited positive predictive value and the absence of significant differences among the candidate algorithms support cautious interpretation. External multicenter evaluation is required before clinical implementation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0361043","type":"journal-article","created":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T17:25:37Z","timestamp":1792085137000},"page":"e0361043","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Predicting lymph node metastasis in non-small cell lung cancer using clinical and PET\/CT-derived morphological features: An interpretable machine-learning study"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-7356-4325","authenticated-orcid":true,"given":"Yunchao","family":"Xu","sequence":"first","affiliation":[{"name":"Central South University\/Hunan Cancer Hospital","place":["Changsha, Hunan, People\u2019s Republic of China"],"department":["The Affiliated Cancer Hospital of Xiangya School of Medicine"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiajun","family":"Hao","sequence":"additional","affiliation":[{"name":"The 970 Hospital of the Joint Logistics Support Force of the People\u2019s Liberation Army of China","place":["Yantai City, Shandong Province, China"],"department":["Department of Radiological Diagnosis"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yujiao","family":"Dou","sequence":"additional","affiliation":[{"name":"Xiangya Hospital, Central South University","place":["Changsha, China"],"department":["Department of Clinical Nursing"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Biao","family":"Zeng","sequence":"additional","affiliation":[{"name":"Central South University\/Hunan Cancer Hospital","place":["Changsha, Hunan, People\u2019s Republic of China"],"department":["The Affiliated Cancer Hospital of Xiangya School of Medicine"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,10,15]]},"reference":[{"issue":"9","key":"pone.0361043.ref001","doi-asserted-by":"crossref","first-page":"1403","DOI":"10.1016\/j.jtho.2017.06.003","article-title":"Clinical and pathological staging validation in the eighth edition of the TNM classification for lung cancer: Correlation between solid size on thin-section computed tomography and invasive size in pathological findings in the new T classification","volume":"12","author":"K Aokage","year":"2017","journal-title":"J Thorac Oncol"},{"issue":"1","key":"pone.0361043.ref002","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1038\/s41416-023-02491-9","article-title":"Circulating tumour cells and PD-L1-positive small extracellular vesicles: The liquid biopsy combination for prognostic information in patients with metastatic non-small cell lung cancer","volume":"130","author":"Z Eslami-S","year":"2024","journal-title":"Br J Cancer"},{"issue":"3","key":"pone.0361043.ref003","doi-asserted-by":"crossref","first-page":"3160","DOI":"10.3390\/curroncol30030239","article-title":"Stage III non-small-cell lung cancer: An overview of treatment options","volume":"30","author":"F Petrella","year":"2023","journal-title":"Curr Oncol"},{"issue":"12","key":"pone.0361043.ref004","doi-asserted-by":"crossref","DOI":"10.1148\/rg.240057","article-title":"Proposed ninth edition TNM staging system for lung cancer: Guide for radiologists","volume":"44","author":"M Klug","year":"2024","journal-title":"Radiographics"},{"key":"pone.0361043.ref005","doi-asserted-by":"crossref","first-page":"17588359241293364","DOI":"10.1177\/17588359241293364","article-title":"The role of baseline 18F-FDG PET\/CT for survival prognosis in NSCLC patients undergoing immunotherapy: A systematic review and meta-analysis","volume":"16","author":"M Huang","year":"2024","journal-title":"Ther Adv Med Oncol"},{"issue":"3","key":"pone.0361043.ref006","doi-asserted-by":"crossref","first-page":"195","DOI":"10.2174\/1874471013666191223153755","article-title":"FDG PET\/CT in the staging of lung cancer","volume":"13","author":"M Farsad","year":"2020","journal-title":"Curr Radiopharm"},{"key":"pone.0361043.ref007","article-title":"Value of preoperative 18F-FDG PET\/CT and HRCT in predicting the differentiation degree of lung adenocarcinoma dominated by solid density","volume":"11","author":"X Chen","year":"2023","journal-title":"PeerJ"},{"issue":"3","key":"pone.0361043.ref008","doi-asserted-by":"crossref","first-page":"1645","DOI":"10.1016\/j.acra.2024.11.037","article-title":"Non-invasive prediction of lymph node metastasis in NSCLC using clinical, radiomics, and deep learning features from 18F-FDG PET\/CT based on interpretable machine learning","volume":"32","author":"F Duan","year":"2025","journal-title":"Acad Radiol"},{"issue":"9","key":"pone.0361043.ref009","doi-asserted-by":"crossref","first-page":"696","DOI":"10.3390\/diagnostics10090696","article-title":"Value of shape and texture features from 18F-FDG PET\/CT to discriminate between benign and malignant solitary pulmonary nodules: An experimental evaluation","volume":"10","author":"B Palumbo","year":"2020","journal-title":"Diagnostics (Basel)"},{"issue":"1","key":"pone.0361043.ref010","doi-asserted-by":"crossref","first-page":"2541316","DOI":"10.1080\/07853890.2025.2541316","article-title":"Machine learning and SHAP value interpretation for predicting the response to neoadjuvant chemotherapy and long-term clinical outcomes in Chinese female breast cancer","volume":"57","author":"Q Yuan","year":"2025","journal-title":"Ann Med"},{"issue":"1","key":"pone.0361043.ref011","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1186\/s13014-024-02523-5","article-title":"Regularity and correlation analysis of regional lymph node metastasis in nonoperative patients with non-small cell lung cancer based on positron emission tomography\/computed tomography images","volume":"19","author":"F Sun","year":"2024","journal-title":"Radiat Oncol"},{"issue":"1","key":"pone.0361043.ref012","doi-asserted-by":"crossref","first-page":"23","DOI":"10.1186\/s13550-022-00895-x","article-title":"Development of a PET\/CT molecular radiomics-clinical model to predict thoracic lymph node metastasis of invasive lung adenocarcinoma\u2009\u2264\u20093 cm in diameter","volume":"12","author":"C Chang","year":"2022","journal-title":"EJNMMI Res"},{"key":"pone.0361043.ref013","doi-asserted-by":"crossref","first-page":"1453232","DOI":"10.3389\/fimmu.2024.1453232","article-title":"Whole slide image-based weakly supervised deep learning for predicting major pathological response in non-small cell lung cancer following neoadjuvant chemoimmunotherapy: A multicenter, retrospective, cohort study","volume":"15","author":"D Han","year":"2024","journal-title":"Front Immunol"},{"key":"pone.0361043.ref014","doi-asserted-by":"crossref","first-page":"1552479","DOI":"10.3389\/fendo.2025.1552479","article-title":"A machine learning-based model for predicting recurrence in intermediate- and high-risk differentiated thyroid cancer: Insights from a retrospective single-center study of 2388 patients","volume":"16","author":"Y Li","year":"2025","journal-title":"Front Endocrinol (Lausanne)"},{"key":"pone.0361043.ref015","doi-asserted-by":"crossref","first-page":"1403392","DOI":"10.3389\/fonc.2024.1403392","article-title":"Construction of a risk prediction model for lung infection after chemotherapy in lung cancer patients based on the machine learning algorithm","volume":"14","author":"T Sun","year":"2024","journal-title":"Front Oncol"},{"issue":"5","key":"pone.0361043.ref016","first-page":"622","article-title":"Predictive model for difficult laryngoscopy using machine learning: Retrospective cohort study","volume":"72","author":"JH Kim","year":"2022","journal-title":"Braz J Anesthesiol"},{"issue":"1","key":"pone.0361043.ref017","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1007\/s00259-020-04924-6","article-title":"Value of 18F-FDG PET\/CT radiomic features to distinguish solitary lung adenocarcinoma from tuberculosis","volume":"48","author":"Y Hu","year":"2021","journal-title":"Eur J Nucl Med Mol Imaging"},{"issue":"10","key":"pone.0361043.ref018","doi-asserted-by":"crossref","first-page":"6272","DOI":"10.1016\/j.acra.2025.06.001","article-title":"A deep learning model for preoperative prediction of lymph node metastasis in cT1-Stage lung adenocarcinoma: A multicenter external validation study","volume":"32","author":"Q Yan","year":"2025","journal-title":"Acad Radiol"},{"issue":"1","key":"pone.0361043.ref019","doi-asserted-by":"crossref","first-page":"12","DOI":"10.21147\/j.issn.1000-9604.2025.01.02","article-title":"CT-based radiomics-deep learning model predicts occult lymph node metastasis in early-stage lung adenocarcinoma patients: A multicenter study","volume":"37","author":"X Yin","year":"2025","journal-title":"Chin J Cancer Res"},{"issue":"7","key":"pone.0361043.ref020","doi-asserted-by":"crossref","first-page":"2384","DOI":"10.1007\/s00259-025-07090-9","article-title":"An explainable transformer model integrating PET and tabular data for histologic grading and prognosis of follicular lymphoma: A multi-institutional digital biopsy study","volume":"52","author":"C Jiang","year":"2025","journal-title":"Eur J Nucl Med Mol Imaging"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0361043","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T17:25:41Z","timestamp":1792085141000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0361043"}},"subtitle":[],"editor":[{"given":"Carmelo","family":"Caldarella","sequence":"first","affiliation":[{"name":"Fondazione Policlinico Universitario Agostino Gemelli IRCCS","place":["ITALY"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,10,15]]},"references-count":20,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2026,10,15]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0361043","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,15]]}}}