{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T16:58:10Z","timestamp":1789664290958,"version":"4.0.1"},"reference-count":28,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,4,18]],"date-time":"2023-04-18T00:00:00Z","timestamp":1681776000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004844","name":"National Cheng Kung University Hospital","doi-asserted-by":"publisher","award":["NCKUH-11103026"],"award-info":[{"award-number":["NCKUH-11103026"]}],"id":[{"id":"10.13039\/501100004844","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004663","name":"Ministry of Science and Technology, Taiwan","doi-asserted-by":"publisher","award":["MOST 110-2314-B-006-103"],"award-info":[{"award-number":["MOST 110-2314-B-006-103"]}],"id":[{"id":"10.13039\/501100004663","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Oncol."],"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>To compare the diagnostic performance of radiomic analysis with machine learning (ML) model with a convolutional neural network (CNN) in differentiating thymic epithelial tumors (TETs) from other prevascular mediastinal tumors (PMTs).<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>A retrospective study was performed in patients with PMTs and undergoing surgical resection or biopsy in National Cheng Kung University Hospital, Tainan, Taiwan, E-Da Hospital, Kaohsiung, Taiwan, and Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan between January 2010 and December 2019. Clinical data including age, sex, myasthenia gravis (MG) symptoms and pathologic diagnosis were collected. The datasets were divided into UECT (unenhanced computed tomography) and CECT (enhanced computed tomography) for analysis and modelling. Radiomics model and 3D CNN model were used to differentiate TETs from non-TET PMTs (including cyst, malignant germ cell tumor, lymphoma and teratoma). The macro F1-score and receiver operating characteristic (ROC) analysis were performed to evaluate the prediction models.<\/jats:p><\/jats:sec><jats:sec><jats:title>Result<\/jats:title><jats:p>In the UECT dataset, there were 297 patients with TETs and 79 patients with other PMTs. The performance of radiomic analysis with machine learning model using LightGBM with Extra Tree (macro F1-Score = 83.95%, ROC-AUC = 0.9117) had better performance than the 3D CNN model (macro F1-score = 75.54%, ROC-AUC = 0.9015). In the CECT dataset, there were 296 patients with TETs and 77 patients with other PMTs. The performance of radiomic analysis with machine learning model using LightGBM with Extra Tree (macro F1-Score = 85.65%, ROC-AUC = 0.9464) had better performance than the 3D CNN model (macro F1-score = 81.01%, ROC-AUC = 0.9275).<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>Our study revealed that the individualized prediction model integrating clinical information and radiomic features using machine learning demonstrated better predictive performance in the differentiation of TETs from other PMTs at chest CT scan than 3D CNN model.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fonc.2023.1105100","type":"journal-article","created":{"date-parts":[[2023,4,18]],"date-time":"2023-04-18T01:11:33Z","timestamp":1681780293000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":12,"title":["Clinical radiomics-based machine learning versus three-dimension convolutional neural network analysis for differentiation of thymic epithelial tumors from other prevascular mediastinal tumors on chest computed tomography scan"],"prefix":"10.3389","volume":"13","author":[{"given":"Chao-Chun","family":"Chang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"En-Kuei","family":"Tang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu-Feng","family":"Wei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chia-Ying","family":"Lin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fu-Zong","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming-Ting","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi-Sheng","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi-Ting","family":"Yen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mi-Chia","family":"Ma","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yau-Lin","family":"Tseng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2023,4,18]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1016\/j.ejro.2014.12.003","article-title":"Anterior mediastinal masses in the framingham heart study: Prevalence and CT image characteristics","volume":"2","author":"Araki","year":"2015","journal-title":"Eur J Radiol Open"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1007\/s11604-020-01031-2","article-title":"Anterior mediastinal lesions: CT and MRI features and differential diagnosis","volume":"39","author":"Nakazono","year":"2021","journal-title":"Japanese J Radiol"},{"key":"B3","unstructured":"Thymomas and thymic carcinomas2022"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2007.10.002","article-title":"Anterior mediastinal tumors: diagnostic accuracy of CT and MRI","volume":"69","author":"Tomiyama","year":"2009","journal-title":"Eur J Radiol"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejca.2011.11.036","article-title":"Radiomics: extracting more information from medical images using advanced feature analysis","volume":"48","author":"Lambin","year":"2012","journal-title":"Eur J Cancer"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.7150\/thno.30309","article-title":"The applications of radiomics in precision diagnosis and treatment of oncology: Opportunities and challenges","volume":"9","author":"Liu","year":"2019","journal-title":"Theranostics"},{"key":"B7","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1016\/j.ejrad.2017.04.017","article-title":"Quantitative computed tomography texture analysis for estimating histological subtypes of thymic epithelial tumors","volume":"92","author":"Yasaka","year":"2017","journal-title":"Eur J Radiol"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1007\/s11547-017-0845-4","article-title":"Analysis of CT features and quantitative texture analysis in patients with thymic tumors: correlation with grading and staging","volume":"123","author":"Iannarelli","year":"2018","journal-title":"Radiol Med"},{"key":"B9","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1186\/s41747-018-0068-z","article-title":"Radiomics: the facts and the challenges of image analysis","volume":"2","author":"Rizzo","year":"2018","journal-title":"Eur Radiol Exp"},{"key":"B10","doi-asserted-by":"publisher","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":"B11","doi-asserted-by":"publisher","first-page":"5097","DOI":"10.3390\/s20185097","article-title":"3D deep learning on medical images: A review","volume":"20","author":"Singh","year":"2020","journal-title":"Sensors (Basel)"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1007\/s13244-018-0639-9","article-title":"Convolutional neural networks: an overview and application in radiology","volume":"9","author":"Yamashita","year":"2018","journal-title":"Insights Imaging"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1148\/rg.2021200180","article-title":"Mediastinal masses in children: Radiologic-pathologic correlation","volume":"41","author":"Biko","year":"2021","journal-title":"RadioGraphics"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.21037\/tcr.2016.06.17","article-title":"Computational resources for radiomics","volume":"5","author":"Court","year":"2016","journal-title":"Trans Cancer Res"},{"key":"B15","doi-asserted-by":"publisher","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":"B16","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-59354-4_15","article-title":"Uniformizing techniques to process CT scans with 3D CNNs for tuberculosis prediction","author":"Zunair","year":"2020"},{"key":"B17","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1007\/s13139-018-0514-0","article-title":"Radiomics and deep learning in clinical imaging: What should we do","volume":"52","author":"Choi","year":"2018","journal-title":"Nucl Med Mol Imaging"},{"key":"B18","doi-asserted-by":"publisher","first-page":"353","DOI":"10.1038\/s41467-020-20655-6","article-title":"Deep learning encodes robust discriminative neuroimaging representations to outperform standard machine learning","volume":"12","author":"Abrol","year":"2021","journal-title":"Nat Commun"},{"key":"B19","doi-asserted-by":"publisher","first-page":"16143","DOI":"10.1038\/s41598-021-95748-3","article-title":"Comparison of machine and deep learning for the classification of cervical cancer based on cervicography images","volume":"11","author":"Park","year":"2021","journal-title":"Sci Rep"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.27900","article-title":"Differentiating benign from malignant renal tumors using T2- and diffusion-weighted images: A comparison of deep learning and radiomics models versus assessment from radiologists","volume":"55","author":"Xu","year":"2022","journal-title":"J Magn Reson Imaging"},{"key":"B21","doi-asserted-by":"publisher","first-page":"12","DOI":"10.3390\/cancers14010012","article-title":"Classification of clinically significant prostate cancer on multi-parametric MRI: A validation study comparing deep learning and radiomics","volume":"14","author":"Castillo","year":"2021","journal-title":"Cancers (Basel)"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2020.00418","article-title":"Comparison and fusion of deep learning and radiomics features of ground-glass nodules to predict the invasiveness risk of stage-I lung adenocarcinomas in CT scan","volume":"10","author":"Xia","year":"2020","journal-title":"Front Oncol"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2021.737368","article-title":"A comparative study of radiomics and deep-learning based methods for pulmonary nodule malignancy prediction in low dose CT images","volume":"11","author":"Astaraki","year":"2021","journal-title":"Front Oncol"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1737","DOI":"10.21037\/atm-21-5999","article-title":"A radiomics model combined with XGBoost may improve the accuracy of distinguishing between mediastinal cysts and tumors: a multicenter validation analysis","volume":"9","author":"Wang","year":"2021","journal-title":"Ann Transl Med"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1007\/s11547-020-01188-w","article-title":"Computed tomography (CT)-derived radiomic features differentiate prevascular mediastinum masses as thymic neoplasms versus lymphomas","volume":"125","author":"Kirienko","year":"2020","journal-title":"Radiol Med"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.21037\/jtd.2019.05.32","article-title":"CT-based radiomics features analysis for predicting the risk of anterior mediastinal lesions","volume":"11","author":"Sui","year":"2019","journal-title":"J Thorac Dis"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-020-05075-4","article-title":"Decoding COVID-19 pneumonia: comparison of deep learning and radiomics CT image signatures","volume":"48","author":"Wang","year":"2021","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2014.11.042","article-title":"High rate of unnecessary thymectomy and its cause. can computed tomography distinguish thymoma, lymphoma, thymic hyperplasia, and thymic cysts","volume":"84","author":"Ackman","year":"2015","journal-title":"Eur J Radiol"}],"updated-by":[{"DOI":"10.3389\/fonc.2023.1220962","type":"corrigendum","label":"Corrigendum","source":"publisher","updated":{"date-parts":[[2023,5,31]],"date-time":"2023-05-31T00:00:00Z","timestamp":1685491200000}}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2023.1105100\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,5,31]],"date-time":"2023-05-31T09:06:26Z","timestamp":1685523986000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2023.1105100\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,4,18]]},"references-count":28,"alternative-id":["10.3389\/fonc.2023.1105100"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2023.1105100","relation":{"corrigendum":[{"id-type":"doi","id":"10.3389\/fonc.2023.1220962","asserted-by":"object"}]},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,4,18]]},"article-number":"1105100"}}