{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T14:30:21Z","timestamp":1791469821270,"version":"4.3.4"},"reference-count":36,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2021,6,23]],"date-time":"2021-06-23T00:00:00Z","timestamp":1624406400000},"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. Oncol."],"abstract":"<jats:sec>\n                    <jats:title>Purpose\/Objectives(s)<\/jats:title>\n                    <jats:p>\n                      Salivary gland tumors are a rare, histologically heterogeneous group of tumors. The distinction between malignant and benign tumors of the parotid gland is clinically important. This study aims to develop and evaluate a deep-learning network for diagnosing parotid gland tumors\n                      <jats:italic>via<\/jats:italic>\n                      the deep learning of MR images.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials\/Methods<\/jats:title>\n                    <jats:p>Two hundred thirty-three patients with parotid gland tumors were enrolled in this study. Histology results were available for all tumors. All patients underwent MRI scans, including T1-weighted, CE-T1-weighted and T2-weighted imaging series. The parotid glands and tumors were segmented on all three MR image series by a radiologist with 10 years of clinical experience. A total of 3791 parotid gland region images were cropped from the MR images. A label (pleomorphic adenoma and Warthin tumor, malignant tumor or free of tumor), which was based on histology results, was assigned to each image. To train the deep-learning model, these data were randomly divided into a training dataset (90%, comprising 3035 MR images from 212 patients: 714 pleomorphic adenoma images, 558 Warthin tumor images, 861 malignant tumor images, and 902 images free of tumor) and a validation dataset (10%, comprising 275 images from 21 patients: 57 pleomorphic adenoma images, 36 Warthin tumor images, 93 malignant tumor images, and 89 images free of tumor). A modified ResNet model was developed to classify these images. The input images were resized to 224x224 pixels, including four channels (T1-weighted tumor images only, T2-weighted tumor images only, CE-T1-weighted tumor images only and parotid gland images). Random image flipping and contrast adjustment were used for data enhancement. The model was trained for 1200 epochs with a learning rate of 1e-6, and the Adam optimizer was implemented. It took approximately 2 hours to complete the whole training procedure. The whole program was developed with PyTorch (version 1.2).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The model accuracy with the training dataset was 92.94% (95% CI [0.91, 0.93]). The micro-AUC was 0.98. The experimental results showed that the accuracy of the final algorithm in the diagnosis and staging of parotid cancer was 82.18% (95% CI [0.77, 0.86]). The micro-AUC was 0.93.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The proposed model may be used to assist clinicians in the diagnosis of parotid tumors. However, future larger-scale multicenter studies are required for full validation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fonc.2021.632104","type":"journal-article","created":{"date-parts":[[2021,6,23]],"date-time":"2021-06-23T03:17:41Z","timestamp":1624418261000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":37,"title":["Deep Learning for Differentiating Benign From Malignant Parotid Lesions on MR Images"],"prefix":"10.3389","volume":"11","author":[{"given":"Xianwu","family":"Xia","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bin","family":"Feng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiazhou","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qianjin","family":"Hua","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yide","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liang","family":"Sheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yonghua","family":"Mou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weigang","family":"Hu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2021,6,23]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1016\/j.critrevonc.2009.10.004","article-title":"Major and Minor Salivary Gland Tumors","volume":"74","author":"Guzzo","year":"2010","journal-title":"Crit Rev Oncol Hematol"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.3390\/jcm8050566","article-title":"Characteristics of 5015 Salivary Gland Neoplasms Registered in the Hiroshima Tumor Tissue Registry Over a Period of 39 Years","volume":"8","author":"Sentani","year":"2019","journal-title":"J Clin Med"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1016\/j.bjoms.2018.01.021","article-title":"Current Thinking About the Management of Recurrent Pleomorphic Adenoma of the Parotid: A Structured Review","volume":"56","author":"Kanatas","year":"2018","journal-title":"Br J Oral Maxillofac Surg"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.oraloncology.2005.06.004","article-title":"Conservative Parotidectomy for the Treatment of Parotid Cancers","volume":"41","author":"Lim","year":"2005","journal-title":"Oral Oncol"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1080\/00015458.2016.1262491","article-title":"Fine-Needle Aspiration Cytology (FNAC): Is it Useful in Preoperative Diagnosis of Parotid Gland Lesions","volume":"117","author":"Correia-S\u00e1","year":"2017","journal-title":"Acta Chir Belg"},{"key":"B6","doi-asserted-by":"publisher","first-page":"398","DOI":"10.1002\/lary.26776","article-title":"Utility of Preoperative Fine Needle Aspiration in Parotid Lesions","volume":"128","author":"Eytan","year":"2018","journal-title":"Laryngoscope"},{"key":"B7","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1177\/0194599815607841","article-title":"Sensitivity, Specificity, and Posttest Probability of Parotid Fine-Needle Aspiration: A Systematic Review and Meta-Analysis","volume":"154","author":"Liu","year":"2016","journal-title":"Otolaryngol Head Neck Surg"},{"key":"B8","doi-asserted-by":"publisher","first-page":"e20170267","DOI":"10.1590\/1678-7757-2017-0267","article-title":"The Value of Fine Needle Aspiration Cytology in the Clinical Management of Rare Salivary Gland Tumors","volume":"26","author":"Mezei","year":"2018","journal-title":"J Appl Oral Sci"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1016\/j.carj.2018.10.010","article-title":"Diagnostic Accuracy of Multiparametric Magnetic Resonance Imaging for Differentiation Between Parotid Neoplasms","volume":"70","author":"Elmokadem","year":"2019","journal-title":"Can Assoc Radiol J"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1007\/s00259-019-04447-9","article-title":"A Multiparametric Analysis Based on DCE-MRI to Improve the Accuracy of Parotid Tumor Discrimination","volume":"46","author":"Xu","year":"2019","journal-title":"Eur J Nucl Med Mol Imaging"},{"key":"B11","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1016\/j.clinimag.2018.05.026","article-title":"Diagnostic Accuracy of Magnetic Resonance Imaging Techniques for Parotid Tumors, a Systematic Review and Meta-Analysis","volume":"52","author":"Liang","year":"2018","journal-title":"Clin Imaging"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1148\/radiology.214.1.r00ja05231","article-title":"Salivary Gland Tumors: Evaluation With Two-Phase Helical CT","volume":"214","author":"Choi","year":"2000","journal-title":"Radiology"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1097\/01.rct.0000236418.82395.b3","article-title":"Dynamic Multislice Computed Tomography Findings for Parotid Gland Tumors","volume":"31","author":"Yerli","year":"2007","journal-title":"J Comput Assist Tomogr"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-011-2101-y","article-title":"Distinguishing Benign From Malignant Parotid Gland Tumours: Low-Dose Multi-Phasic CT Protocol With 5-Minute Delay","volume":"21","author":"Jin","year":"2011","journal-title":"Eur Radiol"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.26534","article-title":"Deep Learning in Radiology: An Overview of the Concepts and a Survey of the State of the Art With Focus on MRI","volume":"49","author":"Mazurowski","year":"2019","journal-title":"J Magn Reson Imaging"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12453","article-title":"A Deep Feature Fusion Methodology for Breast Cancer Diagnosis Demonstrated on Three Imaging Modality Datasets","volume":"44","author":"Antropova","year":"2017","journal-title":"Med Phys"},{"key":"B17","doi-asserted-by":"publisher","first-page":"15415","DOI":"10.1038\/s41598-017-15720-y","article-title":"Searching for Prostate Cancer by Fully Automated Magnetic Resonance Imaging Classification: Deep Learning Versus non-Deep Learning","volume":"7","author":"Wang","year":"2017","journal-title":"Sci Rep"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2017.08.006","article-title":"Co-Trained Convolutional Neural Networks for Automated Detection of Prostate Cancer in Multi-Parametric MRI","volume":"42","author":"Yang","year":"2017","journal-title":"Med Image Anal"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1016\/j.jacr.2017.12.026","article-title":"Artificial Intelligence and Machine Learning in Radiology: Opportunities, Challenges, Pitfalls, and Criteria for Success","volume":"15","author":"Thrall","year":"2018","journal-title":"J Am Coll Radiol"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2889096","article-title":"Hierarchical Fully Convolutional Network for Joint Atrophy Localization and Alzheimer\u2019s Disease Diagnosis Using Structural Mri","volume":"42","author":"Lian","year":"2020","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1126\/science.aaf7894","article-title":"Combining Satellite Imagery and Machine Learning to Predict Poverty","volume":"353","author":"Jean","year":"2016","journal-title":"Science"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1002\/hed.25049","article-title":"Computer-Aided System for Diagnosing Thyroid Nodules on Ultrasound: A Comparison With Radiologist-Based Clinical Assessments","volume":"40","author":"Gao","year":"2018","journal-title":"Head Neck"},{"key":"B23","doi-asserted-by":"crossref","DOI":"10.1109\/ISBI.2016.7493384","article-title":"Describing Ultrasound Video Content Using Deep Convolutional Neural Networks","volume-title":"2016 IEEE 13th International Symposium on Biomedical Imaging (Isbi)","author":"Gao","year":"2016"},{"key":"B24","doi-asserted-by":"publisher","first-page":"5110","DOI":"10.1038\/s41598-017-05300-5","article-title":"Location Sensitive Deep Convolutional Neural Networks for Segmentation of White Matter Hyperintensities","volume":"7","author":"Ghafoorian","year":"2017","journal-title":"Sci Rep"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1016\/j.ebiom.2020.102780","article-title":"Deep Learning-Based Fully Automated Detection and Segmentation of Lymph Nodes on Multiparametric-Mri for Rectal Cancer: A Multicentre Study","volume":"56","author":"Zhao","year":"2020","journal-title":"EBioMedicine"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1038\/srep24454","article-title":"Computer-Aided Diagnosis With Deep Learning Architecture: Applications to Breast Lesions in US Images and Pulmonary Nodules in CT Scans","volume":"6","author":"Cheng","year":"2016","journal-title":"Sci Rep"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1186\/s13550-017-0260-9","article-title":"Comparison of Machine Learning Methods for Classifying Mediastinal Lymph Node Metastasis of non-Small Cell Lung Cancer From\u00a018F-FDG PET\/CT Images","volume":"7","author":"Wang","year":"2017","journal-title":"EJNMMI Res"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1007\/s11801-016-6054-x","article-title":"Performance Analyses of Subcarrier BPSK Modulation Over M Turbulence Channels With Pointing Errors","volume":"12","author":"Ma","year":"2016","journal-title":"Optoelectron Lett"},{"key":"B29","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1158\/1055-9965.EPI-15-0578","article-title":"Global Cancer Incidence and Mortality Rates and Trends\u2013An Update","volume":"25","author":"Torre","year":"2016","journal-title":"Cancer Epidemiol Biomarkers Prev"},{"key":"B30","doi-asserted-by":"publisher","first-page":"10134:2E","DOI":"10.1117\/12.2253982","article-title":"Fine-Tuning Convolutional Deep Features for MRI Based Brain Tumor Classification","author":"Ahmed","year":"2017","journal-title":"Proc SPIE 10134 Med Imaging 2017: Computer-Aided Diagn"},{"key":"B31","volume-title":"Deep Learning","author":"Heaton","year":"2016"},{"key":"B32","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/IJCNN.2019.8852239","article-title":"A Proof of Local Convergence for the Adam Optimizer","volume-title":"2019 International Joint Conference on Neural Networks (Ijcnn)","author":"Bock","year":"2019"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2016.06.037","article-title":"Image Analysis and Machine Learning in Digital Pathology: Challenges and Opportunities","volume":"33","author":"Madabhushi","year":"2016","journal-title":"Med Image Anal"},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.1007\/s00428-019-02594-w","article-title":"Machine Learning Approaches for Pathologic Diagnosis","volume":"475","author":"Komura","year":"2019","journal-title":"Virchows Arch"},{"key":"B35","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep Residual Learning for Image Recognition","volume-title":"IEEE","author":"He","year":"2016"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1109\/ICISC44355.2019.9036442","article-title":"\u201cA Study of the Optimization Algorithms in Deep Learning,\u201d 2019 Third International Conference on Inventive Systems and Control (ICISC)","author":"Zaheer","year":"2019","journal-title":"IEEE"}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2021.632104\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,6,23]],"date-time":"2021-06-23T03:17:45Z","timestamp":1624418265000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2021.632104\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,6,23]]},"references-count":36,"alternative-id":["10.3389\/fonc.2021.632104"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2021.632104","relation":{},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,6,23]]},"article-number":"632104"}}