{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T06:28:38Z","timestamp":1787725718021,"version":"build-2784847793"},"reference-count":31,"publisher":"Public Library of Science (PLoS)","issue":"4","license":[{"start":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T00:00:00Z","timestamp":1775692800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"2024 Zhanjiang Non-funded Science and Technology Research Project","award":["2024B01267"],"award-info":[{"award-number":["2024B01267"]}]},{"DOI":"10.13039\/501100017549","name":"Science and Technology Innovation 2025 Major Project of Ningbo","doi-asserted-by":"publisher","award":["2025B01248"],"award-info":[{"award-number":["2025B01248"]}],"id":[{"id":"10.13039\/501100017549","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100017549","name":"Science and Technology Innovation 2025 Major Project of Ningbo","doi-asserted-by":"publisher","award":["GDMULCJC2025059"],"award-info":[{"award-number":["GDMULCJC2025059"]}],"id":[{"id":"10.13039\/501100017549","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100008596","name":"Batten Institute for Innovation and Entrepreneurship, Darden School of Business, University of Virginia","doi-asserted-by":"publisher","award":["JDXM2024032F"],"award-info":[{"award-number":["JDXM2024032F"]}],"id":[{"id":"10.13039\/100008596","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100021161","name":"Guangdong Key Laboratory for Big Data Analysis and Simulation of Public Opinion","doi-asserted-by":"publisher","award":["no applicable"],"award-info":[{"award-number":["no applicable"]}],"id":[{"id":"10.13039\/501100021161","id-type":"DOI","asserted-by":"publisher"}]}],"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>Accurate prediction of radiation-induced oral mucositis is crucial for personalized treatment in head and neck cancer. However, developing robust predictive models utilizing high-dimensional multimodal data (CT imaging, dose distribution, and clinical features) remains challenging, particularly in cohorts with limited sample sizes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study aimed to rigorously evaluate and compare the multi-class predictive performance of traditional machine learning algorithms and deep learning architectures under a small-cohort setting.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Multimodal data from 108 patients were collected. A comprehensive evaluation framework incorporating nine traditional machine learning algorithms and two deep learning models (a dimensionality-reduced 1D-CNN and a multimodal 3D-CNN) was established. To ensure robust evaluation, a stratified 5-fold cross-validation was employed. Model performance was comprehensively quantified using mean\u2009\u00b1\u2009standard deviation (SD) across multiple metrics, including the Area Under the Curve (AUC), accuracy, and Matthews Correlation Coefficient (MCC).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Inter-rater reliability for RIOM grading was excellent (Cohen\u2019s kappa\u2009=\u20090.82, 95% CI: 0.73\u20130.91). Among traditional machine learning approaches, the Extra Trees (ET) algorithm achieved the highest discriminative capacity (AUC: 0.956\u2009\u00b1\u20090.046), while Logistic Regression (LR) demonstrated optimal overall accuracy (0.832\u2009\u00b1\u20090.155) and stability. Regarding deep learning, the lightweight 1D-CNN utilizing fused low-dimensional features exhibited highly competitive and robust performance (AUC: 0.900\u2009\u00b1\u20090.072; Accuracy: 0.732\u2009\u00b1\u20090.140). In stark contrast, the high-dimensional multimodal 3D-CNN suffered from severe overfitting and mode collapse phenomenon, yielding significantly inferior results (AUC: 0.568\u2009\u00b1\u20090.090; MCC: \u22120.025\u2009\u00b1\u20090.031).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec005\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>For small-cohort radiomics and dosimetric analyses, ensemble learning models (e.g., ET) and appropriately regularized linear models (e.g., LR) remain highly effective. While deep learning holds promise, high-dimensional architectures like 3D-CNNs are highly susceptible to mode collapse without massive datasets. Instead, employing feature dimensionality reduction combined with lightweight networks (1D-CNN) is a vastly superior strategy to extract reliable predictive patterns from limited clinical data.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0346251","type":"journal-article","created":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T17:28:47Z","timestamp":1775755727000},"page":"e0346251","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["Development and evaluation of a multimodal feature-based predictive model for radiotherapy-induced oral mucositis in nasopharyngeal carcinoma"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-0323-0180","authenticated-orcid":true,"given":"Ling","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linke","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruifeng","family":"Guo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shiting","family":"Fang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ke","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ge","family":"Yuan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Danxian","family":"Jiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9820-7648","authenticated-orcid":true,"given":"Jing","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,4,9]]},"reference":[{"issue":"1","key":"pone.0346251.ref001","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1093\/jjco\/hyx151","article-title":"Head and neck cancer in Hong Kong","volume":"48","author":"WT Ng","year":"2018","journal-title":"Jpn J Clin Oncol"},{"issue":"7","key":"pone.0346251.ref002","doi-asserted-by":"crossref","first-page":"840","DOI":"10.1200\/JCO.20.03237","article-title":"Chemotherapy in combination with radiotherapy for definitive-intent treatment of stage II-IVA Nasopharyngeal Carcinoma: CSCO and ASCO Guideline","volume":"39","author":"Y-P Chen","year":"2021","journal-title":"J Clin Oncol"},{"issue":"2","key":"pone.0346251.ref003","first-page":"70","article-title":"Incidence and risk factors for radiotherapy-induced oral mucositis among patients with nasopharyngeal carcinoma: a meta-analysis","volume":"17","author":"J Li","year":"2023","journal-title":"Asian Nurs Res (Korean Soc Nurs Sci)"},{"issue":"7","key":"pone.0346251.ref004","doi-asserted-by":"crossref","first-page":"2464","DOI":"10.3390\/ijerph17072464","article-title":"Oncological-therapy related oral mucositis as an interdisciplinary problem-literature review","volume":"17","author":"A Kusiak","year":"2020","journal-title":"Int J Environ Res Public Health"},{"issue":"5","key":"pone.0346251.ref005","doi-asserted-by":"crossref","first-page":"429","DOI":"10.1016\/j.canrad.2017.11.006","article-title":"Pharmacological modulation of radiation-induced oral mucosal complications","volume":"22","author":"S Bockel","year":"2018","journal-title":"Cancer Radiother"},{"issue":"2","key":"pone.0346251.ref006","article-title":"The association between cumulative radiation dose and the incidence of severe oral mucositis in head and neck cancers during radiotherapy","volume":"4","author":"T Sunaga","year":"2021","journal-title":"Cancer Rep (Hoboken)"},{"issue":"3","key":"pone.0346251.ref007","doi-asserted-by":"crossref","first-page":"565","DOI":"10.4103\/jcrt.JCRT_511_19","article-title":"Correlation of planning target volume with mucositis for head-and-neck cancer patients undergoing chemoradiation","volume":"16","author":"JG Manur","year":"2020","journal-title":"J Cancer Res Ther"},{"key":"pone.0346251.ref008","doi-asserted-by":"crossref","first-page":"790","DOI":"10.3389\/fonc.2020.00790","article-title":"Machine learning-based models for prediction of toxicity outcomes in radiotherapy","volume":"10","author":"LJ Isaksson","year":"2020","journal-title":"Front Oncol"},{"issue":"2","key":"pone.0346251.ref009","doi-asserted-by":"crossref","first-page":"102042","DOI":"10.1016\/j.jormas.2024.102042","article-title":"Analysis of factors related to radiation-induced oral mucositis in patients with head and neck tumors undergoing radiotherapy","volume":"126","author":"X-J Tan","year":"2025","journal-title":"J Stomatol Oral Maxillofac Surg"},{"issue":"6","key":"pone.0346251.ref010","first-page":"4410","article-title":"Artificial intelligence: radiologists\u2019 expectations and opinions gleaned from a nationwide online survey","volume":"31","author":"F Coppola","year":"2021","journal-title":"Eur Radiol"},{"issue":"1","key":"pone.0346251.ref011","doi-asserted-by":"crossref","first-page":"86","DOI":"10.1016\/j.radonc.2010.08.013","article-title":"Characteristics of response of oral and pharyngeal mucosa in patients receiving chemo-IMRT for head and neck cancer using hypofractionated accelerated radiotherapy","volume":"97","author":"SA Bhide","year":"2010","journal-title":"Radiother Oncol"},{"key":"pone.0346251.ref012","doi-asserted-by":"crossref","first-page":"1343170","DOI":"10.3389\/fonc.2024.1343170","article-title":"Predicting radiation pneumonitis in lung cancer: a EUD-based machine learning approach for volumetric modulated arc therapy patients","volume":"14","author":"F Ye","year":"2024","journal-title":"Front Oncol"},{"issue":"23","key":"pone.0346251.ref013","doi-asserted-by":"crossref","first-page":"5705","DOI":"10.3390\/cancers15235705","article-title":"Predictive factors for chemoradiation-induced oral mucositis and dysphagia in head and neck cancer: a scoping review","volume":"15","author":"AJ Nicol","year":"2023","journal-title":"Cancers (Basel)"},{"issue":"3","key":"pone.0346251.ref014","doi-asserted-by":"crossref","first-page":"507","DOI":"10.1016\/j.ijrobp.2015.07.2263","article-title":"Evaluation of the Risk of Grade 3 Oral and Pharyngeal Dysphagia Using Atlas-Based Method and Multivariate Analyses of Individual Patient Dose Distributions","volume":"93","author":"S Otter","year":"2015","journal-title":"Int J Radiat Oncol Biol Phys"},{"issue":"1","key":"pone.0346251.ref015","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1016\/j.radonc.2016.05.015","article-title":"Normal tissue complication probability (NTCP) modelling using spatial dose metrics and machine learning methods for severe acute oral mucositis resulting from head and neck radiotherapy","volume":"120","author":"JA Dean","year":"2016","journal-title":"Radiother Oncol"},{"key":"pone.0346251.ref016","doi-asserted-by":"crossref","first-page":"266","DOI":"10.1016\/j.oraloncology.2018.10.006","article-title":"Multivariable model for predicting acute oral mucositis during combined IMRT and chemotherapy for locally advanced nasopharyngeal cancer patients","volume":"86","author":"E Orlandi","year":"2018","journal-title":"Oral Oncol"},{"issue":"1","key":"pone.0346251.ref017","doi-asserted-by":"crossref","first-page":"219","DOI":"10.3390\/curroncol30010017","article-title":"Predicting nomogram for severe oral mucositis in patients with nasopharyngeal carcinoma during intensity-modulated radiation therapy: a retrospective cohort study","volume":"30","author":"Z Liu","year":"2022","journal-title":"Curr Oncol"},{"key":"pone.0346251.ref018","doi-asserted-by":"crossref","first-page":"596822","DOI":"10.3389\/fonc.2020.596822","article-title":"Predictive model and precaution for oral mucositis during chemo-radiotherapy in nasopharyngeal carcinoma patients","volume":"10","author":"P-J Li","year":"2020","journal-title":"Front Oncol"},{"key":"pone.0346251.ref019","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1016\/j.radonc.2020.03.013","article-title":"Prediction of radiation-induced mucositis of H&N cancer patients based on a large patient cohort","volume":"147","author":"CR Hansen","year":"2020","journal-title":"Radiother Oncol"},{"issue":"531","key":"pone.0346251.ref020","doi-asserted-by":"crossref","first-page":"1406","DOI":"10.1080\/01621459.2019.1632079","article-title":"Optimal Sparse Linear Prediction for Block-missing Multi-modality Data without Imputation","volume":"115","author":"G Yu","year":"2020","journal-title":"J Am Stat Assoc"},{"issue":"12","key":"pone.0346251.ref021","first-page":"2434","article-title":"Performance of multi-modality and multi-classifier fusion models for predicting radiation-induced oral mucositis in patients with nasopharyngeal carcinoma","volume":"44","author":"Y Hu","year":"2024","journal-title":"Nan Fang Yi Ke Da Xue Xue Bao"},{"issue":"7","key":"pone.0346251.ref022","doi-asserted-by":"crossref","first-page":"2032","DOI":"10.3390\/cancers15072032","article-title":"Multimodal Data Integration to Predict Severe Acute Oral Mucositis of Nasopharyngeal Carcinoma Patients Following Radiation Therapy","volume":"15","author":"Y Dong","year":"2023","journal-title":"Cancers (Basel)"},{"key":"pone.0346251.ref023","doi-asserted-by":"crossref","first-page":"107841","DOI":"10.1016\/j.oraloncology.2025.107841","article-title":"Ensemble learning-based radiomics model for predicting radiation-induced temporal lobe injury in nasopharyngeal carcinoma","volume":"173","author":"M Zhang","year":"2026","journal-title":"Oral Oncol"},{"issue":"4","key":"pone.0346251.ref024","doi-asserted-by":"crossref","first-page":"2721","DOI":"10.62347\/SWVQ2652","article-title":"Oral dose-volume parameters as independent predictors of severe radiation-induced mucositis in nasopharyngeal carcinoma patients: a retrospective study","volume":"17","author":"Z Zhang","year":"2025","journal-title":"Am J Transl Res"},{"issue":"24","key":"pone.0346251.ref025","article-title":"Improving robustness of a deep learning-based lung-nodule classification model of CT images with respect to image noise","volume":"66","author":"Y Gao","year":"2021","journal-title":"Phys Med Biol"},{"key":"pone.0346251.ref026","doi-asserted-by":"crossref","first-page":"30","DOI":"10.4103\/jpi.jpi_78_20","article-title":"Improving generalization of deep learning models for diagnostic pathology by increasing variability in training data: experiments on osteosarcoma subtypes","volume":"12","author":"H Tang","year":"2021","journal-title":"J Pathol Inform"},{"issue":"4","key":"pone.0346251.ref027","doi-asserted-by":"crossref","first-page":"277","DOI":"10.1038\/nrc1318","article-title":"The pathobiology of mucositis","volume":"4","author":"ST Sonis","year":"2004","journal-title":"Nat Rev Cancer"},{"issue":"5","key":"pone.0346251.ref028","doi-asserted-by":"crossref","first-page":"1483","DOI":"10.1148\/rg.2017170056","article-title":"CT Texture Analysis: Definitions, Applications, Biologic Correlates, and Challenges","volume":"37","author":"MG Lubner","year":"2017","journal-title":"Radiographics"},{"key":"pone.0346251.ref029","doi-asserted-by":"crossref","first-page":"269","DOI":"10.3389\/fonc.2019.00269","article-title":"Dosiomics: Extracting 3D Spatial Features From Dose Distribution to Predict Incidence of Radiation Pneumonitis","volume":"9","author":"B Liang","year":"2019","journal-title":"Front Oncol"},{"issue":"3","key":"pone.0346251.ref030","doi-asserted-by":"crossref","DOI":"10.1016\/j.heliyon.2024.e24866","article-title":"Predicting severe radiation-induced oral mucositis in head and neck cancer patients using integrated baseline CT radiomic, dosimetry, and clinical features: A machine learning approach","volume":"10","author":"R Agheli","year":"2024","journal-title":"Heliyon"},{"issue":"2","key":"pone.0346251.ref031","doi-asserted-by":"crossref","first-page":"181","DOI":"10.4103\/crst.crst_332_22","article-title":"Artificial intelligence-based prediction of oral mucositis in patients with head-and-neck cancer: A prospective observational study utilizing a thermographic approach","volume":"6","author":"R Thukral","year":"2023","journal-title":"Cancer Research, Statistics, and Treatment"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346251","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,9]],"date-time":"2026-04-09T17:28:53Z","timestamp":1775755733000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0346251"}},"subtitle":[],"editor":[{"given":"Lorenzo","family":"Faggioni","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,4,9]]},"references-count":31,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4,9]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0346251","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,9]]}}}