{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,28]],"date-time":"2026-10-28T00:34:08Z","timestamp":1793147648329,"version":"4.5.0"},"reference-count":38,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T00:00:00Z","timestamp":1782432000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Qinhuangdao S&T Plan Program","award":["202301A212"],"award-info":[{"award-number":["202301A212"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To construct and validate a multi-task deep learning model based on ConvNeXt-Tiny for synchronous prediction of isocitrate dehydrogenase (IDH) mutation status and Ki-67 expression level in gliomas.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Materials and Methods<\/jats:title>\n                    <jats:p>This retrospective multicenter study included adult patients with diffuse glioma from two hospitals, which served as the training set and the independent validation set, respectively. All patients underwent multimodal MRI examinations within 4 weeks before surgery, including T2-weighted imaging (T2WI), T2-fluid attenuated inversion recovery (T2-FLAIR), contrast-enhanced T1-weighted imaging (T1CE), and three-dimensional arterial spin labeling imaging (ASL). MRI data were uniformly preprocessed to extract traditional radiomics features, and a dual-task deep learning model based on ConvNeXt-Tiny was constructed. Clinical characteristics, radiomics features, and deep learning features were multimodally fused to establish the multimodal model and were compared with other prediction models. Model performance and clinical utility were evaluated using receiver operating characteristic (ROC) curves, calibration curves, decision curve analysis, and the net reclassification index (NRI). SHapley additive explanations (SHAP) analysis and gradient-weighted class activation mapping (Grad-CAM) were applied for visual interpretation of model decisions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The multimodal model demonstrated the best diagnostic performance for both IDH mutation status and Ki-67 expression level prediction. For IDH prediction, the multimodal model achieved AUCs of 0.901 and 0.883 in the training set and test set, respectively, while for Ki-67 prediction, the area under the curves (AUCs) reached 0.939 and 0.924. NRI analysis further confirmed that the multimodal model significantly improved case reclassification in both tasks. Calibration curves and the Hosmer\u2013Lemeshow test indicated good model fit, and decision curve analysis showed a higher net clinical benefit. SHAP analysis revealed that model predictions mainly relied on deep learning features, particularly those derived from T1CE, ADC, and cerebral blood flow (CBF) images, whereas radiomics features and clinical variables contributed less. Grad-CAM visualizations demonstrated that the model primarily focused on tumor-related regions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This study developed a ConvNeXt-Tiny-based multi-task model that enables preoperative synchronous prediction of IDH mutation and Ki-67 status in adult diffuse glioma and exhibits robust performance superior to single-task and traditional models.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0351757","type":"journal-article","created":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T17:34:57Z","timestamp":1782495297000},"page":"e0351757","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["A multimodal ConvNeXt-Tiny deep learning model for simultaneous prediction of IDH mutation and Ki-67 expression in gliomas"],"prefix":"10.1371","volume":"21","author":[{"given":"Juan","family":"Du","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linsha","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Duo","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinguo","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuo","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Defeng","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huiling","family":"Yi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Liang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaohan","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qinglei","family":"Shi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4845-1295","authenticated-orcid":true,"given":"Tao","family":"Zheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,6,26]]},"reference":[{"issue":"1","key":"pone.0351757.ref001","doi-asserted-by":"crossref","first-page":"33","DOI":"10.1038\/s41572-024-00516-y","article-title":"Glioma","volume":"10","author":"M Weller","year":"2024","journal-title":"Nature Reviews Disease Primers"},{"issue":"2","key":"pone.0351757.ref002","doi-asserted-by":"crossref","first-page":"126","DOI":"10.1038\/s41374-021-00667-6","article-title":"Grading of adult diffuse gliomas according to the 2021 WHO classification of tumors of the central nervous system","volume":"102","author":"T Komori","year":"2022","journal-title":"Lab Invest"},{"issue":"2","key":"pone.0351757.ref003","doi-asserted-by":"crossref","DOI":"10.1016\/j.ccell.2025.11.002","article-title":"Advancing CNS tumor diagnostics with expanded DNA methylation-based classification","volume":"44","author":"M Sill","year":"2026","journal-title":"Cancer Cell"},{"issue":"7","key":"pone.0351757.ref004","doi-asserted-by":"crossref","first-page":"1442","DOI":"10.1002\/acn3.70081","article-title":"An Analysis for IDH-Mutant Grade 4 Astrocytoma Based on WHO CNS 5: Implication of Clinical Practice","volume":"12","author":"X Qiu","year":"2025","journal-title":"Ann Clin Transl Neurol"},{"key":"pone.0351757.ref005","doi-asserted-by":"crossref","first-page":"580335","DOI":"10.3389\/fimmu.2020.580335","article-title":"The change of soluble programmed cell death-ligand 1 in glioma patients receiving radiotherapy and its impact on clinical outcomes","volume":"11","author":"X-C Ding","year":"2020","journal-title":"Front Immunol"},{"issue":"1","key":"pone.0351757.ref006","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1186\/s41016-022-00307-y","article-title":"The risk factors of hemorrhage in stereotactic needle biopsy for brain lesions in a large cohort: 10 years of experience in a single center","volume":"8","author":"H Li","year":"2022","journal-title":"Chinese Neurosurg J"},{"key":"pone.0351757.ref007","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1016\/j.canlet.2021.12.008","article-title":"Unravelling glioblastoma heterogeneity by means of single-cell RNA sequencing","volume":"527","author":"A Hern\u00e1ndez Mart\u00ednez","year":"2022","journal-title":"Cancer Lett"},{"issue":"7","key":"pone.0351757.ref008","doi-asserted-by":"crossref","first-page":"936","DOI":"10.1093\/neuonc\/noaa041","article-title":"\u201cReal world\u201d use of a highly reliable imaging sign: \u201cT2-FLAIR mismatch\u201d for identification of IDH mutant astrocytomas","volume":"22","author":"R Jain","year":"2020","journal-title":"Neuro Oncol"},{"issue":"20","key":"pone.0351757.ref009","doi-asserted-by":"crossref","first-page":"5120","DOI":"10.3390\/cancers13205120","article-title":"Intra- and interobserver variability in magnetic resonance imaging measurements in rectal cancer patients","volume":"13","author":"P Grimm","year":"2021","journal-title":"Cancers (Basel)"},{"key":"pone.0351757.ref010","doi-asserted-by":"crossref","first-page":"1577816","DOI":"10.3389\/fonc.2025.1577816","article-title":"Noninvasive prediction of Ki-67 expression level in IDH-wildtype glioblastoma using MRI histogram analysis: comparison and combination of MRI morphological features","volume":"15","author":"Q Liang","year":"2025","journal-title":"Front Oncol"},{"issue":"1","key":"pone.0351757.ref011","doi-asserted-by":"crossref","first-page":"13318","DOI":"10.1038\/s41598-025-97247-1","article-title":"Interpretable model based on MRI radiomics to predict the expression of Ki-67 in breast cancer","volume":"15","author":"L Zhang","year":"2025","journal-title":"Sci Rep"},{"key":"pone.0351757.ref012","doi-asserted-by":"crossref","unstructured":"Liu Z, Mao H, Wu CY, Feichtenhofer C, Darrell T, Xie S. A ConvNet for the 2020s. arXiv. 2022.","DOI":"10.1109\/CVPR52688.2022.01167"},{"issue":"4","key":"pone.0351757.ref013","doi-asserted-by":"crossref","first-page":"639","DOI":"10.1093\/neuonc\/noab238","article-title":"Improving the noninvasive classification of glioma genetic subtype with deep learning and diffusion-weighted imaging","volume":"24","author":"J Cluceru","year":"2022","journal-title":"Neuro Oncol"},{"key":"pone.0351757.ref014","doi-asserted-by":"crossref","first-page":"102578","DOI":"10.1016\/j.compmedimag.2025.102578","article-title":"Classification of glioma grade and Ki-67 level prediction in MRI data: A SHAP-driven interpretation","volume":"124","author":"EH Bhuiyan","year":"2025","journal-title":"Comput Med Imaging Graph"},{"key":"pone.0351757.ref015","article-title":"Joint classification and regression with deep multi task learning model using conventional based patch extraction for brain disease diagnosis","volume":"10","author":"P K","year":"2024","journal-title":"PeerJ Comput Sci"},{"issue":"2","key":"pone.0351757.ref016","doi-asserted-by":"crossref","first-page":"398","DOI":"10.1148\/radiol.2021204531","article-title":"Multitask Deep Learning for Segmentation and Classification of Primary Bone Tumors on Radiographs","volume":"301","author":"CE von Schacky","year":"2021","journal-title":"Radiology"},{"key":"pone.0351757.ref017","doi-asserted-by":"crossref","first-page":"109370","DOI":"10.1016\/j.compeleceng.2024.109370","article-title":"A review of Explainable Artificial Intelligence in healthcare","volume":"118","author":"Z Sadeghi","year":"2024","journal-title":"Comput Electrical Eng"},{"issue":"8","key":"pone.0351757.ref018","doi-asserted-by":"crossref","first-page":"1231","DOI":"10.1093\/neuonc\/noab106","article-title":"The 2021 WHO Classification of Tumors of the Central Nervous System: a summary","volume":"23","author":"DN Louis","year":"2021","journal-title":"Neuro Oncol"},{"issue":"4","key":"pone.0351757.ref019","article-title":"Genomic profiling and prognostic factors of H3 K27M-mutant spinal cord diffuse glioma","volume":"33","author":"R-C Chai","year":"2023","journal-title":"Brain Pathol"},{"issue":"11","key":"pone.0351757.ref020","doi-asserted-by":"crossref","DOI":"10.18637\/jss.v036.i11","article-title":"Feature Selection with theBorutaPackage","volume":"36","author":"MB Kursa","year":"2010","journal-title":"J Stat Soft"},{"issue":"15","key":"pone.0351757.ref021","doi-asserted-by":"crossref","first-page":"4625","DOI":"10.3390\/jcm11154625","article-title":"Swin Transformer Improves the IDH Mutation Status Prediction of Gliomas Free of MRI-Based Tumor Segmentation","volume":"11","author":"J Wu","year":"2022","journal-title":"J Clin Med"},{"key":"pone.0351757.ref022","doi-asserted-by":"crossref","first-page":"1134626","DOI":"10.3389\/fonc.2023.1134626","article-title":"Convolutional neural network to predict IDH mutation status in glioma from chemical exchange saturation transfer imaging at 7 Tesla","volume":"13","author":"Y Yuan","year":"2023","journal-title":"Front Oncol"},{"key":"pone.0351757.ref023","doi-asserted-by":"crossref","first-page":"1143688","DOI":"10.3389\/fonc.2023.1143688","article-title":"Deep-learning and conventional radiomics to predict IDH genotyping status based on magnetic resonance imaging data in adult diffuse glioma","volume":"13","author":"H Zhang","year":"2023","journal-title":"Front Oncol"},{"issue":"1","key":"pone.0351757.ref024","article-title":"High-order radiomics features based on T2 FLAIR MRI predict multiple glioma immunohistochemical features: A more precise and personalized gliomas management","volume":"15","author":"J Li","year":"2020","journal-title":"PLoS One"},{"issue":"5","key":"pone.0351757.ref025","doi-asserted-by":"crossref","first-page":"300060520914466","DOI":"10.1177\/0300060520914466","article-title":"Radiomic prediction models for the level of Ki-67 and p53 in glioma","volume":"48","author":"X Sun","year":"2020","journal-title":"J Int Med Res"},{"key":"pone.0351757.ref026","doi-asserted-by":"crossref","first-page":"1676","DOI":"10.3389\/fonc.2020.01676","article-title":"Machine learning-based radiomics predicting tumor grades and expression of multiple pathologic biomarkers in gliomas","volume":"10","author":"M Gao","year":"2020","journal-title":"Frontiers in oncology"},{"issue":"2","key":"pone.0351757.ref027","doi-asserted-by":"crossref","first-page":"963","DOI":"10.1016\/j.acra.2024.09.021","article-title":"Deep learning and habitat radiomics for the prediction of glioma pathology using multiparametric mri: a multicenter study","volume":"32","author":"Y Zhu","year":"2025","journal-title":"Acad Radiol"},{"key":"pone.0351757.ref028","doi-asserted-by":"crossref","first-page":"110560","DOI":"10.1016\/j.ejrad.2022.110560","article-title":"A multitask classification framework based on vision transformer for predicting molecular expressions of glioma","volume":"157","author":"Q Xu","year":"2022","journal-title":"Eur J Radiol"},{"issue":"1","key":"pone.0351757.ref029","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1038\/s41698-024-00670-2","article-title":"Biologically interpretable multi-task deep learning pipeline predicts molecular alterations, grade, and prognosis in glioma patients","volume":"8","author":"X Wu","year":"2024","journal-title":"NPJ Precis Oncol"},{"issue":"6","key":"pone.0351757.ref030","doi-asserted-by":"crossref","first-page":"3400","DOI":"10.21037\/qims-22-887","article-title":"Predicting isocitrate dehydrogenase genotype, histological phenotype, and Ki-67 expression level in diffuse gliomas with an advanced contrast analysis of magnetic resonance imaging sequences","volume":"13","author":"Y Cui","year":"2023","journal-title":"Quant Imaging Med Surg"},{"issue":"5","key":"pone.0351757.ref031","doi-asserted-by":"crossref","first-page":"3187","DOI":"10.1007\/s00330-021-08444-1","article-title":"A nomogram strategy for identifying the subclassification of IDH mutation and ATRX expression loss in lower-grade gliomas","volume":"32","author":"S Wu","year":"2022","journal-title":"Eur Radiol"},{"issue":"2","key":"pone.0351757.ref032","doi-asserted-by":"crossref","first-page":"1202","DOI":"10.1007\/s00330-019-06379-2","article-title":"Arterial spin labeling perfusion-weighted imaging aids in prediction of molecular biomarkers and survival in glioblastomas","volume":"30","author":"R-E Yoo","year":"2020","journal-title":"Eur Radiol"},{"issue":"5","key":"pone.0351757.ref033","doi-asserted-by":"crossref","first-page":"1399","DOI":"10.1002\/jmri.27434","article-title":"Predicting Isocitrate Dehydrogenase (IDH) Mutation Status in Gliomas Using Multiparameter MRI Radiomics Features","volume":"53","author":"H Peng","year":"2021","journal-title":"J Magn Reson Imaging"},{"issue":"6","key":"pone.0351757.ref034","doi-asserted-by":"crossref","first-page":"1444","DOI":"10.3390\/diagnostics12061444","article-title":"Arterial Spin Labeling Perfusion in Determining the IDH1 Status and Ki-67 Index in Brain Gliomas","volume":"12","author":"AI Batalov","year":"2022","journal-title":"Diagnostics (Basel)"},{"key":"pone.0351757.ref035","volume":"1","author":"Q Ostrom","year":"2015"},{"issue":"1","key":"pone.0351757.ref036","article-title":"Interrelationships between molecular subtype, anatomical location, and extent of resection in diffuse glioma: a systematic review and meta-analysis","volume":"1","author":"BI De Leeuw","year":"2019","journal-title":"Neurooncol Adv"},{"issue":"2","key":"pone.0351757.ref037","doi-asserted-by":"crossref","first-page":"304","DOI":"10.1093\/neuonc\/noaa177","article-title":"Fully automated hybrid approach to predict the IDH mutation status of gliomas via deep learning and radiomics","volume":"23","author":"YS Choi","year":"2021","journal-title":"Neuro Oncol"},{"issue":"1","key":"pone.0351757.ref038","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1186\/s12885-019-6511-6","article-title":"Prognostic factors of patients with Gliomas - an analysis on 335 patients with Glioblastoma and other forms of Gliomas","volume":"20","author":"J Liang","year":"2020","journal-title":"BMC Cancer"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0351757","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T17:35:04Z","timestamp":1782495304000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0351757"}},"subtitle":[],"editor":[{"given":"Mario","family":"Tortora","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,26]]},"references-count":38,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6,26]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0351757","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,26]]}}}