{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T16:30:23Z","timestamp":1787675423292,"version":"build-2784847793"},"reference-count":43,"publisher":"Wiley","license":[{"start":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T00:00:00Z","timestamp":1787097600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T00:00:00Z","timestamp":1787097600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["82471675"],"award-info":[{"award-number":["82471675"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004608","name":"Natural Science Foundation of Jiangsu Province","doi-asserted-by":"publisher","award":["BK20241718"],"award-info":[{"award-number":["BK20241718"]}],"id":[{"id":"10.13039\/501100004608","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Magnetic Resonance Imaging"],"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Percutaneous transforaminal endoscopic discectomy (PTED) is a common treatment for lumbar disc herniation (LDH), but postoperative recovery varies greatly.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To develop and validate a deep learning (DL) model integrating MRI and clinical features to estimate 1\u2010year outcomes after PTED.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Study Type<\/jats:title>\n                    <jats:p>Retrospective.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Population<\/jats:title>\n                    <jats:p>\n                      252 LDH patients who underwent single\u2010level PTED, with lumbar spine MRI acquired within 1\u2009month before surgery: training set (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u2009129) and internal test set (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u200955) from Center 1; external test set (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u200968) from Center 2.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Field Strength\/Sequence<\/jats:title>\n                    <jats:p>T2\u2010weighted spin echo sequence at 1.5\u2009T.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Assessment<\/jats:title>\n                    <jats:p>The primary endpoint was achieving the Patient Acceptable Symptom State at 1\u2009year (Oswestry disability index \u2264\u200922). A front\u2010end deep learning model extracted disc and paraspinal muscle features from MRI data. Key features identified by SHAP analysis were incorporated into a DL framework that fused them with ResNet50\u2010derived imaging features and clinical data via multi\u2010head attention. Model performance was compared with conventional machine learning models (using clinical and radiomic features) and a spine surgeon. Sensitivity analyses were carried out using alternative ODI thresholds (\u2264\u200915 and &lt;\u200925).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Statistical Tests<\/jats:title>\n                    <jats:p>\n                      Accuracy with confidence interval (CI), F1 score, and Cohen's kappa (\n                      <jats:italic>\u03ba<\/jats:italic>\n                      ) were calculated to assess model performance. SHAP values ranked feature importance. The DeLong test compared areas under the receiver operating characteristic curves (AUCs) across models. Statistical significance was set at\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.05.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Key factors were paraspinal muscle quality features, including the fatty infiltration and relative cross\u2010sectional area. The DL model achieved 83.6% accuracy internally, compared with XGBoost (76.4%) and the surgeon (76.4%), with an F1 score of 0.769 and\n                      <jats:italic>\u03ba<\/jats:italic>\n                      of 0.644. External validation accuracy was 79.4% and\n                      <jats:italic>\u03ba<\/jats:italic>\n                      was 0.564.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Data Conclusion<\/jats:title>\n                    <jats:p>The DL model showed encouraging performance compared with XGBoost and a spine surgeon for estimating 1\u2010year outcomes after PTED, with paraspinal muscle quality as a key factor, though these preliminary findings require larger multi\u2010center validation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Level of Evidence<\/jats:title>\n                    <jats:p>4.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Technical Efficacy<\/jats:title>\n                    <jats:p>Stage 2.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/jmri.70518","type":"journal-article","created":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T02:55:08Z","timestamp":1787194508000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["A Combined Clinical\/\n                    <scp>MRI<\/scp>\n                    Deep Learning Model for Estimating Outcomes After Percutaneous Transforaminal Endoscopic Discectomy"],"prefix":"10.1002","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0794-2116","authenticated-orcid":false,"given":"Yefu","family":"Xu","sequence":"first","affiliation":[{"name":"Department of Pain Medicine Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University  Nanjing Jiangsu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiangkai","family":"Yu","sequence":"additional","affiliation":[{"name":"Department of Spine Surgery Zhongda Hospital, School of Medicine, Southeast University  Nanjing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yan","family":"Zhou","sequence":"additional","affiliation":[{"name":"Department of Spine Surgery Zhongda Hospital, School of Medicine, Southeast University  Nanjing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-3567-0630","authenticated-orcid":false,"given":"Sangni","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Medicine, Southeast University  Nanjing Jiangsu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chao","family":"Ma","sequence":"additional","affiliation":[{"name":"Department of Pain Medicine Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University  Nanjing Jiangsu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8224-7803","authenticated-orcid":false,"given":"Ying","family":"Han","sequence":"additional","affiliation":[{"name":"Department of Pain Medicine Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University  Nanjing Jiangsu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6608-7984","authenticated-orcid":false,"given":"Daojuan","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Pain Medicine Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University  Nanjing Jiangsu China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinghui","family":"Xie","sequence":"additional","affiliation":[{"name":"Department of Spine Surgery Zhongda Hospital, School of Medicine, Southeast University  Nanjing China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-5543-8989","authenticated-orcid":false,"given":"Gaojian","family":"Tao","sequence":"additional","affiliation":[{"name":"Department of Pain Medicine Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University  Nanjing Jiangsu China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,8,19]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.1002\/jsp2.1289"},{"key":"e_1_2_10_3_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.wnsx.2024.100276","article-title":"Role of Surgery in Primary Lumbar Disk Herniation: WFNS Spine Committee Recommendations","volume":"22","author":"Costa F.","year":"2024","journal-title":"World Neurosurgery X"},{"key":"e_1_2_10_4_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2021-065846"},{"key":"e_1_2_10_5_1","doi-asserted-by":"crossref","DOI":"10.1042\/BSR20181866","article-title":"Percutaneous Endoscopic Transforaminal Discectomy Precedes Interlaminar Discectomy in the Efficacy and Safety for Lumbar Disc Herniation","volume":"39","author":"Chen P.","year":"2019","journal-title":"Bioscience Reports"},{"key":"e_1_2_10_6_1","doi-asserted-by":"crossref","DOI":"10.3389\/fsurg.2022.984868","article-title":"Comparison of Percutaneous Transforaminal Endoscopic Discectomy and Open Lumbar Discectomy for Lumbar Disc Herniations: A Systematic Review and Meta\u2010Analysis","volume":"9","author":"Zhang J.","year":"2022","journal-title":"Frontiers in Surgery"},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2017-018230"},{"key":"e_1_2_10_8_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0316-z"},{"key":"e_1_2_10_9_1","doi-asserted-by":"crossref","DOI":"10.3389\/fpubh.2026.1788454","article-title":"The Role of Multimodality in Clinical Disease Diagnosis: Advances, Challenges, and Opportunities","volume":"14","author":"Nawaz A.","year":"2026","journal-title":"Frontiers in Public Health"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.5435\/JAAOS-D-23-00831"},{"key":"e_1_2_10_11_1","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab569","article-title":"Multimodal Deep Learning for Biomedical Data Fusion: A Review","volume":"23","author":"Stahlschmidt S. R.","year":"2022","journal-title":"Briefings in Bioinformatics"},{"key":"e_1_2_10_12_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.jbi.2023.104482","article-title":"A Scoping Review on Multimodal Deep Learning in Biomedical Images and Texts","volume":"146","author":"Sun Z.","year":"2023","journal-title":"Journal of Biomedical Informatics"},{"key":"e_1_2_10_13_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.compbiomed.2024.108635","article-title":"A Review of Deep Learning\u2010Based Information Fusion Techniques for Multimodal Medical Image Classification","volume":"176","author":"Li Y.","year":"2024","journal-title":"Computers in Biology and Medicine"},{"key":"e_1_2_10_14_1","doi-asserted-by":"crossref","first-page":"503","DOI":"10.1038\/s41746-025-01903-9","article-title":"Multimodal Deep Learning Model for Prognostic Prediction in Cervical Cancer Receiving Definitive Radiotherapy: A Multi\u2010Center Study","volume":"8","author":"Wang W.","year":"2025","journal-title":"npj Digital Medicine"},{"key":"e_1_2_10_15_1","doi-asserted-by":"crossref","first-page":"933","DOI":"10.1186\/s12967-025-06956-8","article-title":"Multimodal Prediction of Tyrosine Kinase Inhibitors Therapy Outcomes in Advanced EGFR\u2010Mutated NSCLC Patients","volume":"23","author":"Chai X.","year":"2025","journal-title":"Journal of Translational Medicine"},{"key":"e_1_2_10_16_1","doi-asserted-by":"crossref","first-page":"9613","DOI":"10.1038\/s41467-024-53450-8","article-title":"An Explainable Longitudinal Multi\u2010Modal Fusion Model for Predicting Neoadjuvant Therapy Response in Women With Breast Cancer","volume":"15","author":"Gao Y.","year":"2024","journal-title":"Nature Communications"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1097\/BRS.0000000000005044"},{"key":"e_1_2_10_18_1","doi-asserted-by":"crossref","first-page":"1221","DOI":"10.1016\/j.spinee.2016.06.010","article-title":"Determination of the Oswestry Disability Index Score Equivalent to a \u201cSatisfactory Symptom State\u201d in Patients Undergoing Surgery for Degenerative Disorders of the Lumbar Spine\u2014A Spine Tango Registry\u2010Based Study","volume":"16","author":"Hooff M. L.","year":"2016","journal-title":"Spine Journal"},{"key":"e_1_2_10_19_1","doi-asserted-by":"publisher","DOI":"10.1097\/BRS.0000000000004479"},{"key":"e_1_2_10_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.spinee.2020.11.008"},{"key":"e_1_2_10_21_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00586-023-07982-0"},{"key":"e_1_2_10_22_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.29403"},{"key":"e_1_2_10_23_1","doi-asserted-by":"publisher","DOI":"10.1093\/biostatistics\/kxj037"},{"key":"e_1_2_10_24_1","doi-asserted-by":"crossref","first-page":"1087","DOI":"10.1007\/s00586-009-1274-4","article-title":"MSU Classification for Herniated Lumbar Discs on MRI: Toward Developing Objective Criteria for Surgical Selection","volume":"19","author":"Mysliwiec L. W.","year":"2010","journal-title":"European Spine Journal"},{"key":"e_1_2_10_25_1","doi-asserted-by":"publisher","DOI":"10.1002\/jsp2.70003"},{"key":"e_1_2_10_26_1","first-page":"4768","article-title":"A Unified Approach to Interpreting Model Predictions","volume":"30","author":"Lundberg S. M.","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_10_27_1","article-title":"Personalized Treatment Recommendation System in Head and Neck Cancer Using Survival Analysis and Deep Learning","volume":"14","author":"Fei X.","year":"2026","journal-title":"Healthcare (Basel, Switzerland)"},{"key":"e_1_2_10_28_1","doi-asserted-by":"crossref","first-page":"475","DOI":"10.1007\/978-3-642-01307-2_43","volume-title":"Proceedings of the 13th Pacific\u2010Asia Conference on Knowledge Discovery and Data Mining","author":"Bunkhumpornpat C.","year":"2009"},{"key":"e_1_2_10_29_1","first-page":"785","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA","author":"Chen T.","year":"2016"},{"key":"e_1_2_10_30_1","first-page":"2623","volume-title":"In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, AK","author":"Akiba T.","year":"2019"},{"key":"e_1_2_10_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2008.239"},{"key":"e_1_2_10_32_1","first-page":"5998","article-title":"Attention Is All You Need","volume":"30","author":"Vaswani A.","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_10_33_1","doi-asserted-by":"crossref","DOI":"10.1007\/s44196-023-00397-1","article-title":"Transformative Breast Cancer Diagnosis Using CNNs With Optimized ReduceLROnPlateau and Early Stopping Enhancements","volume":"17","author":"Mahesh T. R.","year":"2024","journal-title":"International Journal of Computational Intelligence Systems"},{"key":"e_1_2_10_34_1","first-page":"689","volume-title":"Proceedings of the 28th International Conference on Machine Learning (ICML)","author":"Ngiam J.","year":"2011"},{"key":"e_1_2_10_35_1","article-title":"Layer Normalization","author":"Ba J. L.","year":"2016","journal-title":"arXiv Preprint"},{"key":"e_1_2_10_36_1","doi-asserted-by":"crossref","DOI":"10.1080\/07853890.2023.2258922","article-title":"The Association Between Morphological Characteristics of Paraspinal Muscle and Spinal Disorders","volume":"55","author":"Suo M.","year":"2023","journal-title":"Annals of Medicine"},{"key":"e_1_2_10_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.spinee.2024.03.016"},{"key":"e_1_2_10_38_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.wneu.2024.07.074"},{"key":"e_1_2_10_39_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.wneu.2017.03.097"},{"key":"e_1_2_10_40_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.230614"},{"key":"e_1_2_10_41_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00586-021-07020-x"},{"key":"e_1_2_10_42_1","doi-asserted-by":"publisher","DOI":"10.1097\/JS9.0000000000001222"},{"key":"e_1_2_10_43_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41392-024-01932-y"},{"key":"e_1_2_10_44_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.29639"}],"container-title":["Journal of Magnetic Resonance Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/jmri.70518","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/jmri.70518","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/jmri.70518","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T02:55:13Z","timestamp":1787194513000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/jmri.70518"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,19]]},"references-count":43,"alternative-id":["10.1002\/jmri.70518"],"URL":"https:\/\/doi.org\/10.1002\/jmri.70518","archive":["Portico"],"relation":{},"ISSN":["1053-1807","1522-2586"],"issn-type":[{"value":"1053-1807","type":"print"},{"value":"1522-2586","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,19]]},"assertion":[{"value":"2026-04-16","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-13","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-19","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"jmri.70518"}}