{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T19:04:14Z","timestamp":1782846254447,"version":"3.54.5"},"reference-count":29,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Thyroid carcinoma is the most prevalent endocrine malignancy, with a worldwide increasing incidence. Capsular invasion and neural invasion (NI) are pivotal prognostic factors for recurrence and survival; however, their preoperative noninvasive assessment remains challenging.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n We aimed to identify computed tomography (CT) radiomic biomarkers associated with capsular invasion in thyroid carcinoma, construct machine learning models integrating radiomic and clinical data, and evaluate their utility for NI risk stratification.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n In this retrospective cohort, 111 patients with thyroid carcinoma were divided into capsular invasion\u2013positive (n=63) and capsular invasion\u2013negative (n=48) groups, with 37 (33.3%) cases presenting concurrent NI. Radiomic features were extracted from arterial and venous phase CT images at original resolution, including 111 gray-level co-occurrence matrix features. Nine key radiomic features (A1-A9) were selected via least absolute shrinkage and selection operator regression (\u03bb=0.017). To preserve the physical meaning of texture features (eg, spatial correlation and contrast reflecting tumor microstructural heterogeneity), no resampling or scaling was performed on the regions of interest during radiomic feature extraction. Nomogram models and random forest (RF) models were constructed based on clinical indicators (galectin-3, etc) and radiomic features, respectively. Additionally, a neural network (NN) model integrating multimodal data was developed. Model stability was verified using 5-fold cross-validation and 1000-time bootstrap resampling, while performance was evaluated via receiver operating characteristic curves, calibration curves, and decision curve analysis.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n Model performance analysis revealed that among the nomogram models, the clinical indicator-based nomogram achieved an internally estimated area under the curve (AUC) of 0.9418 (95% CI 0.892\u20100.976) in the capsular invasion prediction task. The radiomic-based nomogram had an internally estimated AUC of 0.9334 (95% CI 0.881\u20100.968) in the capsular invasion prediction task and 0.8001 (95% CI 0.663\u20100.898) in the cross-label association analysis task. In RF models, clinical indicator-based and radiomic-based RFs exhibited an AUC of 0.7646 (95% CI 0.651\u20100.857) and 0.8102 (95% CI 0.703\u20100.892) in the cross-label association analysis task, respectively. The NN model performed promisingly, with an AUC of 0.775 (95% CI 0.621\u20100.903) in the cross-label association analysis task and a mean absolute error of <0.05 on the calibration curve.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Capsular invasion is a strong predictor of NI risk in thyroid carcinoma. Radiomic models based solely on preoperative CT images show potential for preoperative NI risk stratification. Models incorporating clinical parameters (obtained from postoperative tissue), including the integrated multimodal model, are more accurately characterized as postoperative risk stratification tools. The NN model, which integrated raw CT images with clinical data, achieved an AUC of 0.775 (95% CI 0.621\u20100.903), underscoring the potential of such multimodal analysis to capture complex relationships between imaging phenotypes and tissue-level biomarkers for enhanced postoperative assessment. This framework\u2019s radiomic component points toward purely image-based, preoperative evaluation tools\u2019 development.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/77349","type":"journal-article","created":{"date-parts":[[2026,3,12]],"date-time":"2026-03-12T21:21:28Z","timestamp":1773350488000},"page":"e77349-e77349","source":"Crossref","is-referenced-by-count":0,"title":["CT Radiomics\u2013Based Machine Learning Model for Predicting Capsular and Neural Invasion in Thyroid Carcinoma: Diagnostic Accuracy Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-3874-5664","authenticated-orcid":false,"given":"Fang-fang","family":"Cong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-8660-1952","authenticated-orcid":false,"given":"Ke","family":"Tian","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-5334-1593","authenticated-orcid":false,"given":"Qian","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-6204-8196","authenticated-orcid":false,"given":"Fulin","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-1484-2660","authenticated-orcid":false,"given":"Peng","family":"Sun","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-5777-9738","authenticated-orcid":false,"given":"Nan","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,3,12]]},"reference":[{"issue":"1","key":"R1","doi-asserted-by":"publisher","first-page":"104","DOI":"10.1016\/j.yexcr.2018.11.012","article-title":"Side population cells in anaplastic thyroid cancer and normal thyroid","volume":"374","author":"Mahkamova","journal-title":"Exp Cell Res"},{"issue":"2","key":"R2","doi-asserted-by":"publisher","first-page":"359","DOI":"10.1007\/s12020-022-02990-4","article-title":"Impact of transforming growth factor beta 1 on normal and thyroid cancer side population cells","volume":"76","author":"Latar","journal-title":"Endocrine"},{"issue":"7","key":"R3","doi-asserted-by":"publisher","first-page":"3112","DOI":"10.1007\/s12325-020-01391-1","article-title":"Papillary thyroid cancer-aggressive variants and impact on management: a narrative review","volume":"37","author":"Coca-Pelaz","journal-title":"Adv Ther"},{"issue":"2","key":"R4","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1007\/s12022-018-9532-9","article-title":"Immunohistochemical biomarkers in thyroid pathology","volume":"29","author":"Baloch","journal-title":"Endocr Pathol"},{"issue":"7","key":"R5","doi-asserted-by":"publisher","first-page":"5004","DOI":"10.1007\/s00330-022-08548-2","article-title":"DCE-MRI based radiomics nomogram for preoperatively differentiating combined hepatocellular-cholangiocarcinoma from mass-forming intrahepatic cholangiocarcinoma","volume":"32","author":"Zhou","journal-title":"Eur Radiol"},{"key":"R6","doi-asserted-by":"publisher","DOI":"10.2196\/78306","article-title":"Diagnostic performance of computed tomography-based artificial intelligence for early recurrence of cholangiocarcinoma: systematic review and meta-analysis","volume":"27","author":"Chen","journal-title":"J Med Internet Res"},{"issue":"4","key":"R7","doi-asserted-by":"publisher","first-page":"432","DOI":"10.21037\/tcr.2016.06.20","article-title":"Prostate cancer radiomics and the promise of radiogenomics","volume":"5","author":"Stoyanova","journal-title":"Transl Cancer Res"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.2196\/56851","article-title":"Development and validation of a computed tomography\u2013based model for noninvasive prediction of the T stage in gastric cancer: multicenter retrospective study","volume":"26","author":"Tao","journal-title":"J Med Internet Res"},{"key":"R9","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2017.12.004","article-title":"Texture analysis and machine learning to characterize suspected thyroid nodules and differentiated thyroid cancer: where do we stand?","volume":"99","author":"Sollini","journal-title":"Eur J Radiol"},{"issue":"11","key":"R10","doi-asserted-by":"publisher","first-page":"2902","DOI":"10.1007\/s12094-024-03736-6","article-title":"SEOM-GETNE-TTCC clinical guideline thyroid cancer (2023)","volume":"26","author":"Alonso-Gordoa","journal-title":"Clin Transl Oncol"},{"issue":"1","key":"R11","doi-asserted-by":"publisher","DOI":"10.1186\/s12880-021-00553-z","article-title":"Radiomics based on multiparametric MRI for extrathyroidal extension feature prediction in papillary thyroid cancer","volume":"21","author":"Wei","journal-title":"BMC Med Imaging"},{"issue":"5","key":"R12","doi-asserted-by":"publisher","first-page":"425","DOI":"10.1001\/jama.2023.26348","article-title":"Thyroid cancer: a review","volume":"331","author":"Boucai","journal-title":"JAMA"},{"issue":"4","key":"R13","doi-asserted-by":"publisher","first-page":"761","DOI":"10.1016\/j.ecl.2022.04.002","article-title":"The patient experience of thyroid cancer","volume":"51","author":"Pitt","journal-title":"Endocrinol Metab Clin North Am"},{"issue":"5","key":"R14","doi-asserted-by":"publisher","first-page":"984","DOI":"10.1002\/ijc.33690","article-title":"Thyroid cancer, recent advances in diagnosis and therapy","volume":"149","author":"Nabhan","journal-title":"Int J Cancer"},{"issue":"2","key":"R15","doi-asserted-by":"publisher","first-page":"185","DOI":"10.1002\/path.3031","article-title":"Interstitial guidance of cancer invasion","volume":"226","author":"Gritsenko","journal-title":"J Pathol"},{"key":"R16","doi-asserted-by":"publisher","DOI":"10.1016\/0167-8140(91)90181-f","article-title":"Measurement of tumor hypoxia by invasive and non-invasive procedures: a review of recent clinical studies","volume":"20 Suppl 1","author":"Chapman","journal-title":"Radiother Oncol"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","DOI":"10.1186\/s43046-025-00277-z","article-title":"Advances in colorectal cancer screening and detection: a narrative review on biomarkers, imaging and preventive strategies","volume":"37","author":"Khan","journal-title":"J Egypt Natl Canc Inst"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.2196\/72420","article-title":"Effectiveness of radiomics-based machine learning models in differentiating pancreatitis and pancreatic ductal adenocarcinoma: systematic review and meta-analysis","volume":"27","author":"Zhang","journal-title":"J Med Internet Res"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2018.10.016","article-title":"A computer-aided diagnosis scheme of breast lesion classification using GLGLM and shape features: combined-view and multi-classifiers","volume":"55","author":"Liang","journal-title":"Phys Med"},{"issue":"8","key":"R20","doi-asserted-by":"publisher","first-page":"1626","DOI":"10.1016\/j.molonc.2015.05.001","article-title":"Nerve fibers infiltrate the tumor microenvironment and are associated with nerve growth factor production and lymph node invasion in breast cancer","volume":"9","author":"Pundavela","journal-title":"Mol Oncol"},{"key":"R21","doi-asserted-by":"publisher","DOI":"10.1016\/j.nicl.2016.03.007","article-title":"Population-based MRI atlases of spatial distribution are specific to patient and tumor characteristics in glioblastoma","volume":"12","author":"Bilello","journal-title":"Neuroimage Clin"},{"issue":"1155","key":"R22","doi-asserted-by":"publisher","first-page":"526","DOI":"10.1093\/bjr\/tqae010","article-title":"Prediction of cervical lymph node metastasis in differentiated thyroid cancer based on radiomics models","volume":"97","author":"Mu","journal-title":"Br J Radiol"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","first-page":"362","DOI":"10.1007\/s12032-012-0362-1","article-title":"Cytokeratin19 expression discriminates papillary thyroid carcinoma from other thyroid lesions and predicts its aggressive behavior","volume":"30","author":"Dencic","journal-title":"Med Oncol"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.18286","article-title":"Expression of CK7, CK19 and p16 in HPV-mediated oropharyngeal squamous cell carcinoma","volume":"12","author":"Yan","journal-title":"PeerJ"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","first-page":"61","DOI":"10.4103\/npmj.npmj_3_25","article-title":"Cross-sectional study to determine the diagnostic accuracy of cytokeratin-19 immunomarker in the diagnosis of thyroid malignancy with histopathology as gold standard","volume":"32","author":"Laishram","journal-title":"Niger Postgrad Med J"},{"issue":"5","key":"R26","doi-asserted-by":"publisher","DOI":"10.1002\/jcla.24341","article-title":"A combination of circulating tumor cells and CA199 improves the diagnosis of pancreatic cancer","volume":"36","author":"Chen","journal-title":"J Clin Lab Anal"},{"issue":"6","key":"R27","doi-asserted-by":"publisher","first-page":"565","DOI":"10.15537\/smj.2024.45.6.20230878","article-title":"Prevalence of CEA, CA 125, and CA 15-3 serum tumour markers in different regions of Saudi Arabia","volume":"45","author":"Ashi","journal-title":"Saudi Med J"},{"issue":"1","key":"R28","first-page":"1","volume":"9","author":"Chen","journal-title":"Am J Cancer Res"},{"key":"R29","doi-asserted-by":"publisher","DOI":"10.2196\/37599","article-title":"Application of virtual and augmented reality technology in hip surgery: systematic review","volume":"25","author":"Sun","journal-title":"J Med Internet Res"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e77349","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:09:29Z","timestamp":1782842969000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e77349"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,12]]},"references-count":29,"URL":"https:\/\/doi.org\/10.2196\/77349","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,12]]},"article-number":"v14i1e77349"}}