AnMRI‐Based Radiomic Nomogram for Discrimination Between Malignant and Benign Sinonasal Tumors

Journal of Magnetic Resonance Imaging - Tập 53 Số 1 - Trang 141-151 - 2021
Han Zhang1,2, Hexiang Wang1,2, Dapeng Hao1, Yaqiong Ge3, Guangyao Wan1, Jun Zhang1, Shunli Liu1, Yu Zhang1, Deguang Xu4
1The Department of Radiology, the Affiliated Hospital of Qingdao University, Qingdao, China
2these authors contributed equally to this work
3GE Healthcare, Shanghai, China
4Huangdao Hospital of Traditional Chinese Medicine, Qingdao, China

Tóm tắt

BackgroundPreoperative discrimination between malignant and benign sinonasal tumors is important for treatment plan selection.PurposeTo build and validate a radiomic nomogram for preoperative discrimination between malignant and benign sinonasal tumors.Study TypeRetrospective.PopulationIn all, 197 patients with histopathologically confirmed 84 benign and 113 malignant sinonasal tumors.Field Strength/SequencesFast‐spin‐echo (FSE) T1‐weighted and fat‐suppressed FSE T2‐weighted imaging on a 1.5T and 3.0T MRI.AssessmentT1and fat‐suppressed T2‐weighted images were selected for feature extraction. The least absolute shrinkage selection operator (LASSO) algorithm was applied to establish a radiomic score. Multivariate logistic regression analysis was applied to determine independent risk factors, and the radiomic score was combined to build a radiomic nomogram. The nomogram was assessed in a training dataset (n =138/3.0T MRI) and tested in a validation dataset (n =59/1.5T MRI).Statistical TestsIndependentt‐test or Wilcoxon's test, chi‐square‐test, or Fisher's‐test, univariate analysis, LASSO, multivariate logistic regression analysis, area under the curve (AUC), Hosmer–Lemeshow test, decision curve, and the Delong test.ResultsIn the validation dataset, the radiomic nomogram could differentiate benign from malignant sinonasal tumors with an AUC of 0.91. There was no significant difference in AUC between the combined radiomic score and radiomic nomogram (P > 0.05), and the radiomic nomogram showed a relatively higher AUC than the combined radiomic score. There was a significant difference in AUC between each two of the following models (the radiomic nomogram vs. the clinical model, allP < 0.001; the combined radiomic score vs. the clinical model,P= 0.0252 and 0.0035, respectively, in the training and validation datasets). The radiomic nomogram outperformed the radiomic scores and clinical model.Data ConclusionThe radiomic nomogram combining the clinical model and radiomic score is a simple, effective, and reliable method for patient risk stratification.Level of Evidence4Technical Efficacy Stage2

Từ khóa


Tài liệu tham khảo

10.1007/s11912-013-0369-6

EggesboHB.Imaging of sinonasal tumours. Cancer Imaging.2012;12:136‐152.

10.1002/(SICI)1097-0347(200003)22:2<164::AID-HED8>3.0.CO;2-#

10.1053/j.ro.2019.03.001

10.3174/ajnr.A2675

10.1053/j.sult.2008.10.013

10.2500/ajra.2013.27.3894

10.1016/j.otohns.2010.05.019

10.4103/0366-6999.151649

10.2214/AJR.12.9432

10.1038/nrclinonc.2012.2

10.3174/ajnr.A1817

10.1016/j.neurad.2009.06.001

10.3390/cancers5030875

10.1055/s-0036-1597323

10.1016/j.mri.2012.06.010

10.1038/nrclinonc.2017.141

10.1148/radiol.2015151169

10.1007/s13244-012-0196-6

10.1158/1078-0432.CCR-14-0990

10.1007/s12105-017-0791-4

10.1016/j.neuroimage.2006.01.015

10.1158/1078-0432.CCR-15-2997

10.21147/j.issn.1000-9604.2019.04.08

10.1002/sim.3148

10.1200/JCO.2007.12.9791

10.3174/ajnr.A5532

10.1002/jmri.25500

Xian J, 2014, Feasibility and value of quantitative dynamic contrast enhancement MR imaging in the evaluation of sinonasal tumors, Chin Med J (Engl), 127, 2259, 10.3760/cma.j.issn.0366-6999.20140712

10.1097/RCT.0000000000000502

10.1016/j.oooo.2018.07.004

10.1016/j.ejrad.2019.07.006

Thông tin
Thông tin xuất bản

Journal of Magnetic Resonance Imaging

Tập 53 Số 1

141-151

Thông tin tác giả