The final, formatted version of the article will be published soon.
ORIGINAL RESEARCH article
Front. Oncol.
Sec. Radiation Oncology
Volume 14 - 2024 |
doi: 10.3389/fonc.2024.1399270
Preoperative MRI-Based Radiomic Nomogram for Distinguishing Solitary Fibrous Tumor from Angiomatous Meningioma: A MulticenterTwo-Center Study
Provisionally accepted- 1 The Affiliated Hospital of Qingdao University, Qingdao, Shandong Province, China
- 2 Shizuishan First People's Hospital, Shizuishan, Ningxia, China
- 3 Qilu Hospital, Shandong University, Jinan, Shandong Province, China
Purpose: This study evaluates the efficacy of radiomics-based machine learning methodologies in differentiating solitary fibrous tumor (SFT) from angiomatous meningioma (AM).A retrospective analysis was conducted on 171 pathologically confirmed cases (94 SFT and 77 AM) spanning from January 2009 to September 2020 across four institutions.The study comprised a training set (n=137) and a validation set (n=34). All patients underwent contrast-enhanced T1-weighted (CE-T1WIl) and T2-weighted(T2WI) MRI scans, from which 1166 radiomics features were extracted. Subsequently, seventeen features were selected through minimum redundancy maximum relevance (mRMR) and the least absolute shrinkage and selection operator (LASSO). Multivariate logistic regression analysis was employed to assess the independence of these features as predictors. A clinical model, established via both univariate and multivariate logistic regression based on MRI morphological features, was integrated with the optimal radiomics model to formulate a radiomics nomogram. The performance of the models was assessed utilizing the area under the receiver operating characteristic curve (AUC), accuracy (ACC), sensitivity (SEN), specificity (SPE), positive predictive value (PPV), and negative predictive value (NPV).The radiomics nomogram demonstrated exceptional discriminative performance in the validation set, achieving an AUC of 0.989. This outperformance was evident when compared to both the radiomics algorithm (AUC= 0.968) and the clinical model (AUC = 0.911) in the same validation sets. Notably, the radiomics nomogram exhibited impressive values for ACC, SEN, and 2 / 28 SPE at 97.1%, 93.3%, and 100%, respectively, in the validation set.The machine learning-based radiomic nomogram proves to be highly effective in distinguishing between SFT and AM.
Keywords: Solitary fibrous tumor, Angiomatous meningioma, Radiomics, nomogram, machine learning
Received: 11 Mar 2024; Accepted: 27 Aug 2024.
Copyright: © 2024 Li, Fu, DU, Han, Duan, Ren, Qiao and Tang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Yande Ren, The Affiliated Hospital of Qingdao University, Qingdao, 266000, Shandong Province, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.