Neuroimaging: BrainA Comparative and Summative Study of Radiomics-based Overall Survival Prediction in Glioblastoma PatientsRuan, Zhuoying MD∗; Mei, Nan MD†; Lu, Yiping MD†; Xiong, Ji MD†; Li, Xuanxuan MD†; Zheng, Weiwei MD‡; Liu, Li MD§; Yin, Bo MD† Author Information From the ∗Shanghai Institute of Medical Imaging †Department of Radiology, Huashan Hospital ‡Department of Environmental Health, School of Public Health §Department of Radiology, Shanghai Cancer Center, Fudan University, Shanghai, China. Received for publication June 5, 2021; accepted October 20, 2021. Z.R., N.M., Y.L., and J.X. contributed equally to the article. They should be regarded as the co-first authors. Correspondence to: Bo Yin, MD, Department of Radiology, Huashan Hospital, Fudan University, 12, Middle Wulumuqi Rd, Jing'an District, Shanghai 200040, China (e-mail: [email protected]); Li Liu, MD, Department of Radiology, Shanghai Cancer Center, Fudan University, 270 Dong'an Rd, Shanghai 200000, China (e-mail: [email protected]). The authors declare no conflict of interest. This work was supported by the Clinical Research Plan of SHDC (grant no. SHDC2020CR4069), the Youth Program of National Natural Science Foundation of China (grant no. 81901697), Shanghai Sailing Program (grant no. 21YF1404800), the Youth Medical Talents–Medical Imaging Practitioner Program (grant no. 2020SHWRS087), the Shanghai Municipal Science and Technology Major Project (no. 2018SHZDZX01), and ZJ Lab. Supplemental digital contents are available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.jcat.org). Journal of Computer Assisted Tomography: 5/6 2022 - Volume 46 - Issue 3 - p 470-479 doi: 10.1097/RCT.0000000000001300 Buy SDC Metrics Abstract Purpose This study aimed to assess different machine learning models based on radiomic features, Visually Accessible Rembrandt Images features and clinical characteristics in overall survival prediction of glioblastoma and to identify the reproducible features. Materials and Methods Patients with preoperative magnetic resonance scans were allocated into 3 data sets. The Least Absolute Shrinkage and Selection Operator was used for feature selection. The prediction models were built by random survival forest (RSF) and Cox regression. C-index and integrated Brier scores were calculated to compare model performances. Results Patients with cortical involvement had shorter survival times in the training set (P = 0.006). Random survival forest showed higher C-index than Cox, and the RSF model based on the radiomic features was the best one (testing set: C-index = 0.935 ± 0.023). Ten reproducible radiomic features were summarized. Conclusions The RSF model based on radiomic features had promising potential in predicting overall survival of glioblastoma. Ten reproducible features were identified. Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.