Comparison of MRI radiomics-based machine learning survival models in predicting prognosis of glioblastoma multiforme

Aug 4, 2026

This study compares radiomics-based machine learning survival models for predicting overall survival in glioblastoma multiforme (GBM) patients. A total of 131 GBM cases were analyzed, with five models evaluated using the concordance index (C-index): the Cox proportional-hazards model and four machine learning approaches (SurvivalTree, Random Survival Forest, DeepSurv, and DeepHit). From 1,792 radiomics features, seven prognosis-associated features were selected via LASSO regression. DeepSurv achieved the highest predictive performance, confirming deep learning radiomics models outperform the traditional method.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Objective

Methods

Study cohort

Models and evaluation metric

Results

Feature screening and selection

Model performance comparison

Conclusion

Keywords

TYPE Original Research PUBLISHED 30 November 2023 DOI 10.3389/fmed.2023.1271687 OPEN ACCESS EDITED BY Gongning Luo, Harbin Institute of Technology, China REVIEWED BY Suyu Dong, Northeast Forestry University, China Shaodong Cao, The Fourth Hospital of Harbin Medical University, China *CORRESPONDENCE Guolin Ma [maguolin1007@qq.com](maguolin1007@qq.com)[ ](maguolin1007@qq.com)Chuanchen Zhang [zhangchuanchen666@163.com](zhangchuanchen666@163.com)[ ](zhangchuanchen666@163.com)†These authors share first authorship RECEIVED 02 August 2023 ACCEPTED 15 November 2023 PUBLISHED 30 November 2023 CITATION Zhang D, Luan J, Liu B, Yang A, Lv K, Hu P, Han X, Yu H, Shmuel A, Ma G and Zhang C (2023) Comparison of MRI radiomicsbased machine learning survival models in predicting prognosis of glioblastoma multiforme. Front. Med. 10:1271687. doi: 10.3389/fmed.2023.1271687 Comparison of MRI radiomics-based machine learning survival models in predicting prognosis of glioblastoma multiforme Di Zhang 1†, Jixin Luan 2, 3†, Bing Liu 2, 3, Aocai Yang 2, 3, Kuan Lv4, Pianpian Hu4, Xiaowei Han 5, Hongwei Yu3, Amir Shmuel 6, 7, Guolin Ma3* and Chuanchen Zhang 1* 1 Department of Radiology, Liaocheng People’s Hospital, Shandong First Medical University & Shandong Academy of Medical Sciences, Liaocheng, Shandong, China, 2China-Japan Friendship Hospital (Institute of Clinical Medical Sciences), Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing, China, 3 Department of Radiology, China-Japan Friendship Hospital, Beijing, China, 4 Peking University China-Japan Friendship School of Clinical Medicine, Beijing, China, 5 Department of Radiology, The Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, China, 6 McConnell Brain Imaging Centre, Montreal Neurological Institute, McGill University, Montreal, QC, Canada, 7 Department of Neurology and Neurosurgery, McGill University, Montreal, QC, Canada Objective: To compare the performance of radiomics-based machine learning survival models in predicting the prognosis of glioblastoma multiforme (GBM) patients. COPYRIGHT © 2023 Zhang, Luan, Liu, Yang, Lv, Hu, Han, Yu, Shmuel, Ma and Zhang. 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) and the copyright owner(s) 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. Methods: 131 GBM patients were included in our study. The traditional Cox proportional-hazards (CoxPH) model and four machine learning models (SurvivalTree, Random survival forest (RSF), DeepSurv, DeepHit) were constructed, and the performance of the five models was evaluated using the C-index. Results: After the screening, 1792 radiomics features were obtained. Seven radiomics features with the strongest relationship with prognosis were obtained following the application of the least absolute shrinkage and selection operator (LASSO) regression. The CoxPH model demonstrated that age (HR = 1. 576, p = 0.037), Karnofsky performance status (KPS) score (HR = 1. 890, p = 0.006), radiomics risk score (HR = 3.497, p = 0.001), and radiomics risk level (HR = 1. 572, p = 0 .043) were associated with poorer prognosis. The DeepSurv model performed the best among the five models, obtaining C-index of 0.882 and 0.732 for the training and test set, respectively. The performances of the other four models were lower: CoxPH (0 .663 training set / 0.635 test set), SurvivalTree (0 .702/0 .655), RSF (0 .735/0 .667), DeepHit (0 .608/0 . 560) . Conclusion: This study confirmed the superior performance of deep learning algorithms based on radiomics relative to the traditional method in predicting the overall surv

FAQs

What is the main objective of this study?

To compare the performance of radiomics-based machine learning survival models in predicting the prognosis of glioblastoma multiforme patients.

Which survival models were compared?

The study compared the Cox proportional-hazards model with four machine learning models: SurvivalTree, Random Survival Forest (RSF), DeepSurv, and DeepHit.

How were radiomics features selected and evaluated?

A total of 1,792 radiomics features were screened, and seven prognosis-related features were selected using LASSO regression. Model performance was assessed using the C-index on training and test sets.

Aug 4, 2026

Upload by Ava Thompson