Evaluation of GBM prognosis based on different tumor components and functional MRI.
Motivation: Glioblastomas with IDH wildtype, being highly aggressive and heterogeneous brain tumors, pose significant challenges to accurate prognosis. Goal(s): We aim to explore the value of radiomics features from different modalities in identifying the different prognosis group of GBM. Approach: Based on overall survival (OS), GBM is categorized into long and short survival groups. Radiomics features are extracted from five tumor areas in each subject using seven MRI modalities for prognosis assessment. Results: The AUC for the prognosis assessment based on the features of the core area in the ADC sequence is 0.758 (0.787 for testing). Impact: Our study further indicates that the ADC value, derived from DWI, along with the core area of the tumor, offers crucial information for evaluating the prediction of survival in GBM patients.
Authors
- Xiaorui Su (ORCID: https://orcid.org/0000-0001-5468-6085)
- Qiang Yue (ORCID: https://orcid.org/0000-0001-7697-060X)
Publication Details
- Journal
- Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition
- Published
- 2025-09-16
- DOI
- https://doi.org/10.58530/2025/5024
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 0.00