Related Experiment Video
Updated: Aug 25, 2025
Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning
Published on: April 12, 2024
Radiomics features based on MRI predict BRAF V600E mutation in pediatric low-grade gliomas: A non-invasive method for
Jiali Xu1, Mingyao Lai2, Shaoqun Li2
1Department of Medical Imaging Centre, the First Affiliated Hospital, Jinan University, Guangzhou, China; Department of Radiology, The First Affiliated Hospital of Bengbu Medical College, Bengbu, Anhui, China.
Objective:
To investigate the clinical application value of radiomics features based on preoperative magnetic resonance imaging for predicting B-Raf proto-oncogene serine/threonine-protein (BRAF) V600E mutation in pediatric low-grade gliomas.
Materials And Methods:
The clinical, imaging, and pathological data from 113 pediatric patients with low-grade gliomas patients were retrospectively analyzed. Using open-source software, three-dimensional imaging features were extracted on the basis of FLAIR sequences, and the radiomics process was analyzed to dichotomize BRAFV600E mutant and wild type. All cases were randomly divided into the training and test sets according to a 7:3 training and test group ratio, and a 5-fold cross-validation was performed on the training set. The optimal hyperparameters were selected to build the prediction model, and the test set was used for external validation to assess the diagnostic value of the model using the receiver operating characteristic curve.
Results:
The training set comprised 79 patients (47 males, 32 females, mean age 9.86 ± 5.20) and the test set comprised 34 patients (20 males, 14 females, mean age 10.97 ± 5.14). Sex, age, and brain side were not significant predictors of BRAF, and tumor location on the supratentorial region was a BRAF predictor (p < 0.05). The radiomics model constructed by principal component analysis for dimensionality reduction, Kruskal-Wallis for filtering of features, and random forest as a classifier performed best. In the training set, the mean area under the curve (AUC) with a five-fold cross-validation was 0.72 ( ± 0.057; 95 % confidence interval (CI), 0.602-0.831) and AUC of the test set was 0.875 ( ± 0.062; 95 % CI, 0.731-0.983).
Conclusion:
The use of a radiomics model based on FLAIR sequences can help predict BRAF V600E mutations in pediatric low-grade gliomas.
