Distinction of pseudoprogression from true progression in glioblastomas using machine learning based on multiparametric magnetic resonance imaging and O6-methylguanine-methyltransferase promoter methylation status
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Distinguishing true progression from pseudoprogression in glioblastoma (GBM) is crucial. Machine learning with multiparametric MRI shows promise for accurate differentiation, aiding treatment decisions.
Area Of Science
- Radiology
- Oncology
- Artificial Intelligence
Background
- Differentiating true progression (TP) from pseudoprogression (PsP) is critical for glioblastoma (GBM) patient management.
- Standard treatment for GBM can lead to imaging changes that mimic tumor recurrence.
Purpose Of The Study
- To investigate the utility of quantitative diffusion and perfusion MRI parameters combined with machine learning for distinguishing TP from PsP in GBM.
- To develop a robust prediction model for accurate classification.
Main Methods
- 75 GBM patients with contrast-enhancing lesions underwent 3T MRI.
- Quantitative MRI parameters were extracted, and machine learning algorithms (random forest, SVM) were applied.
- Data were split for training and testing, with cross-validation and ROC analysis for performance evaluation.
Main Results
- A quadratic support vector machine achieved high accuracy (91% training, 86% cross-validation, 85% testing).
- Receiver operating characteristic (ROC) analysis demonstrated 85% accuracy, 70% sensitivity, and 100% specificity.
Conclusions
- Quantitative multiparametric MRI coupled with machine learning offers a promising non-invasive approach for distinguishing TP from PsP in GBM.
- This method can potentially improve treatment strategies and patient outcomes.
Contact us if these videos are not relevant.
Contact us if these videos are not relevant.
