Preoperative phenotypic stratification of primary central nervous system lymphoma using multiparametric MRI-based radiomics: prediction of germinal center B-cell-like and double-expression status - Summary - MDSpire
Advertisement
Preoperative phenotypic stratification of primary central nervous system lymphoma using multiparametric MRI-based radiomics: prediction of germinal center B-cell-like and double-expression status
To develop and evaluate multiparametric MRI-based radiomics models for preoperative prediction of double-expression lymphoma (DEL) and germinal center B-cell-like (GCB) status in primary central nervous system lymphoma (PCNSL).
Approach:
Study Design: Retrospective study including 160 pathologically confirmed PCNSL patients.
MRI Analysis: Multiparametric MRI sequences analyzed, including T2-weighted, T2-FLAIR, contrast-enhanced T1-weighted, and apparent diffusion coefficient.
Feature Extraction: Radiomics features extracted from enhancing tumor core and peritumoral edema using nnU-NetV2-based models.
Model Training: Six machine-learning classifiers trained and evaluated using ROC and decision curve analysis.
Key Findings:
For DEL classification, the SVM classifier achieved an AUC of 0.807 (95% CI, 0.649–0.936), accuracy of 0.730, sensitivity of 0.714, and specificity of 0.750.
For GCB/non-GCB classification, the Random Forest classifier achieved an AUC of 0.897 (95% CI, 0.796–0.973), accuracy of 0.846, sensitivity of 0.792, and specificity of 0.893.
Interpretation:
Limitations:
Single-center study may limit generalizability.
Retrospective design may introduce selection bias.
Reliance on histopathologic confirmation may not capture all biological variations.
