Article Figures & Data
Figures
- FIG 1.
Flow chart showing the distribution of the patient population in the training (A) and the test (B) sets. IDHwt indicates isocitrate dehydrogenase wild-type.
- FIG 2.
Representative cases of IDH wild-type lower-grade gliomas with (A) and without (B) molecular features of GBM. A, Initial MR imaging of a 49-year-old man with IDH wild-type lower-grade glioma with molecular features of GBM, WHO grade III. MRIs reveal infiltrative T2-hyperintense tumor in the right frontal and parietal lobes extending into the corpus callosum with cortical involvement (left). Focal contrast enhancement is noted in the right frontal lobe (right). B, Initial MRIs of a 26-year-old woman without molecular features of GBM, WHO grade III. MRIs reveal a relatively well-defined enhancing mass with peritumoral edema in the right temporo-occipital lobe without cortical involvement.
- FIG 3.
Heat maps illustrating the predictive performance (AUCs) of the different combinations of feature selection methods (rows) and classifiers (columns) from models 3 and 4 in the test set. LDA indicates linear discriminant analysis; AdaBoost, adaptive boosting.
- FIG 4.
Receiver operating characteristic curves of models 1, 2, 3, and 4 with the highest predictive performance. The combination of SVM and RFE shows the highest predictive performance in both models 3 and 4.
Tables
Clinical Characteristics Training Set Test Set, Institution 2 P Valueb Institution 1 TCGA No. of patients 32 32 57 Age (yr) 49.0 [SD, 19.1] 51.1 [SD, 14.7] 53.4 [SD, 16.2] .291 Sex (male/female) 19:13 17:15 27:30 .901 WHO grade .800 Grade II 16 (50.0%) 9 (28.1%) 23 (40.4%) Grade III 16 (50.0%) 23 (71.9%) 34 (59.6%) Molecular GBM status .053 With molecular features of GBM 21 (65.6%) 26 (81.3%) 32 (56.1%) Without molecular features of GBM 11 (34.4%) 6 (18.7%) 25 (43.9%) - Table 2:
Highest predictive performances of different models in identifying the molecular features of glioblastomas in IDH wild-type lower-grade gliomas in the test set
Model Feature Selection + Classifier AUC P Values for Model Comparisons Model 1 (clinical features) NA 0.514 (0.356–0.672) Model 2 (VASARI features) NA 0.648 (0.511–0.784) Model 3 (VASARI + radiomics features) RFE + SVM 0.854 (0.766–0.941) <.001a .023b Model 4 (clinical + VASARI + radiomics features) RFE + SVM 0.863 (0.778–0.947) <.001a .02b .476c
