Generative Adversarial Networks for Brain MRI Synthesis: Impact of Training Set Size on Clinical Application.

We evaluated the impact of training set size on generative adversarial networks (GANs) to synthesize brain MRI sequences. We compared three sets of GANs trained to generate pre-contrast T1 (gT1) from post-contrast T1 and FLAIR (gFLAIR) from T2. The baseline models were trained on 135 cases; for this study, we used the same model architecture but a larger cohort of 1251 cases and two stopping rules, an early checkpoint (early models) and one after 50 epochs (late models). We tested all models on an independent dataset of 485 newly diagnosed gliomas. We compared the generated MRIs with the original ones using the structural similarity index (SSI) and mean squared error (MSE). We simulated scenarios where either the original T1, FLAIR, or both were missing and used their synthesized version as inputs for a segmentation model with the original post-contrast T1 and T2. We compared the segmentations using the dice similarity coefficient (DSC) for the contrast-enhancing area, non-enhancing area, and the whole lesion. For the baseline, early, and late models on the test set, for the gT1, median SSI was .957, .918, and .947; median MSE was .006, .014, and .008. For the gFLAIR, median SSI was .924, .908, and .915; median MSE was .016, .016, and .019. The range DSC was .625-.955, .420-.952, and .610-.954. Overall, GANs trained on a relatively small cohort performed similarly to those trained on a cohort ten times larger, making them a viable option for rare diseases or institutions with limited resources.

Quantitative muscle mass biomarkers are independent prognosis factors in primary central nervous system lymphoma: The role of L3-skeletal muscle index and temporal muscle thickness.

OBJECTIVE: To investigate the role of quantitative muscle biomarkers assessed with skeletal muscle index at the third lumbar vertebra (L3-SMI) and temporal muscle thickness (TMT) in predicting progression-free and overall survival in patients with primary central nervous system lymphoma (PCNSL) undergoing first-line high-dose methotrexate-based chemotherapy. METHODS: L3-SMI and TMT were calculated on abdominal CT and brain high-resolution 3D-T1-weighted MR images, respectively, using predefined validated methods. Standardized sex-specific cut-off values were used to divide patients in different risk categories. Kaplan-Meier plots were calculated, and survival analysis was performed using log-rank tests, univariate, and multivariable Cox-regression models, calculating hazard ratios (HR) and 95% confidence intervals (CI), also adjusting for potential confounders (age, sex, and performance status). RESULTS: Forty-three patients were included in this study. Median follow-up was 23 months (interquartile range 12-40); at median follow-up, rates of progression-free and overall survival for the cohort were 46% and 57%, respectively. Thirteen (30%) and 11 (26%) patients showed L3-SMI or TMT values below the predefined cut-offs. In Cox-regression multivariable analysis patients with low L3-SMI or TMT showed significantly worse progression-free (HR 4.40, 95% CI 1.66-11.61, p = 0.003; HR 4.40, 95% CI 1.68-11.49, p = 0.003, respectively) and overall survival (HR 3.16, 95% CI 1.09-9.11, p = 0.034; HR 4.93, 95% CI 1.78-13.65, p = 0.002, respectively) compared to patients with high L3-SMI or TMT. CONCLUSIONS: Quantitative muscle mass evaluation assessed by both L3-SMI and TMT is a promising tool to identify PCNSL patients at high risk of negative outcome. Confirmatory studies on larger independent series are warranted.

Comparison of T1 mapping and fixed T1 method for dynamic contrast-enhanced MRI perfusion in brain gliomas.

OBJECTIVES: To compare dynamic contrast-enhanced MRI (DCE-MRI) data obtained using different prebolus T1 values in glioma grading and molecular profiling. METHODS: We retrospectively reviewed 83 cases of gliomas: 46 lower-grade gliomas (LGG; grades II and III) and 37 high-grade gliomas (HGG; grade IV). DCE-MRI maps of plasma volume fraction (Vp), extravascular-extracellular volume fraction (Ve), and tracer transfer constant from plasma to tissue (Ktrans) were obtained using a fixed T1 value of 1400 ms and a measured T1 obtained with variable flip angle (VFA). Tumour segmentations were performed and first-order histogram parameters were extracted from volumes of interest (VOIs) after co-registration with the perfusion maps. The two methods were compared using Wilcoxon matched-pairs signed-rank test and Bland-Altman analysis. Diagnostic accuracy was obtained and compared using ROC curve analysis and DeLong's test. RESULTS: Perfusion parameters obtained with the fixed T1 value were significantly higher than those obtained with the VFA. As regards diagnostic accuracy, there were no significant differences between the two methods both for glioma grading and molecular classification, except for few parameters of both methods. CONCLUSIONS: DCE-MRI data obtained with different prebolus T1 are not comparable and the definition of a prebolus T1 by T1 mapping is not mandatory since it does not improve the diagnostic accuracy of DCE-MRI. KEY POINTS: • DCE-MRI data obtained with different prebolus T1 are significantly different, thus not comparable. • The definition of a prebolus T1 by T1 mapping is not mandatory since it does not improve the diagnostic accuracy of DCE-MRI for glioma grading. • The use of a fixed T1 value represents a valid alternative to T1 mapping for DCE-MRI analysis.