Deep learning-assisted radiomics facilitates multimodal prognostication for personalized treatment strategies in low-grade glioma.

Determining the optimal course of treatment for low grade glioma (LGG) patients is challenging and frequently reliant on subjective judgment and limited scientific evidence. Our objective was to develop a comprehensive deep learning assisted radiomics model for assessing not only overall survival in LGG, but also the likelihood of future malignancy and glioma growth velocity. Thus, we retrospectively included 349 LGG patients to develop a prediction model using clinical, anatomical, and preoperative MRI data. Before performing radiomics analysis, a U2-model for glioma segmentation was utilized to prevent bias, yielding a mean whole tumor Dice score of 0.837. Overall survival and time to malignancy were estimated using Cox proportional hazard models. In a postoperative model, we derived a C-index of 0.82 (CI 0.79-0.86) for the training cohort over 10 years and 0.74 (Cl 0.64-0.84) for the test cohort. Preoperative models showed a C-index of 0.77 (Cl 0.73-0.82) for training and 0.67 (Cl 0.57-0.80) test sets. Our findings suggest that we can reliably predict the survival of a heterogeneous population of glioma patients in both preoperative and postoperative scenarios. Further, we demonstrate the utility of radiomics in predicting biological tumor activity, such as the time to malignancy and the LGG growth rate.

Differences in lipid profiles in two Hispanic ischemic stroke populations.

BACKGROUND AND PURPOSE: The study aims to compare lipid profiles among ischemic stroke patients in a predominantly Caribbean-Hispanic population in Miami and a Mestizo Hispanic population in Mexico City. METHODS: We analyzed ischemic stroke Hispanic patients with complete baseline fasting lipid profile enrolled contemporaneously in the prospective registries of two tertiary care teaching hospitals in Mexico City and Miami. Demographic characteristics, risk factors, medications, ischemic stroke subtype, and first fasting lipid profile were compared. Vascular risk factor definitions were standardized. Multiple linear regression analysis was performed to compare lipid fractions. RESULTS: A total of 324 patients from Mexico and 236 from Miami were analyzed. Mexicans were significantly younger (58 · 1 vs. 67 · 4 years), had a lower frequency of hypertension (53 · 4% vs. 79 · 7%), and lower body mass index (27 vs. 28 · 5). There was a trend toward greater prevalence of diabetes in Mexicans (31 · 5 vs. 24 · 6%, P = 0 · 07). Statin use at the time of ischemic stroke was more common in Miami Hispanics (18 · 6 vs. 9 · 4%). Mexicans had lower total cholesterol levels (169 · 9 ± 46 · 1 vs. 179 · 9 ± 48 · 4 mg/dl), lower low-density lipoprotein (92 · 3 ± 37 · 1 vs. 108 · 2 ± 40 · 8 mg/dl), and higher triglyceride levels (166 · 9 ± 123 · 9 vs. 149 · 2 ± 115 · 2 mg/dl). These differences remained significant after adjusting for age, gender, hypertension, diabetes, body mass index, smoking, ischemic stroke subtype, and statin use. CONCLUSION: We found significant differences in lipid fractions in Hispanic ischemic stroke patients, with lower total cholesterol and low-density lipoprotein, and higher triglyceride levels in Mexicans. These findings highlight the heterogeneity of dyslipidemia among the Hispanic race-ethnic group and may lead to different secondary prevention strategies.