MR textural features (RADIOMICS) for predicting response to treatment in patients with intracranial tuberculoma: A retrospective cross-sectional study

Background and objective:
 MR based radiomics can potentially response to treatment in intracranial tuberculoma, but very scarce literature is available this regard. The purpose of study was determine whether radiomic features be used predict antituberculosis (AT) treatment.
 Methods:
 Data patients with tuberculomas who underwent imaging AT at our institution during the last 10 years analyzed. In each case follow-up performed 6 months post initiation reviewed establish treatment. textural analysis by two consultant neuroradiologists, using open-source software (Lifex) FLAIR coronal image after contrast administration from pretreatment MRI analysis.
 Results: 
 Twenty-four mean age 33.8 were included study. Sixteen responsive group while eight resistant group. Thirty-eight parameters extracted for patient. There a significant difference three out 38 (histogram skewness, GLCM correlation NGLDM Coarseness) amongst groups. Logistic regression model developed these which accurately predicted 83.3% cases according (χ2=11.517, p=0.003). ROC curve histogram skewness showed acceptable discrimination (p=0.037 95% CI =0.577-0.954) predicting Conclusion:
 may as biomarkers tuberculoma.