Abstract #3628

Multiparametric MR radiomics in brain glioma: models comparation to predict biomarker status

Jinlong He¹, Jianlin Wu¹, Yang Gao², Qiong Wu², Jing Shen¹, Shaoyu Wang³, Huapeng Zhang³, Jialiang Ren⁴, and Peng Wang⁵

¹Tianjin Medical University Graduate School, Tianjin, China, ²Department Of Imaging Diagnosis, Affiliated Hospital Of Inner Monglia Medical University, Hohhot, China, ³MR Scientific Marketing, Siemens Healthineers, Shanghai, China, ⁴GE Healthcare, Shanghai, China, ⁵Inner Monglia Medical University, Hohhot, China

Synopsis

Objective:To compare the performance of clinical model, radiomics model, and combined model in predicting biomarker status (IDH, MGMT, TERT, 1p/19q) of glioma.Methods: 81 glioma patients confirmed by histology were enrolled in this study. The predictive performance of each model was validated by receiver operating characteristic curve (ROC) analysis and decision curve analysis (DCA).Results: The mixed model showed the highest performance in each genic phenotype (IDH AUC = 0.93, MGMT AUC=0.88, TERT AUC=0.76, 1p/19q AUC=0.71).Conclusion: The mixed model is an effective tool to distinguish genic phenotype of brain glioma which have highest diagnostic efficiency than other models.

This abstract and the presentation materials are available to members only; a login is required.

Join Here