Meeting Banner
Abstract #3079

A Transfer Learning-based Radiomics Model for Prediction MGMT Promotor Methylation Status in Glioblastoma Multiforme

Xin Chen1, Tianjing Zhang2, Zhongping Zhang2, and Zaiyi Liu3

1Department of Radiology, Guangzhou First People's Hospital, School of Medicine, South China University of Technology, Guangzhou, China, 2Philips Healthcare, Guangzhou, China, 3Department of Radiology, Guangdong General Hospital, Guangzhou, China

Glioblastoma multiforme (GBM) is the most common malignant brain tumor. MGMT promoter methylation is associated with beneficial chemotherapy. We extract deep features from a pre-trained deep neural network model via transfer learning and generate an effective feature vector model together with radiomics features for an optimal pretreatment prediction of MGMT promoter methylation status. The deep feature set achieved the higher predictive accuracy of 0.86 and 0.70 for validation and test group comparing to handcrafted radiomics feature and combined feature sets. The deep feature model may serve as a potential imaging biomarker for pretreatment prediction of MGMT methylation in GBM.

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

Join Here