Deep radiomic features from MRI scans predict survival outcome of\n recurrent glioblastoma

2019/11/15 by Ahmad Chaddad, Saima Rathore, Chaddad, Ahmad +7
Medicine · Engineering · #Radiomics and Machine Learning in Medical Imaging #Glioma Diagnosis and Treatment #Medical Imaging and Analysis

paper · pdf · doi:10.48550/arxiv.1911.06687

Abstract

This paper proposes to use deep radiomic features (DRFs) from a convolutional\nneural network (CNN) to model fine-grained texture signatures in the radiomic\nanalysis of recurrent glioblastoma (rGBM). We use DRFs to predict survival of\nrGBM patients with preoperative T1-weighted post-contrast MR images (n=100).\nDRFs are extracted from regions of interest labelled by a radiation oncologist\nand used to compare between short-term and long-term survival patient groups.\nRandom forest (RF) classification is employed to predict survival outcome\n(i.e., short or long survival), as well as to identify highly group-informative\ndescriptors. Classification using DRFs results in an area under the ROC curve\n(AUC) of 89.15% (p<0.01) in predicting rGBM patient survival, compared to\n78.07% (p<0.01) when using standard radiomic features (SRF). These results\nindicate the potential of DRFs as a prognostic marker for patients with rGBM.\n

Related

• A Deep Learning Model with Radiomics Analysis Integration for Glioblastoma Post-Resection Survival Prediction
• Brain Tumor Recurrence vs. Radiation Necrosis Classification and Patient Survivability Prediction
• The added value for MRI radiomics and deep-learning for glioblastoma prognostication compared to clinical and molecular information
• The stability of oncologic MRI radiomic features and the potential role of deep learning: a review
• Predicting survival of glioblastoma from automatic whole-brain and tumor segmentation of MR images