AUTHOR=de Causans Alix , Carré Alexandre , Roux Alexandre , Tauziède-Espariat Arnault , Ammari Samy , Dezamis Edouard , Dhermain Frederic , Reuzé Sylvain , Deutsch Eric , Oppenheim Catherine , Varlet Pascale , Pallud Johan , Edjlali Myriam , Robert Charlotte 

TITLE=Development of a Machine Learning Classifier Based on Radiomic Features Extracted From Post-Contrast 3D T1-Weighted MR Images to Distinguish Glioblastoma From Solitary Brain Metastasis

JOURNAL=Frontiers in Oncology

VOLUME=Volume 11 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.638262

DOI=10.3389/fonc.2021.638262

ISSN=2234-943X

ABSTRACT=Objectives:To differentiate Glioblastomas (GBM) and Brain Metastases (BM) using a radiomicfeatures-based Machine Learning (ML) classifier trained from post-contrast three-dimensional T1-weighted (post-contrast 3DT1) MR imaging, and compare its performance in medical diagnosisversus human experts, on a testing cohort.
Methods:We enrolled 143 patients (71 GBM and 72 BM) in a retrospective bicentric study fromJanuary 2010 to May 2019 to train the classifier. Post-contrast 3DT1 MR images were performedon a 3-Tesla MR unit and 100 radiomic features were extracted. Selection and optimization ofthe Machine Learning (ML) classifier was performed using a nested cross-validation. Sensitivity,specificity, balanced accuracy, and area under the receiver operating characteristic curve (AUC)were calculated as performance metrics. The model final performance was cross-validated, thenevaluated on a test set of 37 patients, and compared to human blind reading using a McNemar’stest.
Results:The ML classifier had a mean [95% confidence interval] sensitivity of 85% [77; 94], aspecificity of 87% [78; 97], a balanced accuracy of 86% [80; 92], and an AUC of 92% [87; 97] withcross-validation. Sensitivity, specificity, balanced accuracy and AUC were equal to 75%, 86%,80% and 85% on the test set. Sphericity 3D radiomic index highlighted the highest coefficient inthe logistic regression model. There were no statistical significant differences observed betweenthe performance of the classifier and the experts’ blinded examination.
Conclusions:The proposed diagnostic support system based on radiomic features extractedfrom post-contrast 3DT1 MR images helps in differentiating solitary BM from GBM with highdiagnosis performance and generalizability.