AUTHOR=Liu Yukun , Li Tianshi , Fan Ziwen , Li Yiming , Sun Zhiyan , Li Shaowu , Liang Yuchao , Zhou Chunyao , Zhu Qiang , Zhang Hong , Liu Xing , Wang Lei , Wang Yinyan TITLE=Image-Based Differentiation of Intracranial Metastasis From Glioblastoma Using Automated Machine Learning JOURNAL=Frontiers in Neuroscience VOLUME=Volume 16 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.855990 DOI=10.3389/fnins.2022.855990 ISSN=1662-453X ABSTRACT=Purpose: The majority of solitary brain metastases appear similar to glioblastomas on magnetic resonance imaging (MRI). This study aimed to develop and validate an MRI-based model to differentiate intracranial metastases from glioblastomas using automated machine learning. Materials and Methods: Radiomics features from 354 patients with brain metastases and 354 with glioblastomas were used to build prediction algorithms based on T2-weighted images, contrast-enhanced T1-weighted images, or both. The data of these subjects were subjected to a nested 10-fold split in the training and testing groups to build the best algorithms using the Tree-based Pipeline Optimization Tool (TPOT). The algorithms were independently validated using data from 124 institutional patients with solitary brain metastases and 103 patients with glioblastomas from The Cancer Genome Atlas. Results: Three groups of models were developed. The average areas under the receiver operating characteristic curve (AUCs) were 0.856 for contrast-enhanced T1-weighted images, 0.828 for T2-weighted images, and 0.988 for a combination in the testing groups, and the AUCs of the groups of models in the independent validation were 0.687, 0.831, and 0.867, respectively. A total of 149 radiomics features were considered as the most valuable features for the differential diagnosis of glioblastomas and metastases. Conclusions: The models established by TPOT can distinguish glioblastoma from solitary brain metastases well, and its noninvasiveness, convenience, and robustness make it potentially useful for clinical applications.