AUTHOR=Zuo Zhichao , Liu Wen , Zeng Ying , Fan Xiaohong , Li Li , Chen Jing , Zhou Xiao , Jiang Yihong , Yang Xiuqi , Feng Yujie , Lu Yixin 

TITLE=Multiparametric magnetic resonance imaging-derived deep learning network to determine ferroptosis-related gene signatures in gliomas

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 16 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.1082867

DOI=10.3389/fnins.2022.1082867

ISSN=1662-453X

ABSTRACT=Glioma recurrence, progression, and metastasis are the three primary challenges that lead to treatment failure. Recently, molecular markers related to the prognosis and treatment of gliomas have been explored; however, there are still many gaps in the prognostic assessment and treatment of gliomas. Therefore, the identification of new markers remains imperative. Ferroptosis-related gene (FRG) signatures are important for the assessment of novel therapeutic approaches and the prognosis of glioma. We trained a deep learning network to determine FRG signatures using multiparametric magnetic resonance imaging (MRI). FRGs of patients with glioma were acquired from public databases. FRG-related risk score stratifying prognosis was developed from The Cancer Genome Atlas (TCGA) and validated using the Chinese Glioma Genome Atlas. Multiparametric MRI-derived glioma images and the corresponding genomic information were obtained for 122 cases from TCGA and The Cancer Imaging Archive. The deep learning network was trained using 3D-Resnet, and 3-fold cross-validation was performed to evaluate the predictive performance.
The FRG-related risk score was associated with poor clinicopathological features and had a high predictive value for glioma prognosis. Based on the FRG-related risk score, patients with glioma were successfully classified into two subgroups (28 and 94 in the high- and low-risk groups, respectively). The deep learning networks TC (enhancing tumor and non-enhancing portion of the tumor core) mask achieved an average cross-validation accuracy of 0.842 and average AUC of 0.781, while the deep learning networks WT (whole tumor and peritumoral edema) mask achieved an average cross-validation accuracy of 0.825 and average AUC of 0.781. 
Our findings indicate that FRG signature is a prognostic indicator of glioma. In addition, we developed a deep learning network that has high classification accuracy in automatically determining FRG signatures, which may be an important step toward the clinical translation of novel therapeutic approaches and prognosis of glioma.