AUTHOR=Zhang Liqiang , Yao Rui , Gao Jueni , Tan Duo , Yang Xinyi , Wen Ming , Wang Jie , Xie Xiangxian , Liao Ruikun , Tang Yao , Chen Shanxiong , Li Yongmei 

TITLE=An Integrated Radiomics Model Incorporating Diffusion-Weighted Imaging and 18F-FDG PET Imaging Improves the Performance of Differentiating Glioblastoma From Solitary Brain Metastases

JOURNAL=Frontiers in Oncology

VOLUME=11

YEAR=2021

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.732704

DOI=10.3389/fonc.2021.732704

ISSN=2234-943X

ABSTRACT=<sec><title>Background</title><p>The effectiveness of conventional MRI (cMRI)-based radiomics in differentiating glioblastoma (GBM) from solitary brain metastases (SBM) is not satisfactory enough. Therefore, we aimed to develop an integrated radiomics model to improve the performance of differentiating GBM from SBM.</p></sec><sec><title>Methods</title><p>One hundred patients with solitary brain tumors (50 with GBM, 50 with SBM) were retrospectively enrolled and randomly assigned to the training set (<italic>n</italic> = 80) or validation set (<italic>n</italic> = 20). A total of 4,424 radiomic features were obtained from contrast-enhanced T1-weighted imaging (CE-T1WI) with the contrast-enhancing and peri-enhancing edema region, T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI)-derived apparent diffusion coefficient (ADC), and <sup>18</sup>F-fluorodeoxyglucose positron emission tomography (<sup>18</sup>F-FDG PET) images. The partial least squares (PLS) regression with fivefold cross-validation is used to analyze the correlation between different radiomic features and different modalities. The cross-validity analysis was performed to judge whether a new principal component or a new feature dimension can significantly improve the final prediction effect. The principal components with effective interpretation in all radiomic features were projected to a low-dimensional space (2D in this study). The effective features of the new projection mapping were then sent to the random forest classifier to predict the results. The performance of differentiating GBM from SBM was compared between the integrated radiomics model and other radiomics models or nonradiomics methods using the area under the receiver operating characteristics curve (AUC).</p></sec><sec><title>Results</title><p>Through the cross-validity analysis of partial least squares, hundreds of radiomic features were projected into a new two-dimensional space to complete the construction of radiomics model. Compared with the combined radiomics model using DWI + <sup>18</sup>F-FDG PET (AUC = 0.93, <italic>p</italic> = 0.014), cMRI + DWI (AUC = 0.89, <italic>p</italic> = 0.011), cMRI + <sup>8</sup>F-FDG PET (AUC = 0.91, <italic>p</italic> = 0.015), and single radiomics model using cMRI (AUC = 0.85, <italic>p</italic> = 0.018), DWI (AUC = 0.84, <italic>p</italic> = 0.017), and <sup>18</sup>F-FDG PET (AUC = 0.85, <italic>p</italic> = 0.421), the integrated radiomics model (AUC = 0.98) showed more efficient diagnostic performance. The integrated radiomics model (AUC = 0.98) also showed significantly better performance than any single ADC, SUV, or TBR parameter (AUC = 0.57–0.71, <italic>p</italic> &lt; 0.05). The integrated radiomics model showed better performance in the training (AUC = 0.98) and validation (AUC = 0.93) sets than any other models and methods, demonstrating robustness.</p></sec><sec><title>Conclusions</title><p>We developed an integrated radiomics model incorporating DWI and <sup>18</sup>F-FDG PET, which improved the performance of differentiating GBM from SBM greatly.</p></sec>