Search

Abstract Background The purpose of this study was to develop a voxel-wise clustering method of multiparametric magnetic resonance imaging (MRI) and 3,4-dihydroxy-6-[18F]-fluoro-L-phenylalanine (FDOPA) positron emission tomography (PET) images using an unsupervised, two-level clustering approach followed by support vector machine in order to classify the isocitrate dehydrogenase (IDH) status of gliomas. Methods Sixty-two treatment-naïve glioma patients who underwent FDOPA PET and MRI were retrospectively included. Contrast enhanced T1-weighted images, T2-weighted images, fluid-attenuated inversion recovery images, apparent diffusion coefficient maps, and relative cerebral blood volume maps, and FDOPA PET images were used for voxel-wise feature extraction. An unsupervised two-level clustering approach, including a self-organizing map followed by the K-means algorithm was used, and each class label was applied to the original images. The logarithmic ratio of labels in each class within tumor regions was applied to a support vector machine to differentiate IDH mutation status. The area under the curve (AUC) of receiver operating characteristic curves, accuracy, and F1-socore were calculated and used as metrics for performance. Results The associations of multiparametric imaging values in each cluster were successfully visualized. Multiparametric images with 16-class clustering revealed the highest classification performance to differentiate IDH status with the AUC, accuracy, and F1-score of 0.81, 0.76, and 0.76, respectively. Conclusions Machine learning using an unsupervised two-level clustering approach followed by a support vector machine classified the IDH mutation status of gliomas, and visualized voxel-wise features from multiparametric MRI and FDOPA PET images. Unsupervised clustered features may improve the understanding of prioritizing multiparametric imaging for classifying IDH status.

Abstract Purpose To identify clinically relevant magnetic resonance imaging (MRI) features of different types of metastatic brain lesions, including standard anatomical, diffusion weighted imaging (DWI) and dynamic susceptibility contrast (DSC) perfusion MRI. Methods MRI imaging was retrospectively assessed on one hundred and fourteen (N = 114) brain metastases including breast (n = 27), non-small cell lung cancer (NSCLC, n = 43) and ‘other’ primary tumors (n = 44). Based on 114 patient’s MRI scans, a total of 346 individual contrast enhancing tumors were manually segmented. In addition to tumor volume, apparent diffusion coefficients (ADC) and relative cerebral blood volume (rCBV) measurements, an independent component analysis (ICA) was performed with raw DSC data in order to assess arterio-venous components and the volume of overlap (AVOL) relative to tumor volume, as well as time to peak (TTP) of T2* signal from each component. Results Results suggests non-breast or non-NSCLC (‘other’) tumors had higher volume compare to breast and NSCLC patients (p = 0.0056 and p = 0.0003, respectively). No differences in median ADC or rCBV were observed across tumor types; however, breast and NSCLC tumors had a significantly higher “arterial” proportion of the tumor volume as indicated by ICA (p = 0.0062 and p = 0.0018, respectively), while a higher “venous” proportion were prominent in breast tumors compared with NSCLC (p = 0.0027) and ‘other’ lesions (p = 0.0011). The AVOL component was positively related to rCBV in all groups, but no correlation was found for arterial and venous components with respect to rCBV values. Median time to peak of arterial and venous components were 8.4 s and 12.6 s, respectively (p