Whole-tumor radiomics analysis of DKI and DTI may improve the prediction of genotypes for astrocytomas: A preliminary study.

PURPOSE: To test whether the whole-tumor radiomics analysis of DKI and DTI images could predict IDH and MGMTmet genotypes of astrocytomas. METHOD: Sixty-two astrocytomas were enrolled. 364 radiomics features of whole tumor were extracted from mean-kurtosis (MK), and mean-diffusivity (MD) images, respectively. The multivariable logistic regression was used to select the most meaningful radiomics features for predicting IDH and MGMTmet genotypes. A radiomics model was built by logistic linear regression. A combined model was established based on selected radiomic, radiological and clinical features. To assess the difference between the models, the Z-test was performed. RESULTS: The radiomics model built using the three most informative radiomics features for each genotype yielded an AUC of 0.831 ((95 % confidence interval [CI]: 0.721-0.918) for predicting IDH genotype, and 0.835 (95 %CI: 0.686-0.951) for MGMTmet genotype. A combined model for predicting IDH based on the radiomics score, age, and degree of edema reached an AUC of 0.885 (95 %CI: 0.802-0.955) and a combined model for predicting MGMTmet based on radiomics score and edema degree reached an AUC of 0.859 (95 %CI: 0.751-0.945) which was not significantly higher than the radiomics only model (P =  0.081). CONCLUSIONS: The radiomics models via an objective whole-tumor analysis of MK and MD maps were independent imaging biomarkers for predicting IDH and MGMTmet genotypes, and the combined model further improved the performance for IDH, but not for MGMTmet.

Radiomics of 18F Fluorodeoxyglucose PET/CT Images Predicts Severe Immune-related Adverse Events in Patients with NSCLC.

PURPOSE: To investigate the performance of pretreatment fluorine 18 (18F) fluorodeoxyglucose PET/CT radiomics in predicting severe immune-related adverse events (irSAEs) among patients with advanced non-small cell lung cancer (NSCLC) treated with immunotherapy, which is important in optimizing treatment plans and alleviating future complications with early interventions. MATERIALS AND METHODS: The retrospective arm of this study included 146 patients with histologically confirmed stage IIIB-IV NSCLC who were treated with immune checkpoint blockade between June 2011 and December 2017 and who were split into training (n = 97) and test (n = 49) cohorts. A prospective validation arm enrolled 48 patients before initiation of immunotherapy between January 2018 and June 2019 as an independent test cohort. Radiomics features extracted from baseline (preimmunotherapy treatment) PET, CT, and PET/CT fusion images were used to generate a radiomics score (RS) to quantify patient risk for developing irSAEs by an improved least absolute shrinkage and selection operator method. Weighted multivariable logistic regression analysis was then used to develop a nomogram model to predict irSAEs, which was assessed by its calibration, discrimination, and clinical usefulness. RESULTS: The radiomics nomogram, incorporating the RS, type of immune checkpoint blockade, and dosing schedule, was able to predict patients with and without irSAEs with area under the receiver operating characteristic curve of 0.92 (95% confidence interval [CI]: 0.86, 0.98), 0.92 (95% CI: 0.86, 0.99), and 0.88 (95% CI: 0.78, 0.97) in the training, test, and prospective validation cohorts, respectively. Decision curve analysis showed that the radiomics nomogram model had the highest overall net benefit. CONCLUSION: A high RS is a significant risk factor for development of irSAEs, demonstrating the value of PET/CT images in predicting irSAEs. By the identification, at baseline, of patients with NSCLC most likely to have irSAEs, treatment plans can be optimized before initiation of immunotherapy.Supplemental material is available for this article.© RSNA, 2020See also the commentary by Yousefi.

Improving survival prediction of high-grade glioma via machine learning techniques based on MRI radiomic, genetic and clinical risk factors.

OBJECTIVES: To develop a radiomic signature to predict overall survival (OS) for high-grade glioma (HGG), and construct a nomogram by combining selected radiomic, genetic and clinical risk factors to further improve the performance of the risk model. MATERIALS AND METHODS: 147 cases of HGG with MRI images, genetic data, clinical data were studied, wherein 112 patients were used as training cohort, and 35 patients were as independent test cohort. Radiomics features were extracted from tumor area and peritumoral edema area on CE-T1WI and T2FLAIR images. Association between radiomics signature, genetic, clinical risk factors and OS was explored by Kaplan-Meier survival analysis and log rank test. The multivariate Cox regression analysis was trained with radiomic features along with selected genetic and clinical risk factors, which was presented as a nomogram. RESULTS: The radiomic signature constructed by 11 radiomics features stratified patients into low- and high-risk groups, and the C-Index for OS prediction was 0.707 and 0.711 in training and test cohorts, respectively. The multivariable Cox regression analysis identified radiomics signature (hazard ratio (HR): 2.18, P = 0.005), IDH (HR: 0.490, P = 0.007) and age (HR: 1.039, P = 0.005) as independent risk factors. A nomogram combining these independent risk factors further improved the performance for OS estimation (C-index = 0.764 and 0.758 in training and test cohorts, respectively). CONCLUSION: The radiomics signature is a new prognostic biomarker for HGG. A nomogram incorporating radiomics signature, IDH and age improved the performance of OS estimation, which might be a new complement to the treatment guidelines of glioma.

Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels.

PURPOSE: Many radiomics features were originally developed for non-medical imaging applications and therefore original assumptions may need to be reexamined. In this study, we investigated the impact of slice thickness and pixel spacing (or pixel size) on radiomics features extracted from Computed Tomography (CT) phantom images acquired with different scanners as well as different acquisition and reconstruction parameters. The dependence of CT texture features on gray-level discretization was also evaluated. METHODS AND MATERIALS: A texture phantom composed of 10 different cartridges of different materials was scanned on eight different CT scanners from three different manufacturers. The images were reconstructed for various slice thicknesses. For each slice thickness, the reconstruction Field Of View (FOV) was varied to render pixel sizes ranging from 0.39 to 0.98 mm. A fixed spherical region of interest (ROI) was contoured on the images of the shredded rubber cartridge and the 3D printed, 20% fill, acrylonitrile butadiene styrene plastic cartridge (ABS20) for all phantom imaging sets. Radiomic features were extracted from the ROIs using an in-house program. Features categories were: shape (10), intensity (16), GLCM (24), GLZSM (11), GLRLM (11), and NGTDM (5), fractal dimensions (8) and first-order wavelets (128), for a total of 213 features. Voxel-size resampling was performed to investigate the usefulness of extracting features using a suitably chosen voxel size. Acquired phantom image sets were resampled to a voxel size of 1 × 1 × 2 mm3 using linear interpolation. Image features were therefore extracted from resampled and original datasets and the absolute value of the percent coefficient of variation (%COV) for each feature was calculated. Based on the %COV values, features were classified in 3 groups: (1) features with large variations before and after resampling (%COV >50); (2) features with diminished variation (%COV <30) after resampling; and (3) features that had originally moderate variation (%COV <50%) and were negligibly affected by resampling. Group 2 features were further studied by modifying feature definitions to include voxel size. Original and voxel-size normalized features were used for interscanner comparisons. A subsequent analysis investigated feature dependency on gray-level discretization by extracting 51 texture features from ROIs from each of the 10 different phantom cartridges using 16, 32, 64, 128, and 256 gray levels. RESULTS: Out of the 213 features extracted, 150 were reproducible across voxel sizes, 42 improved significantly (%COV <30, Group 2) after resampling, and 21 had large variations before and after resampling (Group 1). Ten features improved significantly after definition modification effectively removed their voxel-size dependency. Interscanner comparison indicated that feature variability among scanners nearly vanished for 8 of these 10 features. Furthermore, 17 out of 51 texture features were found to be dependent on the number of gray levels. These features were redefined to include the number of gray levels which greatly reduced this dependency. CONCLUSION: Voxel-size resampling is an appropriate pre-processing step for image datasets acquired with variable voxel sizes to obtain more reproducible CT features. We found that some of the radiomics features were voxel size and gray-level discretization-dependent. The introduction of normalizing factors in their definitions greatly reduced or removed these dependencies.