Nomogram combining dual-energy computed tomography features and radiomics for differentiating parotid warthin tumor from pleomorphic adenoma: a retrospective study

Background Radiomic features may quantify characteristics present in medical imaging. However, the lack of standardized definitions and validated reference values have hampered clinical use. Purpose To standardize a set of 174 radiomic features. Materials and Methods Radiomic features were assessed in three phases. In phase I, 487 features were derived from the basic set of 174 features. Twenty-five research teams with unique radiomics software implementations computed feature values directly from a digital phantom, without any additional image processing. In phase II, 15 teams computed values for 1347 derived features using a CT image of a patient with lung cancer and predefined image processing configurations. In both phases, consensus among the teams on the validity of tentative reference values was measured through the frequency of the modal value and classified as follows: less than three matches, weak; three to five matches, moderate; six to nine matches, strong; 10 or more matches, very strong. In the final phase (phase III), a public data set of multimodality images (CT, fluorine 18 fluorodeoxyglucose PET, and T1-weighted MRI) from 51 patients with soft-tissue sarcoma was used to prospectively assess reproducibility of standardized features. Results Consensus on reference values was initially weak for 232 of 302 features (76.8%) at phase I and 703 of 1075 features (65.4%) at phase II. At the final iteration, weak consensus remained for only two of 487 features (0.4%) at phase I and 19 of 1347 features (1.4%) at phase II. Strong or better consensus was achieved for 463 of 487 features (95.1%) at phase I and 1220 of 1347 features (90.6%) at phase II. Overall, 169 of 174 features were standardized in the first two phases. In the final validation phase (phase III), most of the 169 standardized features could be excellently reproduced (166 with CT; 164 with PET; and 164 with MRI). Conclusion A set of 169 radiomics features was standardized, which enabled verification and calibration of different radiomics software. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Kuhl and Truhn in this issue. Show more

To compare image quality obtained in phantoms with virtual monochromatic spectral (VMS) imaging with that obtained with conventional 120-kVp computed tomography (CT) for a given radiation dose. Three syringes were filled with a diluted contrast medium (each syringe contained a contrast medium with a different iodine concentration [5, 10, or 15 mg of iodine per milliliter]), and a fourth syringe was filled with water. These syringes were placed in a torso phantom meant to simulate the standard human physique. The phantom was examined with a CT system and use of the fast kilovoltage switching (80 and 140 kVp) and conventional (120 kVp) modes. Image noise and contrast-to-noise (CNR) ratio were analyzed on VMS images and 120-kVp CT images. Image noise on VMS images in the range of 67-72 keV was significantly lower than that on the 120-kVp CT images (P < .014). Image noise was lowest at 69 keV and was 12% lower when compared with that on 120-kVp CT images. CNR on the VMS images was highest at 68 keV. CNR on the VMS images obtained at 68 keV in the syringes filled with diluted contrast material (5, 10, and 15 mg of iodine per milliliter) was 28%, 31%, and 30% higher, respectively, compared with that on the 120-kVp CT images (P < .001). VMS imaging at approximately 70 keV yielded lower image noise and higher CNR than did 120-kVp CT for a given radiation dose. VMS imaging has the potential to replace 120-kVp CT as the standard CT imaging modality, since optimal VMS imaging may be expected to yield improved image quality in a patient with standard body habitus. © RSNA, 2011. Show more