This article was originally published here
Technol Cancer Res treatment. 2022 Jan-Dec;21:15330338221099113. doi: 10.1177/15330338221099113.
Goal: Radiomics involves the extraction of quantitative imaging biomarkers (or radiomics features) believed to provide additional pathophysiological and/or clinical information over qualitative visual observation and interpretation. This retrospective study explores the variability of radiomic characteristics extracted from images acquired with the 0.35 T scanner of an integrated MRI-Linac. We hypothesized that we would be able to identify features with high repeatability and reproducibility under various imaging conditions using phantom and patient imaging studies. We also compared literature results relevant to our results. Methods : Eleven scans of a Magphan® A 13-month RT phantom and 11 scans of an 11-day ViewRay Daily QA phantom constituted the phantom data. The patient datasets included 50 images of ten anonymized stereotactic body radiation therapy (SBRT) patients with pancreatic cancer (50 Gy in 5 fractions). A True Fast Imaging with Steady-State Free Precession (TRUFI) pulse sequence was selected, using a voxel resolution of 1.5 mm × 1.5 mm × 1.5 mm and 1.5 mm × 1, 5 mm × 3.0 mm for phantom and patient data, respectively. A total of 1087 shape-based, first-, second-, and higher-order features were extracted, followed by robustness analysis. Robustness was assessed with the coefficient of variation (CoV Results: We identified 130 robust features in the datasets. Robust features were found in each category except for 2 second-order subgroups, namely, Grayscale Size Area Matrix (GLSZM) and Neighborhood Grayscale Tone Difference Matrix (NGTDM). Additionally, several robust features were consistent with results from other stability assessments or predictive performance studies in the literature. Conclusion: We verified the stability of the 0.35 T scanner of an integrated MRI-Linac for longitudinal radiomic phantom studies and identified robust features under various imaging conditions. We conclude that phantom measurements can be used to identify robust radiomic features. Further research on stability assessment is warranted.
PMID:35521966 | DOI: 10.1177/15330338221099113