AUTHOR=Wang Hao , Zhou Yongkang , Wang Xiao , Zhang Yin , Ma Chi , Liu Bo , Kong Qing , Yue Ning , Xu Zhiyong , Nie Ke 

TITLE=Reproducibility and Repeatability of CBCT-Derived Radiomics Features

JOURNAL=Frontiers in Oncology

VOLUME=Volume 11 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.773512

DOI=10.3389/fonc.2021.773512

ISSN=2234-943X

ABSTRACT=Purpose: To determine the reproducibility and repeatability of CBCT radiomics features.
Methods: Daily CBCT images from 10 H&N cancer patients and 10 pelvic cancer patients were retrospectively collected for this study. Eighteen common radiomics features were extracted from the longitudinal CBCT images using two radiomics packages. The reproducibility of CBCT derived radiomics features was assessed using the first-day image as input and compared across the two software packages. The site- specific intra-class correlation coefficient (ICC) was used to quantitatively assess the agreement between packages. The repeatability of CBCT based radiomics features was evaluated by comparing following days CBCT to the first-day image and quantified using site-specific concordance correlation coefficient (CCC). Furthermore, the correlation with volume for all the features were assessed with linear regression and R2 as correlation parameters.
Results: The first-order histogram-based features as skewness and entropy showed good agreement computed in either software package (ICCs ≥ 0.80), while the kurtosis measurements were consistent in HN patients between two software tools but not in pelvic cases. The ICCs for GLCM-based features showed good agreement (ICCs  ≥  0.80) between packages in both HN and pelvic groups except the GLCM-correction. The GLRLM-based texture features were overall less consistent calculated by two different software packages compared to GLCM-based features. The CCC values of all first-order and second-order GLCM features (except GLCM-energy) were all above 0.80 from the 2-day part test-retest set, while the CCC values all droped below the cut-off after 5-day treatment scans. All first-order histogram based and GLCM-texture based features were not highly correlated with volume, while two GLRLM features, in both HN and pelvic cohorts, showed R2 ≥ 0.8 meaning a high correlation with volume.
Conclusion: The reproducibility and repeatability of CBCT-based radiomics features were assessed and compared as the first time on both HN and pelvic sites. There are overlaps of stable features in both disease sites yet the overall stability of radiomic features may be disease/protocol-specific and a function of the time between scans.