Influence of Feature Calculating Parameters on the Reproducibility of CT Radiomic Features: a Thoracic Phantom Study

Overview

Journal Quant Imaging Med Surg

Specialty Radiology

Date 2020 Sep 4

PMID 32879856

Citations 9

Authors

Ying Li

Guanghua Tan

Mark Vangel

Jonathan Hall

Wenli Cai

Affiliations

Soon will be listed here.

Abstract

Background: Existing studies have demonstrated that imaging parameters may affect radiomic features. However, the influence of feature calculating parameters has been overlooked. The purpose of this study is to investigate the influence of feature calculating parameters (gray-level range and bin size) on the reproducibility of CT radiomic features.

Methods: Thirty-six CT scans from an anthropomorphic thoracic phantom were acquired with different imaging parameters including effective dose, pitch, slice thicknesses and reconstruction kernels. The influence of feature calculating parameters was investigated in terms of three gray-level ranges and eleven gray-level bin sizes. Feature reproducibility was assessed by the intraclass correlation coefficient (ICC) with the cutoff value of 0.8 and the coefficient of variation (CV) with the cutoff value of 20%. The agreements of reproducible features in different ranges and bin sizes were analyzed by Kendall's W test and Kappa test. The proportions of reproducible features, in terms of two calculating, four imaging and two segmentation parameters, were evaluated using Cochran's Q test and Dunn's test.

Results: For the three gray-level ranges, 50% (44/88) of features were reproducible with a perfect agreement (Kendall's W coefficient 0.844, P<0.001). Of the 72 features that may be influenced by gray-level bin size, 33.3% (24/72) were reproducible for 11 bin sizes with a perfect agreement (Kendall's W coefficient 0.879, P<0.001). For the proportions of reproducible features, there was no statistically significant difference among three ranges (P=0.420), but there was among eleven bin sizes (P=0.013). The proportions of reproducible features in feature calculating parameters were statistically significantly lower than those in imaging parameters (adjusted P<0.05).

Conclusions: Feature calculating parameters may have a greater influence than imaging parameters on the reproducibility of CT radiomic features, which should be given special attention in clinical applications.

Citing Articles

Radiomics in Dermatological Optical Coherence Tomography (OCT): Feature Repeatability, Reproducibility, and Integration into Diagnostic Models in a Prospective Study.

Widaatalla Y, Wolswijk T, Khan M, Halilaj I, Mosterd K, Woodruff H Cancers (Basel). 2025; 17(5).

PMID: 40075619 PMC: 11899706. DOI: 10.3390/cancers17050768.

Decoding Radiomics: A Step-by-Step Guide to Machine Learning Workflow in Hand-Crafted and Deep Learning Radiomics Studies.

Ce M, Chiriac M, Cozzi A, Macri L, Rabaiotti F, Irmici G Diagnostics (Basel). 2024; 14(22).

PMID: 39594139 PMC: 11593328. DOI: 10.3390/diagnostics14222473.

Multicentric study on the reproducibility and robustness of PET-based radiomics features with a realistic activity painting phantom.

Kallos-Balogh P, Vas N, Toth Z, Szakall S, Szabo P, Garai I PLoS One. 2024; 19(10):e0309540.

PMID: 39446842 PMC: 11500893. DOI: 10.1371/journal.pone.0309540.

Enhancing the Clinical Utility of Radiomics: Addressing the Challenges of Repeatability and Reproducibility in CT and MRI.

Teng X, Wang Y, Nicol A, Ching J, Wong E, Lam K Diagnostics (Basel). 2024; 14(16).

PMID: 39202322 PMC: 11353986. DOI: 10.3390/diagnostics14161835.

Pretherapy investigations using highly robust visualized biomarkers from CT imaging by multiple machine-learning techniques toward its prognosis prediction for ALK-inhibitor therapy in NSCLC: a feasibility study.

Sun J, Li F, Yang J, Lin C, Zhou X, Liu N J Cancer Res Clin Oncol. 2023; 149(10):7341-7353.

PMID: 36928998 DOI: 10.1007/s00432-023-04615-3.

References

Berenguer R, Pastor-Juan M, Canales-Vazquez J, Castro-Garcia M, Villas M, Mansilla Legorburo F . Radiomics of CT Features May Be Nonreproducible and Redundant: Influence of CT Acquisition Parameters. Radiology. 2018; 288(2):407-415. DOI: 10.1148/radiol.2018172361. View

Lu L, Ehmke R, Schwartz L, Zhao B . Assessing Agreement between Radiomic Features Computed for Multiple CT Imaging Settings. PLoS One. 2016; 11(12):e0166550. PMC: 5199063. DOI: 10.1371/journal.pone.0166550. View

Mackin D, Fave X, Zhang L, Yang J, Kyle Jones A, Ng C . Harmonizing the pixel size in retrospective computed tomography radiomics studies. PLoS One. 2017; 12(9):e0178524. PMC: 5608195. DOI: 10.1371/journal.pone.0178524. View

Orlhac F, Frouin F, Nioche C, Ayache N, Buvat I . Validation of A Method to Compensate Multicenter Effects Affecting CT Radiomics. Radiology. 2019; 291(1):53-59. DOI: 10.1148/radiol.2019182023. View

Goodsitt M, Chan H, Way T, Larson S, Christodoulou E, Kim J . Accuracy of the CT numbers of simulated lung nodules imaged with multi-detector CT scanners. Med Phys. 2006; 33(8):3006-17. PMC: 2742214. DOI: 10.1118/1.2219332. View

Sun R, Limkin E, Vakalopoulou M, Dercle L, Champiat S, Han S . A radiomics approach to assess tumour-infiltrating CD8 cells and response to anti-PD-1 or anti-PD-L1 immunotherapy: an imaging biomarker, retrospective multicohort study. Lancet Oncol. 2018; 19(9):1180-1191. DOI: 10.1016/S1470-2045(18)30413-3. View

Aerts H, Velazquez E, Leijenaar R, Parmar C, Grossmann P, Carvalho S . Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun. 2014; 5:4006. PMC: 4059926. DOI: 10.1038/ncomms5006. View

Szczykutowicz T, DuPlissis A, Pickhardt P . Variation in CT Number and Image Noise Uniformity According to Patient Positioning in MDCT. AJR Am J Roentgenol. 2017; 208(5):1064-1072. DOI: 10.2214/AJR.16.17215. View

Larue R, van Timmeren J, de Jong E, Feliciani G, Leijenaar R, Schreurs W . Influence of gray level discretization on radiomic feature stability for different CT scanners, tube currents and slice thicknesses: a comprehensive phantom study. Acta Oncol. 2017; 56(11):1544-1553. DOI: 10.1080/0284186X.2017.1351624. View

10.

Caramella C, Allorant A, Orlhac F, Bidault F, Asselain B, Ammari S . Can we trust the calculation of texture indices of CT images? A phantom study. Med Phys. 2018; 45(4):1529-1536. DOI: 10.1002/mp.12809. View

11.

Buch K, Li B, Qureshi M, Kuno H, Anderson S, Sakai O . Quantitative Assessment of Variation in CT Parameters on Texture Features: Pilot Study Using a Nonanatomic Phantom. AJNR Am J Neuroradiol. 2017; 38(5):981-985. PMC: 7960393. DOI: 10.3174/ajnr.A5139. View

12.

Midya A, Chakraborty J, Gonen M, Do R, Simpson A . Influence of CT acquisition and reconstruction parameters on radiomic feature reproducibility. J Med Imaging (Bellingham). 2018; 5(1):011020. PMC: 5812985. DOI: 10.1117/1.JMI.5.1.011020. View

13.

Zhang L, Fried D, Fave X, Hunter L, Yang J, Court L . IBEX: an open infrastructure software platform to facilitate collaborative work in radiomics. Med Phys. 2015; 42(3):1341-53. PMC: 5148126. DOI: 10.1118/1.4908210. View

14.

Gillies R, Kinahan P, Hricak H . Radiomics: Images Are More than Pictures, They Are Data. Radiology. 2015; 278(2):563-77. PMC: 4734157. DOI: 10.1148/radiol.2015151169. View

15.

Huang Y, Liang C, He L, Tian J, Liang C, Chen X . Development and Validation of a Radiomics Nomogram for Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer. J Clin Oncol. 2016; 34(18):2157-64. DOI: 10.1200/JCO.2015.65.9128. View

16.

Arslanoglu A, Chalian H, Sodagari F, Seyal A, Tore H, Salem R . Threshold for Enhancement in Treated Hepatocellular Carcinoma on MDCT: Effect on Necrosis Quantification. AJR Am J Roentgenol. 2016; 206(3):536-43. DOI: 10.2214/AJR.15.15339. View

17.

Shafiq-Ul-Hassan M, Zhang G, Latifi K, Ullah G, Hunt D, Balagurunathan Y . Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels. Med Phys. 2017; 44(3):1050-1062. PMC: 5462462. DOI: 10.1002/mp.12123. View

18.

Kumar V, Gu Y, Basu S, Berglund A, Eschrich S, Schabath M . Radiomics: the process and the challenges. Magn Reson Imaging. 2012; 30(9):1234-48. PMC: 3563280. DOI: 10.1016/j.mri.2012.06.010. View

19.

Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout R, Granton P . Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer. 2012; 48(4):441-6. PMC: 4533986. DOI: 10.1016/j.ejca.2011.11.036. View

20.

Ger R, Zhou S, Chi P, Lee H, Layman R, Kyle Jones A . Comprehensive Investigation on Controlling for CT Imaging Variabilities in Radiomics Studies. Sci Rep. 2018; 8(1):13047. PMC: 6115360. DOI: 10.1038/s41598-018-31509-z. View