Back to the list
Congress: ECR25
Poster Number: C-19104
Type: Poster: EPOS Radiologist (scientific)
Authorblock: M. Aymerich, B. González-Oblanca, A. Garcia Baizan, M. Otero Garcia; Vigo/ES
Disclosures:
Maria Aymerich: Nothing to disclose
Beatriz González-Oblanca: Nothing to disclose
Alejandra Garcia Baizan: Nothing to disclose
Milagros Otero Garcia: Nothing to disclose
Keywords: Artificial Intelligence, CT, CT-Quantitative, Computer Applications-General, Experimental investigations, Physics, Kv imaging
Purpose Radiological images contain much information that exceeds what can be visually perceived. In this context, radiomics enables the extraction of detailed features and biomarkers from images using automated algorithms, to serve as decision-support tools in the clinical management of various pathologies [1,2]. Despite being an active field of research for years, radiomics remains in the investigative stage, with the ultimate objective of integrating this discipline into clinical practice. This aligns with the effort in medicine to advance personalized healthcare [3].One...
Read more Methods and materials For this study, a CCR (Credence Cartridge Radiomics) phantom, analogous to the one used by Mackin et al. [8], was used, as shown in Figure 1.  The phantom consists of ten materials enclosed in an acrylic casing, specifically: plaster, wood, cork, rubber, polyurethane, PMMA (a highly transparent thermoplastic polymer), and ABS (Acrylonitrile Butadiene Styrene) at 20%, 30%, 40%, and 50%. [fig 1] Image acquisition was performed using the Gemstone Spectral Imaging (GSI) computed tomography scanner from General Electric. A total of seven...
Read more Results In terms of repeatability, as the energy of the reconstruction increased, the percentage of features with excellent ICC also increased, ranging from 80% to 95% of the analyzed features meeting this criterion. The results are shown in Figure 4.   [fig 4] In general, the ICC was excellent for 76.9% to 95% of the features, both in single-energy acquisitions and in reconstructions derived from dual-energy acquisitions.When comparing kV-Like reconstructions, it can be observed that as the simulated kilovoltage increases, the number of...
Read more Conclusion Following the described methodology, it was determined that the GLSZM, GLDM, and NGTDM groups do not contain any features that meet the established robustness criteria. In contrast, the First-Order and GLRLM groups each had two features that met the criteria in some of the analyses. Overall, radiomic features demonstrated specific repeatability and reproducibility when extracted from a phantom using fsCT. The results obtained are consistent with the reliability of these features in pCT. Additionally, this methodology can be applied to...
Read more References 1. Yip, S. S., & Aerts, H. J. Applications and limitations of radiomics. Physics in Medicine & Biology, 2016; 61(13), R150.2. Rizzo S, Botta F, Raimondi S, Origgi D, Fanciullo C, Morganti AG, et al. Radiomics: the facts and the challenges of image analysis. Eur Radiol Exp. 2018; 2(1):36.3. Mayerhoefer, M. E., Materka, A., Langs, G., Häggström, I., SzczypiÅ„ski, P., Gibbs, P., & Cook, G. Introduction to radiomics. Journal of Nuclear Medicine, 2020; 61(4), 488-495.4. Van Timmeren JE, Cester D,...
Read more
GALLERY