ESTRO 2025 - Abstract Book

S3821

Physics - Radiomics, functional and biological imaging and outcome prediction

ESTRO 2025

Results: Fig. 1 shows mean HU for each cartridge in the original CCR and 3D printed replica after scanning in-house, as well as the original TCIA scan. TCIA and in-house scans of the original CCR matched well. 3D printed attenuation differed from the original by an average of 63 ± 34 HU across all cartridges (range: 16.2 HU [rubber] - 122.1 HU [ABS 50%]). For the original CCR, 74.1% of features had an ICC above 0.8 in test-retest analysis, whereas the 3D printed had a higher number of reliable features (88.9%). ABS 30%, ABS 40%, and particular improvement was observed for plaster (Table 1).

Conclusion: 3D printing is a viable technique for recreating a radiomics phantom due to the ability to replicate cartridges. Insights gained from this study could help standardise future phantom designs for the cross-comparison of radiomics outputs across Institutions.

Keywords: radiomics, 3D printing, phantoms References:

1. Mackin, D., Fave, X., Zhang, L., Fried, D., Yang, J., Brian Taylor, Rodriguez-Rivera, E., Dodge, C., Jones, A. K., & Court, L. (2015). Measuring computed tomography scanner variability of radiomics features. Investigative Radiology, 50 (11), 757–765. https://doi.org/10.1097/RLI.0000000000000180 2. Xue, C., Yuan, J., Lo, G. G., Chang, A. T. Y., Poon, D. M. C., Wong, O. L., Zhou, Y., & Chu, W. C. W. (2021). Radiomics feature reliability assessed by intraclass correlation coefficient: A systematic review. Quantitative Imaging in Medicine and Surgery, 11 (10), 4431–4460. https://doi.org/10.21037/qims-21-86