Purpose: The purpose of this study is to quantify the temporal stability and energy dependence of a compact, iodine inkjet-printed calibration phantom in order to verify the feasibility of producing size-specific anthropomorphic phantoms for optimizing radiography and fluoroscopy imaging of pediatric patients.
Methods: A 1200 mg/mL solution of potassium iodide in deionized water was used in an inkjet-printer to create a 35 square calibration pattern of varying grayscale values. Two x-ray systems, an XRAD 225cx (Precision X-ray) and Discovery XR 656 (GE Healthcare), were used to acquire CT and radiography images of the printed materials. The calibration pattern was imaged at energies 40 - 100 kVp across a 19-week period to determine both energy dependence of the phantom and temporal stability of liquid ink solutions and printed phantoms. To assess the statistical significance of the median differences, a Wilcoxon signed-rank test was performed with a 5% significance level.
Results: Temporal stability of the liquid ink solution was tested after 16 weeks. The original calibration pattern was tested after 19 weeks. Median differences in HU were small (56.65 HU and -46.24 HU respectively). This difference was not statistically significant for the liquid ink (p=0.06), but it was significant for the printed pattern (p<0.05). This indicates that the phantom will need additional protection after printing. The calibration pattern validates the inverse relationship between attenuation and energy.
Conclusion: An inkjet-printed phantom demonstrates time-dependent loss of iodine, requiring additional protection after printing. The attenuation of the printed iodine solution demonstrates the expected relationship between iodine density and energy. These results can be used to swiftly and inexpensively print temporally stable compact anthropomorphic phantoms based on patient radiographs.
Funding Support, Disclosures, and Conflict of Interest: This project was supported by the University of Chicago Radiation Oncology Department, a grant from the Hodges Society of the Department of Radiology, the NIH T32 Training Grant (LD), and Varian Medical Systems (EP). HAH receives royalties and licensing fees for computer-aided diagnosis technology through the University of Chicago.