(Tuesday, 7/14/2020) 2:00 PM - 3:00 PM [Eastern Time (GMT-4)]
Room: Track 1
Purpose: Inkjet-printers can be used to fabricate anthropomorphic phantoms. However, this method has two drawbacks: limited gray-value resolution and printing artifacts due to printer stress. Our purpose was to develop an inkjet-printing based framework to overcome these limitations for 2D x-ray imaging applications.
Methods: A phantom was created from inkjet-printed paper and designed to model specific patient anatomy. Printing was done with a commercial inkjet-printer (CanonPIXMA-G1200) using ink made of KI dissolved in deionized water at 1g/cc. A calibration for the attenuation of ink, printed as pure-black pixels (0 pixel-value) on a single sheet of paper was measured using a commercial x-ray system at 70kVp. A neonate radiograph obtained at 70kVp was converted to an x-ray attenuation map that served as the model for phantom printing. The attenuation map was and converted into a series of binary images using the ink attenuation calibration. Each binary image was printed on 90g/m2 bond paper, 3 passes per page, and then the entire stack imaged at 70kVp. The printing accuracy was evaluated by comparing attenuation values between the printed phantom and the model.
Results: The phantom required 36 pages. The entire manufacturing process was automated and took ~2 hours. Qualitatively, the x-ray image of the printed phantom demonstrates a close resemblance to the original neonate x-ray. No printer stress artifacts were identified in the image. A quantitative comparison between the attenuation (µL) values of the printed phantom and that of the model demonstrate differences between 3 and 12% relative to the phantom maximum attenuation (µL=1.7).
Conclusion: A framework to create compact and anatomically realistic phantoms from patient radiographs has been developed. This method is fast and cost-effective, requiring only a single-value calibration, and can easily be adapted to different models. By printing in binary-mode, it was possible to overcome the limitations described above.
Funding Support, Disclosures, and Conflict of Interest: This project received funding from the Department of Radiology Hodges Society, the Women's Board of the University of Chicago and the NIH T-32 Training Grant T32 EB002103
Keywords
Not Applicable / None Entered.
Taxonomy
IM- X-Ray: Phantoms - physical
Contact Email
