Share Email Print
cover

Proceedings Paper

Second generation anthropomorphic physical phantom for mammography and DBT: Incorporating voxelized 3D printing and inkjet printing of iodinated lesion inserts
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Physical phantoms are needed for the evaluation and optimization of new digital breast tomosynthesis (DBT) systems. Previously, we developed an anthropomorphic phantom based on human subject breast CT data and fabricated using commercial 3D printing. We now present three key advancements: voxelized 3D printing, photopolymer material doping, and 2D inkjet printing of lesion inserts. First, we bypassed the printer’s control software in order to print in voxelized form instead of conventional STL surfaces, thus improving resolution and allowing dithering to mix the two photopolymer materials into arbitrary proportions. We demonstrated ability to print details as small as 150μm, and dithering to combine VeroWhitePlus and TangoPlus in 10% increments. Second, to address the limited attenuation difference among commercial photopolymers, we evaluated a beta sample from Stratasys with increased TiO2 doping concentration up to 2.5%, which corresponded to 98% breast density. By spanning 36% to 98% breast density, this doubles our previous contrast. Third, using inkjet printers modified to print with iopamidol, we created 2D lesion patterns on paper that can be sandwiched into the phantom. Inkjet printing has advantages of being inexpensive and easy, and more contrast can be delivered through overprinting. Printing resolution was maintained at 210 μm horizontally and 330 μm vertically even after 10 overprints. Contrast increased linearly with overprinting at 0.7% per overprint. Together, these three new features provide the basis for creating a new anthropomorphic physical breast phantom with improved resolution and contrast, as well as the ability to insert 2D lesions for task-based assessment of performance.

Paper Details

Date Published: 22 March 2016
PDF: 10 pages
Proc. SPIE 9783, Medical Imaging 2016: Physics of Medical Imaging, 978360 (22 March 2016); doi: 10.1117/12.2217667
Show Author Affiliations
Dhiraj Sikaria, Duke Univ. (United States)
Carl E. Ravin Advanced Imaging Labs., Duke Univ. School of Medicine (United States)
Stephanie Musinsky, Duke Univ. (United States)
Carl E. Ravin Advanced Imaging Labs., Duke Univ. School of Medicine (United States)
Gregory M. Sturgeon, Carl E. Ravin Advanced Imaging Labs., Duke Univ. School of Medicine (United States)
Justin Solomon, Carl E. Ravin Advanced Imaging Labs., Duke Univ. School of Medicine (United States)
Duke Univ. (United States)
Andrew Diao, Duke Univ. (United States)
Michael E. Gehm, Duke Univ. (United States)
Ehsan Samei, Duke Univ. (United States)
Carl E. Ravin Advanced Imaging Labs., Duke Univ. School of Medicine (United States)
Stephen J. Glick, U.S. Food and Drug Administration (United States)
Joseph Y. Lo, Duke Univ. (United States)
Carl E. Ravin Advanced Imaging Labs., Duke Univ. School of Medicine (United States)


Published in SPIE Proceedings Vol. 9783:
Medical Imaging 2016: Physics of Medical Imaging
Despina Kontos; Thomas G. Flohr, Editor(s)

© SPIE. Terms of Use
Back to Top