. This project is motivated by two intersecting new needs. First, rapid advances in imaging technology are outpacing evaluation procedures that ensure quality and safety. Second, radiology research has a newfound focus on quantitative imaging biomarkers and radiomics. For both these trends of evaluating new technology and patient-specific assessments, we need a new generation of imaging phantoms that can provide clinically relevant gold standards. To meet these unmet needs, phantoms should ideally have a range of body habitus representing the diversity of patients, their organs should have realistic shape and texture, and detection tasks such as lesions should similarly be realistic in morphology with known contrast uptake. In other words, we would like to have patient-like anatomy with phantom-like gold standards. We propose a new approach to designing imaging phantoms that will be highly customizable to the unique needs of researchers, clinical physicists, and radiologists. These phantoms will incorporate new rapid prototyping or ?3D printing? techniques, as well as new tissue-equivalent materials with or without uptake of contrast. If successful, these phantoms will boast voxel-level ground truth that would permit quantitative radiomics measurements, task-based assessment of performance, or patient-specific dosimetry. Feasibility of our new approach will be demonstrated with 3 aims: (1) Develop new 3D printing devices and procedures to deliver high resolution, voxelized fabrication using graded blends of multiple materials. (2) Design new 3D printing materials that can match imaging characteristics of normal tissue as well as lesions with contrast uptake. (3) Create prototype phantoms for CT imaging that demonstrate these abilities for a diverse range of modalities including dual energy CT and CT angiography, as well as applications including reconstruction optimization, task-based assessment, and radiomics. This work has many key innovations: voxelized printing of 3D volume, graded blending of multiple materials at the voxel level, new materials by nanoparticle doping or by dissolving salts, practical creation of tissue equivalent materials with common components, printing with 3 or 4 materials simultaneously, and anthropomorphic phantoms based on human subject image data. There will be significant impact by enabling practical yet customized creation of phantoms to suit every researcher?s unique needs. We envision an ?open source? philosophy by embracing publicly available materials, software, and procedures, with the goal of enabling every research group to create highly customizable phantoms. Such technology used to be possible only with very time-consuming, expensive custom jobs achievable only by one or two manufacturers, but we intend to share our innovations to encourage widespread adoption with the research community. This work is a close collaboration between researchers from radiology (MPI Lo and investigator Samei), chemistry (MPI Wiley), and electrical & computer engineering (investigator Gehm).

Public Health Relevance

Computed tomography (CT) and other imaging technology have advanced rapidly, paving the way to transition from merely qualitative descriptions of images toward quantitative imaging biomarkers. We propose a new approach to create ?phantoms? or test instruments that use 3D printing to mimic human anatomy. By making phantoms realistic, customizable, and widely available, our study will allow researchers and physicians to test new imaging systems in ways that are clinically relevant. These phantoms may therefore improve the safety and efficacy of medical imaging, thus directly contributing to the improved diagnosis and treatment of many diseases.

National Institute of Health (NIH)
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Biomedical Imaging Technology Study Section (BMIT)
Program Officer
Shabestari, Behrouz
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Duke University
Schools of Medicine
United States
Zip Code