What Are Phantoms?

Phantoms are physical or computational objects designed to mimic the properties of human tissue for the purpose of testing, calibrating, and validating medical imaging systems and radiation dosimetry equipment. In radiology, nuclear medicine, and radiation therapy, a phantom stands in for a patient during quality-assurance measurements, scanner performance evaluations, and protocol development, eliminating the ethical constraints and biological variability that would complicate direct human measurements. The field draws on medical physics, materials science, and biomedical engineering, and has evolved from simple water-filled tanks to highly detailed anatomical models that replicate the density, composition, and geometry of specific organs or body regions.

Phantoms serve two broad categories of purpose: they provide a reproducible reference object for measuring instrument performance, and they allow researchers to derive absorbed dose or image quality metrics under controlled conditions that can be repeated across sites and time points.

Physical Phantom Construction and Materials

Physical phantoms are three-dimensional objects fabricated from materials chosen to match the radiation-attenuation or acoustic properties of human soft tissue, bone, lung, or fat at clinically relevant energies. Water and water-equivalent plastics such as polymethyl methacrylate (PMMA) and solid water are standard for X-ray and computed tomography work because their mass attenuation coefficients closely match those of muscle tissue. For positron emission tomography (PET) and single-photon emission computed tomography (SPECT), fillable phantoms made of acrylic are loaded with known concentrations of radiotracer so that scanners can be cross-calibrated against dose calibrators. Bone-equivalent and lung-equivalent inserts extend the range of a single phantom to cover heterogeneous anatomy. Geometric phantoms, such as the Jaszczak phantom used in SPECT quality control, contain fillable spheres of varying diameter to test contrast and resolution simultaneously. A 2016 study published in Medical Physics on PET phantom design for detectability assessment demonstrated how sphere-in-cylinder designs can characterize scanner sensitivity to small lesions across a range of signal-to-background ratios.

Computational Phantoms and the XCAT Series

Computational phantoms are mathematical or voxel-based models of the human body used in Monte Carlo radiation transport simulations and image reconstruction research. The eXtended CArdiac-Torso (XCAT) phantom series, developed at Johns Hopkins University, provides a parametric anatomical model spanning newborns through adults, with realistic cardiac and respiratory motion encoded as deformable surfaces. Research from the Johns Hopkins XCAT phantom program shows these models used to benchmark reconstruction algorithms, simulate patient-specific dosimetry, and develop motion-correction methods without any radiation burden to real subjects. Voxel phantoms derived from segmented CT or MRI datasets of individual patients represent a further refinement, allowing dose estimates to reflect actual anatomy rather than an idealized model.

Dosimetry and Imaging Quality Assurance

Dosimetry is the primary driver of phantom use in clinical settings. Regulatory standards and accreditation programs require that PET/CT systems maintain calibration accuracy within defined tolerances, typically 5–10% for common radionuclides, and phantoms loaded with a precisely measured radioactivity provide the reference against which scanner output is compared. The European Federation of Organisations for Medical Physics (EFOMP) protocol for PET/CT and PET/MR quality control specifies uniform cylindrical phantoms for cross-calibration, image uniformity checks, and system sensitivity verification. In diagnostic X-ray, anthropomorphic phantoms simulate patient geometry for organ dose estimation under clinical acquisition protocols.

Applications

Phantoms have applications in a range of fields, including:

• Radiation therapy planning, where patient-specific phantoms verify dose delivery before treatment
• PET and SPECT scanner acceptance testing and routine quality control
• X-ray and CT imaging protocol optimization and dose reduction research
• Training radiologists and technologists on equipment operation
• Development and validation of image reconstruction and AI-based analysis algorithms