biomimetics

Article
3D Printing Materials Mimicking Human Tissues after Uptake of
Iodinated Contrast Agents for Anthropomorphic
Radiology Phantoms
Peter Homolka 1, * , Lara Breyer 2 and Friedrich Semturs 1

1 Center for Medical Physics and Biomedical Engineering, Medical University of Vienna, 1090 Vienna, Austria
2 Department of Biomedical Imaging and Image-Guided Therapy, Medical Imaging Cluster (MIC),
Medical University of Vienna, 1090 Vienna, Austria
* Correspondence: [Link]@[Link]

Abstract: (1) Background: 3D printable materials with accurately defined iodine content enable the
development and production of radiological phantoms that simulate human tissues, including lesions
after contrast administration in medical imaging with X-rays. These phantoms provide accurate,
stable and reproducible models with defined iodine concentrations, and 3D printing allows maximum
flexibility and minimal development and production time, allowing the simulation of anatomically
correct anthropomorphic replication of lesions and the production of calibration and QA standards
in a typical medical research facility. (2) Methods: Standard printing resins were doped with an
iodine contrast agent and printed using a consumer 3D printer, both (resins and printer) available
from major online marketplaces, to produce printed specimens with iodine contents ranging from
0 to 3.0% by weight, equivalent to 0 to 3.85% elemental iodine per volume, covering the typical
levels found in patients. The printed samples were scanned in a micro-CT scanner to measure the
properties of the materials in the range of the iodine concentrations used. (3) Results: Both mass
density and attenuation show a linear dependence on iodine concentration (R2 = 1.00), allowing
highly accurate, stable, and predictable results. (4) Conclusions: Standard 3D printing resins can
be doped with liquids, avoiding the problem of sedimentation, resulting in perfectly homogeneous
prints with accurate dopant content. Iodine contrast agents are perfectly suited to dope resins with
Citation: Homolka, P.; Breyer, L.; appropriate iodine concentrations to radiologically mimic tissues after iodine uptake. In combination
Semturs, F. 3D Printing Materials
with computer-aided design, this can be used to produce printed objects with precisely defined
Mimicking Human Tissues after
iodine concentrations in the range of up to a few percent of elemental iodine, with high precision
Uptake of Iodinated Contrast Agents
and anthropomorphic shapes. Applications include radiographic phantoms for detectability studies
for Anthropomorphic Radiology
and calibration standards in projective X-ray imaging modalities, such as contrast-enhanced dual
Phantoms. Biomimetics 2024, 9, 606.
[Link]
energy mammography (abbreviated CEDEM, CEDM, TICEM, or CESM depending on the equipment
biomimetics9100606 manufacturer), and 3-dimensional modalities like CT, including spectral and dual energy CT (DECT),
and breast tomosynthesis.
Academic Editor: Erina Baynojir Joyee

Received: 30 August 2024 Keywords: radiology phantoms; medical imaging; contrast-enhanced mammography; computed
Revised: 27 September 2024 tomography; quality control; iodine contrast; anthropomorphic phantoms; tissue mimicking materials;
Accepted: 2 October 2024 additive manufacturing; 3D printing
Published: 8 October 2024

1. Introduction
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland. Phantoms serve as physical [1] or computerized models [2,3] that simulate human
This article is an open access article tissues and organs in various applications from hardware evaluation, optimization of proce-
distributed under the terms and dures to system testing and quality assurance. These specialized objects also play a crucial
conditions of the Creative Commons role in various aspects of medical imaging, including research, development, and regular
Attribution (CC BY) license (https:// quality control, all serving vital roles in patient care optimization. Especially when the
[Link]/licenses/by/ phantom design involves realistic structures and details, additive manufacturing and 3D
4.0/). printing are indispensable, offering the ability to produce phantoms with patterns with a

Biomimetics 2024, 9, 606. [Link] [Link]

Biomimetics 2024, 9, 606 2 of 12

high degree of realism [4–6]. Setting it apart from a simple test object, a phantom is defined
as a model of the human body or body part designed to replicate specific features or inter-
actions relevant to the specific application, e.g., a medical imaging procedure. 3D-printed
phantoms are used across all diagnostic and therapeutic radiologic modalities, includ-
ing ultrasound, MRI, CT, mammography, nuclear medicine, and radiation therapy [5,7,8].
The primary purpose of phantoms in medical imaging with ionizing radiation and radi-
ation therapy is to evaluate and optimize imaging devices and procedures, facilitating
virtual trials, managing protocols, assessing diagnostic performance, conducting dosimetry
studies [9], and ensuring that radiation safety standards are met [10,11].
Phantoms can range from simple to highly complex, depending on their intended
application. Simple phantoms include basic slabs or stylized objects that mimic typical
patient attenuation in medical X-ray fields, while anthropomorphic phantoms are complex
task-specific models designed to study the effects of dose level variation, radiation qual-
ity, or system specifications on diagnostic performance. Since no real anthropomorphic
phantom that mimics all the relevant properties is yet available, usually, procedure/task
specific pseudo-anthropomorphic phantoms are designed and used, mimicking the most
relevant properties with a wide range of finesse. Digital phantoms serve as computerized
models used for dosimetry, in silico testing, and virtual imaging trials [2,12,13]. Another
dedicated family of phantoms is represented by calibration phantoms, which are utilized
for quantitative imaging to calibrate results, and/or perform accuracy tests of clinical
data [14–19].
The construction of phantoms involves materials that mimic the properties of human
tissues or well-defined materials in relation to the specific imaging modality. For X-ray
imaging, materials must replicate the modification of radiation fields caused by absorption
and scattering in body tissues [20,21].
As imaging technologies advance, there is a growing need for more sophisticated
“super phantoms” that can reproduce in vivo imaging features with high fidelity, combine
digital and physical elements for comprehensive testing, serve as ground truth for various
properties and imaging biomarkers, and facilitate the development and validation of
machine-learning algorithms in medical imaging.
Additive manufacturing, particularly 3D printing, has revolutionized phantom pro-
duction by offering flexibility in design and cost efficiency, and an accurate representation
of the anatomy and tissue properties [22]. One of the most significant constraints associated
with the 3D printing of radiographic phantoms, test objects, and calibration standards is the
restricted range of printable polymers. In addition, these commercially available materials
are usually designed for different applications, with optimization mainly occurring in terms
of mechanical properties, user-friendliness or the visual appearance of the printed objects.
A variety of studies has been undertaken to test and establish the radiologic equivalence
of available printing materials with tissues and standard materials like liquid water for a
variety of X-ray photon energies, from mammography to CT and therapeutic photon ener-
gies in radiotherapy [23–31]. Nevertheless, challenges remain in developing 3D-printable
materials that accurately mimic tissue properties across a wide range of imaging modalities
and energy levels.
Iodine plays a key role in X-ray imaging as a contrast agent, primarily due to its
high atomic number and K-edge at 33.2 keV closely matching the energy range of X-rays
used in diagnostic imaging, resulting in strong X-ray attenuation due to the photoelectric
effect. Iodine-based contrast agents are used in a variety of imaging modalities, including
computed tomography (CT), fluoroscopy, angiography, and interventional radiology. By
combining images taken at different photon energies in DECT (dual energy CT) [32] and
CEDEM (contrast enhanced dual energy mammography [33]), iodine-enhanced tissues or
lesions can be isolated from the anatomical background and/or iodine uptake quantified.
Typical iodine uptake in CT examinations depends on the contrast medium application
protocol, as well as patient-related factors, such as age, gender, and body mass index [34]. In
healthy subjects, typical iodine concentrations in the tissues of interest in standard contrast-
Biomimetics 2024, 9, 606 3 of 12

enhanced CT scans range from typically 1 to 6 mg iodine per ml tissue [34]. Depending
on the scan protocol (kV used in standard scans, or keV reconstruction energy in pseudo-
monochromatic reconstructions), the voxel value increases by typically >20 (140 kV) to
>45 HU (80 kV) per mg Iodine/mL, with a value of approximately slightly above 25 HU
per mg/mL iodine in standard scan conditions (120 kV) [35]. Thus, the range of iodine
concentrations in typical phantoms is expected to be in the range of <1 to a few percent
of iodine.
In contrast to CT, where iodine content is quantified volumetrically (in mg/mL),
in 2D imaging modalities (like general radiography, dual energy mammography except
breast tomosynthesis, and fluoroscopy including angiography), only total attenuation
is represented in the pixel values. Therefore, iodine content needs to be specified as
the iodine area weight (also referred to as area density, corresponding to the product
of volume density and thickness) typically provided in mg/cm2 . While in angiography
and interventional cardiology, high iodine concentrations are used to visualize dynamic
processes that are best replicated in phantom models by using liquid iodine solutions,
contrast-enhanced mammography (CEDEM) represents a modality where very accurate
low iodine concentrations in phantoms are necessary. The field of applications in CEDEM
ranges from system testing and the optimization of image-processing algorithms [36] to
regular quality control. Necessary iodine concentrations for CEDEM phantoms typically
range from 0.25 to 2 mg/cm2 iodine [19,37]. If these phantom details were to be printed
with 1 mm of thickness, the corresponding area weights necessitate an iodine concentration
(w/w) of approximately 0.2 to 1.6 mg/mL. In patient images, iodine enhancement in lesions
was shown to exceed 0.5 mg/mL iodine with a peak at 1 mg/mL in the great majority of
clinical cases, with only a few single cases exhibiting enhancements greater or equal to
3 mg/mL [38].
Phantoms containing iodine in appropriate concentrations thus are not only important
for research and development, but will find their application in the regular quality assur-
ance and continuing optimization of both acquisition protocols and imaging devices. Most
of these phantoms use cavities filled with liquid solutions containing the desired iodine
concentration(s) [39–43]. However, solid iodine contrast phantoms offer several advantages
over traditional liquid-filled phantoms. While liquid-filled phantoms or phantom inserts
allow for adjustable iodine concentrations, they present challenges such as potential evapo-
ration, leakage, and bubble formation during refilling. These issues can limit long-term
stability and complicate use when concentration adjustments are not required.
To address these limitations, researchers have explored solid phantom designs in-
corporating iodinated materials. One approach is to embed iodobenzene in an epoxy
resin matrix [44]. Production involves handling toxic and potentially difficult-to-access
chemicals (like iodobenzene). Another option is to incorporate iodine (as Iohexol powder)
into paraffin-based formulations that mimic tissue- and water-equivalent materials [45].
Epoxy resin-based phantom materials have a long history in tissue- and water-equivalent
materials for imaging and radiotherapy applications [46]. However, the production method
requires careful manufacturing techniques to prevent air bubble formation and sedimenta-
tion, which are typically achieved by vacuum mixing and the control of viscosity. When
produced under optimum conditions, both of these, epoxy- and paraffin-based phantoms
proved effective for various imaging applications, including CT and radiotherapy. How-
ever, none of these materials can be used for additive manufacturing in combination with
3D printing.
By providing multiple fixed iodine concentrations in a single stable unit, solid io-
dine phantoms provide a convenient and reliable tool for imaging research and quality
assurance [19]. These phantoms eliminate concerns about evaporation, leakage, and air
entrapment associated with liquid-filled designs, while still allowing the simultaneous
imaging of multiple iodine concentrations.
The aim of this work was to develop, test, and verify a method to produce easily
available printing materials with exactly definable iodine concentrations, that are chemically
Biomimetics 2024, 9, 606 4 of 12

stable, cheap and readily available, can easily be produced in average medical research
settings, and can be printed with consumer grade 3D printers.

2. Materials and Methods

Commercially available 3D-printing resins, which are aimed at the mass market of
hobbyists and are available in a wide range on internet marketplaces such as Amazon, have
been doped with an iodine contrast agent commonly used in diagnostic X-ray imaging
and printed on a consumer grade resin printer to produce samples with appropriate iodine
concentrations mimicking those of tissues after contrast agent administration. Samples
were then evaluated in a micro-CT scanner to measure the X-ray attenuation dependent on
iodine concentrations. The reproducibility and stability of the process was evaluated by
reprinting samples with a different batch of the printing resin.

2.1. Doping of Printing Polymers

Standard 3D-printing polymers targeting professional and hobbyist users were ob-
tained from major online marketplaces. Resins were mixed with 5 to 10 percent contrast
media and allowed to settle overnight in a first test run, and for extended time periods in
a second test round. The resins that either gelled as a reaction to the contrast agent intro-
duced or where the iodine complex in the contrast agent precipitated as a white powder
were discarded.
As contrast agent, the formulation with the highest available iodine concentration
used at our university hospital was chosen (Iomeron 400, Bracco Imaging S.p.A, Milan,
Italy). It contains Iomeprol (C17 H22 I3 N3 O8 ) with 400.0 mg iodine per mL solution. From
the chemical composition, the elemental iodine content in Iomeprol is calculated as 48.992%.
To calculate mixing ratios with the resin, the mass density of the liquid contrast agent
was measured using a self-calibrating analytical laboratory balance (Sartorius, Göttingen,
Germany) and a calibrated pycnometer with a nominal volume of 25.131 mL. The volume
of the pycnometer was verified by measuring deionized water at 22 ◦ C, resulting in a
correction of 0.1%. The density of the contrast agent was determined at the laboratory
temperature of 22 ◦ C as 1.450 ± 0.0014 [Link]−3 . Using this density and the elemental iodine
content of the contrast agent, the iodine content translates from 400.0 mg/mL to 0.2759 g
iodine per g Iomeron 400. Using these numbers, the mixing rations of the base resin with
the contrast agent were calculated. The doping process in the chemical lab was performed
by weighing the desired amount of resin into 500 mL light–tight wide neck PE laboratory
bottles and adding the liquid contrast agent with a Pasteur pipette on the balance with sub
mg accuracy. Then, the bottles were closed and agitated by hand to ensure the appropriate
mixing of the two liquids.

2.2. Determination of Mass Density of Printed Samples

In volumetric imaging modalities like CT, iodine concentrations are specified in mg
iodine per ml (or cm3 ). However, the volume of uncured resin does not exactly equal the
volume of the cured resin, since polymerization during the printing process is associated
with (minimal) volume loss due to shrinkage. In addition, resin/contrast agent mixtures
are prepared and specified as iodine per weight (% w/w) rather than per volume, because
mixing components in percent by weight implies a much smaller uncertainty than mixing
as mass per volume (% w/v) in a research laboratory setting.
To determine the mass density as a function of iodine content, three cylinders each with
a 1.5 cm diameter and a 2 cm height were printed with iodine concentrations of 0, 1, 1.5, and
2% iodine by weight, corresponding to 0 to 7.250% w/w of liquid Iomeron 400, respectively.
To calculate the mass densities, the weight was determined using a self-calibrating analytical
laboratory balance. To calculate the volume of the cylinders, the height and diameter
were measured using a digital caliper (Mitutoyo, Kawasaki, Japan). Measurements were
taken 3 times at different positions (height) and 6 times (diameter) for each sample at
perpendicular angles and different heights (top/middle/bottom), then averaged.
Biomimetics 2024, 9, 606 5 of 12

2.3. Samples for Measurement of Attenuation vs. Iodine Content

Since the samples printed for the determination of mass density (up 2% iodine content
w/w) printed identically to the undoped resin, the iodine concentration was increased to
2.5% (corresponding to 9.063% Iomeron 400), and further to 3% iodine (10.875% Iomeron
400), respectively. Cylindrical samples with three sections, each 10 mm long, and 8, 5, and
3 mm in diameter, respectively, were printed with all the iodine concentrations studied. For
the determination of the dependence of the measured attenuation on iodine content in the
micro-CT scanner, the rod sections with 8 mm diameter and 10 mm length were used. These
were produced in 2 printing runs; in the first printing run, the samples for scanning were
printed together with the larger samples used for the determination of mass density (0 to
2% iodine content). In a second run, 2% iodine was repeated using a different (re-ordered)
batch of the base resin to test the reproducibility and 2.5 and 3% iodine containing samples
were added. For each concentration except 2% iodine, two samples were scanned in the
micro-CT scanner. For 2% iodine, 4 samples were used; two from each printing run were
scanned to compare the results from the two different mixing and printing cycles using
two different batches of resin.
Samples were printed on a consumer-grade (M)SLA printer (Elegoo Mars 4 Ultra,
Elegoo Inc., Shenzhen, China), employing a 7-inch 9K Mono LCD screen with a resolution
of 8520 × 4320 pixels. Printing parameters were determined with the undoped and doped
resin, applying the Photonsters Validation Matrix V2 [47]. However, no changes in the
exposure parameters between the doped and the stock resin were found; thus, exposure
parameters for the stock resin were used.
After printing, the samples were spray-cleaned with a sequence of 100% IPA (isopropyl
alcohol) followed by 30% IPA and tap water, then washed with liquid hand soap to remove
the last traces of uncured resin, followed by rinsing off with deionized water. After drying
for a minimum 24 h, the samples were post-cured in a curing chamber (Wash and Cure,
Anycubic, Shenzhen Anycubic Technology Co. Ltd., Shenzhen, China) for 3 min.
Printing models were designed in Fusion 360 (Autodesk Inc., San Francisco, CA,
USA) and STL files were sliced with Chitubox V1.9.5 (Shenzhen CBD Technology Co. Ltd.,
Shenzhen, China).

2.4. CT Scanning of Iodine Samples

For scanning, the printed samples were mounted into cylindrical phantoms 6 cm
in diameter and 7 cm in length, respectively (Figure 1, panel (a)). This phantom was
printed with a PLA filament (ecoPLA Tough transparent, 3DJake, Niceshops GmbH, Paldau,
Austria) with an FDM printer (Flsun V400, Zhengzhou Chaokuo Electronic Technology
Co., Ltd., Zhengzhou, China) using a 0.8 mm nozzle and 30% gyroid infill, guaranteeing
the structural stability of the phantom while allowing for minimal X-ray attenuation in the
scanning process, still ensuring correct working of the beam-hardening correction. In the
center, a 2 mL syringe filled with deionized water for the comparison of image noise and
normalization of CT voxel values was placed.
CT scanning was performed in a Siemens Inveon Trimodality scanner (Inveon Tri-
modality imaging system, Siemens Healthineers, Erlangen, Germany), applying a CCD
detector combined with a variable focal spot X-ray source, producing an image volume
of 1024 × 1024 × 1024 isotropic voxels with 0.09753 mm3 size. The scan was performed
with 70 kV and 500 µA tube current in Hires Mode with a static table. The reconstruction
was performed using the Feldkamp cone beam algorithm, employing a Shepp filter with a
cut-off at the half-Nyquist frequency. Since image values are not provided in Hounsfield
units (HU) with zero indicating no attenuation rather than water, voxel values of water
were measured in the water-filled syringe in the center of the phantom. Additionally, a
ROI (number 9 in Figure 1b) outside the phantom was placed in the air to verify the scaling
of voxel values. The voxel value histogram measured in the air exhibited a strong peak at
zero, followed by a half-sided Gaussian distribution corresponding to the normalization of
the voxel values to zero at no measured attenuation.
 Biomimetics
Biomimetics 2024, 9, x9,FOR
2024, 606 PEER REVIEW 6 of 612
of 12

(a) (b)

(c) (d) (e)

Figure 1. (a) PLA phantom of 6 cm diameter used for scanning of printed samples. View from above
Figure 1. (a) PLA phantom of 6 cm diameter used for scanning of printed samples. View from above
and below showing individual samples, and a 2 mL syringe in center filled with deionized water.
and below showing individual samples, and a 2 mL syringe in center filled with deionized water.
(b) Single slice of the micro-CT image volume, indicating placement of ROIs used for determination
(b) Single
of voxel slice(c–e):
values. of theTransverse,
micro-CT image volume,
coronal, indicating
and sagittal placement
reformat of CTofvolume
ROIs used for determination
showing samples
of voxel values. (c–e): Transverse, coronal, and sagittal reformat of CT volume
inside phantom. (c) Transversal reformat through section showing 6 cylinders with 8 mm diameter showing samples
inside
used phantom. (c)voxel
for determining Transversal
values reformat through
with iodine section showing
concentrations from 0 6tocylinders withand
2.5% iodine, 8 mm diameter
water in
theused
center.
for In (d,e) the water-filled
determining voxel valuessyringe is shown
with iodine centrally. Note:
concentrations from 0phantom sections
to 2.5% iodine, andcontaining
water in the
thinner rods
center. (5 and
In (d,e) the3 water-filled
mm in diameter) seen
syringe in coronal
is shown and sagittal
centrally. reformats
Note: phantom were not
sections used in thinner
containing this
work.
rods (5 and 3 mm in diameter) seen in coronal and sagittal reformats were not used in this work.

[Link] Analysis
scanning was performed in a Siemens Inveon Trimodality scanner (Inveon Tri-
modality Theimaging
micro-CT system,
imageSiemens
volume Healthineers,
stored as 16-bit Erlangen, Germany),
integer values applying
was loaded intoa the
CCDAna-
detector combined
lyze image with a software
processing variable focal
packagespot[48]
X-ray source, 12.0),
(Analyze producing an image
Biomedical volume
Image of
Resource
1024 × 1024
(Mayo × 1024
Clinic, isotropicNY,
Rochester, voxels
USA)with
as raw0.0975
data3 mm size. The scan
and3reformatted to awas
sliceperformed
thickness of with
1 mm
70 for
kV evaluation.
and 500 µAVoxel tube current
values inin the
Hires Mode were
samples with measured
a static table. Theimage
in the reconstruction was
slices containing
performed using the
the cylindrical Feldkamp
samples withcone
an 8beam algorithm,The
mm diameter. employing
sample asections
Shepp filter
of 5 with
and 3a mmcut- in
offdiameter were not used,
at the half-Nyquist since no
frequency. beam-hardening
Since image values effects
are notfrom the iodine
provided attenuation
in Hounsfield were
units
seen
(HU) for all
with zerotheindicating
iodine concentrations
no attenuation studied,
ratherthat
thanwould
water,have necessitated
voxel a reduction
values of water were of
the iodine
measured in content in the scan
the water-filled cross-section
syringe to derive
in the center exact
of the attenuation
phantom. measurements.
Additionally, a ROI
To avoid partial volume effects, circular regions of interest with
(number 9 in Figure 1b) outside the phantom was placed in the air to verify the scaling of a 65 voxel (6.3 mm)
diameter slightly smaller than the cylinders were applied (Figure
voxel values. The voxel value histogram measured in the air exhibited a strong peak at 1b), and the top and
bottom
zero, slicesby
followed were omitted. Gaussian
a half-sided Depending on the sample,
distribution 8 or 9 artefact
corresponding freenormalization
to the slices remained,
andvoxel
of the the voxel
valuesvalues of the
to zero individual
at no measured slices were averaged. ROIs were also placed into the
attenuation.
filled syringe at the center to measure the voxel value of water for comparison. In addition
[Link]
the Analysis
average values, the STDs of average voxel values in the individual slices were also
determined
The micro-CT and uncertainties
image volumewere calculated.
stored as 16-bit integer values was loaded into the An-
Data evaluation was performed in Microsoft Excel, Ver.16.88 Standard for Mac 2021
alyze image processing software package [48] (Analyze 12.0), Biomedical Image Resource
(Microsoft Corp., Redmont, WA, USA).
(Mayo Clinic, Rochester, NY, USA) as raw data and reformatted to a slice thickness of 1
mm for evaluation. Voxel values in the samples were measured in the image slices con-
taining the cylindrical samples with an 8 mm diameter. The sample sections of 5 and 3
mm in diameter were not used, since no beam-hardening effects from the iodine attenua-
tion were seen for all the iodine concentrations studied, that would have necessitated a
 eter are shown in Figure 2. A linear dependence (R2 = 0.9980) of density versus iodine con-
tent demonstrates the density increase with increasing iodine content due to the higher
mass density of contrast media compared to the base resin. Using the equation derived
from the regression analysis, the iodine content per mass translates to the iodine content
Biomimetics 2024, 9, 606 7 of 12
per volume.
Figure 3 indicates the voxel values measured in the micro-CT scanner vs. the iodine
content by weight (panel (a)), and by volume (panel (b)). The linear regression of voxel
3. Results
values versus iodine content per volume indicates an almost perfect (R2 = 0.9965) linear
Mass Densityof
dependence of iodine
Printedcontent
Iodine Samples and Conversion
and attenuation, to % w/v
indicating an accurate and stable produc-
Mass densities
tion process. determined
The voxel value offrom the test cylinders
the demineralized 2 cmisinshown
water heightas
and 1.5 cm
a blue in diameter
data point in
are shown in Figure 2. A linear dependence (R2 = 0.9980) of density versus iodine content
the figure. Owned to the higher mass density of the printing resin, the attenuation of the
demonstrates
resin is slightlythe density
above increase
water. with
In the increasing
printed iodine
samples, content dueof
no indications toinhomogeneities
the higher mass
density of contrast media compared to the base resin. Using the equation
were present; the average standard deviation of voxel values within each sample was derived from5.63
the
regression
and thus notanalysis,
larger the
thaniodine
in thecontent
water inper
themass translates
centrally to the
placed iodine
syringe content per volume.
(6.65).

(a) (b)
Figure
Figure 2.
2. (a)
(a) Mass
Mass density
density measured
measured in
in the test cylinders
the test cylinders as a function
as a function of
of the
the iodine
iodine content.
content. Points
Points
shown correspond to measurements from individual cylinders; (b) conversion of iodine content per
shown correspond to measurements from individual cylinders; (b) conversion of iodine content per
weight (% w/w) to iodine content per volume (% w/v) as used in tomographic imaging. Points shown
weight (% w/w) to iodine content per volume (% w/v) as used in tomographic imaging. Points shown
correspond to the calculated values using linear regression from (a).
correspond to the calculated values using linear regression from (a).

Figure 3 indicates the voxel values measured in the micro-CT scanner vs. the iodine
content by weight (panel (a)), and by volume (panel (b)). The linear regression of voxel
values versus iodine content per volume indicates an almost perfect (R2 = 0.9965) linear
dependence of iodine content and attenuation, indicating an accurate and stable production
process. The voxel value of the demineralized water is shown as a blue data point in the
figure. Owned to the higher mass density of the printing resin, the attenuation of the resin
is slightly above water. In the printed samples, no indications of inhomogeneities were
Biomimetics 2024, 9, x FOR PEER REVIEW
present; the average standard deviation of voxel values within each sample was 5.63 8 of 12
and
thus not larger than in the water in the centrally placed syringe (6.65).

(a) (b)
Figure 3. Voxel values of the samples printed with iodine concentrations from zero (base polymer)
Figure 3. Voxel values of the samples printed with iodine concentrations from zero (base polymer)
to 3% elemental iodine (w/w) corresponding to 3.85% w/v. Blue data point at zero iodine content
to 3% elemental iodine (w/w) corresponding to 3.85% w/v. Blue data point at zero iodine content
corresponds to the voxel value of deionized water. 1σ error bars are smaller than symbol size. (a)
corresponds
Voxel to iodine
value vs. the voxel value
content byof deionized
weight, water.
and (b) 1σ error
vs. iodine bars per
content are volume,
smaller converted
than symbol size.
accord-
(a) Voxel value
ing to Figure 2b. vs. iodine content by weight, and (b) vs. iodine content per volume, converted
according to Figure 2b.
With 2% w/w iodine content, four cylinders (two each) were produced with two dif-
ferent batches of resin, with different production dates indicated on the resin bottles in
independent production cycles several weeks apart. The average voxel values measured
in the individual CT slices in the cylinders produced with the different batches are not
statistically different (p > 0.40, Cohen s d < 0.15). Thus, the production process is classified
Biomimetics 2024, 9, 606 8 of 12

With 2% w/w iodine content, four cylinders (two each) were produced with two
different batches of resin, with different production dates indicated on the resin bottles in
independent production cycles several weeks apart. The average voxel values measured
in the individual CT slices in the cylinders produced with the different batches are not
statistically different (p > 0.40, Cohen’s d < 0.15). Thus, the production process is classified
as stable and independent of the resin batch.

4. Discussion
Standard printing polymers, used for either fused deposition modeling (FDM) printers
printing with solid filaments, stereolithography (SLA, including masked SLA or MSLA,
also termed LCD printing), digital light processing (DLP), or polyjet technology, which
utilizes liquid printing resins solidified in the printing process by irradiation with visible
or UV light, can be doped with functional additives. Doping printing polymers involves
adding specific functional substances to modify their properties for custom applications. In
general-purpose and industrial 3D printing, examples include adding carbon nanofibers,
nanotubes, or graphene to increase the electrical conductivity or mechanical strength,
ceramic powders to increase the thermal resistance, flame retardants, or UV resistant agents.
In medical-imaging applications, resins doped with radioactive tracers are used in nuclear
medicine phantoms [49,50] mimicking patient tissue after the uptake of radioactivity. In
X-ray imaging, several high Z materials are typically added to increase X-ray attenuation.
Bismuth-doped ABS is suggested for radiation shielding or X-ray phantoms [51] mimicking
different body tissues from soft tissues to bone. Printing iodine-doped ink with a consumer
grade inkjet printer on paper and stacking up the printed sheets may not be considered
3D printing, but it represents another possibility for producing 3D phantoms with doped
printing inks [52].
When doping printing polymers or resins with additives, there are several potential
problems that need to be addressed. While it is relatively easy to mix solid polymer pellets
or powders with the desired additives to produce custom filaments for FDM printing, this
process usually requires industrial equipment such as mills and extrusion machines, and
also requires larger batches. In most cases, these issues preclude the option to produce
custom filaments for research groups involved in the development of phantoms in a medical
environment [53].
Doping liquid resins with functional additives, on the other hand, can normally be
done in a small lab without the need of specialized equipment. Historically, epoxy resin
systems have been used to produce numerous phantom materials mimicking different body
tissues very well by mixing the resins with mineral components like CaCO3 , MgO, TiO2,
and other materials to compensate for the typically low effective atomic number of the base
resin [46]. To mimic the X-ray attenuation properties of soft tissues and liquid water, the
mass density of the resulting material needed to be lowered, which was typically achieved
by adding gas-filled phenolic microballoons. The two main issues in the production of
these phantom materials are the separation of additives, since the heavier powders tend
to sink to the bottom and the microspheres tend to float. To avoid or at least minimize
these separation effects, viscosity has to be controlled. This can be done by adding fine
polymer powders like polyethylene. The other issue, especially since the viscosity was
often increased to avoid sedimentation and the separation of the microballoons, is the
trapping of air bubbles during the mixing process. These were, in a next step, removed
in a vacuum chamber [46,54,55]. To avoid these issues, a production method based on
mixing and sintering fine polymer powders with additives was developed [56,57]. It should
be noted that these formulations, however, could easily be used—possibly with minor
adaptions to the base polymers—for the production of radiologically tissue-equivalent
filaments for FDM printing.
Modern 3D-printing resins for DLP/(M)SLA printers possess very low viscosity
to allow fast printing with high spatial resolution. As a result, mixing solid powders
into these resins tends to result in sedimentation. In standard prints, this is observed
Biomimetics 2024, 9, 606 9 of 12

with color pigments and minimized by agitating the resin before and, in some printers,
also during the printing process. While a minor issue in the additive manufacturing of
functional prototypes or decorative items, the sedimentation of powders added to control
X-ray attenuation will result in inhomogeneous phantoms exhibiting a gradient in X-ray
attenuation properties, resulting in inaccurate and inconsistent properties prohibiting
clinical use. To avoid the problem of separation, possible solutions include the use of
dopants that either dissolve in the liquid resin, or are liquid themselves and chemically
inert to the resin. In this work, it was shown that standard printing resins can be doped
with liquid iodine solutions to avoid concentration inhomogeneities and inaccuracies
due to separation issues from adding iodine compounds in powder form, like Iopamidol
powder [19], without negatively influencing their printing properties. However, it needs to
be noted that a verification of the homogeneous distribution of the additive—iodine, in this
case—needs to be verified, as not all commercial 3D-printing resins allow the formation of
a homogeneous mixture with aqueous iodine-containing solutions, such as X-ray contrast
media, that will print appropriately.
During the development of these iodine-doped resins, we also encountered difficulties
because suppliers changed the chemical formulations of the resins with new batches. This
happened with the first resin identified as suitable for iodine contrast agent doping. The re-
ordered resin did not form a homogeneous mixture with the contrast agent and gelled after
a short time, preventing printing samples with accurate and homogeneously distributed
iodine content. Therefore, each batch should be checked before printing phantoms or
phantom parts. If this is verified, doping resins with liquid iodine contrast agents intended
for patient use provides a very accurate and appropriate way of adding iodinated contrasts
or details with very well-defined iodine content to custom phantoms.

Limitations, Future Work, and Outlook

In this work, the resulting iodine concentrations have not been verified by independent
measurements and rely on the accuracy of the iodine content in the contrast medium stated
by the producer as 400.0 mg/mL. While extensive quality assurance is mandatory for
medically used drugs like contrast agents, and the product data sheet specifies an accuracy
better than ±0.05 mg/mL (400.0), larger uncertainties are involved when mixing calculated
amounts of the contrast medium into the liquid resin manually. However, the almost perfect
linearity of the measured mass density and voxel values in the scan indicates that the latter
has been performed with appropriate accuracy, an external test would be favorable. Further
work will compare attenuation in medical CT scanners with standardized iodine solutions
prepared from non-ionic iodine compounds dissolved in deionized water to verify the
iodine content in the printed samples.
Another limitation is the constancy of the supply of the resins used, as manufacturers
can (and have) changed the chemical formulations of the resins without notice to the end
user in order to improve the overall printing process and results. Therefore, when using a
new batch, the resin properties must be thoroughly checked.
With the material developed in this work, calibration plates and quality control phan-
toms for contrast-enhanced mammography with exact iodine area weight are currently
developed. Future envisaged developments include printing realistic tumor models in
anthropomorphic phantoms, including appropriate “anatomical noise”, i.e., tissue back-
ground structures, for optimization and detectability studies in different 2D and 3D modali-
ties, mainly in mammography and CT, including spectral and dual energy CT applications.

Author Contributions: Conceptualization, P.H.; methodology, P.H. and L.B.; validation, P.H.; formal
analysis, P.H. and F.S.; investigation, all authors; resources, P.H. and L.B.; data curation, P.H.; writing—
original draft preparation, all authors; writing—review and editing, P.H. and F.S.; visualization, P.H.
All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Biomimetics 2024, 9, 606 10 of 12

Data Availability Statement: Data are contained within the article.

Acknowledgments: We would like to express our sincere gratitude to Claudia Kuntner and
Thomas Wanek for their invaluable support and for providing access to the preclinical imaging
lab infrastructure.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Wegner, M.; Gargioni, E.; Krause, D. Classification of phantoms for medical imaging. Procedia CIRP 2023, 119, 1140–1145.
[CrossRef]
2. Petoussi-Henss, N.; Bolch, W.E.; Eckerman, K.F.; Endo, A.; Hertel, N.; Hunt, J.; Menzel, H.G.; Pelliccioni, M.; Schlattl, H.; Zankl,
M. ICRP Publication 116—The first ICRP/ICRU application of the male and female adult reference computational phantoms.
Phys. Med. Biol. 2014, 59, 5209–5224. [CrossRef] [PubMed]
3. Bolch, W.; Lee, C.; Wayson, M.; Johnson, P. Hybrid computational phantoms for medical dose reconstruction. Radiat. Environ.
Biophys. 2010, 49, 155–168. [CrossRef] [PubMed]
4. Tino, R.; Yeo, A.; Leary, M.; Brandt, M.; Kron, T. A Systematic Review on 3D-Printed Imaging and Dosimetry Phantoms in
Radiation Therapy. Technol. Cancer Res. Treat. 2019, 18, 1533033819870208. [CrossRef] [PubMed]
5. Filippou, V.; Tsoumpas, C. Recent advances on the development of phantoms using 3D printing for imaging with CT, MRI, PET,
SPECT, and ultrasound. Med. Phys. 2018, 45, e740–e760. [CrossRef]
6. Solomon, J.; Ba, A.; Bochud, F.; Samei, E. Comparison of low-contrast detectability between two CT reconstruction algorithms
using voxel-based 3D printed textured phantoms. Med. Phys. 2016, 43, 6497. [CrossRef]
7. Irnstorfer, N.; Unger, E.; Hojreh, A.; Homolka, P. An anthropomorphic phantom representing a prematurely born neonate for
digital x-ray imaging using 3D printing: Proof of concept and comparison of image quality from different systems. Sci. Rep. 2019,
9, 14357. [CrossRef]
8. Schopphoven, S.; Cavael, P.; Bock, K.; Fiebich, M.; Mader, U. Breast phantoms for 2D digital mammography with realistic
anatomical structures and attenuation characteristics based on clinical images using 3D printing. Phys. Med. Biol. 2019, 64, 215005.
[CrossRef]
9. Homolka, P.; Figl, M.; Wartak, A.; Glanzer, M.; Dunkelmeyer, M.; Hojreh, A.; Hummel, J. Design of a head phantom produced on
a 3D rapid prototyping printer and comparison with a RANDO and 3M lucite head phantom in eye dosimetry applications. Phys.
Med. Biol. 2017, 62, 3158–3174. [CrossRef]
10. IAEA. Radiation Protection and Safety in Medical Uses of Ionizing Radiation; IAEA: Vienna, Austria, 2018; Vol. IAEA Specific Safety
Guide No. 46.
11. IAEA. Handbook of Basic Quality Control Tests for Diagnostic Radiology; IAEA: Vienna, Austria, 2023; Vol. IAEA Human Health
Series No. 47.
12. Abadi, E.; Segars, W.P.; Tsui, B.M.W.; Kinahan, P.E.; Bottenus, N.; Frangi, A.J.; Maidment, A.; Lo, J.Y.; Samei, E. Virtual clinical
trials in medical imaging: A review. J. Med. Imaging 2020, 7, 042805. [CrossRef] [PubMed]
13. Kiarashi, N.; Nolte, A.C.; Sturgeon, G.M.; Segars, W.P.; Ghate, S.V.; Nolte, L.W.; Samei, E.; Lo, J.Y. Development of realistic
physical breast phantoms matched to virtual breast phantoms based on human subject data. Med. Phys. 2015, 42, 4116–4126.
[CrossRef] [PubMed]
14. Nakao, M.; Ozawa, S.; Miura, H.; Yamada, K.; Habara, K.; Hayata, M.; Kusaba, H.; Kawahara, D.; Miki, K.; Nakashima, T.;
et al. Development of a CT number calibration audit phantom in photon radiation therapy: A pilot study. Med. Phys. 2020, 47,
1509–1522. [CrossRef] [PubMed]
15. Constantinou, C.; Harrington, J.C.; DeWerd, L.A. An electron density calibration phantom for CT-based treatment planning
computers. Med. Phys. 1992, 19, 325–327. [CrossRef]
16. Cockmartin, L.; Bosmans, H.; Marshall, N.W. Investigation of test methods for QC in dual-energy based contrast-enhanced digital
mammography systems: I. Iodine signal testing. Phys. Med. Biol. 2023, 68. [CrossRef] [PubMed]
17. Jiang, X.; Yang, X.; Hintenlang, D.E.; White, R.D. Effects of Patient Size and Radiation Dose on Iodine Quantification in Dual-Source
Dual-Energy CT. Acad. Radiol. 2021, 28, 96–105. [CrossRef]
18. Euler, A.; Solomon, J.; Mazurowski, M.A.; Samei, E.; Nelson, R.C. How accurate and precise are CT based measurements of iodine
concentration? A comparison of the minimum detectable concentration difference among single source and dual source dual
energy CT in a phantom study. Eur. Radiol. 2019, 29, 2069–2078. [CrossRef]
19. Leithner, R.; Knogler, T.; Homolka, P. Development and production of a prototype iodine contrast phantom for CEDEM. Phys.
Med. Biol. 2013, 58, N25–N35. [CrossRef]
20. McGarry, C.K.; Grattan, L.J.; Ivory, A.M.; Leek, F.; Liney, G.P.; Liu, Y.; Miloro, P.; Rai, R.; Robinson, A.P.; Shih, A.J.; et al. Tissue
mimicking materials for imaging and therapy phantoms: A review. Phys. Med. Biol. 2020, 65. [CrossRef]
21. ICRU. ICRU Report 44. Tissue Substitutes in Radiation Dosimetry and Mesurement; ICRU Report 44; International Commission on
Radiation Units and Mesurement: Bethesda, MD, USA, 1989.
22. Okkalidis, N. 3D printing methods for radiological anthropomorphic phantoms. Phys. Med. Biol. 2022, 67, 15TR04. [CrossRef]
Biomimetics 2024, 9, 606 11 of 12

23. Ma, X.; Figl, M.; Unger, E.; Buschmann, M.; Homolka, P. X-ray attenuation of bone, soft and adipose tissue in CT from 70 to 140
kV and comparison with 3D printable additive manufacturing materials. Sci. Rep. 2022, 12, 14580. [CrossRef]
24. Ahmed, A.M.M.; Buschmann, M.; Breyer, L.; Kuntner, C.; Homolka, P. Tailoring the Mass Density of 3D Printing Materials for
Accurate X-ray Imaging Simulation by Controlled Underfilling for Radiographic Phantoms. Polymers 2024, 16, 1116. [CrossRef]
[PubMed]
25. Ma, X.; Buschmann, M.; Unger, E.; Homolka, P. Classification of X-Ray Attenuation Properties of Additive Manufacturing and
3D Printing Materials Using Computed Tomography From 70 to 140 kVp. Front. Bioeng. Biotechnol. 2021, 9, 763960. [CrossRef]
[PubMed]
26. Kalidindi, Y.; Ganapathy, A.K.; Nayak, Y.; Elumalai, A.; Chen, D.Z.; Bishop, G.; Sanchez, A.; Albers, B.; Shetty, A.S.; Ballard, D.H.
Computed Tomography Attenuation of Three-Dimensional (3D) Printing Materials-Depository to Aid in Constructing 3D-Printed
Phantoms. Micromachines 2023, 14, 1928. [CrossRef] [PubMed]
27. Fonseca, G.P.; Rezaeifar, B.; Lackner, N.; Haanen, B.; Reniers, B.; Verhaegen, F. Dual-energy CT evaluation of 3D printed materials
for radiotherapy applications. Phys. Med. Biol. 2023, 68, 035005. [CrossRef] [PubMed]
28. Bibb, R.; Thompson, D.; Winder, J. Computed tomography characterisation of additive manufacturing materials. Med. Eng. Phys.
2011, 33, 590–596. [CrossRef]
29. Okkalidis, N.; Marinakis, G. Technical Note: Accurate replication of soft and bone tissues with 3D printing. Med. Phys. 2020, 47,
2206–2211. [CrossRef]
30. Fredenberg, E.; Dance, D.R.; Willsher, P.; Moa, E.; von Tiedemann, M.; Young, K.C.; Wallis, M.G. Measurement of breast-tissue
x-ray attenuation by spectral mammography: First results on cyst fluid. Phys. Med. Biol. 2013, 58, 8609–8620. [CrossRef]
31. Cockmartin, L.; Marshall, N.W.; Zhang, G.; Lemmens, K.; Shaheen, E.; Van Ongeval, C.; Fredenberg, E.; Dance, D.R.; Salvagnini,
E.; Michielsen, K.; et al. Design and application of a structured phantom for detection performance comparison between breast
tomosynthesis and digital mammography. Phys. Med. Biol. 2017, 62, 758–780. [CrossRef]
32. McCollough, C.H.; Leng, S.; Yu, L.; Fletcher, J.G. Dual- and Multi-Energy CT: Principles, Technical Approaches, and Clinical
Applications. Radiology 2015, 276, 637–653. [CrossRef]
33. Jochelson, M.S.; Lobbes, M.B.I. Contrast-enhanced Mammography: State of the Art. Radiology 2021, 299, 36–48. [CrossRef]
34. Zopfs, D.; Graffe, J.; Reimer, R.P.; Schafer, S.; Persigehl, T.; Maintz, D.; Borggrefe, J.; Haneder, S.; Lennartz, S.; Grosse Hokamp, N.
Quantitative distribution of iodinated contrast media in body computed tomography: Data from a large reference cohort. Eur.
Radiol. 2021, 31, 2340–2348. [CrossRef] [PubMed]
35. Braeutigam, M.; Huebner-Steiner, U.; Pietsch, H.; Speck, U. Dual Energy CT Imaging; Seidensticker, P.R., Hofmann, L.K., Eds.;
Springer: Heidelberg, Gernamy, 2008.
36. Knogler, T.; Homolka, P.; Hornig, M.; Leithner, R.; Langs, G.; Waitzbauer, M.; Pinker-Domenig, K.; Leitner, S.; Helbich, T.H.
Contrast-enhanced dual energy mammography with a novel anode/filter combination and artifact reduction: A feasibility study.
Eur. Radiol. 2016, 26, 1575–1581. [CrossRef] [PubMed]
37. Klausz, R.; Rouxel, M.; Mancardi, X.; Carton, A.K.; Jeunehomme Patoureaux, F. Introduction of a comprehensive phantom for the
quality control of contrast enhanced spectral mammography. In Proceedings of the European Congress of Radiology, Vienna,
Austria, 28 February–4 March 2018.
38. Jeunehomme, F. Mammographie Numérique Avec Injection de Produit de Contraste; Université de Paris-Sud: Paris, France, 2005.
Available online: [Link] (accessed on 29 August 2024).
39. Baldelli, P.; Bravin, A.; Di Maggio, C.; Gennaro, G.; Sarnelli, A.; Taibi, A.; Gambaccini, M. Evaluation of the minimum iodine
concentration for contrast-enhanced subtraction mammography. Phys. Med. Biol. 2006, 51, 4233–4251. [CrossRef] [PubMed]
40. Jong, R.A.; Yaffe, M.J.; Skarpathiotakis, M.; Shumak, R.S.; Danjoux, N.M.; Gunesekara, A.; Plewes, D.B. Contrast-enhanced digital
mammography: Initial clinical experience. Radiology 2003, 228, 842–850. [CrossRef]
41. Palma, B.A.; Rosado-Méndez, I.; Villaseñor, Y.; Brandan, M.E. Phantom study to evaluate contrast-medium-enhanced digital
subtraction mammography with a full-field indirect-detection system. Med. Phys. 2010, 37, 577. [CrossRef]
42. Takanami, K.; Higano, S.; Takase, K.; Kaneta, T.; Yamada, T.; Ishiya, H.; Mori, I.; Takahashi, S. Validation of the use of calibration
factors between the iodine concentration and the computed tomography number measured outside the objects for estimation of
iodine concentration inside the objects: Phantom experiment. Radiat. Med. 2008, 26, 237–243. [CrossRef]
43. Arvanitis, C.D.; Speller, R. Quantitative contrast-enhanced mammography for contrast medium kinetics studies. Phys. Med. Biol.
2009, 54, 6041–6064. [CrossRef]
44. Hill, M.L.; Mainprize, J.G.; Mawdsley, G.E.; Yaffe, M.J. A solid iodinated phantom material for use in tomographic x-ray imaging.
Med. Phys. 2009, 36, 4409. [CrossRef]
45. Shikhaliev, P.M. Dedicated phantom materials for spectral radiography and CT. Phys. Med. Biol. 2012, 57, 1575–1593. [CrossRef]
46. White, D.R.; Martin, R.J.; Darlison, R. Epoxy resin based tissue substitutes. Br. J. Radiol. 1977, 50, 814–821. [CrossRef]
47. X3msnake. Photonsters Validation Matrix v2. Available online: [Link] (accessed on 25
September 2024).
48. Robb, R.A.; Hanson, D.P.; Karwoski, R.A.; Larson, A.G.; Workman, E.L.; Stacy, M.C. Analyze: A comprehensive, operator-
interactive software package for multidimensional medical image display and analysis. Comput. Med. Imaging Graph. 1989, 13,
433–454. [CrossRef] [PubMed]
Biomimetics 2024, 9, 606 12 of 12

49. Lappchen, T.; Meier, L.P.; Furstner, M.; Prenosil, G.A.; Krause, T.; Rominger, A.; Klaeser, B.; Hentschel, M. 3D printing of
radioactive phantoms for nuclear medicine imaging. EJNMMI Phys. 2020, 7, 22. [CrossRef]
50. Gillett, D.; Marsden, D.; Ballout, S.; Attili, B.; Bird, N.; Heard, S.; Gurnell, M.; Mendichovszky, I.A.; Aloj, L. 3D printing (18)F
radioactive phantoms for PET imaging. EJNMMI Phys. 2021, 8, 38. [CrossRef] [PubMed]
51. Ceh, J.; Youd, T.; Mastrovich, Z.; Peterson, C.; Khan, S.; Sasser, T.A.; Sander, I.M.; Doney, J.; Turner, C.; Leevy, W.M. Bismuth
Infusion of ABS Enables Additive Manufacturing of Complex Radiological Phantoms and Shielding Equipment. Sensors 2017, 17,
459. [CrossRef] [PubMed]
52. Jahnke, P.; Schwarz, F.B.; Ziegert, M.; Almasi, T.; Abdelhadi, O.; Nunninger, M.; Hamm, B.; Scheel, M. A radiopaque 3D printed,
anthropomorphic phantom for simulation of CT-guided procedures. Eur. Radiol. 2018, 28, 4818–4823. [CrossRef] [PubMed]
53. Mirón, V.; Ferrándiz, S.; Juárez, D.; Mengual, A. Manufacturing and characterization of 3D printer filament using tailoring
materials. Procedia Manuf. 2017, 13, 888–894. [CrossRef]
54. Homolka, P.; Gahleitner, A.; Nowotny, R. Temperature dependence of HU values for various water equivalent phantom materials.
Phys. Med. Biol. 2002, 47, 2917–2923. [CrossRef]
55. Hatamikia, S.; Oberoi, G.; Unger, E.; Kronreif, G.; Kettenbach, J.; Buschmann, M.; Figl, M.; Knausl, B.; Moscato, F.; Birkfellner, W.
Additively Manufactured Patient-Specific Anthropomorphic Thorax Phantom With Realistic Radiation Attenuation Properties.
Front. Bioeng. Biotechnol. 2020, 8, 385. [CrossRef]
56. Homolka, P.; Nowotny, R. Production of phantom materials using polymer powder sintering under vacuum. Phys. Med. Biol.
2002, 47, N47–N52. [CrossRef]
57. Homolka, P.; Gahleitner, A.; Prokop, M.; Nowotny, R. Optimization of the composition of phantom materials for computed
tomography. Phys. Med. Biol. 2002, 47, 2907–2916. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.