3D Printing Materials for Physical Breast Phantoms:

Monte Carlo Assessment and Experimental Validation

R. M. Tucciariello

, P. Barca

, D. Caramella

, R. Lamastra

, A. Retico

, A. Traino

and M. E. Fantacci

1,3 c

Department of Physics, University of Pisa, Pisa, Italy

Unit of Medical Physics, Pisa University Hospital “Azienda Ospedaliero-Universitaria Pisana”, Pisa, Italy

INFN, Pisa Section, Pisa, Italy

Department of Radiology, Pisa University Hospital “Azienda Ospedaliero-Universitaria Pisana”, Pisa, Italy

alessandra.retico@pi.infn.it, c.traino@ao-pisa.toscana.it, maria.evelina.fantacci@unipi.it

Keywords: 3D-printing Materials, Breast Phantoms, Digital Mammography, Monte Carlo Simulations, Dosimetry,

GEANT4, Radiochromic Films, XR-QA2, RADIOMA.

Abstract: The aim of this work is to characterize 3D-printing materials to be used for breast physical phantoms in

mammography and digital breast tomosynthesis QA procedures or research. Our approach involves both Monte

Carlo (MC) calculations and experimental measurements. Using a GEANT4-based application, MC

simulations are involved in order to compare transmission properties of the digital “standard breast”, which is

composed by the external skin layer and the breast tissue inside, with those of typical printable materials.

Substitute materials for skin layer and breast tissue have been identified and a 3D-printed physical breast

phantom has been derived. Finally, a comparison between MC results and experimental measurements has been

performed with the Hologic Selenia® Dimensions® mammography unit using XR-QA2 radiochromic films.

1 INTRODUCTION

In the last decades breast cancer screening programs

have been introduced by public health services of

many countries, highlighting an increasing

involvement on early detection of breast masses.

Indeed, breast cancer is the leading cause of cancer

deaths in female subjects and tumour detection in an

early stage ensures greater possibilities of treatment

cures. Early detection and accurate diagnosis are

carried out, in the last decades, with Digital

Mammography (DM) and, in the last few years, with

Digital Breast Tomosynthesis (DBT), a new pseudo-

3D imaging modality (Sechopoulos 2013a, 2013b).

X-ray mammography and breast tomosynthesis

provide radiographic images of the compressed

breast. In the first case two images for each breast are

acquired (cranio-caudal and medio-lateral-oblique

views), while in DBT the X-ray tube moves in an arc

over the compressed breast and multiple projections

https://orcid.org/0000-0001-9600-4177

https://orcid.org/0000-0003-3521-6293

https://orcid.org/0000-0003-2130-4372

are acquired and then reconstructed by a computer,

forming pseudo-three-dimensional images. The

purpose of screening programs is to reduce breast

cancer mortality by ensuring high quality services and

optimized X-ray mammography units. This can be

reached first of all with quality assurance (QA)

protocols, which guarantee optimized equipment, and

with training and research activities. Since both

investigations use ionizing radiation, dosimetry

assessment is mandatory.

Breast physical phantoms, which are test objects,

represent fundamental tools used to perform quality

assurance (QA) procedures and allow the calculation

of useful parameters for imaging and radiation

dosimetry. QA procedures and research are usually

performed using polymethyl-metacrilate phantoms

(PMMA), or other tissues simulating breast

composition, which generally include objects

representing mammographic lesions (tumour masses,

fibers, microcalcifications), resolution patterns and

254

Tucciariello, R., Barca, P., Caramella, D., Lamastra, R., Retico, A., Traino, A. and Fantacci, M.

3D Printing Materials for Physical Breast Phantoms: Monte Carlo Assessment and Experimental Validation.

DOI: 10.5220/0009162302540262

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 1: BIODEVICES, pages 254-262

ISBN: 978-989-758-398-8; ISSN: 2184-4305

step wedges for assessing spatial resolution and

contrast for the image quality assessment (Barca et al.

2019a; Barca et al. 2019b). The most used

commercially physical phantoms for QA procedures

are TORMAM (www.leedstestobjects.com) CDMAM

(www.artinis.com), ACR (www.cirsinc.com). It is

commonly assumed that a uniform PMMA block 45

mm thick is equivalent in absorption to a standard

breast, which is a 5 cm thick compressed breast. It

consists in a 40 mm thick central region comprising a

certain mixture by weight of adipose tissue and

glandular tissue (dependent on compressed breast

thickness and age) surrounded by a 5 mm thick

superficial layer of adipose tissue, simulating skin

absorption (Perry et al. 2008).

Since breast glandularity

can vary from 0 to

100% and it strongly affects MGD, there is the need

to consider this variable in physical phantoms, as well

as in the MC simulations. Nevertheless, skin layer,

not included in commercial phantoms, influences

MGD and attenuation properties (Massera and Tomal

2018, Tucciariello et al. 2019).

The spread of the 3D-printing technology in the

last years and the relatively inexpensive materials

have led research groups to include printing materials

in the context of medical physics and radiotherapy,

for research, QA procedures and patient treatments

(Ferreira et al. 2010; Madamesila et al. 2016).

Nevertheless, 3D-printing is challenging due to

variability of materials and printing methods, and an

accurate characterization of printing materials is

needed. Ivanov and colleagues (2018) explored 3D-

printing materials exposing step-wedge phantoms

with monochromatic beams at ESRF in Grenoble, in

order to characterize attenuation coefficients.

The purpose of this study is to explore different

3D-printing materials which could be employed in the

creation of new physical phantoms for DM and DBT

which better represent both breast anatomy and X-ray

attenuation properties. We propose the method used

by our research group to define X-ray transmission

properties of different materials using a DM X-ray

source, widespread in clinics, and we introduce an

experimental 3D-printed physical phantom. We made

use of Monte Carlo (MC) simulations

and validated

our method with experimental measurements using

GAFchromic™ films.

The term glandularity means the percentage of glandular tissue

respect to the adipose tissue.

The Monte Carlo method refers to a set of computational methods

based on the use of artificially generated random numbers for

solving phenomenon under investigation. In this case, photons

2 MATERIALS AND METHODS

Best practice in dosimetry purposes is to consider

glandular tissue (a complex system of branched ducts

that develop from the inside of the breast to the

nipple) as the radiosensitive tissue in the breast. Thus,

Mean Glandular Dose (MGD) is the parameter used

to assess dose delivered to the glandular tissue. Since

MGD is not a physical quantity, radiation dosimetry

is performed using MC simulations thanks to the

ability to estimate quantities that are challenging to

measure empirically. This kind of approach makes

use of certain geometry assumptions that depend on

breast characteristics and allows to digitally

reproduce a breast phantom model.

We investigated 3D-printing materials for

physical breast phantoms, using the geometry

assumptions followed by research groups whose

works have been milestones for international

dosimetry protocols (Boone 1999; Dance 1990;

Dance, Young, and Van Engen 2011). Our approach

involves both MC simulations as well as experimental

measurements to validate our method.

2.1 Monte Carlo Model

Using the GEANT4 toolkit

, which is a C++ object-

oriented toolkit for the simulation of particle through

matter, we developed a MC code (Tucciariello et al.

2019) that reproduces mammographic and

tomosynthesis investigations, with the same

geometry assumptions (Figure 1) used for validation

purposes (AAPM Task Group 2015). According to

the prescriptions provided by the report of AAPM, the

Option4 PhysicsList was used in GEANT4, for the

constructors and instances, designed for high

accuracy in low-energy physics processes.

In MC models, breast digital phantom is modelled

as a semi-cylinder with an outer layer of skin made

by adipose tissue while the inner part is a

homogeneous mixture of adipose and glandular

tissues. Hammerstein et al. (1979) derived weight

fraction of elements and total tissue density of both

tissues (Table 1). Glandularities ranging from 0 to

100% are composed by mixing properly glandular

and adipose tissues.

Polychromatic X-ray source has been

implemented referring to the Hologic Selenia®

Dimensions® mammography unit, with which

emitted by the X-ray source and interacting with the breast tissue

are traced and all the interactions and dose deposits are registered.

https://geant4.web.cern.ch/

3D Printing Materials for Physical Breast Phantoms: Monte Carlo Assessment and Experimental Validation

255

experimental measurements have been executed. An

algorithm for tungsten anode spectral model has been

involved, dubbed TASMIP

(Boone, Fewell, and

Jennings 1997), based on experimental measurements

of mammography-energy X-ray spectra.

Figure 1: Schematic drawing of the (a) acquisition

geometry for the cranio-caudal view in DM, and (b) breast

phantom geometry adopted.

Table 1: Elemental composition and density of the two

main constituents of the breast tissue.

Tissue H C N O P density

(g/cm

)

glandular 0.102 0.184 0.032 0.677 0.005 1.04

adipose 0.112 0.619 0.017 0.251 0.001 0.93

Since X-ray imaging is a transmission-based

technique, X-ray transmission properties have been

investigated involving Air Kerma (K) estimates. K can

be easily defined in both experimental measurements,

using e.g. an ionization chamber or radiochromic films,

and in MC simulations. Air Kerma is the reference

physical quantity for MGD evaluation purpose.

Indeed, glandular dose estimates start from incident air

kerma and then multiplying it for dedicated conversion

factor from K to MGD, with the surface S for air kerma

scoring placed under the compression paddle and on

the upper surface of the breast. This formalism has

been well defined in literature (Sarno et al. 2019) and

is not the intend of this work.

For air kerma scoring, in our code we use the

formalism provided by Sarno and colleagues (Sarno,

Mettivier, and Russo 2017) using





















Scos













(1)

where E



is the energy of the ith incident photon

passes through the scoring surface S, 













the air mass energy absorption coefficient at the

www.3ntr.net

energy E



(Hubbell and Seltzer 1995) and 



is the

angle between the photon direction and the direction

perpendicular to S.

In MC code, in order to define transmission

properties of different materials, we simply place the

air kerma scoring surface S inside a reference

phantom 5 cm thick under the skin layer, or under the

whole phantom, respectively to evaluate the influence

of a certain material, e.g. skin layer or breast tissue.

Data will be provided in terms of mGy per event, for

both monoenergetic and polychromatic

investigations.

In order to compare MC simulations results with

experimental ones, we need to normalize both for a

reference measurement that is the air kerma incident

on the top of the phantom. The ratios in eq. (2) will

be compared



,







,



≅



,







,





(2)

which provides the transmission factors using MC

simulations and GAFchromic films, where 

,







is the air kerma at a given depth d in the medium m,

and 

,

is the incident air kerma on the upper

side. In the paragraph 2.3 the formalism for obtaining







will be presented.

Since MC results uncertainties are evaluated

following Sempau et al. (2001), air kerma estimations

are performed for monochromatic beams with 10

incident photons, while for polychromatic beams 10

primary photons are involved. These numbers let to

obtain uncertainty on air kerma respectively tree and

four orders of magnitude less. Uncertainties have not

been introduced in all figures because they would not

be visible.

2.2 3D-printing Materials

Breast tissue substitutes have been searched from few

commercial low-cost 3D-printing materials. PLA,

PET-G, ABS, PCABS, CARBON PA, GLASS PA,

ASA have been investigated.

Since MC code needs for each material both

elemental composition and density, using a A2v4 3D-

printer

we printed test objects for each one in order

to define the printing precision and density

We used two simple parallelepiped solids of

10580 mm

and 40405 mm

, and for each of

them three copies were printed. Solids dimensions

and weights were then measured and densities

Is commonly known that after 3D-printing phase material density

can change due to the extrusion printing procedure.

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

256

estimated. Due to the low printing reliability,

CARBON PA and GLASS PA were rejected, while

ASA material, despite the good printing quality, was

avoided since it is a copolymer and its chemical

formula can be different depending on the modality

the three monomers repeat on the structure (Liu et al.

2011). A separate evaluation has been achieved for

PCABS; because of not negligible variations in the

test objects, we used a bigger test object to assess

material density.

Elemental compositions of PLA, ABS, and PET-

G have been taken from Alssabbagh et al. (2017),

while for PCABS, polycarbonate composition is

available on GEANT4 database (Table 2)

Composition and density obtained were involved in

MC simulations for evaluating transmission

properties.

In the subsequent chapter the results that led to

choose materials for the final physical phantom will

be presented.

Table 2: Percentage elemental compositions of 3D-printing

materials evaluated in our work.

Tissue H C

O S K

PLA 0.053 0.519 - 0.426 0.001 0.001

ABS 0.075 0.855 0.053 0.016 0.001 -

PET‐G

0.075 0.652 - 0.271 0.002 -

Poly‐

carbonate

0.055 0.756 0 0.189 0 -

PCABS

80 % polycarbonate, 20% ABS

2.3 Experimental Verification

Experimental verification of transmission properties

for the 3D-printed materials was executed using

GAFchromic

XR-QA2 films. Radiochromic films

are well suited for radiographic QA tests and research

in dosimetry, thanks to the self-developing of the

response after the irradiation process. XR-QA2 are

designed for energies ranging in radiology, with

anode tube potential ranging from 20 to 200 kVp.

XR-QA2 films are sensitive in the dose range 1-

200 mGy and an increasing change in optical

reflectance occurs with increasing doses.

2.3.1 Film Calibration and Digitization

Response of radiochromic films must be assessed

with an accurate calibration in order to obtain a

calibration curve, expressed in terms of air kerma

versus reflectance change R. Since Di Lillo et al.

PCABS is composed by polycarbonate and ABS. The percentage

of ABS can be different depending on manufactures.

Scans for irradiated samples have been executed 24h after the

exposition.

(2016) showed energy dependant dose-response

curves for XR-QA2 using synchrotron radiation, we

proceeded to realize three calibration curves for 25,

30 and 35 kVp with the Hologic Selenia Dimensions

in mammography modality. For each calibration

curve 12 points were used, each of them is from the

average value of 3 different radiochromic samples.

Since X-ray mammography units permits low-

doses irradiations, in order to observe XR-QA2 dose

range, a Radcal 20X6-60E ionisation chamber

coupled with the 2026C dosimeter was used to choose

correct mAs tube loading values and air kerma

exposures, chosen in an optimal range from about 1

mGy to a maximum value depending on the

kilovoltage applied.

The formalism defined by Tomic et al. (2010) and

Di Lillo et al. (2016) was used and discussed above.

From original XR-QA2 1012'' sheets, samples of

33 cm

have been cut to be used for calibrations and

measurements. For calibration, 5 samples were used

as “control films” to quantify background radiation

and 36 samples for each calibration curve.

Using a flatbed scanner (Epson Expression

10000XL), samples were scanned before and after

the exposition, in 48-bit RGB mode, at 150 dpi, in the

same position of the scanner surface and saved as

TIFF image file format. Multiple scans for each

sample were executed. Raw images have been

analysed with the open software ImageJ

. Formalism

provides the film response in terms of reflectance

change R



using the 16-bit red channel in a

ROI of 11 cm

in the center of each sample (Figure

2).

R













(3)











































(4)

where 



and 



are the mean pixel value

of samples respectively before and after the X-ray

exposition, and 





and 





the standard

deviations.

Statistical uncertainty due to scanner response in

multiple scans is included in





. The final value

is considered

net



R



R







(5)

with the relative uncertainty

https://imagej.nih.gov/ij/

3D Printing Materials for Physical Breast Phantoms: Monte Carlo Assessment and Experimental Validation

257























R







R







(6)

Calibration data points have been fitted using the

commercial analysis software Origin 9

and using the

exponential function ∙∙



Figure 2: Scheme of radiochromic samples before and after

the exposure, and ROI used for the pixel values estimation.

3 RESULTS

3.1 Density Assessment

Our first purpose was to evaluate the change in

density for different materials from the coil nominal

density value to the post-printed value. Test objects

reported variation between -9% and -14% and

reported in Table 3. PCABS final density has been

evaluated with a bigger test object which presented a

different value in density respect to the smaller

objects, suggesting a “form factor” influencing

printed objects density depending on dimensions.

Table 3: Change in density for the materials under

investigation. The uncertainty on the estimated density is

0.01 g/cm

[g/cm

] PLA ABS PET‐G PCABS ASA

Nominal

density

1.24 1.05 1.27 1.13 1.07

Estimated

density

1.12 0.92 1.09 0.99 0.97

variation

-10% -12% -14% -12% -9%

3.2 MC Assessment

Data provided in tables 1-3 were used in MC

simulations in order to compare transmission

www.originlab.com

properties of adipose skin layer and breast tissue with

those of 3D-printing material.

3.2.1 Skin Layer

Using the simulation setup shown in Figure 1, a

reference phantom 5 cm thick, 50% glandular, and a

skin thickness of 5 mm made by adipose tissue has

been adopted. We placed a scoring surface below the

skin and evaluated the air kerma transmission curve

through the skin due to monoenergetic X-ray beams,

from 8 to 40 keV at 1 keV steps. Using this approach,

we replaced adipose tissue composing the skin layer

with 3D-printing materials with density correction.

MC simulations were performed with 10

incident

photons for each energy beam. Results are shown in

Figure 3. At lower photon energies, the skin “shields”

the breast tissue and air kerma values are low; with

increasing photon energies X-ray beam penetrates the

skin layer up to a maximum value after which K

decreases due to the decreasing energy absorption

coefficient. Simulations suggest a better behaviour by

PCABS as skin layer respect to the other material, for

both low- and high-energies.

Since DM and DBT use polychromatic X-ray

source, we investigated polyenergetic X-ray beams @

27, 31 and 35 kV, in W/Rh anode/filter combination.

In Figure 4 is shown air kerma transmission through

the skin layer using polychromatic beams. MC

simulations were performed using 10

incident

photons. Air kerma values percentage variations,

respect to the adipose skin layer, suggest also in this

case that PCABS is a well substitute as skin layer.

Figure 3: Air kerma transmission through the skin layer due

to monoenergetic beams.

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

258

Figure 4: Air Kerma transmission through the skin layer

due to polychromatic beams. Percentage values in figure

refer to the variation respect to the adipose tissue.

3.2.2 Breast Tissue

Using the same approach of the previous paragraph,

we investigated 3D-printing materials for the inner

part of the breast phantom assuming PCABS as the

preferred skin layer material. We used, as usual, the

reference breast phantom cited previously. Figure 5

shows that both PLA and PETG can be used for the

inner part of the breast phantom, coupled with 5 mm

thick PCABS skin layer.

Since in literature there are studies involving the

amount of glandular fraction of women breast, we

noted a publication of Yaffe et al. (2009) whose work

found that, out of 2831 women, 95% were below the

45% glandularity. We show 0% glandularity

transmission curve in order to define the “range” of

transmission which 3D-materials have to follow.

Polychromatic beams have been investigated and

results are reported in Figure 6.

Figure 5: Air kerma transmission through the whole

phantom for monochromatic beams.

Figure 6: Air kerma transmission through the whole

phantom for polychromatic beams. Percentage values in

figure refer to the variation respect to the breast 50%

glandular.

3.2.3 Film Calibration Curves

Results from calibration procedure of GAFchromic

films XR-QA2 free-in-air are shown in Figure 7.

Energy dependant response curve is slightly marked,

where the beam mean energy is for 25, 30 and 35 kV

respectively 18.4, 19.1 and 20.0 keV. Points in the

graph are scattered from about 1 to 2 mGy and follow

the fit curve for doses higher than 2 mGy.

Even if digital mammography is a low-exposure

procedure, it is worth noting that high milliamperages

have been used in order to have a major statistic in the

calibration curves, which otherwise would be affected

too much by the low exposure fluctuations.

Figure 7: Calibration curves @ 25, 30 and 35 kV. In figure

is shown a magnification for low-doses exposures.

3D Printing Materials for Physical Breast Phantoms: Monte Carlo Assessment and Experimental Validation

259

3.3 3D-printed Phantom Dosimetry

Assessment

Results from the previous investigations led to create

a breast phantom with an outer layer made by PCABS

material and an inner part of PLA material.

Despite of PETG curve is between those of

glandularities 0% and 50% (Figure 5), we decide to

characterize experimentally PLA because during the

3D-printing phase, density can be further changed

with the infill option, that is the percentage of air

filling (Madamesila et al. 2016). This can cause a

major X-ray transmission depending on the infill

percentage. Using this approach, a greater number of

glandularities can be explored, from more than 50%

to 0%, incrementing the infill option. Infill option has

not been investigated yet, and our purpose is first of

all to evaluate experimentally our method and

PCABS+PLA physical phantom.

Using the open software FreeCAD

, we

developed a modular 3D phantom made by few slices

in order to compose different breast phantom

thicknesses. Each component of the phantom is

exported in STL file format used by the 3D printer.

Phantom is composed by external layers (in grey,

Figure 8) printed in PCABS, and PLA inner

components (in orange). As reported in Figure 1,

radius semi-cylinder is 10 cm, while thickness is

variable depending on how many slices are used.

Slices let to perform dosimetry with radiochromic

films by inserting below each layer a 33 cm

GAFchromic sample (Figure 9).

Figure 8: 3D project of the modular phantom created.

In order to demonstrate the agreement with

transmission properties of the physical phantom with

the expected values from MC simulations, a

comparison has been performed. Following equation

(2), experimental measurements have been performed

https://www.freecadweb.org/

@ 35 kV by inserting radiochromic film samples

under each slice of the phantom for obtaining



,



 at a given depth , normalized by 

,



obtained placing samples between the phantom

surface and the compression paddle.

Figure 9: Example of positioning of radiochromic sample

inside the phantom. The photograph refers to the



,



0.5 value obtained under the upper PCABS skin

layer which has been temporary removed for placing the

sample.

MC simulations refer to air kerma estimates at the

same depth in the phantom. Results in Table 4 show

a good agreement in transmission properties of the

skin layer (depth 0.5 mm) and the firsts two PLA

layers (depth 1.3 and 2.1 mm); in the last layers the

discrepancy is greater, because of the low doses

reached. Indeed, air kerma estimates derived from the

fitting curve show values from about 2 mGy to 0.8

mGy, not considered reliable. This is a limitation due

to characterizing low dose in mammography,

especially in the phantom lower layers, with

subsequent low dose exposures in radiochromic

films. This does not allow to complete the percentage

depth dose curve.

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

260

Table 4: Comparison between experimental data with MC

simulations. Variations in the last column have been

obtained using equation 2. The first row (in grey) represents

the transmission due to the 5 mm thick PCABS layer, while

the other (orange) rows the transmission due to the 8 mm

thick PLA layers composing the phantom inner part.

Depth

d(mm)

net



R









(mGy)









(mGy/ev)

variation



0.1461 13.22

3.78 E-12

0.5

0.1205 10.19

2.98 E-12

-2.5%

1.3

0.0806 6.14

1.77 E-12

-1.1%

2.1

0.0530 3.75

1.09 E-12

-1.7%

2.9

0.0329 2.15

6.73 E-12

-9.7%

3.7

0.0207 1.35

4.23 E-12

-9.9%

4.5

0.0132 0.84

2.74 E-12

-14.4%

4 DISCUSSION

We presented the method used by our research group

to characterize 3D-printing materials to be used for

the creation of breast physical phantoms, having the

same transmission properties of the digital breast

phantoms used in MC breast dosimetry, whose have

dedicated elemental compositions for adipose and

glandular tissues. The approach of involving physical

phantoms with dedicated materials for the skin and

the for the breast tissue could be useful in research or

QA procedures for DM and DBT investigations,

where, until now, polymethyl-metacrilate

homogeneous phantoms are used. We involved MC

simulations to investigate transmission properties of

some 3D-printing materials, whose densities have

been corrected in the MC code, since the final density

of the printed object can vary during the printing

phase. Based on the MC results, performed over

monoenergetic and polyenergetic beams, a physical

breast phantom has been created. Results deriving

from densities estimation and MC simulations led to

consider PCABS material as a well substitute for the

5 mm thick skin layer and PLA material as substitute

for the inner breast tissue.

In order to validate our test phantom,

experimental measurements with GAFchromic XR-

QA2 films were performed, which results confirmed

an agreement with transmission estimations in MC

results for both PCABS layer and for the inner PLA

material, supporting our method, which uses

relatively low-cost equipment and procedures.

Since it is known that mostly of women breast

glandularities ranging from 0% to about 50%, we

decided to adopt PLA material, to support our next

step, which will be to characterize the infill option

that allows to decrease voluntarily the material

density during the printing phase. With this approach

various breast glandularities from 0 to about 50% can

be reached in synthetic phantoms, leading to perform

the image quality assessment for different synthetic

breast anatomies.

ACKNOWLEDGEMENTS

The presented work is part of the RADIOMA project

which is partially funded by "Fondazione Pisa",

Technological and Scientific Research Sector, Via

Pietro Toselli 29, Pisa (Italy). The authors would like

to thank Fondazione Pisa for giving the opportunity

to start this study.

REFERENCES

Alssabbagh, M., Tajuddin, A. A., Abdulmanap, M. and

Zainon, R., 2017. Evaluation of 3D Printing Materials

for Fabrication of a Novel Multi-Functional 3D

Thyroid Phantom for Medical Dosimetry and Image

Quality. Radiation Physics and Chemistry 135: 106–12.

http://dx.doi.org/10.1016/j.radphyschem.2017.02.009.

Barca, P. et al. 2019. Image Quality Comparison between

Digital and Synthetic 2D Mammograms: A Qualitative

and Quantitative Phantom Study. BIOIMAGING 2019

- 6th International Conference on Bioimaging,

Proceedings; Part of 12th International Joint

Conference on Biomedical Engineering Systems and

Technologies, BIOSTEC 2019: 129–36.

Barca, P., et al. 2019. Comprehensive Assessment of Image

Quality in Synthetic and Digital Mammography : A

Quantitative Comparison. Australasian Physical &

Engineering Sciences in Medicine (Cd).

Boone, J. M., 1999. Glandular Breast Dose for

Monoenergetic and High-Energy x-Ray Beams: Monte

Carlo Assessment. Radiology 213(1): 23–37.

Boone, J. M., Fewell, T. R. and Jennings, R. J., 1997.

Molybdenum, Rhodium, and Tungsten Anode Spectral

Models Using Interpolating Polynomials with

Application to Mammography. Medical Physics 24(12):

1863–74.

Dance, D. R., 1990. Monte-Carlo Calculation of

Conversion Factors for the Estimation of Mean

Glandular Breast Dose. Physics in Medicine and

Biology 35(9): 1211–19.

Dance, D. R., Young, K. C. and Van Engen, R. E., 2011.

Estimation of Mean Glandular Dose for Breast

Tomosynthesis: Factors for Use with the UK, European

and IAEA Breast Dosimetry Protocols. Physics in

Medicine and Biology 56(2): 453–71.

Ferreira, C. C. et al., 2010. Evaluation of Tissue-Equivalent

Materials to Be Used as Human Brain Tissue Substitute

in Dosimetry for Diagnostic Radiology. Nuclear

Instruments and Methods in Physics Research, Section

B: Beam Interactions with Materials and Atoms

3D Printing Materials for Physical Breast Phantoms: Monte Carlo Assessment and Experimental Validation

261

268(16): 2515–21. http://dx.doi.org/10.1016/j.nimb.

2010.05.051.

Hammerstein, G. R. et al., 1979. Absorbed Radiation Dose

in Mammography. Radiology 130(2): 485–91.

Hubbell, J. H., and Seltzer, S. M., 1995. Tables of X-Ray

Mass Attenuation Coefficients and Mass Energy-

Absorption Coefficients 1 KeV to 20 MeV for Elements

Z=1 to 92 and 48 Additional Substances of Dosimetric

Interest. Http://Physics.Nist.Gov/Xaamdi 5632.

http://www.nist.gov/pml/data/xraycoef/index.cfm.

Ivanov, D. et al., 2018. Suitability of Low Density Materials

for 3D Printing of Physical Breast Phantoms. Physics

in Medicine and Biology 63(17): aad315.

https://doi.org/10.1088/1361-6560/aad315.

Di Lillo, F. et al., 2016. Energy Dependent Calibration of

XR-QA2 Radiochromic Film with Monochromatic and

Polychromatic x-Ray Beams. Medical Physics 43(1):

583–88.

Liu, S. et al., 2011. Reactive Compatibilization of Poly

(Butylene Terephthalate)/Acrylonitrile- Styrene-

Acrylate Blends by Epoxy Resin: Morphology and

Mechanical Properties. Journal of Macromolecular

Science, Part B: Physics 50(9): 1780–90.

Madamesila, J., McGeachy P., J. Barajas, E. J. and Khan,

R., 2016. Characterizing 3D Printing in the

Fabrication of Variable Density Phantoms for Quality

Assurance of Radiotherapy. Physica Medica 32(1):

242–47. http://dx.doi.org/10.1016/j.ejmp.2015.09.013.

Massera, R. T. and Tomal, A., 2018. Skin Models and Their

Impact on Mean Glandular Dose in Mammography.

Physica Medica 51: 38-47.

Perry, N. et al., 2008. European Guidelines for Quality

Assurance in Breast Cancer Screening and Diagnosis.

Fourth Edition - Summary Document. Annals of

Oncology 19(4): 614–22. http://europa.eu.int/comm/

dgs/health_consumer/index_en.htm.

Sarno, A. et al., 2019. Monte Carlo Calculation of

Monoenergetic and Polyenergetic DgN Coefficients for

Mean Glandular Dose Estimates in Mammography

Using a Homogeneous Breast Model. Physics in

medicine and biology 64(12): 125012.

Sarno, A., Mettivier, G. and Russo, P., 2017. Air Kerma

Calculation in Monte Carlo Simulations for Deriving

Normalized Glandular Dose Coefficients in

Mammography. Physics in Medicine and Biology.

Sechopoulos, I., 2013a. A Review of Breast Tomosynthesis.

Part I. The Image Acquisition Process. Medical

Physics.

———. 2013b. A Review of Breast Tomosynthesis. Part II.

Image Reconstruction, Processing and Analysis, and

Advanced Applications. Medical Physics.

Sempau, J. et al., 2001. Monte Carlo Simulation of Electron

Beams from an Accelerator Head Using PENELOPE.

Physics in Medicine and Biology 46(4): 1163–86.

Task Group, Aapm. 2015. Monte Carlo Reference Data

Sets for Imaging Research. Medical Physics 42.

http://dx.doi.org/10.1118/1.4928676.

Tomic, N., Devic, S., Deblois, F. and Seuntjens, J., 2010.

Reference Radiochromic Film Dosimetry in

Kilovoltage Photon Beams during CBCT Image

Acquisition. Medical Physics 37(3): 1083–92.

Tucciariello, R. M. et al., 2019. Monte Carlo Methods for

Assessment of the Mean Glandular Dose in

Mammography: Simulations in Homogeneous

Phantoms. BIOINFORMATICS 2019 - 10th

International Conference on Bioinformatics Models,

Methods and Algorithms, Proceedings; Part of 12th

International Joint Conference on Biomedical

Engineering Systems and Technologies, BIOSTEC

2019: 242–49.

Yaffe, M. J. et al., 2009. The Myth of the 50-50 Breast.

Medical Physics 36(12): 5437–43.

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

262