View Full Version : Help regarding CT phantom - Responses

12-01-2009, 06:42 AM
Dear Biomch-L subscribers,

You will find below responses to my post
regarding methods for creating hydroxyapatite
calibration phantoms. Thanks to all those who
responded. Several labs choose to use
commercially available phantoms to avoid issues
pertaining to accuracy, precision and
homogeneity. I have divided the responses up into
those recommending commercial phantoms and those
regarding methodology for custom phantoms.



Original Post:

Our lab is interested in creating a calcium
hydroxyapatite calibration phantom with a range
of densities so that we may relate CT attenuation
data to equivalent hydroxyapatite density. As an
initial step, we suspended hydroxyapatite powder
in a gelatin solution. Unfortunately, the
hydroxyapatite settled out before the gelatin set
in. We would greatly appreciate any information
pertaining to 1) substances for hydroxyapatite
powder suspension, 2) ensuring uniform density,
and 3) any other helpful tips or tricks you have
picked up through experience. References on the topic would be helpful as well.

Responses (commercial phantoms):

* I am not very familiar with the technology,
but I know that Scanco makes pre-fabricated
phantoms for use in their CT scanners. They have
a range of densities that I think are equivalent
to particular densities of hydroxyapatite. Might be worth a look.

Sean Osis
Human Performance Lab
University of Calgary

* Are you aware of the commercial phantoms
available this (Image Analysis, Mindways, plus
many others)? If not, I'd recommend going with
that route unless you have scientific reasons to
not use these phantoms. This is because precision
and stability are important issues in all this
and these commercial phantoms have evolved over
the years to address these issues.
Good luck with your project.
Tony Keaveny

* This company makes commercial calcium
hydroxyapatite phantoms. The old one had 0, 50,
100 and 200 mg/cc concentrations. New ones might
be different. I bet they could find an old one on
ebay from a center that no longer does Coronary
calcium scans with the phantom underneath the

Responses (methods for custom phantoms):

* A good strategy is to keep the gel moving
during solidification, by using a roller mixer or
a 3d mixer. In our group
(http://oyenlab.org) we
managed to get uniform dispersions of ceramic
microspheres in gelatin gels by rolling the
suspensions in a tin until the gelatin
solidified. It is important to fill the tin
completely and to roll it slowly to prevent the
formation of bubbles in the gel. To give you an
idea we rolled a 0.5 l tin for about 3 hours.

Best regards,

* I've never worked with hydroxyapatite
before, but I have created gels for MRI phantoms,
which simulate electrical properties of human
tissue. We use HEC-based gel recipes
(hydroxyethyl cellulose). I don't know if this
will give you the type of gel you need, but it
might be worth looking into. We make liquid gels
with a viscosity of 12000 - 17000 cP at room
temperature. HEC powder turns to gel of this
viscosity quite quickly (within 10 minutes of
mixing) when mixed with water. Roughly 2% HEC to DI water by weight.

Kieran Coghlan, BSME, MSES, MSBME

* You can also use test tubes containing
different concentrations of di-potassium hydrogen
phosphate solution (e.g. 64%, 32%, 16% and 4%)
prior to scanning in order to calibrate the grey
values of the CT-scan images. The solutions do
not interfere with the attenuation of X rays of
the tissues and there is a direct relationship
between the Hounsfield units and the
concentration of the dipotassium hydrogen
phosphate solution. After scanning, the average
values of the Hounsfield units corresponding to
each test-tube can be noted. The calibration of
di-potassium hydrogen phosphate solution will
make it possible to convert the grey values at
different locations of the images to Hounsfield
units, based on the linear relationship.

Good luck.

Rajshree Mootanah, Ph.D, MBA
Visiting Scientist, Leon Root, M.D. Motion Analysis Laboratory
Hospital for Special Surgery
Director, Medical Engineering Research Group, Anglia Ruskin University
535 East 70th Street
New York, N.Y. 10021
Lab: (212) 606-1215
Fax: (212) 774-7859
Emails: rajshree.mootanah@anglia.ac.uk

* Encountering the same problems as described
in your mail we used coconut oil instead of
gelatin solution. Being solid, or at least
viscous, at room temperature it could be molten
easily and returns to the solid state after
cooling, while x-ray absorption is close to zero.
Careful stirring during the cooling process keeps
the mixture as homogeneous as possible (depending
on your HA grain size) and careful centrifugation
removes remaining voids. We used small Eppendorf
tubes for each HA concentration so the phantom
can easily adapted to the scanned material. And
don't forget to put something no x-ray
translucent into your Phantom for orientation and
identification of the test volumina...

I hope this was a little help for you,


* Available hydroxyapatite phantoms are
mostly based on water-equivalent epoxy resins.
The tricky secret is the way you disperse
uniformly the powder within the resins, which is
not so easy. After some experiences in home-made
resins phantoms for morphometric microCT (Perilli
E et al, A physical phantom for the calibration
of three-dimensional X-ray microtomography
examination, J Microsc. 2006 May;222(Pt
2):124-34) we decided to buy our densitometric
phantoms from specialized providers of quality control tools in radiology:
* European Spine Phantom for clinical CT
* QRM phantoms for in-vitro microCT (http://www.qrm.de )
Phantoms are normally provided with accuracy
certifications in hydroxyapatite content.

Hope it helps, Fabio Baruffaldi

* Here you find work done at our institute
some while ago: - Nazarian A, Snyder BD,
Zurakowski D, Muller R. Quantitative
micro-computed tomography: a non-invasive method
to assess equivalent bone mineral density. Bone
43: 302-311 (2008). - Schweizer S, Hattendorf B,
Schneider P, Aeschlimann B, Gauckler L, Müller R,
Gunther D. Preparation and characterization of
calibration standards for bone density
determination by micro-computed tomography. Analyst 132: 1040-1045 (2007).

Best regards

* I carried out some work on the creation of
high solid content low viscosity hydroxyapatite
slurries as part of my PhD studies. The work
involved both the manipulation of particle size
distribution and the search for an optimum
dispersant to aid both steric and electrostatic
repulsion. In particular the dispersant results
may be of interest to you. In the end we found
the optimum dispersant for our HA powder was
Ammonium polyacrylate. The work was published and
can be found by searching for:
‘High-solid-content hydroxyapatite slurry for the
production of bone substitute scaffolds’.
Proceedings of the Institution of Mechanical
Engineers, Part H: Journal of Engineering in
Medicine, 2009. 223(6): p. 727-737.If this info
can be of any use and you have difficulty finding
the paper I can send it to you. If you have any further queries please ask.

Best regards

Dr. Eoin Cunningham
Research Assistant
School of Mechanical & Aerospace Engineering
Queen's University
Ashby Building Stranmillis Road
E-mail: ecunningham07@qub.ac.uk
Telephone: 02890 974782

W. Brent Edwards, Ph.D.
Postdoctoral Research Fellow
Department of Kinesiology and Nutrition
University of Illinois at Chicago
1919 W. Taylor Street
650 AHSB, M/C 517
Chicago, IL 60612

Tel. 312 996 1582