Inhalt des Dokuments
Development of a MR-Phantom and acurracy testing with a magnetic resonance tomograph
Martin TettkeAnna-Luisa Uhlitz
Abstract – This paper presents a MRI kneephantom for the evaluation of the measurement techniques in the research project “Functional assessment of orthopaedic aids using openvertical MRIs”.The phantom has a know geometry and simulatesthe T1- and T2-values of muscle, bone and fat inthe knee. It is also flexible and adjustable to different angles.An open low-field MRI was used to scan the phantom. The bone was then segmented and a 3D-model was generated. The phantom can generally be used to evaluate the accuracy, although a few changes should be made.
Keywords – relaxation times, MRI, knee phantom
This paper presents the work of a student bachelor thesis. It is part of the research project “Functional assessment of orthopaedic aids using open vertical MRIs”. This project aims to find evaluation parameters for orthopaedic devices using a vertical low-field MRI. For the evaluation of the developed MRI acquisition and following data processing a MRI phantom simulating the human “femurotibial” joint was needed. Important parameters for that were a known geometry and T1- and T2-values like in the human knee joint. The phantom was supposed to simulate fat, muscle and bone and be able to be adjustable to different flexion angles. Evaluation of the measurement techniques can then be done by comparing the know geometry to the geometry, which was imaged by the MRI and processed.
II. Material and Methods
A. Phantommaterial Simulation of different tissues was already done by ,  and . The different ingredients were mainly taken from those papers, but adapted a little. The phantom contained agar as a T2-modifier, Prohance as a T1-modifier (Bracco Imaging), potassium sorbate and phenoxyethanol as antiseptics, carrageenan as a gelling agent and distilled water. In the beginning testphantoms were fabricated to test
the T1- and T2-values, the procedure and the alteration of the ingredients.
B. Phantom construction
The gel was imbedded into a shell of Plexiglas, which is MR-compatible. The geometry of the phantom is shown in Fig.1. This unusual geometry was chosen because this way the bones are always as close together as possible. The angle can be locked every 7.5°. The phantom was cut into two halves. Thus, the bone could be fabricated extra and could not under any circumstances change its form. The bone-geometry was measured with a structured light 3D-scanner in order to be able to compare it with the MRI-scan later. The phantom was scanned just like the testphantoms with the
low-field MRI (0.25 T, Esaote GScan) every 15°. The acquisition data was then segmented to generate a 3Dmodel for every angle. The femora of these models were registered to cover each other. After that the axes of the phantom could be calculated.
With the presented substances it was possible to simulate the T1- and T2-values of fat, muscle and bone. Two major problems occurred. First, the musclemixture hardened too fast. The result were gabs between muscle and bone. Second, there was a great distance between femur and tibia, which is not physiological.
To solve the two problems it should be tested whether the phantom could be build without cutting it. This would not only solve the problem with the gaps, the phantom would also be more stable and the fabrication-process easier. If the process is altered in that way the tibia should also have the form of a cuboid. Additionally, the placeholder-bones in the muscle-mixtures should be positioned differently, so that the bones are closer together.
All in all, an evaluation of the measurement techniques should be possible, although a new phantom with attention to the changes that should be
made needs to be built.
 Kato, H.; Kuroda, M.; Yoshimura, K.;
Yoshida, A.; Hanamoto, K.; Kawasaki,
S.; Shibuya, K. & Kanazawa, S.:
Composition of MRI phantom
equivalent to human tissues; Med
Phys (2005) 32: pp. 3199-3208.
 Ohno, S.; Kato, H.; Harimoto, T.;
Ikemoto, Y.; Yoshitomi, K.; Kadohisa,
S.; Kuroda, M. & Kanazawa, S.:
Production of a human-tissueequivalent
MRI phantom: optimization
of material heating; Magn Reson Med
Sci (2008) 7: pp. 131-140.
 Yoshimura, K.; Kato, H.; Kuroda, M.;
Yoshida, A.; Hanamoto, K.; Tanaka, A.;
Tsunoda, M.; Kanazawa, S.; Shibuya,
K.; Kawasaki, S. & Hiraki, Y.:
Development of a tissue-equivalent
MRI phantom using carrageenan gel;
Magn Reson Med (2003) 50