gms | German Medical Science

48th Meeting of the Particle Therapy Co-Operative Group

Particle Therapy Co-Operative Group (PTCOG)

28.09. - 03.10.2009, Heidelberg

Quality Assurance of Hounsfield Values for Therapy Planning

Meeting Abstract

Search Medline for

  • J. Engelke - German Cancer Research Center, Heidelberg
  • O. Jäkel - German Cancer Research Center, Heidelberg

PTCOG 48. Meeting of the Particle Therapy Co-Operative Group. Heidelberg, 28.09.-03.10.2009. Düsseldorf: German Medical Science GMS Publishing House; 2009. Doc09ptcog055

doi: 10.3205/09ptcog055, urn:nbn:de:0183-09ptcog0557

Published: September 24, 2009

© 2009 Engelke et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.



Background: Heavy ions are known for their characteristic depth dose curve. Accurate prediction of the position of the high-dose peak is crucial for precise treatment. During therapy planning the range of heavy ion in tissue is calculated using X-ray computed tomography (CT) images. Dedicated CT scan protocols for head and body scans are used for therapy planning at the Heidelberg Ion Beam-Therapy facility (HIT) in order to assure the image quality in terms of Hounsfield unit (HU). In clinical routine, however, changes of the CT image due to software or hardware modifications of the CT system can in principle affect the HUs and thus introduce uncertainties in the dose calculation. Therefore a quality assurance (QA) procedure is needed to control the imaging quality; currently CT images of a cylindrical phantom with tissue equivalent inserts are analyzed manually [1]. This method is very time consuming and is subject to individual variations in the data analysis. The aim of this study was to design a fast and automated CT-QA procedure, which allows for HU based monitoring of the constancy of the CT protocols.

Material and methods: In phantom studies specific inserts with a defined HU are placed inside a phantom. When plotting a histogram of the frequency of HUs in a CT, very narrow peaks are expected at particular HU-values for each of the different insert materials. In order to determine the peak position (in HU) a Gaussian is fitted to the peaks in the CT histogram using a least square fit function. An algorithm for automated read-out of the peak position from a special setup of Gammex inserts (Electron Density CT Phantom, GAMMEX-RMI GmbH) in a phantom was developed. For this purpose the suitability of various phantom setups was investigated. Therefore three available polymethyl methacrylate (PMMA) phantoms and a Gammex phantom with different combination of Gammex inserts were evaluated.

Results: In order to enable automated detection of the peaks belonging to different materials, the HUs of the inserts have to be well separated. Using this criteria the following inserts have been chosen: Lung, Adipose, Muscle, True Water, Air, CB2 30%, CB2 50% and Cortical Bone. A number of tests revealed that best results were obtained using a small cylindrical (head) PMMA phantom and a (pelvic) PMMA phantom, representing the geometry of the particular body region, with the above mentioned inserts in a defined arrangement.

Conclusion: The developed software for CT protocol monitoring allows to accurately detect changes of the CT soft- and hardware for a given CT protocol. The software tool has been designed with a fast and automated read-out, a user-friendly interface and a setup, which provides reproducible HUs of the used materials. This tool is going to replace the old manual read-out of HUs. It will abbreviate the time needed for the CT-QA and significantly improve the efficiency of the procedure.


Qamhiyeh S. A Monte Carlo study of the accuracy of CT-numbers for range calculations in Carbon ion therapy [PhD thesis]. Heidelberg: Ruperto-Carola University of Heidelberg; 2007.