Connect
MJA
MJA

A magnetic resonance imaging-based workflow for planning radiation therapy for prostate cancer

Peter B Greer, Jason A Dowling, Jonathon A Lambert, Jurgen Fripp, Joel Parker, James W Denham, Chris Wratten, Anne Capp and Olivier Salvado
Med J Aust 2011; 194 (4): S24. || doi: 10.5694/j.1326-5377.2011.tb02939.x
Published online: 21 February 2011

Abstract

Standard treatment planning for EBRT

In EBRT for prostate cancer, high-energy x-ray beams from multiple directions deposit energy (dose) within the tumour (prostate) to destroy cancer cells. The standard process for EBRT treatment planning is shown in Box 1(A). The first step is patient imaging (a CT scan sometimes combined with MRI). The treatment targets (the prostate and sometimes the seminal vesicles) and important normal tissues (the rectum, bladder, and femoral heads) are then manually defined from the scans (Box 2). Modern radiotherapy machines offer improved treatment accuracy through better visualisation and the correction of errors in patient setup, making target delineation the most significant uncertainty in radiotherapy planning.

In standard treatment planning, if MRI is used for visualising the prostate, then the scans from MRI and CT are aligned to transfer the structure contours defined by MRI to the CT scans for accurate dose calculation. The defined prostate volume is then expanded to become the larger, planning target volume for treatment. This allows for uncertainties in delineation and patient setup, and for prostate movement. The next step is the use of computer planning tools to determine the directions, strengths and shapes of the treatment beams used to deliver the prescribed dose to the defined target, while minimising the dose to normal tissues. Finally, the patient is carefully positioned and the treatment is delivered. The optimal way to align the patient for treatment is to use small implanted fiducial markers in the prostate. These are visible under x-ray imaging and show the precise position of the prostate within the body. Image guidance is used to align the treatment target each day for the entire radiotherapy course.

CT v MRI for radiation therapy planning

CT scans are acquired using low-energy x-ray beams. An image is created, with each pixel value assigned a CT number. This number can be easily related by a calibration process to the density of electrons within the tissues. The CT scan converted to electron densities can then be used to calculate the dose to be delivered to a patient from a radiotherapy x-ray beam. Energy is deposited in tissues mainly by the Compton scattering process. The likelihood of dose deposition depends on local electron densities; hence, this information is obtained from CT scans to calculate radiation dose.

MRI scans are acquired by a completely different physical process, which is related to the magnetic properties of tissues within the body. Tissues with slightly different magnetic properties (eg, because of their water content) will give different MRI values. This means that MRI scans cannot be calibrated to electron density. MRI scans must be manually aligned to CT scans to accurately transfer the contour of the organ or tissue delineated on the MRI scan, and then dose calculation can be performed using the CT scan.

MRI scans have vastly superior soft-tissue contrast compared with CT scans, with better visualisation and more accurate delineation of the border of the prostate. Prostate borders delineated on MRI scans show greater consistency between radiation oncologists,5,6 and prostate size is smaller than that determined by CT scan.7 Using MRI, the size of the extra margins to be added to account for delineation uncertainties is reduced, and less normal tissue is irradiated, thus reducing treatment toxicity.

A proposed new workflow for prostate radiation therapy planning using MRI scans only is shown in Box 1(B). The prostate and organs would be automatically defined on a high-contrast MRI scan using computer segmentation (organ delineation) algorithms. These algorithms use an “average” MRI scan that has been created from a set of patient MRI scans, commonly known as an “atlas”. This atlas is deformably “warped” (registered) to the patient’s MRI scan until all the tissues line up with the patient’s MRI scan. As the organ boundaries are known in the MRI atlas, these boundaries are also warped onto the patient’s MRI scan and hence the patient’s organ boundaries are determined.

The automatically defined prostate borders would then be inspected and adjusted if necessary by the radiation oncologist. Automatic delineation of organs minimises interobserver variation and uncertainty. To calculate dose, a pseudo-CT scan would be created automatically from the MRI scan, with electron densities mapped to the tissues by means of a CT electron densities atlas that corresponds exactly with the MRI atlas. As the deformation for the MRI atlas to the patient’s MRI scan is known, the same deformation will work for the CT electron densities atlas. The result is a pseudo-CT scan with electron densities mapped to the patient’s MRI scan.

Patients and feasibility study design

The feasibility of this MRI-based workflow is being investigated through a collaborative project between the clinical and medical physics group at the Calvary Mater Newcastle Hospital/University of Newcastle and the biomedical imaging processing group at the CSIRO (Commonwealth Scientific and Industrial Research Organisation) Australian e-Health Research Centre.

Thirty-nine patients undergoing prostate radiotherapy have had whole pelvic T2-weighted MRI scans, in addition to standard small-field-of-view MRI scans and conventional radiation therapy CT scans. The contours of the bladder, rectum, bone and prostate were manually delineated on each MRI and CT dataset.

Probabilistic atlas-based organ segmentation methods were developed to automatically segment the organs of interest (prostate, rectum, bladder, and femoral heads) from the MRI scans. After the scans were preprocessed (including correction of intensity inhomogeneity8 and patient movement artefacts9), the images and manually defined contours for each patient were combined into a single reference atlas by image registration methods. This atlas can then be warped onto the MRI scans of a new patient, automatically contouring the organs of interest (Box 3). A matching CT atlas enables the automatic estimation of electron densities from MRI scans by the same method.

The differences between the original CT scan and the MRI-based pseudo-CT scans for each of the 39 patients were validated by two methods.

Another important part of the new workflow process is the production of customised software that has been written and validated (i) for information transfer from the Varian Eclipse and Philips Pinnacle treatment planning systems at Calvary Mater Newcastle (Digital Imaging and Communications in Medicine – Radiotherapy [DICOM-RT] and Radiation Therapy Oncology Group [RTOG]),10 and (ii) for transfer of automatically defined contours and pseudo-CT images back into each system for dose planning.

Results

As shown in Box 4, for the main organs of interest for the 39 patients, HU values of the generated pseudo-CT scans and the original CT scans were very similar, with no significant differences observed (P > 0.09). The mean ratios (SDs) of physical path length and water equivalent path length for the left lateral beams were 1.068 (0.02) and 1.089 (0.02) (P < 0.001), and for the right lateral beams were 1.074 (0.02) and 1.084 (0.03) (P = 0.09). These results indicate that pseudo-CT scan HU values approximate CT scan HU values well, particularly for bone. A significant difference observed for the left lateral beams is not enough to result in dose-calculation differences.

Conclusion

Atlas-based segmentation methods provide the ability to automatically define anatomy on MRI scans of the pelvis, as well as to map electron densities to MRI scans for radiotherapy dose planning. This project combines the expertise of a major radiation therapy and medical physics clinical research group with the computer algorithm and software development expertise of the CSIRO’s Australian e-Health Research Centre. The work is ongoing and currently involves further validation of the pseudo-CT scan and automatically defined contours against standard CT scan-based treatment planning, and the automatic identification of fiducial markers from MRI.

We have demonstrated the feasibility of MRI-based radiotherapy, and the results of this work can be applied to cancers in other locations, such as breast or head and neck cancers.

  • Peter B Greer1,2
  • Jason A Dowling3
  • Jonathon A Lambert2
  • Jurgen Fripp3
  • Joel Parker1
  • James W Denham1,2
  • Chris Wratten1,2
  • Anne Capp1,2
  • Olivier Salvado3

  • 1 Calvary Mater Newcastle Hospital, Newcastle, NSW.
  • 2 University of Newcastle, Newcastle, NSW.
  • 3 Biomedical Imaging, Australian e-Health Research Centre, CSIRO ICT Centre, Brisbane, QLD.


Correspondence: jason.dowling@csiro.au

Acknowledgements: 

This work was partially funded by Cancer Council NSW, Project Grant RG 07-06. We gratefully acknowledge the expert assistance received during this project from: Jacqueline Patterson and Dr Colin Tang (Calvary Mater Newcastle Hospital, Newcastle); Dr Peter Lau, Kristen Fisher and Cynthia Hughes (Hunter Imaging Group, Maitland Private Hospital, Newcastle); Dr Pierrick Bourgeat (CSIRO Australian e-Health Research Centre); Dr Martin Ebert (Department of Radiation Oncology, Sir Charles Gairdner Hospital, Perth); and Dr Sebastien Ourselin (Centre for Medical Image Computing, University College London, UK).

Competing interests:

None identified.

  • 1. Australian Institute of Health and Welfare. Australian cancer incidence and mortality (ACIM) books. Canberra: AIHW, 2007. http://www.aihw.gov.au/cancer/data/acim_books/ (accessed Nov 2010).
  • 2. Australian Institute of Health and Welfare and Australasian Association of Cancer Registries. Cancer in Australia: an overview, 2008. Canberra: AIHW/AACR, 2008. (Cancer Series No. 46; Cat. No. CAN 42.) http://www.aihw.gov.au/publications/can/ca08/ca08.pdf (accessed Nov 2010).
  • 3. Lavelle P. Prostate cancer [online article]. ABC Health and Wellbeing. 2003 (last updated 19 Jan 2007). http://www.abc.net.au/health/library/stories/2003/05/15/1831519.htm (accessed Nov 2010).
  • 4. National Health and Medical Research Council. Clinical practice guidelines: evidence-based information and recommendations for the management of localised prostate cancer. Canberra: NHMRC, 2002. http://www.nhmrc.gov.au/_files_nhmrc/file/publications/synopses/cp88.pdf (accessed Nov 2010).
  • 5. Roach M 3rd, Faillace-Akazawa P, Malfatti C, et al. Prostate volumes defined by magnetic resonance imaging and computerized tomographic scans for three dimensional conformal radiotherapy. Int J Radiat Oncol Biol Phys 1996; 35: 1011-1018.
  • 6. Debois M, Oyen R, Maes F, et al. The contribution of magnetic resonance imaging to the three-dimensional treatment planning of localized prostate cancer. Int J Radiat Oncol Biol Phys 1999; 45: 857-865.
  • 7. Rasch C, Barillot I, Remeijer P, et al. Definition of the prostate in CT and MRI: a multi-observer study. Int J Radiat Oncol Biol Phys 1999; 43: 57-66.
  • 8. Dowling J, Bourgeat P, Raffelt D, et al. Non-rigid correction of interleaving artefacts in pelvic MRI. In: Pluim JPW, Dawant BM, editors. Proceedings of SPIE Medical Imaging 2009: Image Processing (Proceedings Volume 7259, Article No. 72592P).
  • 9. Salvado O, Hillenbrand C, Zhang S, Wilson DL. Method to correct intensity inhomogeneity in MR images for atherosclerosis characterization. IEEE Trans Med Imaging 2006; 25: 539-552.
  • 10. Dowling J, Malaterre M, Greer PB, Salvado O. Importing contours from DICOM-RT structure sets. The Insight Journal 2009; Jul–Dec. http://www.insight-journal.org/browse/publication/701 (accessed Nov 2010).

Author

remove_circle_outline Delete Author
add_circle_outline Add Author

Comment
Do you have any competing interests to declare? *

I/we agree to assign copyright to the Medical Journal of Australia and agree to the Conditions of publication *
I/we agree to the Terms of use of the Medical Journal of Australia *
Email me when people comment on this article

Online responses are no longer available. Please refer to our instructions for authors page for more information.