A new approach to expand the field of view for MV cone beam computed tomographyby A. Janse van Rensburg

Physica Medica

About

Year
2015
DOI
10.1016/j.ejmp.2015.07.104
Subject
Biophysics / Physics and Astronomy (all) / Radiology Nuclear Medicine and imaging

Similar

3D ROI imaging for cone-beam computed tomography

Authors:
Jens Wiegert, Matthias Bertram, Jörg Wulff, Dirk Schäfer, Jürgen Weese, Thomas Netsch, Hermann Schomberg, Georg Rose
2004

A New Scatter Correction Method for KV and MV Cone Beam CT

Authors:
X. Li, T. Li, Y. Yang, D. Heron, M. Huq
2008

A comparison of a new limited cone beam computed tomography machine for dental use with a multidetector row helical CT machine

Authors:
Koji Hashimoto, Yoshinori Arai, Kazuo Iwai, Masao Araki, Shoji Kawashima, Masaaki Terakado
2003

Text

O.19

A NEW APPROACH TO EXPAND THE FIELD OF VIEW FOR MV CONE BEAM

COMPUTED TOMOGRAPHY

A. Janse van Rensburg. Department of Medical Physics, Steve Biko Academic

Hospital, Pretoria, South Africa

Introduction: The field of view inMVCBCT is limited by the size of the electronic portal imaging device. This severely hampers the use of these images for planning purposes. The author proposes a novel new non-isocentric image acquisitionmethodology that by-passes this limitation. The basic idea has been patented by Siemens. Unfortunately the hardware was not able to support a full implementation of the technique resulting in inferior image quality. The original idea will be presented in this paper.

Materials and Methods: Images are acquired while the gantry rotates around the patient; at the same time the couch moves in a coordinated fashion to describe an ellipse. In this way the edge of the patient’s body is included in every frame. Radio-opaque markers are placed on the patient to assist in correcting for any inaccuracies in the couch movement as well as patient motion during the acquisition.

Results: A careful examination of the geometry of the images reveals that, shifting the center of the images and using a novel differential magnification algorithm, the set can be converted to the equivalent isocentric case.

Since the position of each marker should fall on a simple sinus curve, small positional adjustments can be made to correct for the errors in the couch movement and even patient motion. The reconstruction of the image set can then proceed as normal and no complicated algorithms have to be invented.

Conclusion: This approach can be implemented on a number of different linear accelerators andwould increase the FOV to the point where the images can be used for planning on virtually any body part.

Keywords: FOV, MVCBCT, Non-isocentric, CT, Reconstruction

O.20

THE IMPACT OF MODELLING PARAMETER VARIATIONS IN A VIRTUAL

SOURCE MODEL BASED TREATMENT PLANNING SYSTEM

L. Koen *,a,b, F.C.P. Du Plessis b, W. Shaw b. a Department of Medical Physics,

Equra Health, Cape Town, South Africa; b Department of Medical Physics,

University of the Free State, Bloemfontein, South Africa

Introduction: The Monaco Treatment Planning System (TPS) utilizes a

Virtual Source Model (VSM) described by analytical functions. The aim of this study was to evaluate the dosimetric impact of some modelling parameters by deliberate alteration and observing the effect in the TPS calculated dose distribution.

Materials and Methods: 5 different VSM parameters were evaluated: 3 energy spectrum (primary photon contribution, maximum energy and the “b” exponent value); 2 geometry (primary- and scatter sigma). The effect of each parameter on the calculated TPS dose for small, medium and large field sizes was compared to a reference VSM dataset in terms of percentage depth doses (PDD), dose profiles and output factors. Gamma analysis (1%/1 mm) was performed, where 95% of points passing this criterion were deemed acceptable.

Results: A significant difference was found if the primary photon contribution was outside the range of 0.70 and 0.73 (reference = 0.72). Most pronounced differences in dose distributions were seen at deeper depths for large fields. The maximum energy parameter (reference = 6.6 MeV) was acceptable in the range of 6.5–6.6 when comparing doses at dmax and 6.6– 6.7 at 10 cm. The b-value (reference = 0.756443) proved a more sensitive parameter for large fields. Geometry parameters had a profound influence at a depth of dmax, even though small changes in the sigma value did not influence the penumbra significantly.

Conclusion: The VSMmodelling parameters were successfully quantified for a specificmachinemodel in terms of their range of variabilitywithin themodel.

Changing a single parameter can significantly change energy and fluence. The scope of acceptable variation in these parameters could potentially be exploited by deriving a single VSM for various beammatched linacs.

Keywords: Treatment Planning System, Virtual Source Model, Parameters, variation

O.21

INTRODUCTION OF A PROPER INTERNAL DOSIMETRY PROTOCOL

IN THE NUCLEAR MEDICINE DEPARTMENT AT GROOTE SCHUUR

HOSPITAL

N. Joubert *, C.J. Trauernicht. Department of Medical Physics, University of Cape

Town/Groote Schuur Hospital, Cape Town, South Africa

Introduction: Internal dosimetry in Nuclear Medicine is currently done on an ad-hoc basis at Groote Schuur Hospital (GSH); however, the need has been identified to implement a structuredway of doing it to improve patient care. It is important to determine which patients will benefit from internal dosimetry. The current therapeutic applications at GSH include I-131 therapy, both for benign and malignant thyroid disease, I-131 MIBG treatments, Y-90 synovectomies and the potential for Lu-177 therapy for neuroendocrine tumours. Internal dosimetry is also used in cases where there was misadministration of radioactivity, e.g. in a pregnant or breastfeeding woman or when the wrong radiopharmaceutical was injected. The aim of this paper is to introduce imaging protocols and investigate options of optimising dose calculations.

Materials and Methods: A protocol was set up, initially for a single site, which includes the required imaging, sensitivity and also dead-time corrections from the gamma-camera. The gamma-camera count-rate performance was evaluated and the MIRD formalism was used in individual cases when sufficient imaging was available. As further enhancement it will have to be determined whether it is more accurate to use counts from static images for organs or for the whole body in calculations. Further investigations include dose measurements in phantom studies, simulating biological half-lives, and attenuation corrections.

Results: Sensitivities for different collimator/radionuclide combinations were determined to convert image counts to activity, and were measured as follows: 88 counts/s/MBq for LEHR/Tc-99m; 133 counts/s/MBq for LEAP/