<abstract xmlns="http://www.w3.org/1999/xhtml"><p>Better knowledge of CT number values and their uncertainties can be applied to improve heavy ion treatment planning. We developed a novel method to calculate CT numbers for a computed tomography (CT) scanner using the Monte Carlo (MC) code, BEAMnrc/EGSnrc. To generate the initial beam shape and spectra we conducted full simulations of an X-ray tube, filters and beam shapers for a Siemens Emotion CT. The simulation output files were analyzed to calculate projections of a phantom with inserts. A simple reconstruction algorithm (FBP using a Ram-Lak filter) was applied to calculate the pixel values, which represent an attenuation coefficient, normalized in such a way to give zero for water (Hounsfield unit (HU)). Measured and Monte Carlo calculated CT numbers were compared. The average deviation between measured and simulated CT numbers was 4 ± 4 HU and the standard deviation σ was 49 ± 4 HU. The simulation also correctly predicted the behaviour of H-materials compared to a Gammex tissue substitutes. We believe the developed approach represents a useful new tool for evaluating the effect of CT scanner and phantom parameters on CT number values.</p></abstract>

Better knowledge of CT number values and their uncertainties can be applied to improve heavy ion treatment planning. We developed a novel method to calculate CT numbers for a computed tomography (CT) scanner using the Monte Carlo (MC) code, BEAMnrc/EGSnrc. To generate the initial beam shape and spectra we conducted full simulations of an X-ray tube, filters and beam shapers for a Siemens Emotion CT. The simulation output files were analyzed to calculate projections of a phantom with inserts. A simple reconstruction algorithm (FBP using a Ram-Lak filter) was applied to calculate the pixel values, which represent an attenuation coefficient, normalized in such a way to give zero for water (Hounsfield unit (HU)). Measured and Monte Carlo calculated CT numbers were compared. The average deviation between measured and simulated CT numbers was 4 ± 4 HU and the standard deviation σ was 49 ± 4 HU. The simulation also correctly predicted the behaviour of H-materials compared to a Gammex tissue substitutes. We believe the developed approach represents a useful new tool for evaluating the effect of CT scanner and phantom parameters on CT number values.

Monte Carlo calculated CT numbers for improved heavy ion treatment planning

West German Proton Therapy Centre Essen (WPE), Hufelandstraße 55, 45147 Essen, Germany

Division of Accelerator Physics, National Centre for Nuclear Research (NCBJ), 7 Andrzeja Soltana Str., 05-400 Otwock/Świerk, Poland, Tel.: +48 22 718 0423, Fax: +48 22 779 3481

Heidelberg Ion-Beam Therapy Centre HIT, Im Neuenheimer Feld 450, 69120 Heidelberg, Germany

Nukleonika

{"article-title":"Monte Carlo calculated CT numbers for improved heavy ion treatment planning"}

Better knowledge of CT number values and their uncertainties can be applied to improve heavy ion treatment planning. We developed a novel method to calculate...

Monte Carlo calculated CT numbers for improved heavy ion treatment planning

Publié en ligne: 25 mars 2014

Pages: 15 - 23

DOI: https://doi.org/10.2478/nuka-2014-0002

This content is open access.