Survey on fan-beam computed tomography for radiotherapy: Imaging for dose calculation and delineation

From computed tomography innovation to routine clinical application in radiation oncology – A joint initiative of close collaboration

“…Hence, variations in CT numbers of the same material directly translate into uncertainties in dose calculation. Several CT innovations, which improve the stability of CT numbers, have been widely implemented by now [16] : Iterative image reconstruction for reducing image noise [17] ; Metal artefact reduction for mitigating streak artefacts caused by beam hardening, scattering and photon starvation near metal objects [18] ; Beam hardening correction for reducing cupping artefacts and size dependencies of CT numbers in high-attenuating materials like bone or iodine [19] . However, the current inter-centre variation in stopping-power prediction from single-energy CT (SECT) using a Hounsfield look-up table (HLUT) still adds up to 2.5–3.0% in proton range as assessed by the European particle therapy network (EPTN) [20] , [21] .…”

“…However, the clinical transition from conventional CT to DECT is not only a simple change of scan protocols, but it impacts the entire RT chain from delineation and treatment planning to image registration at the treatment machine before dose delivery. Consequently, it is rarely used in routine RT workflows up to now [16] , [21] . Most RT departments still use one single image dataset for all required RT tasks, whereas DECT can provide a huge variety of different datasets tailored to each clinical step [27] : A virtual monoenergetic image (VMI) of low energy can enhance the tissue contrast for tumour and organ segmentation; A VMI of high energy can be beneficial to mitigate artefacts caused by beam hardening (e.g., weakly attenuating metallic implants); An iodine enhancement image can be used for perfusion assessment in tumour and healthy tissue; Material parameters such as relative electron density or stopping-power ratio can be directly derived from DECT and used for dose calculation.…”

“…In addition to CT innovations driven by diagnostic requirements, several CT applications with RT purpose were introduced recently. Special reconstruction kernels for SECT provide images with information on mass density or electron density independent from the selected tube voltage, which can be directly used for photon dose calculation without the need of voltage-specific HLUTs [16] , [36] , [37] , [38] . This allows for optimising the image contrast by automatically selecting the most suited tube voltage for each patient and body region examined.…”