ESTRO 36 Abstract Book

S232 ESTRO 36 _______________________________________________________________________________________________

The phantom was scanned at different tube potentials (80 kV, 120 kV and 140 kV) with a novel SOMATOM Confidence® RT Pro scanner (Siemens Healthcare GmbH, Germany). Images were reconstructed both into HU and ED for each tube potential. Next, the usability of the reconstruction algorithm was evaluated in a clinical workflow. Five patients with an abdominal lesion (e.g. rectal or prostate cancer) were scanned using the clinically used tube potential of 120 kV and an additional dual-spiral dual-energy CT acquisition was made at 80 kV and 140 kV. Dose distributions (Eclipse TM , Varian, USA) of the ED images of the 80 kV, 120 kV, 140 kV acquisitions using the novel reconstruction algorithm were then compared with the clinical plan based on the 120 kV acquisition using the clinical CT to ED curve with the standard HU image of the 120 kV scan. The difference in mean doses delivered to the planning target volume were quantified (i.e. relative difference ± 1 SD). Results The CT to ED conversion curve for the HU images depended on the tube potential of the CT scanner. The novel reconstruction algorithm produced ED values that had a residual from the identity line of -0.1% ± 2.2% for all inserts and energies and is shown in Figure 1. The dose distributions between the standard and the novel reconstruction algorithm were compared for different energies. The relative differences in target dose ranges were small and ranged from -0.2% to 0.7% for 80 kV, -0.1% to 1.1% for 120 kV, and 0.1 to 1.0% for 140 kV.

Results The average beam-on delivery time was 3min, ranging from 1.22min to 8.82min. The average ɣ (3%,3mm) passing rate for film measurements was 97.0 ±1.6% (range from 93.3 to 98.8%), while the Delta 4 measurements presented an average ɣ (2%,2mm) of 97.7±1.4% (range from 95.3 to 99.6%) respectively. The average isocentre point dose ratio was 99.9±1.2% (range from 98.0 to 102.8%). For the lung SBRT verifications with the CIRS phantom, an average local ɣ of 97.0±1,0% and 93.1±2.0% was obtained during the coronal and sagittal film analysis, whereas the average isocentre point dose ratio was 100.0±1.4%. An overall mean gantry-ring geometric deviation of 0.04° ± 0.46° and 0.19° ± 0.26° was obtained, respectively. Conclusion DWA has been successfully added to the non-coplanar rotational IMRT techniques’ arsenal, allowing additional flexibility in dose shaping while preserving dosimetrically robust delivery. In a short period of time, it has become a standard treatment technique on the Vero system in our department. OC-0440 Characterization and clinical evaluation of a novel CT reconstruction to derive electron densities B. Van der Heyden 1 , M. Ollers 1 , C. Loon Ong 1 , F. Verhaegen 1 , W. Van Elmpt 1 1 School for Oncology and Developmental Biology- Maastricht University Medical Centre, Department of Radiation Oncology MAASTRO- GROW, Maastricht, The Netherlands Purpose or Objective Radiotherapy dose calculations are almost exclusively performed on CT images. In a clinical workflow, the Hounsfield Units (HU) are converted into electron density (ED) by using a CT to ED conversion curve calibrated for a typically fixed tube potential (e.g. 120 kV). Recently, a novel CT image reconstruction algorithm (DirectDensity TM , Siemens Healthcare GmbH, Germany) was developed that directly reconstructs the ED, independent of the tube potential of the CT scanner. This allows the elimination of a conversion curve for each tube potential. Our study describes the accuracy in terms of dose calculation of the reconstruction algorithm based on phantom measurements and shows the application in a clinical radiation therapy workflow for different tube potentials. Material and Methods The accuracy of the novel reconstruction algorithm to derive ED was validated in a Gammex RMI 467 phantom (Gammex, USA) using different tissue mimicking inserts.

Figure 1: The linear conversion curve (fitted) of the novel reconstruction algorithm. Conclusion A novel reconstruction algorithm to derive directly relative electron density irrespective of the tube potential of the CT scanner was evaluated. A single identity curve for the CT to ED could be used in the treatment planning system. This reconstruction algorithm may enhance the clinical workflow by selecting an optimal tube potential for the individual patient examination that is not restricted to the commonly used 120 kV tube potential. OC-0441 Dose Prescription Function from Tumor Voxel Dose Response for Adaptive Dose Painting by Number D. Yan 1 , S. Chen 2 , G. Wilson 1 , P. Chen 1 , D. Krauss 1 1 Beaumont Health System, Radiation Oncology, Royal Oak MI, USA 2 Beaumont Health System, Radiation Oncology, Royal Oak, USA Purpose or Objective Dose-painting-by-number (DPbN) needs a novel Dose Prescription Function (DPF) which provides the optimal clinical dose to each tumor voxel based on its own dose response. To obtain the DPF for adaptive DPbN, a voxel- by-voxel tumor dose response matrix needs to be constructed during the early treatment course. The study demonstrated that the voxel-by-voxel tumor dose

Made with