ESTRO 2020 Abstract book

S953 ESTRO 2020

1. Intra Bin Variation (IBV): interquartile range of the diaphragm position within the bin-slice combination, averaged over three central slices and ten respiratory bins. Peak-to-Peak value (PKP): included amplitude range after OR. Reconstruction Completeness (RC) : Percentage of bin-slice combinations containing at least one image after binning. To determine the ISV, the mean and standard deviation (SD) of a QP over six 4DMRIs was evaluated for each volunteer, see Fig. 1B. From this, distributions from the SD over all volunteers were generated. A paired Wilcoxon’s signed rank test was used to test differences in ISV over all volunteers. Results Fig. 1B displays a typical case of a volunteer where applying Min95 OR resulted in less variation in PKP with SDs of 10.7 mm for MaxIE and 6.3 mm for Min95. Fig. 2 shows boxplots of the SDs over all volunteers per OR strategy. Average differences in PKP were significant between Phase and MaxIE, and Min95 (7.6 mm and 4.4 mm respectively) as well as between Min95 and MeanIE (4.4 mm and 3.7 mm). IBV only showed a significant average difference between Phase and Min95 (2.0 mm and 1.5 mm respectively). For RC, differences were significant between Phase and Min95 (1.0% for Phase, 5.4% for Min95). Conclusion The Min95 strategy has a lower intersession variability compared to applying no outlier rejection and the MeanIE strategy. 2. 3.

Conclusion CBCT scans, IGRT reviewed after regular clinic hours, by non-treating physicians, by physicians with less IGRT images to review were associated with increased rates of IGRT rejection. This data can help radiation oncologists know factors that improve the quality of IGRT review. In other words, it appears that more precise volumetric imaging, focused time with minimal distractions, “new eyes” and fewer cases to review were all important in image rejection, along with the disease site and proclivity of the physician. PO-1644 Inter-session variability of 4DMRI image quality after outlier rejection J.K. Veldman 1 , D. Den Boer 2 , A. Bel 2 , Z. Van Kesteren 2 ; 2 Amsterdam UMC - Location AMC, Department of Radiotherapy, Amsterdam, The Netherlands Purpose or Objective For the radiotherapy treatment of tumours in the abdomen, 4DMRI can be used to determine the influence of respiration. Irregularities in respiration may cause artefacts in 4D reconstructions, introducing inaccuracies in treatment. To further improve 4DMRI reconstruction and to be able to handle irregularities in respiration, outlier rejection has been proposed. For radiotherapy purposes, robustness of image quality is an important aspect as, ideally, a single planning image represents the following treatment fractions adequately. Our study aimed to analyze the inter-session variability (ISV) of 4DMRI image quality after outlier rejection (OR), by means of reconstruction quality parameters. Material and Methods Ten healthy volunteers were included, all scanned for two sessions with at least one week in between, each session consisting of three consecutive 4D acquisitions (total of 60 acquisitions). During each acquisition, 11 2D coronal slices were acquired repetitively (60 times) during free- breathing, using a T2W-TSE sequence. Prior to each slice a navigator signal was acquired (position of the diaphragm) and used to bin the 2D slices in ten bins. Subsequently, a 4DMRI was reconstructed using four OR strategies: discarding 5% of the extreme navigator positions whilst minimizing the amplitude range (Min95), phase binning without OR (Phase), amplitude binning without OR (MaxIE) and amplitude binning with thresholds at mean inhale and exhale levels (MeanIE), see Fig. 1A. We compared ISV based on three quality parameters (QP):