ESTRO 2022 - Abstract Book

S1475

Abstract book

ESTRO 2022

Figure 1: Three different stent geometries: A) straight, B), curved and C) branched stent.

Table 1: Global gamma index (±3mm / ±3%) of the three geometries for the calculated and measured CyberKnife and Brachytherapy treatments. CyberKnife Brachytherapy Stationary phantom Moving phantom Stationary phantom Pass % Average Pass % Average Pass % Average Straight 97.7 0.37 96.3 0.37 96.1 0.33 Curved 96.1 0.42 94.9 0.44 68.4 0.79 Branched 84.3 0.60 93.1 0.44 93.8 0.42

Conclusion The CK system was able to track and treat the moving 3D phantoms resembling bile duct obstruction after stent placement using fiducial markers. CK was able to produce accurate dose distributions in the moving phantom and could be an option for the palliative treatment of bile duct obstruction. The dose delivery accuracy was also comparable to BT.

PO-1676 Comprehensive evaluation of ProtegeAI Prostate 2.0 auto-segmentation: time-gain and accuracy

N. Jullian 1 , Z. Paquier 2,3 , M. Burghelea 2,3 , D. Van Gestel 1 , N. Reynaert 2,3 , A. Gulyban 2,3

1 Institut Jules Bordet, Université Libre de Bruxelles (ULB), Radiation Oncology, Brussels, Belgium; 2 Institut Jules Bordet, Medical Physics, Brussels, Belgium; 3 Université Libre de Bruxelles (ULB), Radiophysics and MRI physics laboratory, Brussels, Belgium Purpose or Objective The aim of this study was to evaluate the time gain and accuracy of the MIM ProtegeAI 2.0 auto-segmentation solution (version 7.1.5, MIM software Inc, Cleveland OH, USA). A second objective was to assess intra-observer variability and familiarization bias when using auto-segmentation. Materials and Methods Twenty-five patients with prostate cancer were included. For each case a planning CT scan (from vertebrae L1/2 to 3cm below the ischial tuberosity, 3mm slice thickness) was performed, followed by auto-segmentation using the ProtegeAI Prostate 2.0 model ( AI ) and manual delineation by a single observer ( Manual ). Femur_L/_R, PenileBulb, Rectum, SeminalVes, Bladder were evaluated; while another five AI-generated OARs did not match our institutional template, hence were not evaluated. Time of AI delineation scoring ( AIscor : major/minor/no correction needed), AI correction ( AIcor ), total AI (=AIscor+AIcor) and manual delineation was measured. Time gain was also calculated per individual OAR. Half of the cohort started with AIscor and AIcor followed by Manual , while the other half started with Manual , followed by AIscor