ESTRO 2023 - Abstract Book

S1400

Digital Posters

ESTRO 2023

Results The network segmented well real CT and CBCT images with DSC ranging 0.89 – 0.98 for all tissue types (Tab 1, “Validation”). The automatic contours of GI gas separated individual gas pockets more commonly than ground-truth segmentations. Contrast enhanced vessels were sometimes mislabelled as skeleton. The network allowed us to quantitatively compare the quality of different synCTs, which were compared against ground-truth segmentations on CBCT (Tab 1, “Application”). In general, synCT AI performed the best in terms of structure consistency. While both methods performed similarly in the lungs (HD={3.5, 3.5} mm for {DIR, AI}), the AI method was advantageous to DIR in GI gas regions (HD={24.0, 6.1} mm for {DIR, AI}). This indicates the need for additional strategies to account for non-deformable anatomical change when using DIR. Note that GI gas was segmented correctly as suggested by the similarity in HUs between CT and synCTs (HU={-805, - 762, -759} for {CT, DIR, AI}).

Conclusion The proposed automated tissue segmentation method is well-suited to evaluate the structure consistency of synCTs. It can used to quantitatively compare different methodologies and identify their limitations. It also has potential application in automatic quality assurance of synCTs in clinical workflows.

PO-1690 The influence of CBCT reconstruction or sCT generation on bony anatomy based position verification.

Y. de Hond 1 , C. Hurkmans 2 , R. Tijssen 2 , P. van Haaren 2

1 Catharina Hospital , Radiotherapy, Eindhoven, The Netherlands; 2 Catharina Hospital, Radiotherapy, Eindhoven, The Netherlands Purpose or Objective CBCT image quality is less compared to planning-CT(pCT). To improve CBCT quality, several iterative reconstruction methods and several deep-learning networks have been introduced. The CBCT is used to assure patient positioning and verify if the patients anatomy is still valid for the given treatment plan prior to treatment. The geometric accuracy of the