ESTRO 2021 Abstract Book

S705

ESTRO 2021

23.62 vs 24.71) and for the thyroid (DICE: 0.85 vs 0.74; HD: 8.43 vs 13.73; HD 95 Fig.1. Average DICE per organ obtained with different segmentation algorithms.

: 3.04 vs 8.00).

Conclusion The ADMIRE multi-ABAS was capable of segmenting OARs in HN region with comparable or superior accuracy to a DL model. Further work is in progress to evaluate the clinical value of these results by assessing the impact of contouring inaccuracies with respect to dose coverage, and by evaluating the time needed for manual correction of the automatically delineated contours. PD-0869 influence of chest geometry on treatment technique selection for breast cancer radiotherapy D. Mychko 1 , P. Dziameshka 2 1 Minsk City Clinical Oncologic Dispensary, Radiotherapy Department, Minsk, Belarus; 2 N.N. Alexandrov National Сancer Centre of Belarus, Radiation Therapy Department, Lesnoy, Belarus Purpose or Objective Chest metrics were found to be a predictor for left-sided breast cancer (LSBC) RT heart toxicity [Pili et al. 2011; Mendez et al. 2018]. The study aims to evaluate chest geometry influence on LSBC RT technique selection. Materials and Methods 80 LSBC patients were enrolled in retrospective study. 3D-CRT, IMRT, VMAT free-breathing (FB) and DIBH plans (50Gy in 2Gy per fx, 5fx/week) were generated for each patient. The OAR difference value in pairs 3D-CRT/IMRT, 3D-CRT/VMAT, IMRT/VMAT, DIBH/BH techniques were calculated for 480 plans. Statistical tests (Friedman, Nemenyi) were conducted on comparisons (p<0.05). Clinically significant differences of studied parameters were defined for the heart and left ventricle (threshold values for doses difference: D mean >2 Gy, D max >10%, V 25Gy >10% of volume), LAD-region (threshold values: D max >10%, V 305Gy >10% of volume). Chest geometry (l max , h max , X/Y, α) was measured on 160 series of planning CT scans for FB and DIBH RT (picture 1).

Picture 1. – Evaluated planimetric parameters for LSBC RT planning In order to determine the influence of planimetric parameters on the OAR dose difference value in pairs of the techniques, 288 scatter plots were assessed. Planimetric parameters threshold values when clinically significant difference between RT plans is observed were determined using ROC-analysis.