ESTRO 2025 - Abstract Book

S637

Clinical - Breast

ESTRO 2025

Keywords: Cognitive-Behavioural Therapy, Menopause, CBT

References: 1. Newson, L. (n.d.). Menopause Symptom Sheet. Balance Menopause. Retrieved from https://www.balance menopause.com/menopause-library/menopause-symptom-sheet/ 2. Brady MJ et al (1997) Reliability and validity of the functional assessment of cancer therapy – breast quality of life instrument. J Clin Onc 15(3):974-986. 3. Simard S, Savard J (2015) Screening and co morbidity of clinical levels of fear of cancer recurrence. J Cancer Surv 9:481-491. 4. Spitzer RL, Kroenke K, Williams JBW et al (2006) A brief measure for assessing generalized anxiety disorder. JAMA Internal Medicine 166:1092-1097 5. Kroenke K, Spitzer RL, Williams JB (2001) The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 16:606-13. Digital Poster Bee colony optimization and automation of initial beam positioning in postoperative breast irradiation Neda Milosavljević 1,2 , Tatjana Stojanović 3 , Darko Stojanović 2 , Marko Spasić 4,5 , Marija Živković Radojević 1,2 1 Department of Clinical Oncology, University of Kragujevac, Faculty of Medical Sciences, Kragujevac, Serbia. 2 Centre for radiation oncology, University Clinical Centre Kragujevac, Kragujevac, Serbia. 3 Faculty of Science, University of Kragujevac, Kragujevac, Serbia. 4 Department of Surgery, University of Kragujevac, Faculty of Medical Sciences, Kragujevac, Serbia. 5 Clinic for General Surgery, University Clinical Centre Kragujevac, Kragujevac, Serbia Purpose/Objective: Radiation therapy is an integral part of breast cancer (BC) treatment, with 3DCRT (3D conformal radiation therapy) considered the standard of care, which is time consuming. Templates saves time, without providing adequate values, requiring additional manual corrections. Objective is to generate an original AI (artificial intelligence) optimization tool for placing and optimizing initial beams for 3DCRT planning, facilitating following criteria: to work for as a ESAPI (Eclipse Scripting Application Programming Interface) script for VMS (Varian Medical Systems) Eclipse TPS (Treatment Planning System), for both breast with beams minimizing inclusion of organs at risk. Material/Methods: Bee Colony Optimization (BCO) is a stochastic, population-based algorithm effective in solving hard optimization problems and continuous optimization. We used an enhanced version - BCOi, developed as a script for VMS Eclipse TPS version 16.1, written in C# using Microsoft Visual Studio 2022 and VMS ESAPI libraries. The retrospective study was performed in two phases. First phase included 20 postoperative BC patients, equally distributed within breast sides and type of surgery (breast conserving surgery and mastectomy). The AI created 10 plans for each patient. Five Qualified Medical Physicists (QMP) created one plan each, creating additional structures and placing initial fields, timing the process. The solutions created by AI were compared to those acquired from QMPs. Second phase included testing AI solution on a larger patient set, comparing approved solution to the one suggested by AI. Results: Phase one showed good match between field parameters (gantry and collimator angle, field size and isocenter position) given by AI and QMPs. The AI time is more consistent and shorter than QMPs. Time for QMP to set up the initial fields, without adding contours, vary between 3-22 minutes, while AI time needed to achieve a solution varied from 3.5-4.5 minutes, independent on the anatomical features, laterality, or the need to provide additional contours. 4618