S1893
Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning
ESTRO 2026
Digital Poster 2657
Increasing workflow efficiency and safety using the scripting interface of the treatment planning system Johanna Rieke 1 , Stefan Menkel 1,2 , Maria Tschiche 1,2 , Falk Tillner 1,2 , Lena Nenoff 1 , Christian Richter 1,2 1 OncoRay - National Center for Radiation Research in Oncology, Medical Faculty and University Hospital Carl Gustav Carus, TUD Dresden University of Technology, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany. 2 Department of Radiotherapy and Radiation Oncology, Medical Faculty and University Hospital Carl Gustav Carus, TUD Dresden University of Technology, Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany Purpose/Objective: To improve patient safety and treatment efficiency of our clinical workflow we implemented scripts for partial automation in our treatment planning system (TPS). Material/Methods: We identified workflow steps suitable for automation within the TPS and developed and validated five scripts to improve the workflow (Fig. 1). The first script checks the contours for suspicious features such as contour fragments, inconsistent target volumes, or missing interpolations. A second script creates a blank plan with pre-filled fraction and plan options and automatically imports and aligns support structures. The third script provides treatment field sorting and naming. The fourth script executes plausibility checks on the final plan to mitigate known pitfalls. Finally, a fifth script supports plan documentation and archiving and performs all exports to systems needed for the treatment. For validation, we checked if intentionally added errorswere correctly identified (scripts 1, 4, 5). To assess the efficiency gain for scripts 2, 3, and 5 we measured time and number of mouse clicks for the workflow step execution manually and with script for 15, 16, and 21 plans, respectively. Results: The scripts correctly detected all manipulations: script 1 (contour check) found missing interpolations, contour fragments, and logical inconsistencies. Script 4 (plausibility checks) detected e.g. wrong dose grids, fields with too low monitor units, missing final doses, erroneous bolus, and wrong beam models. With script 5 (plan report) we detected various added errors such as a missing reference point.The scripts showed a substantial efficiency gain (Fig. 2):
Script 2 (plan creation) reduced the time by 18s and the number of mouse clicks by 27. Script 3 (field sorting and naming) showed a reduction of 65s and 27 clicks. Together with script 4 they mitigate discrepancies from our internal conventions that occurred frequently in manual planning. This prevents manual corrections and double-work at the earliest possible workflow step.Plan documentation includes report creation, plan export, and archiving. Although the TPS provides report templates, script 5 reduced the time by 336s and the clicks by 112.Altogether, the scripts saved 419s (-47%) and 166 mouse clicks (-64%) necessary for these workflow steps. Moreover, the additional and redundant checks improve plan consistency and provide enhanced safety against plan deficiencies. Conclusion: We successfully integrated five scripts for automation in our clinical routine. This improves workflow efficiency, reduces error probability and mitigates known pitfalls, all enhancing patient safety. Keywords: Treatment planning, automation, scripting A Multicenter Knowledge-based approach for prediction of dose gradient and guidance of plan optimization in robotic intracranial SRS/SRT. Sara Broggi 1 , Marcello Serra 2 , Raffaella Doro 3 , Anna Stefania Martinotti 4 , Irene Redaelli 4 , Maria Cristina Frassanito 5 , Carmelo Ragusa 6 , Giulia Rambaldi Guidasci 7 , Federica Murtas 7 , Rita Buono 1 , Claudio Fiorino 1 , Laura Masi 3 1 Medical Physics, IRCCS San Raffaele Scientific Institute, Milano, Italy. 2 Radiation Oncology, Istituto Nazionale Tumori -IRCCS-Fondazione G. Pascale, Napoli, Italy. 3 Medical Phyisics and Radiation Oncology, IFCA, Firenze, Italy. 4 Cyberknife, Centro Diagnostico Italiano, Milano, Italy. 5 Cyberknife Center, Mater Dei Hospital, CBH Città di Bari Hospital, Bari, Italy. 6 Medical Physics, A.O.U. Policlinico G. Martino, Messina, Italy. 7 UOC di Radioterapia Oncologica, Ospedale Isola Tiberina-Gemelli Isola, Roma, Italy Purpose/Objective: This study aimed to develop and validate a multi- center knowledge based (KB) model to predict dose gradient of CyberKnife (CK) brain SRS/SRT plans to Digital Poster 2673
Made with FlippingBook - Share PDF online