ESTRO 2021 Abstract Book

S740

ESTRO 2021

Conclusion All MLC models achieved reasonable accuracy in clinical plans. The prototypes slightly improve the agreement and are easier to configure with the SG tests. This could improve the standardisation and consistency across centres, as it would be less likely for individuals to take different ‘paths’ to the final values. References [1] Glenn et al, Med Phys 2019. [2] Saez et al, PMB 2020. PD-0901 Is it possible to predict QA failures using plan complexity metrics? S. Russo 1 , G. Della Gala 2 , S. Bettarini 3 , A. Ghirelli 1 , M. Esposito 1 , S. Pini 1 , H.O. Ghafour 4 , V. Hernandez 5 1 Azienda USL Toscana Centro, Medical Physics Unit - Sede di Firenze, Firenze, Italy; 2 IRCCS Azienda Ospedaliera Universitaria di Bologna, Medical Physics Unit, Bologna, Italy; 3 Università degli Studi di Firenze, Specialization School in Medical Physics, Firenze, Italy; 4 Ministry of Health KRG / Sulaymani Directorate of Health, Zhianawa Cancer Center, Zhianawa Cancer Center, Sulaymaniyha, Iraq; 5 Hospital Sant Joan de Reus, Department of Medical Physics , Tarragona, Spain Purpose or Objective To evaluate the potential of the main reported complexity metrics for predicting failures from different patient-specific QA systems Materials and Methods Forty VMAT lung (non SBRT) plans were elaborated with TPS Elekta Monaco 5.12 for an Elekta Synergy. Pre- treatment verifications were performed using three PTW devices: Octavius 729, Octavius 1500 and Octavius 1600SRS inserted in the Octavius 4D phantom for 3D dose distribution measurements. Following AAPM TG-218 recommendations, 3D absolute dose gamma analysis with global normalization (90% of maximum dose) and 3%/2mm criteria was performed (via PTW VeriSoft software 6.1). Local action limits (AL) and control limits (CL) were computed using the Statistical Process Control (SPC) methodology. Results were compared with paired t-Test and statistical significance was considered for p<0.05. Plan complexity evaluation was performed using several complexity metrics (CM): Modulation Complexity Score (MCS), Total Modulation Index (MI Total ), Edge Metric, Leaf Travel, Plan Irregularity, Plan modulation, Average leaf pair opening (mean MLC gap) and its first quartile (Q1gap) and mean TGi. Spearman’s correlation coefficients between CM and gamma passing rate were computed, and statistical significance was assessed