ESTRO 2024 - Abstract Book

S6037

RTT - Treatment planning, OAR and target definitions

ESTRO 2024

multiple dose-limiting OARs would not be suitable for the ULTRAS protocol. This may aid clinicians' in assessing patient suitabilty for ULTRAs trial accrual, and future guidlines will be based on diagnostic imaging that can be used during early patient consultations.

Keywords: high-dose, liver-mets, MRL

3159

Digital Poster

Implementation of AI-based DLC contours in the pelvic lymph region of prostate tumor patients

Adam Miovecz 1,2,3 , Daniel Gugyeras 1 , Andrea Farkas 1 , Akos Gulyban 4 , Ferenc Lakosi 1,2

1 Moritz Kaposi General Hospital, Dr. Jozsef Baka Center, Department of Radiation Oncology, Kaposvar, Hungary. 2 University of Pécs, Faculty of Health of Sciences, Department of Medical Imaging, Kaposvar, Hungary. 3 University of Pécs, Faculty of Health Sciences, Doctoral School of Health Sciences, Pecs, Hungary. 4 Institut Jules Bordet, Medical Physics Department, Bruxelles, Belgium

Purpose/Objective:

Our goal was an objective and subjective comparison of the pelvic lymph regions segmented contours by three DLC based models in prostate tumor patients using two AI softwares (Mirada*, Mvison**).

Material/Methods:

We selected 10 prostate cases randomly whom were treated at our institute. According to the planned CT institute protocol, with a slice thickness of 3 mm, starting from the L1/2 vertebra and caudal to 3 cm from the ischial tuberosity. For the objective comparison of the three segmentations (Mvision: RTOG, Pivotal and Mirada), we used volumetric (Dice similarity coefficient, "DSC") and distance-based (Median surface distance "MSD", 95% Hausdorff distance, "HD95%)" indices. Subjective evaluation was classified into major (significant) and minor (insignificant) categories according to the degree of modification. In addition, we recorded the duration of the modification and manual contouring.

Results:

Based on the pairwise comparison, the value of the Dice similarity coefficient showed a similar value: RTOG vs. For Pivotal 0.90 (interval 0.89-0.92), Pivotal vs. 0.85 (0.82-0.91) for Mirada, while RTOG vs. In the case of Mirada, it was 0.81 (0.79-0.84). The median surface distance remained below 3.5 mm on average, considering all cases and comparisons: RTOG vs. Pivotal 1.9 mm (1.7-2.2), Pivotal vs. Mirada 2.3 mm (1.6-2.6), RTOG vs. Mirada 2.6 mm (2.3-3.4). Hausdorff distance 95% sym resulted in a difference of up to nearly 15 mm: RTOG vs. Pivotal 10.6 mm (9.1-12.8), Pivotal vs. Mirada 8.1mm (5.3-10.2), RTOG vs. Formed according to Mirada 11.1 (8.5-14.7). In the case of subjective