ESTRO 2025 - Abstract Book

S312

Brachytherapy – Head & neck, skin, eye

ESTRO 2025

Material/Methods: 10pts underwent sBT using FreiburgFlap applicator were included. The FreiburgFlap was equipped with 2÷6 catheters to cover adequately the target and secured to patient specific mask for reproducible positioning. CBCT-based plans were computed by TPS OncentraBrachy. The applicator geometry was manually reconstructed by tracing X-ray marker within the catheter's lumen, with the indexer calculated indirectly via TPS and double-checked using ruler. For each patient, a pre-treatment planQA was created with the original reconstruction and the first marker of all channels activated for 30s. Plans were delivered using 192 Ir source via Flexitron afterloader. For each plan, multiple CBCT (q-CT) were acquired synchronized to the source's pre-programmed dwell positions. The q-CT were transferred to OncentraBrachy for image registration. The planning CBCT (p-CT) was used as reference and fused with q-CT via landmark-based registration, selecting the coordinates of three marker-points placed on the mask. Alignment between planned and actual dwell positions was verified by superimposing the registered image datasets.

Results: The matches between q-CT and p-CT were evaluated in terms of least square fit of the two sets of user-defined pair points. All registrations showed an average squared error <0.1 cm and have been accepted. The alignment of the marker with the actual source position was verified, confirming the accuracy of applicator reconstruction for all patients.

Conclusion: An online method has been proposed to verify simultaneously the correctness of manual reconstruction, source position and indexer length, minimizing any uncertainties associated with manual intervention. The opportunity to conduct in-room imaging at the moment of the source delivery is a precious tool for plan QA in sBT.

Keywords: Surface Brachytherapy, QA, Image fusion