ESTRO 2025 - Abstract Book

S3029

Physics - Image acquisition and processing

ESTRO 2025

2899

Digital Poster Frame induced MR distortion in Gamma Knife radiosurgery: assessment at 1.5T and 3T for two frame models Stefanos Theocharis 1 , Eleftherios P Pappas 1 , Efi Koutsouveli 2 , Georgios Kollias 2 , Konstantinos Konstantinidis 3 , Kostas I Lampropoulos 2 , Nikolaos Yakoumakis 2 , Chryssa Paraskevopoulou 2 , Ioannis Seimenis 1 , Pantelis Karaiskos 1,2 1 Medical Physics Laboratory, Medical School, National and Kapodistrian University of Athens, Athens, Greece. 2 Medical Physics and Gamma Knife Department, Hygeia Hospital, Marousi, Greece. 3 Spine Action, Ltd, Athens, Greece Purpose/Objective: In frame-based Stereotactic Radiosurgery (SRS) using the Gamma Knife (GK) platform (Elekta, Sweden), the target is commonly defined on Magnetic Resonance (MR) images acquired with the stereotactic frame mounted on the patient. The aim of this study is to assess the magnitude and range of the MR geometric distortion induced by the stereotactic frame for two frame models, at 1.5T and 3T magnetic field strengths, that could deteriorate target localization accuracy [1].

Material/Methods:

Figure 1: (a) The custom 3D distortion detection phantom with a stereotactic frame. (b) The two clinically used stereotactic frames in Gamma Knife SRS with the corresponding MR localization boxes. A custom 3D distortion detection phantom (Fig 1a), made of acrylic, was utilized throughout this study. It incorporates 11 planes on which a total of 1978 holes were drilled, the centroids of which serve as control points (CPs) for distortion detection [2]. The phantom’s size and shape were designed to cover most of the available space within the two frame models used in clinical practice: the Leksell G-frame and the recently-introduced Vantage frame (Elekta, Sweden) which hereinafter will be referred to as frame A and B, respectively (Fig. 1b). The phantom was MR scanned at 1.5T and 3T employing the clinically-used sequences for multiple brain metastases. The analysis for CP localization and distortion detection has been presented elsewhere [2,3]. Results: At 1.5T, frame A induces distortion which may exceed 2mm in the vicinity of the frame base, but reduces however to less than 0.5mm near the phantom’s centroid. Image artifacts are also evident (Fig. 2b). Distortion magnitude close to the frame base is reduced at 3T (Fig 2c) but (compared to 1.5T) higher mean distortion magnitude was detected in