A Novel Approach to Utilization of Magnetic Resonance Imaging (MRI) in High-Dose-Rate (HDR) Brachytherapy
High-dose-rate (HDR) brachytherapy is a commonly used radiotherapy technique for treating gynecological and skin cancers. Localized delivery of the radiation is regulated by moving the radioactive isotope through catheters on various types of applicators. Accurate identification of catheter position is crucial for creating a clinical plan and delivering the intended dose to the target. Hence, imaging has become a critical component in the process of HDR brachytherapy. Computer tomography (CT) has been the principal choice of imaging modality for catheter reconstruction. CT, however, lacks soft tissue information that is useful for optimizing plan quality. Magnetic resonance imaging (MRI) provides soft tissue information but lacks the catheter/applicator contrast needed for reconstruction. Current HDR brachytherapy workflow for skin cancer does not include MRI. It is solely reliant on CT for reconstruction and optimizes plan to a single target depth, instead of an image-based optimization. Gynecological cancer workflow involves both CT and MRI, for catheter reconstruction and tissue segmentation, respectively. Compared to a single-imaging approach, hybrid imaging resulting to increased procedural time, use of hospital resources, and potential sources of errors. Ability to reconstruct catheters on MRI may eliminate CT from the current workflow and provide means to perform both catheter reconstruction and tissue segmentation with a single imaging modality. In this work, a clinically available MRI sequence – Pointwise encoding time reduction with radial acquisition (PETRA) -- was optimized for catheter reconstruction. Initial evaluation of reconstruction and dosimetric accuracies of a plan generated using PETRA images suggested this MRI-only approach shows promising clinical applicability. Evaluation on its clinical implantation feasibility, along with challenges, of MRI-only brachytherapy with PETRA is discussed.
