The 3 D dose distribution was collected from each course of treatment, including external beam radiotherapy and brachytherapy as described in our previous article [24]. For each patient, the different treatment planning images were registered to the most recent image set, which served as a reference image. A deformable registration was performed in Raystation using a mutual-information-based B-spline algorithm. The registration was visually validated for all patients, and a quantitative validation was performed for a subgroup of 14 patients. The latter consisted of manual delineation of three OARs (larynx, mandible, and carotids) in all treatment-planning images, for each patient, followed by a comparison of dose/volume statistics associated with the propagated OAR from deformable registration and the reference image structures, respectively.
The planned dose distribution from each course of treatment was converted to EQD2, using an in-house tool. This was followed by the dose accumulation from all courses of treatment on the reference CT series, facilitated by the deformable registration. Accumulated dose/volume statistics in EQD2 were then extracted for the OARs for each patient.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.