2d-3d image registration for image guided intraoperative radiotherapy

Resumen

Intraoperative electron radiotherapy (IOERT) is a technique that combines the surgical resection of a tumour with the therapeutic radiation of the surgically opened tumour bed or of an unresected tumour avoiding irradiating healthy tissue. To avoid transferring the patient from the operating room to the radiation room, a mobile linear accelerator (LINAC) is used. The radiation is delivered through a specific applicator docked to the collimator and the dosimetry is pre-planned with the help of a treatment planning system. As anatomical modifications occur due to the tumour resection process or movements of patient, dosimetry planning needs to be updated prior to radiation delivery. A major concern is the placement of the applicator over the tumour bed in accurate position and orientation (pose), which can be done with the help of intraoperative imaging of the actual scenario during the treatment. Thus, an acquisition protocol and image processing workflow are needed to update the IOERT spatial planning during surgery to ensure an accurate estimation of dose distribution and treatment verification. To achieve this goal, we propose a specific acquisition protocol and an image processing workflow that updates the IOERT planning refreshing the pose of a virtual applicator with respect to the preoperative CT with the actual pose prior to radiation delivery. The image processing workflow relies on a robust registration of the preoperative computed tomography (CT) and intraoperative projection radiographs acquired with a C-arm system. The workflow initially performs a geometric calibration of the C-arm using fiducials placed on the applicator. In the next step, a point-based 2D-3D registration based on fiducials positioned on the patient’s skin is performed, followed by an intensity-based registration that refines the point-based registration result. To assess the feasibility, we performed a preliminary study with a physical plastic bone phantom. The performance of the workflow has been evaluated using a realistic physical phantom consisting of a pig lower limb. The accuracy has been measured with respect to the applicator origin and axis before and after the registration refinement process. Error analysis revealed angular accuracy of 0.9 ± 0.7 degrees and translational accuracy of 1.9 ± 1 mm. Our experiment demonstrated that the proposed workflow can achieve subdegree angular accuracy in locating the applicator with respect to the preoperative CT to update and supervise the IOERT planning right before radiation delivery. An experiment conducted using patient data has also been performed as a proof of concept of the proposed workflow in the clinical scenario, that resulted in a mean final error in translation of 1.2 ± 1 mm and in rotation of 0.5 ± 0.7 degree. The proposed workflow could be easily implementable in clinical routine to improve the quality assurance during IOERT procedures.