Abstract

Purpose Prostate cancer radiotherapy workflows solely based on magnetic resonance imaging (MRI), are now being clinically used. Intra-prostatic gold fiducial markers (GFM), appear as signal voids in MRI-images while calcifications and bleedings show similar signal behavior. Achieving accurate identification of GFM in an MRI-based workflow thereby constitute a major challenge. Several identification methods have been presented but none has reached an accuracy of 100%. C-arm X-ray images (CkV-images), acquired at GFM implantation, could provide GFM position information and be used to indicate correct identification. This would require negligible GFM migration. The aim of this study was to: 1) study the GFM migration 2) show the feasibility of using CkV-images as an indicator of accurate GFM identification. Methods Frontal digitally reconstructed radiographs (DRR)- and CkV-images were acquired two weeks apart for 31 patients in a CT-based radiotherapy workflow and for 16 patients in an MRI-based radiotherapy workflow. A common image geometry was defined between the DRR- and CkV-image for each patient. For each image, a point cloud was defined from the GFM center of mass coordinates. A rigid registration between the point clouds was performed and the distance between each of the GFM in the DRR- and registered CkV-image was calculated. The distance calculated in the CT-based patient cohort was considered a measure of GFM migration. A statistical test was performed to assess any difference between the cohorts. Results The mean GFM migration assessed in the CT-based patient cohort was 1.2 ± 0.7 mm. The mean absolute distance difference for the GFM center of mass in the MRI-based cohort was 1.7 ± 1.4 mm. No significant difference between the measured total distances of the two patient cohorts could be detected (p-value = 0.37). Conclusions A C-Arm X-ray image acquired from the GFM implantation procedure could be used as an indicator of accurate GFM identification. GFM migration was present but did not constitute a problem for the proposed method.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call