Room: Karl Dean Ballroom C
Purpose: We propose a deformable registration algorithm for prostate specific membrane antigen (PSMA) PET/CT and transrectal ultrasound (TRUS) fusion. PSMA enables highly specific and sensitive identification of prostatic tumors. Accurate registration of PSMA PET to intraoperative TRUS will allow physicians to customize dose planning based on regions of involvement.
Methods: The input to the registration algorithm is the PET/CT and TRUS volumes as well as prostate segmentations. PET/CT and TRUS volumes are first rigidly registered by maximizing the overlap between the segmented prostate binary masks. Three-dimensional anatomical landmarks are then automatically extracted from the boundary as well as within the prostate, both from the CT and TRUS prostate masks. Deformable registration is performed using thin plate-based spline (TPS) where the extracted landmarks are used as control points. Finally, the PET/CT images are deformed to the TRUS space by using the computed TPS transformation.
Results: The proposed algorithm was evaluated on 10 prostate cancer patients treated by low-dose-rate permanent brachytherapy. Each patient had an intraoperative TRUS and six x-ray images, and a post-implant CT taken 1 day after the implantation. Intraoperative seed locations were reconstructed from the x-rays and registered to TRUS, and post-implant seeds were segmented from the CT. We registered the post-implant CT to TRUS using the proposed algorithm and computed target registration errors (TREs) by comparing the seed locations. Our method outperforms state-of-the-art methods with significantly lower (meanÂ±standard deviation) TRE of 1.98Â±1.22 mm. The registration takes approximately 40 seconds.
Conclusion: We propose a landmark-based PET/CT-TRUS deformable registration algorithm that is simple, computationally efficient, and capable of producing quality registration of prostate boundary as well as internal gland. The proposed method enables fast and accurate intraoperative PSMA PET-TRUS fusion, thus allowing for PET-determined focal prostate brachytherapy.
Funding Support, Disclosures, and Conflict of Interest: Funding source: This work was supported by NIH under grants R44CA099374 and R01CA151395. COI for Junghoon Lee: A US patent (No. 8,233,686) has been filed for the fluoroscopy-based implant reconstruction algorithm.
PET, Registration, Brachytherapy