Integration of Intraoperative Ultrasound Volume Into CT-Based Treatment Planning in Prostate HDR Brachytherapy: A Pilot Study

Abstract

Purpose/Objective(s): The main challenge for CT-based prostate HDR treatment planning is to accurately define target (prostate) volume in CT images due to the poor soft-tissue contrast. To overcome this limitation, we propose a novel approach that integrates an intra-operative ultrasound (US)-based prostate volume into treatment planning through US-CT coregistration based on the catheters (needle tracks). Materials/Methods: In CT-based prostate HDR brachytherapy, needle placement is commonly performed under the guidance of intra-operative transrectal ultrasound, while treatment planning is based on post-operative prostate CT images. However, prostate delineation using CT images is difficult. We have developed a technology that integrates an intraoperative US-based prostate volume into HDR treatment planning. Our approach requires an additional 3D US images acquired in the operating room right after the catheters are inserted, which takes less than 5 minutes. These US images are then used to create prostate contours. The catheters are reconstructed on intra-operative US and post-operative CT images, and subsequently used as landmarks for the US-CT image registration using a fuzzy-to-deterministic algorithm. The intra-operative US-based prostate volume is then deformed using B-splines to the CT image to obtain the accurate CT prostate volume, which is used for treatment planning. Three fiducial (gold) markers are implanted into the prostate and the displacements between their positions on the postregistration US and CT images are used to evaluate the accuracy of our technique. This technique was first tested using a prostate phantom, and further investigated in a prospective study of 10 consecutive patients (age: 59 10) who had received HDR monotherapy or boost brachytherapy for prostate cancer. Results: For the prostate phantom study, 14 catheters and 3 gold markers were implanted. The mean displacement of the 3 gold markers between the post-registration US and CT images was 0.41 0.11 mm, which demonstrated the precision of our intra-operative US-based prostate volume approach. In the pilot study of 10 patients, 14-18 catheters were implanted and 9 out of 10 patients had 3 gold markers implanted. US-CT registration was successfully performed for all 10 patients. The mean displacement of 27 gold markers of the 9 patients was 1.09 0.25 mm. This further demonstrated the feasibility and accuracy of our approach. Conclusions:We have introduced a novel approach that integrates intraoperative ultrasound-based prostate volume into CT-based prostate HDR treatment planning, and demonstrated it clinical feasibility. This technique will improve prostate delineation, enable accurate dose planning and delivery, and potentially enhance prostate HDR treatment outcome. Author Disclosure: X. Yang: None. P. Rossi: None. T. Ogunleye: None. W.J. Curran: None. T. Liu: None.