The effect of catheter displacement and anatomical variations on the dose distribution in MRI-guided focal HDR brachytherapy for prostate cancer.

PURPOSE: The aim of this study was to analyze the effect of catheter displacement and anatomical variations of prostate and organs at risk on dose distribution in MRI-guided 19 Gy single fraction focal high-dose-rate brachytherapy (HDR-BT) of the prostate. METHODS AND MATERIALS: Seventeen patients with localized prostate cancer were enrolled in a prospective trial investigating focal HDR-BT in a 1.5 T MRI-HDR-BT facility. The diagnostic MRI delineations were registered with intraoperative MR scan, and a single fraction of 19 Gy was applied to the visible tumor. Self-anchoring umbrella catheters were used for HDR-BT delivery. A 1.5 T MRI was performed directly after ultrasound (US)-guided catheter placement for treatment planning. After treatment and before removal of catheters, a posttreatment 1.5 T MRI was performed. Regions of interest were also delineated on the posttreatment MR images and the catheters of 17 patients were reconstructed. The dose plan was constructed for the posttreatment MRI scan to assess the influence of catheter migration and anatomical variation on the dose delivered to the target and the organs at risk. Also on the posttreatment MRI, the complete catheter reconstruction was reassessed, to correct for, for example, bending of the catheters. The displacement of catheters between the MRI scans was determined by comparing the catheter tip positions on the treatment planning and posttreatment 1.5 T MRI scans. RESULTS: The displacements of 241 catheters were investigated. Average (range) displacements of the umbrella catheters are 0.6 (0-2.9) mm in the x-direction, 0.5 (0-2.1) mm in the y-direction, and 0.9 (0-5.5) mm in the z-direction. In 3 patients, the displacement was >4 mm and up to 5.5 mm. This occurred in respectively 1/13, 1/16, and 1/18 catheters in these patients. The dosimetric differences between the intraoperative treatment and the posttreatment plans were in most patients less than 1.5 Gy. In 4 patients, a dose difference in clinical target volume D95 of >2 Gy up to 5.8 Gy was reported. No discrimination can be made between dose differences due to catheter displacement and/or organ movement/anatomy changes. CONCLUSIONS: In general, catheter displacements were in the order of a mm and differences in dose to the clinical target volume and the organs at risk between the treatment and posttreatment plans smaller than 1.5 Gy. In some patients, dose differences up to 5.8 Gy were determined, due to either individual larger catheter displacement and/or anatomy changes. A longer followup is necessary to assess the clinical implications of individual large dose differences.

Fiber Bragg gratings-based sensing for real-time needle tracking during MR-guided brachytherapy.

PURPOSE: The development of MR-guided high dose rate (HDR) brachytherapy is under investigation due to the excellent tumor and organs at risk visualization of MRI. However, MR-based localization of needles (including catheters or tubes) has inherently a low update rate and the required image interpretation can be hampered by signal voids arising from blood vessels or calcifications limiting the precision of the needle guidance and reconstruction. In this paper, a new needle tracking prototype is investigated using fiber Bragg gratings (FBG)-based sensing: this prototype involves a MR-compatible stylet composed of three optic fibers with nine sets of embedded FBG sensors each. This stylet can be inserted into brachytherapy needles and allows a fast measurement of the needle deflection. This study aims to assess the potential of FBG-based sensing for real-time needle (including catheter or tube) tracking during MR-guided intervention. METHODS: First, the MR compatibility of FBG-based sensing and its accuracy was evaluated. Different known needle deflections were measured using FBG-based sensing during simultaneous MR-imaging. Then, a needle tracking procedure using FBG-based sensing was proposed. This procedure involved a MR-based calibration of the FBG-based system performed prior to the interventional procedure. The needle tracking system was assessed in an experiment with a moving phantom during MR imaging. The FBG-based system was quantified by comparing the gold-standard shapes, the shape manually segmented on MRI and the FBG-based measurements. RESULTS: The evaluation of the MR compatibility of FBG-based sensing and its accuracy shows that the needle deflection could be measured with an accuracy of 0.27 mm on average. Besides, the FBG-based measurements were comparable to the uncertainty of MR-based measurements estimated at half the voxel size in the MR image. Finally, the mean(standard deviation) Euclidean distance between MR- and FBG-based needle position measurements was equal to 0.79 mm(0.37 mm). The update rate and latency of the FBG-based needle position measurement were 100 and 300 ms, respectively. CONCLUSIONS: The FBG-based needle tracking procedure proposed in this paper is able to determine the position of the complete needle, under MR-imaging, with better accuracy and precision, higher update rate, and lower latency compared to current MR-based needle localization methods. This system would be eligible for MR-guided brachytherapy, in particular, for an improved needle guidance and reconstruction.