Center for Interventional Oncology, Radiology and Imaging Sciences, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.
Urologic Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.
Eur Radiol Exp. 2020 Apr 16;4(1):24. doi: 10.1186/s41747-020-00147-4.
Uncorrected patient or prostate motion may impair targeting prostate areas during fusion image-guided procedures. We evaluated if a prototype "tracked Foley catheter" (TFC) could maintain fusion image alignment after simulated organ motion.
A pelvic phantom model underwent magnetic resonance imaging (MRI), and the prostate was segmented. The TFC was placed in the phantom. MRI/ultrasound (US) fusion was performed. Four trials were performed varying motion and TFC presence/absence: (1) TFC/no-motion, (2) TFC/motion, (3) no-TFC/no-motion, and (4) no-TFC/motion. To quantify image alignment, screen captures generated Dice similarity coefficient (DSC) and offset distances (ODs) (maximal US-to-MRI distance between edges on fusion images). Three anatomical targets were identified for placement of a needle under fusion guidance. A computed tomography scan was used to measure system error (SE), i.e., the distance from needle tip to intended target.
The TFC presence improved MRI/US alignment by DSC 0.88, 0.88, 0.74, and 0.61 in trials 1, 2, 3, and 4, respectively. Both OD (trial 2 versus trial 4, 4.85 ± 1.60 versus 25.29 ± 6.50 mm, p < 0.001) and SE (trial 2 versus trial 4, 6.35 ± 1.31 versus 32.16 ± 6.50 mm, p < 0.005) were significantly lower when the TFC was present after artificial motion, and significantly smaller OD when static (trial 1 versus trial 3, 4.29 ± 1.24 versus 6.42 ± 2.29 mm, p < 0.001).
TFC provided better image alignment with or without simulated motion. This may overcome system limitations, allowing for more accurate fusion image alignment during fusion-guided biopsy, ablation, or robotic prostatectomy.
在融合图像引导手术中，未经校正的患者或前列腺运动会影响前列腺区域的靶向定位。我们评估了一种原型“跟踪 Foley 导管”（TFC）在模拟器官运动后是否能保持融合图像的对准。
一个骨盆模型接受了磁共振成像（MRI），并对前列腺进行了分割。将 TFC 放置在模型中。进行了 MRI/超声（US）融合。进行了四种试验，改变了运动和 TFC 的存在/不存在：（1）TFC/无运动，（2）TFC/运动，（3）无 TFC/无运动，和（4）无 TFC/运动。为了量化图像对准，屏幕捕获生成了骰子相似系数（DSC）和偏移距离（OD）（融合图像上边缘之间的最大 US 到 MRI 距离）。在融合引导下，确定了三个解剖目标来放置一根针。使用计算机断层扫描（CT）扫描测量系统误差（SE），即针尖到目标的距离。
TFC 的存在分别将 MRI/US 对准提高了 0.88、0.88、0.74 和 0.61，在试验 1、2、3 和 4 中。OD（试验 2 与试验 4，4.85 ± 1.60 与 25.29 ± 6.50 毫米，p < 0.001）和 SE（试验 2 与试验 4，6.35 ± 1.31 与 32.16 ± 6.50 毫米，p < 0.005）在人工运动后 TFC 存在时显著降低，在静态时 OD 显著减小（试验 1 与试验 3，4.29 ± 1.24 与 6.42 ± 2.29 毫米，p < 0.001）。
TFC 提供了更好的图像对准，无论是否有模拟运动。这可能克服系统限制，在融合引导活检、消融或机器人前列腺切除术期间允许更精确的融合图像对准。
