Robotics and Control Laboratory, Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC V6T 1Z4, Canada.
IEEE Trans Biomed Eng. 2013 Sep;60(9):2663-72. doi: 10.1109/TBME.2013.2262499. Epub 2013 May 13.
Robot-assisted laparoscopic radical prostatectomy (RALRP) using the da Vinci surgical system is the current state-of-the-art treatment option for clinically confined prostate cancer. Given the limited field of view of the surgical site in RALRP, several groups have proposed the integration of transrectal ultrasound (TRUS) imaging in the surgical workflow to assist with accurate resection of the prostate and the sparing of the neurovascular bundles (NVBs). We previously introduced a robotic TRUS manipulator and a method for automatically tracking da Vinci surgical instruments with the TRUS imaging plane, in order to facilitate the integration of intraoperative TRUS in RALRP. Rapid and automatic registration of the kinematic frames of the da Vinci surgical system and the robotic TRUS probe manipulator is a critical component of the instrument tracking system. In this paper, we propose a fully automatic registration technique based on automatic 3-D TRUS localization of robot instrument tips pressed against the air-tissue boundary anterior to the prostate. The detection approach uses a multiscale filtering technique to identify and localize surgical instrument tips in the TRUS volume, and could also be used to detect other surface fiducials in 3-D ultrasound. Experiments have been performed using a tissue phantom and two ex vivo tissue samples to show the feasibility of the proposed methods. Also, an initial in vivo evaluation of the system has been carried out on a live anaesthetized dog with a da Vinci Si surgical system and a target registration error (defined as the root mean square distance of corresponding points after registration) of 2.68 mm has been achieved. Results show this method's accuracy and consistency for automatic registration of TRUS images to the da Vinci surgical system.
机器人辅助腹腔镜前列腺根治术（RALRP）结合达芬奇手术系统是目前局限于临床前列腺癌的治疗选择。鉴于 RALRP 手术部位的视野有限，一些研究小组提出将经直肠超声（TRUS）成像集成到手术流程中，以协助准确切除前列腺并保留神经血管束（NVB）。我们之前引入了一种机器人 TRUS 操作器和一种自动跟踪达芬奇手术器械与 TRUS 成像平面的方法，以便于在 RALRP 中集成术中 TRUS。达芬奇手术系统和机器人 TRUS 探头操作器运动学框架的快速和自动配准是仪器跟踪系统的关键组成部分。在本文中，我们提出了一种完全自动的注册技术，该技术基于机器人器械尖端在前列腺前的空气 - 组织边界上按压时的自动 3-D TRUS 定位。该检测方法使用多尺度滤波技术来识别和定位 TRUS 体积中的手术器械尖端，并且也可以用于检测 3-D 超声中的其他表面基准点。实验已经使用组织体模和两个离体组织样本进行，以证明所提出方法的可行性。此外，还对一只使用达芬奇 Si 手术系统的麻醉狗进行了系统的初步体内评估，目标注册误差（定义为注册后对应点的均方根距离）为 2.68 毫米。结果表明，该方法在自动注册 TRUS 图像到达芬奇手术系统方面具有准确性和一致性。
