Automated Realtime Imaging for Laparoscopic Interventions

腹腔内的微创（腹腔镜）介入越来越多地优于开放式介入。这些手术期间的术中成像主要通过专门的内窥镜（腹腔镜）进行。然而，腹腔镜仅为外科医生提供表面图像，即，在介入过程中，关于手术区域内部的信息仅是有限的。这种信息的缺乏可能会导致不必要的高失血，如果血管受伤或不完全的肿瘤切除。超声检查提供了手术野内部的术中实时成像，因此可以克服前述问题。然而，现有的超声方法是不适合这个应用程序。本项目的目标是提供外科医生从手术领域内的实时图像，而不增加人力需求。在项目的第一阶段，我们通过成功设计和评估机器人概念为项目奠定了基础。该机器人在两个平移自由度上移动经皮超声探头。在项目的下一阶段，我们的目标是在三个领域扩展现有的概念：机器人，机器人控制和可视化。为了移动机器人超声探头类似于手引导探头，我们必须通过两个额外的旋转自由度来扩展运动学。为了确保顺利集成到操作工作流程中，必须为机器人开发一个可行的集成概念。我们的目标是开发一种快速，准确的转向方法，以最小的额外复杂性。因此，需要评估不同的手动（遥控）和自动控制概念，以优化机器人的转向。通过位置或速度控制，手动转向应该是可能的。自动方法尝试将超声探头自动与手术区域对齐。为了简化超声图像的解释，需要开发新的可视化方法。超声图像总是在解剖学背景下进行解释。因此，我们希望将超声图像叠加在腹腔镜图像上，以便为外科医生可视化记录位置。为了帮助外科医生自动检测解剖危险区，它们将在超声图像中突出显示。为了改善外科医生的人体工程学，我们希望研究新获得的图像信息的最佳位置，以最大限度地减少外科医生在不同成像模式之间切换的努力。

Minimal invasive (laparoscopic) interventions within the abdominal cavity are more and more preferred to open interventions. Intraoperative imaging during these surgeries is primarily performed via specialized endoscopes (laparoscopes). However, laparoscopes provide the surgeon only with superficial images, i.e., there is only limited information available about the interior of the surgical field during the intervention. This lack of information may lead to un-necessary high blood loss if blood vessels are injured or to incomplete tumor resections. Ultra-sonography provides intraoperative real-time imaging of the interior of the surgical field and thus may overcome afore described problems. Nevertheless, existing sonographic approaches are not suitable for this application.The goal of this project is to provide the surgeon with real-time images from inside the surgical field without increased manpower requirements. In the first phase of the project we created the basis for the project by successfully designing and evaluating a robotic concept. The robot moves a transcutaneous ultrasound probe in two translational degrees of freedom. It can be integrated with the operating room table.In the next phase of the project our goal is to extend the existing concept in the three areas: robot, robot control, and visualization. To move the robot-s ultrasound probe similar to a hand guided probe, we have to extend the kinematics by two additional rotational degrees of free-dom. To assure a smooth integration into the operational workflow, a feasible integration concept for the robot must be developed. Our goal is to develop a fast and accurate steering method with minimal additional complexity. Hence, different manual (telemanipulation) and automatic control concepts need to be evaluated to optimize the steering of the robot. Manual steering should be possible by position or speed control. Automatic approaches try to align the ultrasound probe automatically with the surgical field. To ease the interpretation of the ultra-sound images, new visualization methods need to be developed. Ultrasound images are always interpreted within the anatomical context. Therefore, we want to overlay the ultrasound images on the laparoscopic images to visualize the recording position for the surgeon. To help the surgeon to automatically detect anatomic danger zone they will be highlighted within the ultra-sound images. To improve the ergonomics of the surgeon we want to investigate the optimal location for the newly gained image information to minimize the effort of the surgeon to switch between different imaging modalities.