Simulation based surgical training for high risk low resourced procedures using a novel simulator with smart mentoring

摘要 腹腔镜胆囊切除术（LC）医源性并发症的一个重要来源是胆道损伤， 胆囊管和胆囊动脉的错误识别在每年进行的约75万例LC中，胆管损伤 发生在高达0.15%的病例中，伴随死亡率为20.8%。其中约30%发生在手术中 由经验丰富的外科医生进行，他们已经完成了200多例LC手术。的批判性观点 安全性（CVS），由美国胃肠和内窥镜外科医师协会（SAGES）任务确定 力，列出了具体的程序步骤，以避免胆损伤LC期间。虚拟现实（VR）的初步数据 在英国和瑞典的CVS培训表明，它导致更少的手术错误。在这 研究我们建议开发和验证VR CVS课程的原型便携式摄像机辅助 模拟器（PortCAS）使用基于Web的交互式虚拟智能指导培训协议。这项研究将 有两个具体目标。第一个目标是开发PortCAS的硬件、软件、虚拟仿真， 以及基于云的交互式智能指导培训环境。第二个目的是测试 PortCAS培训平台，由经验丰富的外科医生虚拟和真实的获取CVS。 PortCAS通过使用计算机视觉来跟踪腹腔镜的运动，将功能和低成本结合起来 工具，弥合低技术模拟器（低保真度，无反馈，低成本）和高技术模拟器之间的差距。 模拟器（高保真度，一些反馈，最少的分析，高成本）。PortCAS的愿景是提供 最便宜的方式来自动评估模拟（反馈给学员）手动或运动- 基础任务。除了腹腔镜手术外，PortCAS还可以扩展到许多其他培训学科。 PortCAS的小尺寸和便携性使其可用于指导式智能培训 随时随地为忙碌的外科医生和处于学习曲线任何阶段的受训者提供帮助。摄像头跟踪是 在客户端进行，以最大限度地减少延迟，而智能训练算法将位于云中。 这允许远程访问培训方案，并降低了对培训方案复杂性的要求。 客户级别。这种策略是创新的，因为它降低了成本，使重量轻的便携式训练设备， 并将学习模式和个人表现以及量身定制的智能指导集中在云中 server.培训将评估技术掌握和实践中专家判断的进展 外科医生使用CVS的术中图像。与培训工具箱或视频培训师不同，基于VR的 模拟器使受训者接触罕见的不良事件，允许根据个人需求进行定制， 微妙的测量受训者的表现，并提供非常详细和客观的反馈，在真实的 时间，独立于普罗克特。我们希望有效地培训外科医生在术中达到 CVS，使用正常和不寻常的血管纤毛解剖模拟，将减少发病率， 医源性胆道损伤导致的死亡率。

ABSTRACT A significant source of iatrogenic complication in laparoscopic cholecystectomy (LC) is biliary injury due to misidentification of the cystic duct and artery. Of the ~750,000 LCs performed annually, bile duct injury occurs in up to 0.15% of cases, with a concomitant mortality rate of 20.8%. Of these ~30% occur in procedures performed by experienced surgeons, who have completed more than 200 LC surgeries. The Critical View of Safety (CVS), established by a Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) task force, lays out specific procedural steps to avoid biliary injury during LC. Preliminary data on virtual reality (VR) training of CVS in the United Kingdom and Sweden have indicated that it results in fewer surgical errors. In this study we propose to develop and validate a VR CVS curriculum on a prototype portable camera aided simulator (PortCAS) using a web-based interactive, virtual, smart mentored training protocol. The study will consist of two specific aims. The first aim is to develop the PortCAS hardware, software, the virtual simulation, and an interactive cloud-based smart mentoring training environment. The second aim is to test the efficacy of the PortCAS training platform on virtual and real obtainment of the CVS by experienced surgeons. PortCAS combines function and low cost by using computer vision to track the motion of laparoscopic tools, bridging the gap between low-tech simulators (low fidelity, no feedback, low cost) and high-tech simulators (high fidelity, some feedback with minimal analytics, high cost). The vision for PortCAS is to provide the least expensive way to automate assessment of simulation (feedback to the trainee) for manual or motion- based tasks. PortCAS can be extended across many other training disciplines besides laparoscopic surgery. The small size and portability of PortCAS makes it available for mentored smart training anywhere/anytime for busy surgeons, and for trainees at any step of the learning curve. Camera tracking is conducted at the client side to minimize latency while the smart training algorithms will be located in the cloud. This allows training regimens to be accessed remotely and reduces the requirements for complexity at the client level. This strategy is innovative in that it reduces the cost, enables a lightweight portable training device and centralizes the learning paradigms and individual performance and tailored smart mentoring in the cloud server. The training will evaluate progress in both technical mastery and expert judgement in practicing surgeons using intra-surgical images of the CVS. Unlike training toolboxes or video trainers, VR-based simulators expose trainees to rare adverse events, allow customization based on individual needs, allow for subtle measurement of the trainee's performance, and provide very detailed and objective feedback in real time, independent of a proctor. We expect that effectively training surgeons in intra-operative attainment of CVS, using both normal and unusual vasculobiliary anatomy simulations, will reduce the incidence and mortality caused by iatrogenic biliary injury.