CHS: Small: Collaborative Research: Teleoperation with Passive, Transparent Force Feedback for MR-Guided Interventions

磁共振成像(MRI)是一种广泛使用的诊断工具，它为医生提供了显著的自然视力扩展，提供了无与伦比的高清晰度图像，使软组织病理生理诊断、病变勾画和治疗监测成为可能，而不需要电离辐射。越来越多的医生希望使用MRI不仅用于诊断，而且还用于活检或肿瘤消融等指导性手术，以获得更高的准确性。然而，磁共振孔的几何形状迫使医生站在外面，将运动和力远程传输到内部对患者进行操作的工具上。因此，尽管MR提供了卓越的成像能力，但缺乏触觉。为了克服这一缺陷并实现远程临场感，医生需要一个高保真度的力反射遥操作系统。MR的强磁场增加了挑战；必须避免铁磁性材料和电流流动的电子产品，甚至有色金属也会产生成像伪影，这限制了执行器、变速器和传感器的选择。PI在这个项目中的目标是通过将实时3D组织成像与用于物理交互的动觉和力反馈相结合，使医生能够像与患者直接接触一样操作。项目成果有可能直接影响大量人口，因为这里开发的高保真力量传输将适用于其他图像引导干预措施，如药物输送和超声波。混合静液传动将适用于(实际上是最初设想用于)互动式人类安全机器人；在使其适应磁共振引导干预方面的进展将使技术成熟，尺寸和成本降低，这将有助于将其推向更多领域，如医疗机器人和双边遥操作。这项研究代表了机器人、触觉和介入放射学专家之间的合作。这项工作将建立和扩展一种新型的基于静液压和气动元件的双边遥控操作器，带有滚动的隔膜致动器，提供了低惯性、被动、高刚性和透明度以及可忽略的摩擦和间隙的独特组合，并且非常适合在MR孔外的医生和患者体内的手术工具之间提供运动和力反馈，使医生能够感觉到组织特性的变化。如果被动遥操作机器人足够坚硬和轻便，就可以实现敏感、灵活的任务。由于MR的特殊限制，MR引导的干预是提议的混合传输的一个引人注目的应用，如上所述，这排除了许多其他技术。这项研究解决的一个关键问题是如何将单轴原型的良好性能扩展到能够执行MR引导程序的复杂多轴系统。PI将运动学和动力学分析与用户的人体工程学测试相结合，以确保它支持直观的运动并提供透明的反馈，同时适合MR孔的受限空间。他们将通过兼容的传感器和成像基准将远程操作系统的运动与MR图像相结合，以提供视觉反馈，防止意外侵入不希望看到的区域，而医生则专注于工具提示的交互。新奇的力反射变速器、运动机构、传感器和软件共同构成了一个具有前所未有能力的赛伯人系统。这个远程呈现系统将是一个理想的平台，可以扩大对传输透明度为MR引导的干预提供的影响的科学理解。

Magnetic resonance imaging (MRI) is a widely used diagnostic tool that provides physicians with a remarkable extension to their natural vision, offering unparalleled high-definition visuals which enable soft tissue pathophysiology diagnosis, lesion delineation, and therapy monitoring without ionizing radiation. Increasingly, physicians would like to use MRI not only for diagnosis but also for guided procedures like biopsy or tumor ablation, for greater accuracy. However, the MR bore's geometry compels the physician to stand outside and transmit motions and forces remotely to tools operating on the patient within. So although MR provides superior imaging, the sense of touch is absent. To overcome this deficiency and achieve telepresence, physicians require a high-fidelity force-reflecting teleoperation system. Added challenges are imposed by MR's intense magnetic field; ferromagnetic materials and electronics with current flow must be avoided, and even non-ferrous metals can produce imaging artifacts, which constrains the choice of actuators, transmissions, and sensors. The PIs' goal in this project is to empower physicians to operate as if they were directly in contact with their patients, by integrating real-time 3D tissue imaging with kinesthetic and force feedback for physical interactions. Project outcomes have the potential to directly affect a large population, because the high-fidelity force transmission developed here will be applicable to other image-guided interventions such as drug delivery and ultrasound. The hybrid hydrostatic transmission will be applicable to (and indeed was initially conceived for) interactive human-safe robots; advances made in adapting it to MR-guided interventions will allow maturation of the technology and reductions in size and cost that will help push it into additional fields like medical robotics and bilateral teleoperation. This research represents a collaboration among experts in robotics, haptics and interventional radiology. The work will build upon and extend a novel bilateral teleoperator based on hydrostatic and pneumatic elements with rolling diaphragm actuators that provides a unique combination of low inertia, passivity, high stiffness and transparency, and negligible friction and backlash, and which is ideally suited to provide kinesthetic and force feedback between a physician outside the MR bore and tools operating on a patient within, allowing physicians to feel tissue property variations, for example. Sensitive, dexterous tasks will be realizable with a passive teleoperator if it is sufficiently stiff and light. MR-guided interventions are a compelling application for the proposed hybrid transmission because of MR's particular constraints, which as noted above rule out many other technologies. A key question this research addresses is how to scale the promising performance of single-axis prototypes to a complex multi-axis system able to perform MR-guided procedures. The PIs will combine kinematic and dynamic analyses with user tests for ergonomics to ensure that it supports intuitive motions and provides transparent feedback while fitting inside the MR bore's constrained space. They will integrate the teleoperated system's motions with MR images via compatible sensors and imaging fiducials to provide visual feedback and prevent accidental intrusion into undesirable regions while the physician focuses on tool tip interactions. Together, the novel force-reflecting transmission, kinematic mechanism, sensors, and software constitute a cyber-human system with unprecedented capabilities. This telepresence system will be an ideal platform to expand scientific understanding of the impact that transmission transparency provides for MR-guided interventions.