Ebrahimi Ali, Alambeigi Farshid, Zimmer-Galler Ingrid E, Gehlbach Peter, Taylor Russell H, Iordachita Iulian
Laboratory for Computational Sensing and Robotics at the Johns Hopkins University, Baltimore, MD, 21218, USA.
Department of Mechanical Engineering at the University of Texas at Austin, Austin, TX, 78712, USA.
Rep U S. 2019 Nov;2019:7075-7082. doi: 10.1109/IROS40897.2019.8967806. Epub 2020 Jan 27.
When robotic assistance is present into vitreoretinal surgery, the surgeon will experience reduced sensory input that is otherwise derived from the tool's interaction with the eye wall (sclera). We speculate that disconnecting the surgeon from this sensory input may increase the risk of injury to the eye and affect the surgeon's usual technique. On the other hand, robot autonomous motion to enhance patient safety might inhibit the surgeons tool manipulation and diminish surgeon comfort with the procedure. In this study, to investigate the parameters of patient safety and surgeon comfort in a robot-assisted eye surgery, we implemented three different approaches designed to keep the scleral force in a safe range during a synergic eye manipulation task. To assess the surgeon comfort during these procedures, the amount of interference with the surgeons usual maneuvers has been analyzed by defining quantitative comfort metrics. The first two utilized scleral force control approaches are based on an adaptive force control method in which the robot actively counteracts any excessive force on the sclera. The third control method is based on a virtual fixture approach in which a virtual wall is created for the surgeon in the unsafe directions of manipulation. The performance of the utilized approaches was evaluated in user studies with two experienced retinal surgeons and the outcomes of the procedure were assessed using the defined safety and comfort metrics. Results of these analyses indicate the significance of the opted control paradigm on the outcome of a safe and comfortable robot-assisted eye surgery.
当机器人辅助应用于玻璃体视网膜手术时,外科医生将体验到感觉输入的减少,而这种感觉输入原本来自手术工具与眼壁(巩膜)的相互作用。我们推测,切断外科医生的这种感觉输入可能会增加眼睛受伤的风险,并影响外科医生的常规技术。另一方面,机器人的自主运动以提高患者安全性,可能会抑制外科医生对工具的操作,并降低外科医生对手术过程的舒适度。在本研究中,为了探究机器人辅助眼科手术中患者安全性和外科医生舒适度的参数,我们实施了三种不同的方法,旨在在协同眼部操作任务期间将巩膜力保持在安全范围内。为了评估这些手术过程中外科医生的舒适度,通过定义定量舒适度指标,分析了对外科医生常规操作的干扰程度。前两种利用巩膜力控制的方法基于自适应力控制方法,其中机器人主动抵消施加在巩膜上的任何过大的力。第三种控制方法基于虚拟夹具方法,其中在不安全的操作方向为外科医生创建一个虚拟壁。在有两位经验丰富的视网膜外科医生参与的用户研究中评估了所采用方法的性能,并使用定义的安全性和舒适度指标评估了手术结果。这些分析结果表明,所选控制范式对安全、舒适的机器人辅助眼科手术的结果具有重要意义。