Bio-impedance Sensor Technology for Robot-Assisted Vitreo-retinal Surgery

At present, vitreoretinal surgery is performed purely manually by highly specialized, welltrained surgeons. The surgeons need superior handeye coordination and precise positioning. When applying even small forces tangential to the surface of the eye, e.g. by means of the inserted instruments, the eye rotates and a different part of the eye becomes visible under the microscope. Microsurgeons are trained to pivot the instruments around the incision point so as to keep the eye from rotating. However, there is a limit to the surgeon's ability to perform pure pivotal motion. Furthermore small vibrations by the surgeon's hand (physiological tremor) are amplified by the leverage effect, resulting in larger vibrations of the instrument tip. It is thus extremely difficult to maintain a stable position down at the retinal surface. All targets are extremely small - ranging from 5 to 500 microns - and fragile. Tremor not only complicates reaching the required precision, it also causes the eye to rotate undeliberately, which results in surgeons aiming on a moving target. Furthermore, the surgeon is forced to work under very tight spatial constraints. Overall the surgeons are under very stringent requirements on safety and reliability as damage to the delicate retina leads to blind spots in the patient's vision and causes partial or full blindness. Within the framework of this PhD robotassisted technology will be developed to improve vitreoretinal surgery in terms of reliability and performance. More in particular the focus is on the development of sensorbased control algorithms that making use of sensory inputs such as e.g. force, proximity, oct, stereocamera, provide various operating modes that allow microsurgeons to operate in a comfortable and reliable manner. Shared control algorithms will be developed specifically for this pplications. Through these methods guidance and motion compensation is offered to the surgeon. Maximal use of surgical skill is possible as the envisioned shared control methods are fully compatible with surgeon input commands. Apart from sensordevelopment and sensorintegration, this work will progress humanrobot interaction and surgical skill in a realistic and demanding scenario. The work foresees in ample clinical validation on exvivo setups.

Jaar van publicatie:2019

Toegankelijkheid:Open