Sensorised Micro-Surgical Robot for Safe Regenerative Retinal Therapy Delivery

Research question Gene and cellular therapies are emerging transformative treatments for blinding retinal diseases.

Treatment delivery into delicate retinal tissue layers, some as thin as 10-20um, requires precision and force sensing beyond human capabilities, and calls for robotic assistance.

Therefore, we must create systems that augment surgical delivery by offering dexterous assistance and information about tool/tissue interaction.

Background Research on robot-assisted vitreoretinal surgery originally focused on vessel cannulation and epiretinal membrane peeling.

Research has now pivoted to regenerative therapy delivery, where robotic assistance is critical due to the precision required.

However, retrofitting existing systems to adhere to the stricter requirements of therapy delivery cannot lead to optimal performance.

The majority of robotic systems under research and development are straight-tool holders that follow conventional surgical protocols.

Further, they do not offer simultaneous force and depth-of-implantation sensing due to the challenge of incorporating multi-modal sensors within a sub-millimetre tool. On the contrary, our research focuses from the onset on cellular and gene therapy delivery.

Our prototyped telemanipulated system can improve clinical performance near the retina due to its snake-like flexibility, offering both precision and dexterity. Aims and Objectives We will produce a sensorised flexible micro-surgical robot to deliver retinal therapies by: 1. Sensorising a flexible robot with fibre-based force sensing and high-resolution imaging at its tip. 2.

Developing algorithms for retina tracking and semi-automated therapy delivery. 3. Compiling documentation in support of first-in-human evaluation of a Minimum Viable Product (MVP). Methods and Timelines The project lasts 12 Quarters (Q).

The objectives will be met via 5 development work-packages: WP1 (Q1-Q4) will deliver a flexible submillimetre core that will house five sensing fibres, a therapy delivery channel and a forceps. The core will be embedded within the robot body to maintain the separation of the fibres. It will be created through fibre-pulling technology.

WP2 (Q2-Q10) will deliver two optical imagers based on single-fibre OCT. The “S” profiler provides an A-scan that will enable the robot to maintain a safe distance from the retina.

The “I” imager will produce a detailed B-scan (series of A-scans) to guide injection thanks to the innovative use of a rotating cleaved fibre. WP3 (Q1-Q10) will develop the first ever Fibre-Bragg grating based force/shape sensor for submillimetre robots. System calibration will be achieved through a bespoke cable-driven force applicator and tailored deep-learning models.

WP4 (Q5-Q12) builds on our software to develop “operating room ready” retina and tool tracking algorithms.

By detecting the tools and motion of the retina, our robot will be stabilised in lateral dimensions to perform minute-long therapy injection.

WP5 (Q1-Q12) will identify desired features of an MVP through interviews with surgeons, redevelop our robot prototype under a Quality Management System, and submit to regulators application for first-in-human consideration.

Impact and Dissemination The project de-risks our innovation while generating publications, new knowledge, and IP in support of a multi-stage commercialisation strategy.

Robust mechanisms to engage and be informed by public and patients will advise the presentation and orientation of the research, while also broadly disseminating our results.