Safe Robotic Capsule Cleaning with Integrated Transpupillary and Intraocular Optical Coherence Tomography
Lai, Yu-Ting, Foroutani, Yasamin, Barzelay, Aya, Tsao, Tsu-Chin
–arXiv.org Artificial Intelligence
--Secondary cataract is one of the most common complications of vision loss due to the proliferation of residual lens materials that naturally grow on the lens capsule after cataract surgery. A potential treatment is capsule cleaning, a surgical procedure that requires enhanced visualization of the entire capsule and tool manipulation on the thin membrane. This article presents a robotic system capable of performing the capsule cleaning procedure by integrating a standard transpupillary and an intraocular optical coherence tomography probe on a surgical instrument for equatorial capsule visualization and real-time tool-to-tissue distance feedback. Using robot precision, the developed system enables complete capsule mapping in the pupillary and equatorial regions with in-situ calibration of refractive index and fiber offset, which are still current challenges in obtaining an accurate capsule model. T o demonstrate effectiveness, the capsule mapping strategy was validated through five experimental trials on an eye phantom that showed reduced root-mean-square errors in the constructed capsule model, while the cleaning strategy was performed in three ex-vivo pig eyes without tissue damage. Capsule cleaning is a potential treatment for eliminating blindness due to residual lens materials that develop around the capsular bag after cataract surgery [1]. The procedure requires precise instrument maneuvers and timely sensing of the environment to obtain successful surgical outcomes. Although transpupillary optical coherence tomography (OCT) and the digital microscope exhibit sufficient resolution to visualize the posterior capsule (PC) and other tissues, the shadowing effect created by the iris limits the visibility of the equatorial region and the amount of residual lens or tissue location remain unknown (Figure 1) [2]. Although polishing is theoretically feasible, many surgeons choose to skip it to avoid increased risks of capsule rupture [3], possibly due to uncharacterized equatorial regions and inaccurate manual manipulation on the thin capsule membrane (error approximately 200-350 µm) [4], [5]. Unlike human intervention, accurate tooltip positioning and enhanced sensing can be achieved with a robotic system that has the potential to assist and enable the polishing procedure. This work was supported by U.S. NIH/R01EY029689 and NIH/R01EY030595. Y u-Ting Lai, Y asamin Fouroutani, and Tsu-Chin Tsao are with the Department of Mechanical and Aerospace Engineering, University of California, Los Angeles, CA, USA.
arXiv.org Artificial Intelligence
Jul-21-2025
- Country:
- North America > United States > California > Los Angeles County > Los Angeles (0.88)
- Genre:
- Research Report > New Finding (0.48)
- Industry:
- Health & Medicine
- Surgery (1.00)
- Therapeutic Area > Ophthalmology/Optometry (1.00)
- Health & Medicine
- Technology: