Summary: UCLA researchers have developed an intraocular OCT-based fiber that provides real-time tissue classification during eye surgery, improving safety and precision for both manual and robotic procedures. Background: Eye surgeries—particularly those involving the back of the eye—demand extraordinary precision. Surgeons operate within microscale, multilayered tissues where even the slightest error can result in severe complications or permanent vision loss. While intraoperative imaging tools such as optical coherence tomography (OCT) have advanced visualization, they remain inherently limited. During surgery, OCT provides only a narrow field of view, its images can be obscured by instruments, and it offers no immediate feedback about the specific tissue the surgeon is contacting. This absence of real-time, localized information makes it challenging to accurately distinguish between delicate ocular tissues during a live procedure. There is a critical unmet need for a technology that extends beyond conventional imaging—one that can instantly identify and differentiate eye tissues in real time. Such a capability would dramatically enhance surgical safety, precision, and efficiency, empowering both manual and robotic-assisted ophthalmic procedures and improving patient outcome. Innovation: UCLA researchers have developed a technology that transforms OCT from a simple imaging tool into a real-time tissue classification system. Instead of simply showing structural images, the system interprets signals from OCT scans to recognize tissues such as the sclera, iris, lens, posterior capsule, and retina. This classification occurs in milliseconds, providing surgeons immediate, actionable feedback during surgery. The technology has demonstrated 98% accuracy, with a distance prediction root mean square error of 5.06 µm. The technology can be integrated within robotic surgical platforms or handheld tools, and feedback can be delivered through visual overlays, sound cues, or gentle tactile alerts. By confirming in real time which tissue the surgeon is interacting with, the system reduces uncertainty and enables safer, more precise surgical maneuvers. Because the design is flexible and software-driven, it can be trained to recognize additional tissue types over time. Importantly, it builds on existing OCT hardware already used in ophthalmology, extending its value without requiring entirely new equipment. This is the first known approach to repurpose intraocular OCT signals for real-time tissue classification, offering a practical path toward intelligent, feedback-driven eye surgery with improved outcomes. Potential Applications: • Retinal surgery – safer subretinal injections and repairs • Cataract surgery – guidance for capsule polishing and lens work • Glaucoma/iris surgery – precise tissue manipulation • Robotic-assisted surgery – smarter, feedback-driven platforms • Handheld tools – real-time support in manual operations • Surgical training – instant feedback for education Advantages: • Real-time tissue recognition in milliseconds • Extends existing OCT systems • Improves safety of delicate eye surgery • Works with robotic or handheld tools • Easily retrained for new tissue types • Multiple feedback options (visual, audio, haptic) • First to use OCT for tissue classification State of Development: Initial description and first successful demonstration have been completed.
Publications:
Intraocular Tissue Detection Using Bio-Impedance Under Real-time Surgical Settings, nvestigative Ophthalmology & Visual Science June 2026, Vol.67, 4026 Reference: UCLA Case No. 2026-023 Lead Inventors: Tsu-Chin Tsao, Distinguished Professor, Mechanical Engineering; Aya Barzelay Wollman, Assistant Professor, Jules Stein Institute