AI helps clarify facial expressions for LASIK patients

Millions of people have undergone LASIK eye surgery since LASIK became commercially available in 1989, but patients may develop cataracts later in life and need new corrective lenses implanted in their eyes. . As more intraocular lens options become available, scientists have developed computational simulations to help patients and surgeons find the best option.

in study at Journal of Cataract and Refractive Surgeryresearchers at University of Rochester Create a computational eye model that includes a patient’s cornea after LASIK surgery to study how standard intraocular lenses and lenses designed to increase depth of focus perform in the operated eye. did. Susana Marcosdirected by David R. Williams Visual Science Center and professor nicholas george, optics and Rochester Eye researchers say a computational model that uses anatomical information about a patient’s eye provides important guidance to surgeons about the optical quality to expect postoperatively.

“Currently, the only preoperative data used to select lenses is essentially corneal length and curvature,” says Marcos, a co-author of the study. “This new technology allows us to reconstruct the eye in three dimensions, giving us the entire shape of the cornea and lens into which the intraocular lens will be implanted. You will be much better off choosing the lens that produces the best images.”

The future of optical coherence tomography

Marcos and collaborators at the Visual Science Center and the University of Rochester Fraum Eye Research Institute and Gergen Data Science Instituteconducted a large-scale study to quantify images of the eye in three dimensions using an optical coherence tomography quantification tool they developed to find broader trends. They use machine learning algorithms to find relationships between pre- and post-surgery data, providing parameters that can inform the best outcome.

We’ve also developed technology that allows patients to see for themselves what different lens options look like.

“What we see is not strictly an image projected onto the retina,” says Marcos. “It requires all kinds of visual processing and perception. When a surgeon plans a surgery, it’s very difficult to tell the patient what they’re going to look like. A calculated, personalized eye model will tell you which lenses We’ll show you what’s best for the patient’s eye anatomy, but patients want to see for themselves.”

The researchers used an optical bench to manipulate the eye’s optical system in the same way as an intraocular lens, using techniques originally developed for astronomy, such as adaptive optics mirrors and spatial light modulators. Masu. This approach allows Marcos and his collaborators to conduct basic experiments and collaborate with industry partners to test new products. Marcos also helped develop a commercial headset version of his device called SimVis Gekko, which allows patients to see the world around them as if they were undergoing surgery.

In addition to studying techniques to help treat cataracts, researchers are also applying their methods to study other major eye diseases, such as presbyopia and myopia.