The Edward S. Harkness Eye Institute at Columbia University is testing “tele-ophthalmology” to screen for the four leading causes of blindness–cataracts, glaucoma, age-related macular degeneration (AMD), and diabetic retinopathy. Screening for visual acuity, non-contact tonometry, frequency doubling technology (FDT) perimetry, fundus imaging, and optical coherence tomography are loaded into a small truck that travels to screening sites.
“We don’t have an ophthalmologist on the truck,” Dr. Al-Aswad said. “We send results to a central reading center. If the doctor decides the patient has potential for eye disease, we teleconference on the spot. It is a highly efficient use of resources to screen for eye disease and raise awareness.”
In London, Visulytix is going one step farther. Instead of sending test data to an ophthalmologist for review, the company is developing an artificial intelligence (AI) system to screen test results.
“The net result is above-human levels of diagnostic accuracy in a fraction of the time and at a fraction of the cost of a typical provider,” explained Sameer Trikha, MBA, FRCOphth, founder and chief medical officer. “We are developing a single algorithm, a single solution to screen for multiple eye diseases, including glaucoma, AMD, and diabetic retinopathy. It is quite conceivable that you can reach accuracies of between 85% and 99%, vastly superior to primary eye providers and equivalent to human experts.”
AI won’t replace clinicians, Dr. Trikha said. AI will handle the initial screening and reduce the incidence of false positives, streamlining workflow, and enabling ophthalmologists to focus on patients who need treatment.