Tel Aviv University and Sheba AI spots anemia from a 10-second video of the eye

Researchers at Tel Aviv University and Sheba Medical Center have developed an artificial intelligence system that can detect anemia and estimate key blood values from a short video of the blood vessels in a person's eye. The proof-of-concept study, published in npj Digital Medicine and announced on June 25, 2026, could eventually lead to quick screening tests that need no needle.

The work was led by Tamir Denis, a master's graduate at Tel Aviv University, with Prof. Haim Suchowski of the School of Physics and Astronomy and Prof. Lior Wolf of the Blavatnik School of Computer Science and AI, together with Prof. Ygal Rotenstreich and Dr. Ifat Sher-Rosenthal of Sheba Medical Center's retinal research laboratories.

The method, called Video-to-Vessels, uses a high-magnification camera to record ten seconds of video of the conjunctiva, the thin membrane covering the white of the eye. Software stabilizes the footage for eye movement, removes noise and turns it into a digital map of the vessels and of the blood moving through them. An AI model then relates those features to hemoglobin levels and red blood cell counts. In a study of 224 people who also had standard blood tests, the system identified anemia with 82.8 percent accuracy and its estimates tracked closely with laboratory results. The narrowest vessels gave the best predictions of hemoglobin.

Anemia affects roughly 30 percent of the world's population and is especially common in places with limited access to laboratories. The researchers envision a compact handheld device that could be used in clinics, community settings or even at home. They stress that larger and more diverse studies are needed before the technology can be used in patient care. The research was supported in part by Israel's Ministry of Innovation, Science and Technology.