TUESDAY, April 7, 2026 (HealthDay News) -- An ultra-low-cost platform is capable of automated, quantitative, and anatomically calibrated anterior segment imaging for eye disease screening, according to a study published online March 17 in Scientific Reports.Neelam Kaushik, Ph.D., from the Tohoku University Graduate School of Medicine in Sendai, Japan, and colleagues introduced an artificial intelligence (AI)-integrated portable scanning slit-light device that delivers quantitative anterior-segment biometry at a cost of less than $500. A motorized slit-scanning mechanism was combined with synchronized imaging and an on-device deep-learning model to segment corneal and iris reflections, pupil boundaries, and corneal surfaces. A calibrated estimation of anterior chamber depth (ACD) was enabled as the primary quantitative output; exploratory estimates of central corneal thickness can be derived from the same scan.The researchers found that ACD showed excellent agreement with costly and clinic-bound anterior segment optical coherence tomography (OCT) in a clinical study of 170 participants (Pearson's r ≈ 0.92; concordance correlation coefficient ≈ 0.90), indicating near-clinical interchangeability for ACD in the context of screening. Clear visualization of anterior-segment features associated with narrow angles, cataract, corneal opacity, and keratoconic ectasia were illustrated from a single scanning-slit video in representative cases. A typical 51-frame video can be fully processed in 18.5 seconds on a Jetson Orin Nano, supporting compact, battery-powered deployment."It is in a patient's best interests to undergo regular check-ups, but this isn't always easy," lead author Toru Nakazawa, M.D., Ph.D., also from the Tohoku University Graduate School of Medicine, said in a statement. "The instruments needed to conduct these exams are expensive, bulky, and largely confined to clinical settings. Patients in rural areas or with low mobility may not be able to access these vital screening tools -- leaving them in the dark."Abstract/Full Text.Sign up for our weekly HealthDay newsletter