Early detection of eye diseases remains a global challenge, particularly in regions with limited access to specialist care. A research team led by Toru Nakazawa at Tohoku University has developed a portable, AI-powered eye scanner designed to make anterior eye screening more accessible and affordable.
Conditions affecting the anterior segment of the eye, such as cataracts, remain among the leading causes of visual impairment worldwide. Despite the importance of early diagnosis, many patients are only screened once symptoms become severe, sometimes too late to prevent lasting damage.
Portable, low-cost alternative
Traditional diagnostic tools, including anterior-segment optical coherence tomography (AS-OCT), are expensive and largely confined to hospitals and specialized clinics. This creates barriers for people in rural areas, elderly populations, and individuals with limited mobility.
The newly developed device offers a compact and cost-effective alternative. Built around a scanning slit-light system, it enables clinicians to assess key eye structures, including the cornea, iris, lens, and ocular surface, using true-color imaging. The results show strong agreement with conventional AS-OCT scans, making it suitable for screening purposes.
Importantly, the system can also help identify the risk of angle-closure glaucoma, a condition that can lead to sudden and severe vision loss if left undetected.
AI at the edge
A distinguishing feature of the platform is its embedded lightweight AI model, which can segment anatomical structures and support disease classification directly on the device. This on-device processing eliminates the need for cloud connectivity, improving data privacy, reducing latency, and enabling use in low-resource settings.
By analyzing a single scanning video, the system provides both quantitative measurements and qualitative assessments, reducing operator dependency and simplifying workflows.
More accessible eyecare
With its combination of low cost, portability, and AI-driven analysis, the device opens the door to new screening environments, ranging from community health centers and pharmacies to elderly care facilities and mobile screening units.
The researchers believe that bringing diagnostic capabilities closer to patients could significantly improve early detection rates and help prevent avoidable vision loss, while easing pressure on specialized healthcare services. The study was recently published in Scientific Reports.
Diagnosing demetia from retinal scan
Last year researchers at the National University of Singapore developed an AI-driven biomarker that can predict the risk of cognitive decline and dementia using a simple retinal scan. The technology, RetiPhenoAge, determines the “biological age” of the retina, which is linked to aging processes in the brain.
In a study involving over 500 patients, a higher retinal age was found to be associated with a 40% higher risk of cognitive decline within five years; these results were confirmed in a large-scale British dataset. The system is applicable in primary care because it utilizes existing equipment. As such, it offers a scalable, non-invasive, and cost-effective method for the early detection and prevention of dementia.
