A new AI-based Glaucoma Screening (AI-GS) network could screen for glaucoma in seconds and its large-scale implementation could revolutionise glaucoma care.
Glaucoma is the leading cause of irreversible blindness in Japan and worldwide. Early detection is critical, as the disease progresses silently, slowly constricting one’s peripheral field of vision.
Patients often don’t notice this loss of vision at first, which means that extensive and irreversible damage can occur before a patient even thinks about booking a doctor’s appointment.
As a result, many cases remain undiagnosed due to the limited availability of ophthalmologists and the challenges of conducting mass screenings, particularly in resource-limited regions.
“This is why we developed a new, quick, portable testing method. It analyses multiple key indicators of glaucoma, integrates the findings, and determines the presence of the disease with unprecedented precision,” said Professor Toru Nakazawa of Tohoku University.
The AI-GS network was tested on a dataset of 8,000 fundus images of the back of the eye (where glaucomatous damage occurs), achieving an impressive 93.52 per cent sensitivity at 95 per cent specificity – a level comparable to expert ophthalmologists.
Unlike traditional AI models, this system excels at detecting early-stage glaucoma, even in cases where fundus abnormalities are subtle and difficult to recognise.
A major challenge in AI-driven healthcare is its lack of interpretability – the so-called “black box” problem where it’s unclear what steps the AI made to come to a conclusion. AI-GS solves this by providing numerical values for each diagnostic feature, allowing ophthalmologists to understand and verify its decision-making process. This transparency enhances trust and facilitates seamless integration into clinical practice.
Another important aspect of making practical implementation as simple as possible was size. At just 110 MB, the AI-GS network is designed for portability and efficiency. It requires minimal computational power and delivers diagnostic results in under a second.
“AI-GS brings expert-level glaucoma screening to your pocket, complementing specialist evaluations,” says associate professor Parmanand Sharma of Tohoku University.
“It can be run on a mobile device and used in all sorts of public places because of its portability. You can run screenings at train stations or even remote regions that otherwise have limited access to ophthalmologists.”
“This AI technology bridges a critical gap in glaucoma detection by making specialist-level diagnostics accessible to underserved communities,” remarks Professor Nakazawa.
“By enabling early detection on a large scale, we have the potential to prevent blindness for millions worldwide.”
With its high accuracy, AI explainability, and lightweight design, the AI-GS network represents a major breakthrough in AI-driven ophthalmology, bringing glaucoma screening out of hospitals and into everyday life.