Thanks to the generous support of NIDEK, six hospitals in Nam Dinh, Nghe An and Can Tho were equipped with modern fundus cameras and connected to Cybersight AI, enabling faster and more accurate screening for diabetic retinopathy. More than 14,000 people have been screened across all sites so far, helping thousands get the timely referrals and follow up care they need to protect their sight.
14,000 screened for diabetic retinopathy using AI in Vietnam
Over the past two years, Orbis has been working with partners across Vietnam to bring sight-saving technology and training to hospitals treating people living with diabetes.
A Focus on Nghe An
Every year, thousands of people living with diabetes in Vietnam are at risk of losing their sight to diabetic retinopathy—an eye condition that can lead to blindness if left untreated. For patients in Nghe An, access to early screening and treatment has not always been easy.
Dr. Huong, an ophthalmologist at the Ophthalmology Department of Nghe An Hospital of Endocrinology (NAHE), knows these challenges all too well.
“Until recently, we only had an old fundus camera that required eye drops to dilate the pupils,” she explains. “The images then had to be reviewed manually by an eye doctor. This meant long waiting times and temporary blurry vision for patients, many of whom were older people who needed help from family members just to make the trip.”
For many patients, this extra burden led them to delay or even skip eye exams altogether, putting their sight at risk.
Patients are rapidly screened for retinal diseases using AI-powered cameras.
New Solution with NIDEK and Cybersight AI
In mid-2024, with support from NIDEK through Orbis Vietnam, NAHE and five other health facilities in Nam Dinh, Nghe An, and Can Tho were equipped with modern non-mydriatic NIDEK fundus cameras and connected to Cybersight AI, Orbis’s artificial intelligence platform for retinal disease detection.
The impact has been immediate.
“The NIDEK fundus camera helps us keep records of patients’ retinal images and track their progress over time,” says Dr. Huong. “With Cybersight AI, results are generated within minutes. Patients spend less time waiting, they are happier with the service, and our team can focus on providing care rather than being overwhelmed by workload.”
Between July 2024 and June 2025, nearly 4,000 patients were screened for diabetic retinopathy in NAHE using the new system. Almost 30% of cases were referred to ophthalmologists for further evaluation and treatment, ensuring patients at risk got the care they urgently needed.
Building Local Expertise
Technology alone is not enough—it must go hand in hand with training. In October 2024, with support from Orbis, Dr. Huong joined a one-week training session in the treatment of diabetic retinopathy at NAHE. Under the guidance of Dr. Vu Tuan Anh, a retina specialist from the Vietnam National Eye Hospital, she and her colleagues strengthened their skills in anti-VEGF injection and laser treatment.
“The training helped me gain the skills needed to treat patients here at our hospital, rather than referring them elsewhere,” says Dr. Huong. “We are now in the process of obtaining approval from the Ministry of Health to provide diabetic retinopathy treatment directly at NAHE.”
Local doctors train together using AI-powered cameras.
A Model for the Future
The success of NAHE’s diabetic retinopathy program—combining modern technology, AI-powered diagnosis, and hands-on training—is now serving as a model for other hospitals across Vietnam. Plans are already underway to scale up the approach and harness the power of AI so more communities can access better diagnostics and treatment.
Thanks to the generous support of NIDEK, more patients are being screened, more cases are being caught before vision loss occurs, and more doctors like Dr. Huong are empowered to provide treatment.
We’d also like to say a big well done to our team in Vietnam and our local partners for their hard work and dedication to improving eye care—and of course, our supporters, without whom none of this would be possible.
Donate today
Help fund more AI initiatives.