[ad_1]
Identifying eye diseases is a tedious task. Trained ophthalmologists sit in dark rooms examining retina images on screen. They are looking for unusual things that could indicate diabetic retinopathy, one of the leading causes of blindness in Singapore. The work is very tedious, shifts usually last only 4 hours.
A beginner wants to change this. Singapore-based health technology company Iris is using artificial intelligence (AI) to diagnose 3 types of eye diseases: Disease Retinopathy, Glaucoma Suspicion and Age-related Macular Dysfunction.
This is done by Selena + in-depth learning software that can automatically detect these diseases from retinal images. When medical professionals upload an image of the retina, the software compares it to a set of data that contains about 500,000 images of the retina with about 3,000 diseases. It will produce a test result in a few minutes.
Selena + was developed by the Institute of Ophthalmology of Singapore and the School of Computer Science at Singapore National University. “It can reduce the workload of students in the professional class by 75%, allowing them to focus on more complex and complex tasks,” said Dr. Daniel Ting, an assistant consultant at Singapore National Eye. CASIA.
Some clinics and hospitals in Malaysia, Vietnam, and the Philippines are using the software, and footprints are on the rise. Under an agreement with Japan’s Topcon Healthcare Solutions, it will soon be available in 18 Southeast and South Asian countries.
Other organizations are similarly working with advanced learning tools to identify eye diseases. American researchers have developed a system that predicts the risk of retinopathy in diabetics over a two-year period, according to a study on The Lancet.
At Mount Sinai, New York’s ophthalmologists have developed an AI algorithm to detect age-related macular degeneration. The system evaluates the patient if he has the disease and his weight.
The technology
Selena + trained some of the participants in the Singapore Integrated Sugar Retinopathy Program (SiDRP), a local polyclinic screening project, Dr. Ting said. Therefore, the software produces images of Singapore, Chinese, Malay and Indian retina. This does not mean that it is impossible for non-Asians to know.
Based on data from the Latino, African-Caribbean and Caucasian communities, it is still possible to distinguish between moderate to severe diabetic retinography. The color of the fund — the inner eye — is important because it varies by race. “The white Caucasian population has a lighter color compared to Asians and Africans,” said Dr. Ting.
The most important thing is that the AI system feeds on a large amount of information. “It is important to ensure that the training data sets include representative information and strong facts,” said Dr. Ting. “Testing the AI system on multiple independent data sets is equally important if not essential.”
Looking ahead
Dr. Ting Selena + suggests that mobile retina cameras have not yet been evaluated. The US Food and Drug Administration is still awaiting confirmation of the safety and effectiveness of medical equipment.
EyRis is currently developing a classifier for various eye conditions, Dr. Ting added. And in addition, AI can also be used to diagnose eye conditions other than diabetic retinopathy, suspected glaucoma, and age-related macular degeneration.
“That being said, we can’t call those new algorithms Selena +,” said Dr. Ting. The new version of the product may be in use.
[ad_2]
Source link
