How VisionQuest is Using AI to Screen for Diabetic Retinopathy

Even though my boyfriend’s mom works at an eye doctor, I’m still kind of afraid to get eye exams – and I’m not alone. Unfortunately, most people do not receive frequent enough eye exams, even though these are especially important for those with other conditions. For example, eye screenings and exams can be helpful for patients with diabetes, who can later develop various ocular problems or diseases. As described by a news release from VisionQuest Biomedical Inc. (“VisionQuest”), only 20% of patients with diabetes in Mexico receive annual diabetic eye examinations. However, VisionQuest is using EyeStar artificial intelligence (AI) software to improve care for these patients. Specifically, the EyeStar AI technology has screened over 40,000 patients at the Clínicas del Azúcar for diabetic retinopathy.

EyeStar Software

For nearly 14 years, VisionQuest has been developing unique AI solutions for the medical realm. One of these solutions was EyeStar, which VisionQuest describes as:

a patented software application that uses Artificial Intelligence to provide point-of-care screening for diabetic eye disease in real time. EyeStar uses a neural network that has been trained on hundreds of thousands of retinal images from people of multiple ethnicities.

Outside of diabetic retinopathy, VisionQuest also uses EyeStar to screen patients for macular edema. However, for the purpose of this article, we will focus on diabetic retinopathy. So far, VisionQuest has been using EyeStar in Mexico for over 5 years. During that time, over 40,000 patients have been screened. Ultimately, more than 4,000 of these patients were diagnosed with eye conditions which could cause vision loss without early treatment and detection.

At the Clínicas del Azúcar, which is located in Monterrey, Mexico, patients arrive at the clinic to record their vitals. Next, nurses or technicians take photographs of a patient’s retina. During this time, dilation is not required. Next, these photographs are uploaded to the Cloud. At this point, EyeStar uses deep learning to:

  • Identify any potential issues
  • Tell doctors whether a patient should be referred to an ophthalmologist for further assistance

Because EyeStar determines these results in under one minute, it offers fast and efficient help. Additionally, the program provides results in around 98% of cases, meaning it provides an effective diagnostic solution.

Diabetic Retinopathy

Patients with diabetes may develop diabetic retinopathy, a complication which occurs when high blood sugar levels damage retinal blood vessels. When the blood vessels become blocked, leaks or bleeds occur. Next, the body attempts to produce new blood vessels, but these are often abnormal. Currently, diabetic retinopathy is the most common type of vision loss associated with diabetes. In many cases, diabetic retinopathy is asymptomatic in early stages. However, as the condition progresses, symptoms include:

  • Vision loss or blindness
  • Seeing “washed out” or faded colors
  • Increasing number of “floaters” in field of vision
  • Difficulty reading or seeing far away objects
  • Poor night vision
  • Dark spots/streaks in vision
  • Neovascular glaucoma (complication)
  • Retinal detachment (complication)
  • Diabetic macular edema (complication)

Learn more about diabetic retinopathy.

Jessica Lynn

Jessica Lynn

Jessica Lynn has an educational background in writing and marketing. She firmly believes in the power of writing in amplifying voices, and looks forward to doing so for the rare disease community.

Share this post

Follow us