Tech Brings Big Changes to Vision Care

Technology is rapidly transforming the practice of medicine as we’ve known it. For one, we’re doing more and more of it from a distance. Even before the pandemic, virtual medical appointments had risen 33% over the previous year, and today, 83% of patients expect to use telemedicine after the pandemic goes away. Experts predict that the telehealth global market will reach nearly $186 billion by 2026, up from $34.28 billion in 2018.

Unexpected expansion of telehealth into new medical specialties is helping fuel that growth – in particular, glaucoma management, which is saving time and money, and reducing patient backlogs that have increased due to the pandemic.

The American Academy of Ophthalmology gives guidance on audio/video visits (telephone, Zoom, Skype, etc.), wherein a doctor can:

• Document patient consent;
• Complete a medical history and history of present illnesses to update medical records; and
• Actually perform visual acuity testing, using a chart or Web-based system; have the patient position his or her camera to view the external structures of the eyes; and attempt to visualize the red reflection from the back of the eye (red reflex test).

Patients can also do 5-10 minute virtual check-ins; text message updates; and so-called asynchronous visits, in which the doctor reviews photos and other images offline and then replies to the patient through a Web portal or email.

Of course, tech driven changes in medical practice go well beyond telehealth. For example, machine learning, aka artificial intelligence (AI), is rapidly assuming a substantial role in diagnostics, first in cardiology and oncology, and now, ophthalmology.

In August 2020, Eyenuk, Inc., a global artificial intelligence company, received a US Food & Drug Administration (FDA) clearance from the FDA to market EyeArt, an autonomous AI screening system for both mild and severe diabetic retinopathy. The company tested the system on more than a half-million patients around the world.

The system works in three steps:

• Capture color internal images of the backs of patient’s eyes;
• Submit images to the cloud for analysis; and
• Download diabetic retinopathy screening results that export to a PDF report.

According to the Centers for Disease Control & Prevention (CDC), an estimated 4.1 million and 899,000 Americans are affected by mild retinopathy or vision-threatening retinopathy, respectively. The number of vision-threatening diabetic retinopathy patients around the world will reach 56.3 million by 2030. The EyeArt system represents a major advance in getting these patients into effective treatment.

On the AI research front, Dr. Andres A. Bustamante, MD, and colleagues recently attempted to train various AI algorithms to discriminate between healthy and diseased corneas by evaluating corneal images taken with spectral-domain optical coherence tomography (SD-OCT), a non-invasive imaging test that uses light waves to take cross-section retinal pictures.

After processing the images, the investigators entered the resulting data into four different machine learning algorithms, and realized success with each.

Bustamante’s work adds another facet to a rapidly expanding field of AI applications that include age-related macular degeneration, glaucoma, diabetic retinopathy, and keratoconus.