A comparison of the same patch of retina labeled with indocyanine green and visualized three different ways [Image courtesy of NIH]
The team says it transformed a device designed to see tissues in the back of the eye into one sharp enough to make out individual cells. According to a post on the NIH website, this creates imaging resolution that compares to the most advanced devices available. However, this imaging method proves cheaper and faster while not requiring specialized equipment or expertise.
NIH says the innovation could enhance early disease detection and treatment response monitoring “by making what was once invisible now visible.”
“AI potentially puts next-generation imaging in the hands of standard eye clinics. It’s like adding a high-resolution lens to a basic camera,” said Johnny Tam, investigator at NIH’s National Eye Institute and senior author of the study report, which published in Communications Medicine.
A look at the NIH’s AI research
Tam and collaborators developed a custom AI system to digitally enhance images of a layer of tissue beneath light-sensing photoreceptors in the retina’s pigmented epithelium (RPE). First, they taught the system to recognize image quality as poor, moderate or good. To achieve this, they fed the system more than 1,400 images from different areas of the retina. They obtained these images through adaptive-optics ophthalmoscopy.
Next, the team used corresponding images from the same retinal locations but obtained through standard ophthalmoscopy. An image sharpness dest demonstrated a an eightfold improvement in clarity with AI.
“Our system used what it learned from rating the images obtained from adaptive optics to digitally enhance images obtained with standard ophthalmoscopy,” said Tam. “It’s important to point out that the system is not creating something from nothing. Features that we see in RPE cells with standard imaging are there, they’re just unclear.”
The technique employed by Tam and the team utilized an injection of indocyanine green (ICG) into the bloodstream. This increases the contrast of anatomical features. In the eye, ICG usually aids in the imaging of the blood vessels of the eye.
“Our ICG imaging strategy allows RPE cells to be quickly and routinely assessed in the clinic,” said Joanne Li, first author of the report and a biomedical engineer in Tam’s lab. “With AI, high quality images of the RPE cells can be obtained in a matter of seconds, using standard clinical imaging instruments.”
NIH says AI-enhanced ICG ophthalmoscopy puts RPE imaging “within reach of the typical eye clinic.”
