NIH AI Sharpens Eye Scans For Better Disease Detection

A groundbreaking development from National Institutes of Health (NIH) researchers is set to transform how we see the back of the eye. They've engineered a custom artificial intelligence (AI) system capable of digitally enhancing images to reveal individual retinal pigment epithelium (RPE) cells. This breakthrough has significant implications for detecting eye diseases earlier and more effectively monitoring how patients respond to treatment.

Image Credit: AdobeStock/rdkcho

AI Revolutionizes Retinal Imaging

Details of this innovative work, titled “Artificial intelligence-assisted clinical fluorescence imaging achieves in vivo cellular resolution comparable to adaptive optics ophthalmoscopy,” were recently published in the journal Communications Medicine.

Dr. Johnny Tam, PhD, an investigator at NIH’s National Eye Institute and senior author of the study report, highlighted the transformative potential of this AI. “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,” he stated in the release.

How the AI Achieves Unprecedented Clarity

The study involved gathering data from 26 healthy eyes across patients aged 22 to 63 years. The AI system was meticulously trained by being fed over 1400 images from different retinal areas taken by adaptive-optics ophthalmoscopy, learning to distinguish image quality as poor, moderate, or good. Researchers then provided the AI with corresponding images from the same retinal areas taken by standard ophthalmoscopy.

After applying an image sharpness test, the results were striking: AI-improved images were found to be eight times clearer than the original photos taken by standard ophthalmoscopy.

“Our system used what it learned from rating the images obtained from adaptive optics to digitally enhance images obtained with standard ophthalmoscopy,” Dr. Tam explained. “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.”

To further aid in sharpening the images, researchers utilized the injection of indocyanine green (ICG), a dye that increases the contrast of anatomical features.

“Our ICG imaging strategy allows RPE cells to be quickly and routinely assessed in the clinic,” said Joanne Li, PhD, 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.”

Significant Findings and Future Implications

Beyond image enhancement, the study yielded other important findings. While no differences in RPE cell parameters due to age were found, and no RPE differences were noted across ethnicities, a notable distinction emerged based on sex. RPE spacing was found to be larger, with a smaller density, in all eccentricities in female eyes when compared to male eyes. Specifically, spacing values in female eyes were on average 8% larger, and RPE density was on average 21% lower than in male eyes.

The study authors stated, “Results show that both the RPE cell-to-cell spacing and density measurements based on [adaptive optics-indocyanine green, or] AO-ICG images are consistent with the values previously published by other imaging studies and histology, and the highest RPE density is observed in the fovea as previously described.”

They further added, “In addition, our results demonstrate an in vivo relationship between RPE packing and age that expands upon published literature. To our knowledge, we present the largest in vivo normative dataset for RPE cell structure in living human eyes, which will be particularly important for future studies using both AO and non-AO imaging to assess the RPE.”

References

1. NIH researchers supercharge ordinary clinical device to get a better look at the back of the eye. News release. National Institutes of Health. April 23, 2025. Accessed May 7, 2025. Read more
2. Li J, Liu J, Das V, et al. Artificial intelligence assisted clinical fluorescence imaging achieves in vivo cellular resolution comparable to adaptive optics ophthalmoscopy. Comm Med. 2025;5:105. View study