NIH says use of AI in retinal imaging will help evaluate AMD, other retinal conditions

The National Institutes of Health (NIH) used artificial intelligence (AI) in the retinal imaging process and found that it was 100 times faster and improved image contrast 3.5-fold compared with manual assessments. The researchers said this approach will help to evaluate age-related macular degeneration (AMD) and other retinal conditions.

Johnny Tam, PhD, of the Clinical and Translational Imaging Section at NIH’s National Eye Institute, and his team developed a novel AI method called parallel discriminator generative adversarial network (P-GAN). They entered 6,000 manually analyzed adaptive optics (AO)-optical coherence tomography-acquired images of human retinal pigment epithelium into the P-GAN network and trained it to identify and recover speckle-obscured cellular features.

P-GAN reported outperformed other AI techniques. It was estimated that P-GAN reduced imaging acquisition and processing time by 100-fold and yielded greater contrast at about 3.5-fold.

“Our P-GAN artificial intelligence will make AO imaging more accessible for routine clinical applications and for studies aimed at understanding the structure, function, and pathophysiology of blinding retinal diseases,” said Dr. Tam. “Thinking about AI as a part of the overall imaging system, as opposed to a tool that is only applied after images have been captured, is a paradigm shift for the field of AI.”

Reference

AI makes retinal imaging 100 times faster, compared to manual method. National Institutes of Health. April 10, 2024. Accessed April 29, 2024. https://www.nih.gov/news-events/news-releases/ai-makes-retinal-imaging-100-times-faster-compared-manual-method