Researchers at the U.S. National Institutes of Health (NIH) have achieved a breakthrough in retinal imaging by applying artificial intelligence (AI) to accelerate the process by a factor of 100. This advancement not only speeds up imaging but also enhances image contrast by 3.5 times, offering a significant tool for evaluating retinal diseases such as age-related macular degeneration.
“Artificial intelligence helps overcome a key limitation of imaging cells in the retina, which is time,” said Johnny Tam, leader of the Clinical and Translational Imaging Section at NIH’s National Eye Institute. Tam is developing adaptive optics (AO) technology to enhance imaging devices based on optical coherence tomography (OCT), a noninvasive and standard equipment in most eye clinics.
Tam and his team introduced a novel AI-based method called parallel discriminator generative adversarial network (P-GAN), a deep learning algorithm. By training the P-GAN with nearly 6,000 manually analyzed AO-OCT images of the human retinal pigment epithelium—paired with their speckled originals—the network learned to identify and recover cellular features obscured by speckle noise.
The improved imaging technique promises to advance the understanding and diagnosis of retinal diseases by providing clearer and faster images of retinal cells. This development holds potential benefits for patients worldwide, including in Asia, where retinal conditions are a growing concern due to aging populations.
The application of AI in medical imaging exemplifies the transformative impact of technology on healthcare. As researchers continue to refine these methods, such innovations may lead to earlier detection and better management of visual impairments, contributing to improved quality of life for millions.
Reference(s):
cgtn.com
