Researchers on the National Institutes of Health applied artificial intelligence (AI) to a method that produces high-resolution images of cells in the attention. They report that with AI, imaging is 100 times faster and improves image contrast 3.5-fold. The advance, they are saying, will provide researchers with a greater tool to guage age-related macular degeneration (AMD) and other retinal diseases.
“Artificial intelligence helps overcome a key limitation of imaging cells within the retina, which is time,” said Johnny Tam, Ph.D., who leads the Clinical and Translational Imaging Section at NIH’s National Eye Institute.
Tam is developing a technology called adaptive optics (AO) to enhance imaging devices based on optical coherence tomography (OCT). Like ultrasound, OCT is noninvasive, quick, painless, and standard equipment in most eye clinics.
Imaging RPE cells with AO-OCT comes with recent challenges, including a phenomenon called speckle. Speckle interferes with AO-OCT the way in which clouds interfere with aerial photography. At any given moment, parts of the image could also be obscured. Managing speckle is somewhat just like managing cloud cover. Researchers repeatedly image cells over an extended time frame. As time passes, the speckle shifts, which allows different parts of the cells to turn out to be visible. The scientists then undertake the laborious and time-consuming task of piecing together many images to create a picture of the RPE cells that is speckle-free.
Tam and his team developed a novel AI-based method called parallel discriminator generative adverbial network (P-GAN) — a deep learning algorithm. By feeding the P-GAN network nearly 6,000 manually analyzed AO-OCT-acquired images of human RPE, each paired with its corresponding speckled original, the team trained the network to discover and get well speckle-obscured cellular features.
When tested on recent images, P-GAN successfully de-speckled the RPE images, recovering cellular details. With one image capture, it generated results comparable to the manual method, which required the acquisition and averaging of 120 images. With a wide range of objective performance metrics that assess things like cell shape and structure, P-GAN outperformed other AI techniques. Vineeta Das, Ph.D., a postdoctoral fellow within the Clinical and Translational Imaging Section at NEI, estimates that P-GAN reduced imaging acquisition and processing time by about 100-fold. P-GAN also yielded greater contrast, about 3.5 greater than before.
“Adaptive optics takes OCT-based imaging to the subsequent level,” said Tam. “It’s like moving from a balcony seat to a front row seat to image the retina. With AO, we will reveal 3D retinal structures at cellular-scale resolution, enabling us to zoom in on very early signs of disease.”
While adding AO to OCT provides a a lot better view of cells, processing AO-OCT images after they have been captured takes for much longer than OCT without AO.
Tam’s latest work targets the retinal pigment epithelium (RPE), a layer of tissue behind the light-sensing retina that supports the metabolically lively retinal neurons, including the photoreceptors. The retina lines the back of the attention and captures, processes, and converts the sunshine that enters the front of the attention into signals that it then transmits through the optic nerve to the brain. Scientists are all for the RPE because many diseases of the retina occur when the RPE breaks down.
By integrating AI with AO-OCT, Tam believes that a significant obstacle for routine clinical imaging using AO-OCT has been overcome, especially for diseases that affect the RPE, which has traditionally been difficult to image.
“Our results suggest that AI can fundamentally change how images are captured,” said Tam. “Our P-GAN artificial intelligence will make AO imaging more accessible for routine clinical applications and for studies geared toward understanding the structure, function, and pathophysiology of blinding retinal diseases. Excited about AI as an element of the general imaging system, versus a tool that is just applied after images have been captured, is a paradigm shift for the sphere of AI.”
