Ted Smith, MD, PhD: Learning More From Hyperspectral Autofluorescence Imaging

R. Theodore Smith MD, PhD, discusses the concept of hyperspectral autofluorescence imaging and how clinicians can learn more from this method than standard autofluorescence imaging.

At the American Academy of Ophthalmology Annual Meeting 2018, R. Theodore Smith, MD, PhD, sat down with MD Magazine® to discuss how hyperspectral autofluorescence imaging is building upon the standard technique that is utilized in diagnosing retinal issues and creating a more comprehensive view for clinicians.

Dr. Smith, a professor of ophthalmology at Icahn School of Medicine at Mount Sinai, also spoke with MD Magazine® about improving retina imaging to learn more about disease processes and collaboration between ophthalmologists and other physicians to identify patients at a higher risk for age-related macular degeneration.

MD Magazine®: Can you explain the concept of hyperspectral autofluorescence imaging?

Smith: “Sure, autofluorescence imaging is a standard technique we used in retinal diagnosis and imaging of the retina. So, it’s in widespread use, particularly in degenerations like age-related macular degeneration, to judge the health of the retina and the health of the underlying structure such as the pigment epithelium. So, it’s used to diagnose, for example, conditions like geographic atrophy which is the advanced form of dry AMD and other types of AMD.

However, the autofluorescence image is simply a monochromatic grayscale image that doesn’t give us any molecular information about the compounds that are emitting these signals. To will understand the disease better we would like to be able to take this image and these signals and split them apart into their components and assign them to their molecular origins and that way we’d have a better idea what is making this disease progress or not.

So, the hyperspectral image essentially takes the fluorescent image from the autofluorescence camera and we now break it up into 30 color challenges and these are basically the colors we are all familiar with in everyday life; it’s visible light. And we break it up into blue, green, yellow orange red, all the way out and 30 subdivisions. Now, you get this beautiful color image and it has a lot of information in it.

There’s now a specific way that we can attack all that information to pull out the characteristic spectral signatures of the compounds, the actual molecules, that are causing these emissions. And so, there are certain mathematical techniques and 1 applies and from that entire signal we can break it down into 4 or 5 major components.

For example, the first 3 components might be coming from the retinal pigment epithelium itself and then the most important 1 for us right now as we discovered there’s an additional new component which comes only from drusen and their precursors. And the drusen’s are the fundamental lesion of age-related macular degeneration. So, the point being that if we can successfully detect the precursors in the earliest stages of age-related macular degeneration, this would be a great clinical benefit.”