Evaluating the potential of hyperspectral imaging for detecting and monitoring geographic atrophy
Accurate monitoring of the sight-threatening stage of AMD, known as geographic atrophy (GA) is important to identify the people who are most likely to benefit from new treatments. Our team has discovered a new way to monitor GA. In GA, the light sensing cells of the eye, plus a key group of supporting cells, gradually die. Existing eye scans are good at detecting affected areas once these cells have died, however some types of these scans are not widely available and are not well tolerated by patients. We have developed a new eye camera that uses the same principles as sophisticated cameras that were first used by NASA to image the Earth from satellites. The camera is like a standard retinal camera, but instead of using a white flash and capturing a single photograph our camera uses a series of 30 flashes with different colours of light, capturing a photograph with each flash. Remarkably, this all happens in a quarter of a second. When these 30 images are combined, we gain incredibly detailed information about the eye. We plan to use artificial intelligence to delve into the rich information provided by these scans to see if they will provide a better way to monitor the disease, and potentially identify people who are at risk of vision loss before it happens, when treatments may be the most effective.