The neurologist and assistant professor at the University of Toronto provided insight on the clinical use of retinal imaging tools like RetiSpec and the need for further validation of these approaches in Alzheimer disease. [WATCH TIME: 2 minutes]
WATCH TIME: 2 minutes
"The way the analysis goes, it’s completely objective. There’s no radiologist thinking if there’s amyloid in the brain or not and having to make a decision. In the machine learning algorithm, it gets better with the more cases it assesses.”
Although there have been increased efforts to detect the pathobiology of Alzheimer disease (AD), a lot of the conventional methods are invasive, expensive, and not widely available. The retina, a protrusion of the central nervous system, has emerged as a prominent site of AD pathology. RetiSpec, a medical imaging company, harnesses hyperspectral imaging with artificial intelligence to allow for the rapid, simple, and cost-effective detection of AD biomarkers.
A cross-sectional study presented at the 14th Clinical Trials on Alzheimer’s Disease Conference (CTAD), November 9-12, 2021, evaluated the accuracy of RetiSpec’s hyperspectral retinal imaging system in predicting individual brain amyloid-ß (Aß) status, as compared to clinical gold standards of PET and/or cerebrospinal fluid assessment. Among a cohort of 108 participants who had a mean age of 70 years, diagnostic performance of RetiSpec in predicting Aß was demonstrated by an area under the curve of 0.88, which corresponded with 86% sensitivity and 80% specificity.
Although investigational, lead author Sharon Cohen, MD, believes there is great potential for retinal approaches such as those being taken by RetiSpec. Cohen, a neurologist and assistant professor at the University of Toronto, sat down with NeurologyLive® to discuss the advantages this type of thinking brings to the AD field, along with the need to further validate these results in additional studies.