Tech & Innovation in Healthcare

Answer: Researchers are exploring the potential of using artificial intelligence (AI) to analyze eye scans to identify several conditions. One way researchers are using AI with eye scans is to evaluate patients for possible Parkinson’s disease.

In a study published to Neurology in August 2023, researchers applied AI to identify Parkinson’s disease biomarkers in retinal images up to seven years before patients exhibit symptoms. Researchers used data from the AlzEye and UK Biobank datasets. The AlzEye dataset was used to detect retinal markers in prevalent Parkinson’s disease, and contained a selection of 154,830 patients aged 40 years and older from ophthalmic hospitals in London, UK from 2008 to 2018. The UK Biobank dataset was used to evaluate retinal markers in incident Parkinson’s disease, and this dataset was made up of “a prospective population-based cohort where 67,311 volunteers aged 40-69 years were recruited between 2006 and 2010 and underwent retinal imaging,” the researchers wrote.

The eye scans use optical coherence tomography (OCT) to capture 3D cross-sectional images of the retina. The extremely high-resolution images can measure the retinal nerves down to a thousandth of a millimeter. OCT scans are valuable to viewing cell layers below the surface of the skin.

In the study, researchers found that a thickness reduction in the cell layers can be associated with an increased risk in developing Parkinson’s disease. Researchers also found using OCT scans and AI to evaluate for Parkinson’s disease is less invasive, lower cost, offers greater scalability, and is faster than scanning the brain to evaluate for the disease.