AI can detect Parkinson's 7 years before clinical diagnosis

High-resolution 3D retina images have specific markers that can indicate the risk of Parkinson's in a person. A new AI program can identify these markers and tell whether or not you have the disease.
Rupendra Brahambhatt
An illustration of 3D retinal scan
An illustration of 3D retinal scan


Although Parkinson’s disease (PD) is incurable, a non-profit National Council of Aging report suggests that early detection and treatment could help patients live a long and productive life even with the disease.

However, in reality, even by the age of 50, less than 10 percent of patients are diagnosed. In fact, most PD patients found out about the condition in their 60s, and by then, it is too late for any treatment to work effectively. 

Researchers from University College London and Moorfields Eye Hospital realized this problem, and in their latest study, they propose an AI-based solution that can detect PD in patients seven years before the current diagnostic methods.

The study reveals that human eyes have markers for Parkinson’s disease. Their AI program can identify such markers in 3D retinal scans and provide insights into a person's potential risk for the disease.

“Finding signs of a number of diseases before symptoms emerge means that, in the future, people could have the time to make lifestyle changes to prevent some conditions arising, and clinicians could delay the onset and impact of lifechanging neurodegenerative disorders,” said Dr. Siegfried Wagner, lead researchers, and an eye specialist at UCL.

Your eyes are really the window to your body

The current study is not the first to report that neurodegenerative diseases like Parkinson’s can be identified by looking at retinal scans. 

Researchers in the past have used optical coherence tomography (OCT) and high-resolution 3D retina scans to detect eye-related problems and neurodegenerative disorders like schizophrenia, Alzheimer’s disease, and multiple sclerosis.

OCT allows scientists to zoom in at even one thousandth millimeter area inside eyes and study inconsistencies and anomalies in different cell layers. The current study's authors even claim that OCT is a better, cheaper, and quicker way to detect PD than brain scans.

“These images are extremely useful for monitoring eye health, but their value goes much further, as a retina scan is the only non-intrusive way to view layers of cells below the skin’s surface,” the researchers note.

For instance, a study published in 2015 hints that if the OCT scan of an individual’s eyes reveals that their retina is thinner than normal GCIPL (ganglion cell–inner plexiform layer), they are at higher risk of developing PD.  

Another study published in 2021 suggests that people with PD are more likely to have retinas with a thinner inner nuclear layer (INL) when compared to those who are in good health. 

However, none of the previous research works suggested a way to diagnose PD several years before clinical diagnosis. The current study confirms these findings and reveals an AI-driven method to detect the disease early using OCT.

AI detects PD without any hassle

To identify the specific PD markers in the eye, the researchers developed an AI program and then trained it using two large datasets. First, the AI examined the information from AlzEye, the world’s largest retinal image database, which comprises over 1.5 OCT and a total of 6.2 million retinal images.

Next, the researchers ran the OCT data of 85,000 patients from the UK Biobank through their AI. Using these two extensive and powerful datasets, the team was able to find subtle markers in the retina linked to Parkinson's disease. 

“This work demonstrates the potential for eye data, harnessed by the technology to pick up signs and changes too subtle for humans to see. We can now detect very early signs of Parkinson’s, opening up new possibilities for treatment,”  said Alistair Denniston, one of the study authors and a consultant opthalmologist at the University of Birmingham.

The researchers suggest that not just PD but many other diseases can be predicted at a very early stage if OCT data of a large population is available. 

This AI-driven technique could emerge as a fast, scalable, and non-invasive way to diagnose diseases. Plus, it also has the potential to significantly lower the burden of diseases like PD on both patients and hospitals. However, the researchers admit that this is just the beginning.