- Age-related macular degeneration (AMD) often results in permanent loss of central vision.
- A team of researchers in the UK developed an AI-based eye test to predict wet AMD in patients before the onset of symptoms.
- They published their findings in the journal Expert Review of Molecular Diagnostics
Researchers at University College London (UCL), in collaboration with Western Eye Hospital, trained a newly developed artificial intelligence (AI) algorithm to analyze retinal scans. Using the AI-based eye test, the researchers aimed to predict wet AMD in patients years before the development of symptoms.
Age-related macular degeneration (AMD) is the leading cause of severe, permanent vision loss in adults aged over 60 years. Symptoms such as blurry vision or loss of central vision occur as a result of damage to the macula of the retina. The medical condition has two main types: wet and dry. Although dry is more common, it can often progress to the wet form.
In wet AMD, abnormal blood vessels grow underneath the macula. These newly formed vessels are fragile and cause fluid and blood to leak. Thus, resulting in scarring and eventually permanent loss of central vision.
The introduction of new treatments has led to much improved results for patients, for a disease that over 20 years ago was regarded as untreatable. However, patient outcomes could be even better if treatment was started in the very earliest stages of the disease.
University College London
The AI-Based ‘Pioneering’ Eye Test
The first part of the ‘pioneering’ test involves DARC, a test initially developed for glaucoma patients. DARC (Detection of Apoptosing Retinal Cells) involves injecting a fluorescent dye into the bloodstream of patients. The dye attaches to retinal cells and illuminates areas where cells are undergoing stress or death. The more the illuminated area, the higher number of stressed retinal cells. Thus, highlighting the possibility of progression to wet AMD.
However, often at times, different specialists analyze the scan differently. Thus, resulting in varying conclusions and diagnoses. Therefore, the researchers included an AI algorithm to compare thousands of these scans and detect disease-related abnormalities. Using their newly trained AI, the researchers assessed 19 patients with early signs of AMD.
A 3-Year Head Start
Results showed that DARC particularly highlighted endothelial cells that line blood vessels. Under stress, these cells generally result in the formation of leaky blood vessels, characteristic of wet AMD. Using conventional eye scans with Optical Coherence Tomography (OCT), doctors detected these leaky vessels in patients three years later.
Our results are very promising as they show DARC could be used as a biomarker for wet AMD when combined with the AI-aided algorithm. Our new test was able to predict new wet AMD lesions up to 36 months in advance of them occurring and that is huge – it means that DARC activity can guide a clinician into treating more intensively those patients who are at high risk of new lesions of wet AMD and also be used as a screening tool.
Professor Francesca Cordeiro, lead researcher
Next, the researchers aim to conduct further clinical trials and develop the test for other eye diseases.
Reference:
UCL. AI-Supported Test Predicts Eye Disease Three Years before Symptoms. 18 Dec. 2020, www.ucl.ac.uk/news/2020/dec/ai-supported-test-predicts-eye-disease-three-years-symptoms.
