Test and AI Algorithm Predict Development of Wet AMD up to 3 Years Before Symptoms Develop

Test and AI Algorithm Predict Development of Wet AMD up to 3 Years Before Symptoms Develop

Scientists at UCL have developed a new eye test and AI-based algorithm that allows the development of wet age-related macular degeneration (AMD) to be detected up to three years before symptoms develop.

AMD is the leading cause of permanent vision loss in the western world and the third most common cause globally. AMD results in the loss of central vision due to damage to the macula. Wet AMD is caused by abnormal blood vessels forming under the macular which cause damage to its cells.

Around 600,000 people in the United Kingdom have some degree of AMD, with the condition more common in women than men. Currently, wet AMD is usually only detected when patients complain of distorted vision. Without treatment the condition can deteriorate very quickly, leading to total loss of central vision.

Wet AMD can be treated with medications that prevent further abnormal blood vessels from developing, and laser surgery can be performed to destroy the abnormal blood vessels under the macula. These treatments will not improve vision that has already deteriorated, but they will prevent the condition from deteriorating. It is therefore important to provide treatment as soon as possible, so early detection is critical.

The new test – Detection of Apoptosing Retinal Cells (DARC) – was developed by UCL researchers in collaboration with London’s Western Eye Hospital and involves injecting a fluorescent dye into the bloodstream via an injection in the arm. The dye travels to the eyes where it attaches to retinal cells and shows cells that are under stress or in the process of apoptosis, which appear bright white. The number of damaged cells is then reflected in the DARC count.

The researchers developed an AI algorithm to assist eye specialists with identifying wet AMD from the DARC test. The new test was evaluated in a trial conducted on 19 patients with symptoms of wet AMD, with the data was used to train the AI algorithm to detect leaking of blood vessels and the formation of new, abnormal blood vessels. The trial showed that DARC can identify endothelial cells under stress, which can be used to predict the development of wet AMD in the future. The test could therefore potentially be used for screening people at risk of developing wet AMD.

“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,” said lead researcher Professor Francesca Cordeiro.

The researchers plan to conduct a clinical trial on a larger group of individuals and will also evaluate the effectiveness of the AI-aided test at detecting and predicting other eye diseases that can lead to sight loss.

You can read more about the study in the paper – A CNN-aided method to predict glaucoma progression using DARC (Detection of Apoptosing Retinal Cells) – which was recently published in Expert Review of Molecular Diagnostics. DOI: 10.1080/14737159.2020.1758067

Leave a Reply