A novel artificial intelligence (AI) algorithm-supported test can detect glaucoma progression 18 months earlier than current gold standard methods, according to a new UK phase 2 clinical study.
The technology could help accelerate clinical trials, and eventually be used in detection and diagnostics, the study published in Expert Review of Molecular Diagnostics suggests.
The test, Detection of Apoptosing Retinal Cells (DARC), involves intravenous injection (via the arm) of a fluorescent dye that attaches to retinal cells, illuminating those in the process of apoptosis. The damaged cells appear bright white when viewed in eye examinations – the more damaged cells detected, the higher the DARC count.
The researchers acquired anonymised DARC images from healthy control participants (n=40) and glaucoma patients (n=20) from which observers manually counted spots.
The convolutional neural network-aided algorithm was trained and validated using manual counts from control participants and then was tested on glaucoma eyes. The algorithm had 97.0 per cent accuracy, 91.1 per cent sensitivity and 97.1 per cent specificity to spot detection when compared with manual grading of 50 per cent control participants.
It was next tested on glaucoma patient eyes defined as progressing or stable based on a significant (P<.05 rate of progression using optical coherence tomography-retinal nerve fibre layer measurements at months. it demonstrated per cent specificity with area under the curve a significantly greater darc count in those patients who later progressed.>
The AI-supported technology was recently approved by the UK Medicines and Healthcare products Regulatory Agency and US Food and Drug Administration as an exploratory endpoint for a new glaucoma drug clinical trial.
The researchers are also applying the DARC test to rapidly detect cell damage caused by other neurodegenerative conditions involving loss of nerve cells, including age-related macular degeneration, multiple sclerosis and dementia.
They are also assessing its use in lung disease and hope that by year end, it may help assess people with breathing difficulties from COVID-19.