University College London and Moorfields Eye Hospital researchers have created a ground-breaking deep-learning AI tool. This tool might completely change how we identify Retinopathy of prematurity (ROP), the primary cause of childhood blindness. The technology automatically identifies infants in danger of the disorder, which affects preterm babies.
In ROP, the retina develops abnormal blood vessels that can bleed or leak. Thus, harming the retina and possibly resulting in blindness. Although less severe cases don’t require treatment, healthcare professionals must quickly address more serious issues to prevent blindness. Unfortunately, ROP is responsible for causing blindness in 50,000 children globally.
Also Read: Machine Learning & AI for Healthcare in 2023
AI Model Trained to Detect Infants at Risk of ROP
Using a sample of 7,414 photographs of the eyes of 1,370 babies admitted to Homerton Hospital in London and evaluated for ROP by ophthalmologists, the University College London and Moorfields Eye Hospital team trained their AI model. The program’s performance was then assessed using 200 more photos and compared to the opinions of senior ophthalmologists. Additionally, the researchers used their tool to validate it on US, Brazil, and Egypt datasets.
AI Tool Proven Effective in Diagnosing ROP
The findings demonstrated that the AI tool was as proficient as senior pediatric ophthalmologists. It helped distinguish between retinal pictures with typical ROP and those that might end in blindness. This observation is crucial since ROP symptoms are challenging to identify without adequate infrastructure. Additionally, it cannot be seen with the unaided eye for thorough prenatal and postnatal care.
Lead author Dr. Konstantinos Balaskas is the director of the Moorfields Ophthalmic Reading Centre & Clinical AI Lab. He said, “Retinopathy of prematurity is becoming increasingly common as survival rates of premature babies improve across the globe. And it is now the leading cause of childhood blindness in middle-income countries and the US.”
He added, “As many as 30% of newborns in sub-Saharan Africa have some degree of ROP, and while treatments are now readily available, it can cause blindness if not detected and treated quickly. This is often due to a lack of eye care specialists—but, given it is detectable and treatable, no child should be going blind from ROP.”
AI Tool as a Code-Free Deep Learning Platform
The creators made the AI tool a code-free deep learning platform, enabling individuals without coding skills to modify it for various contexts easily. The researchers find it encouraging that they discovered the tool is still influential on other continents, even though they designed it for the UK population.
The study’s first author, Dr. Siegfried Wagner, said, “Our findings justify the continued investigation of AI tools to screen for ROP. We are now further validating our tool in multiple hospitals in the UK. We seek to learn how people interact with the AI’s outputs, to understand how we could incorporate the tool into real-world clinical settings.”
Learn More: A Guide on Deep Learning: From Basics to Advanced Concepts
Our Say
The creation of this AI technology represents a significant advancement in the fight against ROP-related infant blindness. We can ensure that even infants in low-resource situations receive the care they require to avoid blindness by automating the diagnosis of ROP.
