Artificial Intelligence Advances in Predicting and Preventing Vision Loss in Young Adults

Researchers have developed an innovative application of artificial intelligence (AI) to identify young adult patients at risk of developing severe corneal conditions that could lead to vision loss. Presented at the 43rd Congress of the European Society of Cataract and Refractive Surgeons (ESCRS), this study highlights AI's potential to revolutionize eye care by predicting which patients with keratoconus—a progressive eye disease affecting the cornea—require urgent treatment.

Keratoconus often begins in adolescence and early adulthood, impairing vision and, in severe cases, necessitating a corneal transplant. Currently, the primary method to determine treatment needs involves ongoing monitoring over years, including repeated imaging and assessments. However, this process can be resource-intensive and challenging.

The research team, led by Dr. Shafi Balal at Moorfields Eye Hospital and University College London, employed AI algorithms to analyze over 36,000 optical coherence tomography (OCT) images from more than 6,600 patients. By integrating these images with patient data, the AI was able to accurately forecast which individuals would experience disease progression. Remarkably, the study demonstrated that using initial visit data, the AI could predict disease trajectory with up to 90% success when additional visit information was included.

This predictive capability allows clinicians to identify high-risk patients early, enabling timely intervention with procedures like corneal cross-linking using ultraviolet light and riboflavin. This treatment can stabilize the cornea, with a success rate exceeding 95%, potentially avoiding the need for transplant surgery. Conversely, low-risk patients can be spared frequent monitoring, optimizing healthcare resources.

The algorithm showed promise in categorizing patients into low- and high-risk groups, thus improving personalized treatment plans. Although the current model is tailored to a specific OCT device, the research team plans to adapt and validate it for broader clinical use. Ultimately, this AI-driven approach aims to facilitate early, targeted interventions, preventing the progression to blindness and reducing the burden of invasive surgeries.

Dr. José Luis Güell from the Institute of Microcirurgia Ocular commented on the significance of this work, noting that AI could enhance early diagnosis and treatment decisions, especially for managing keratoconus in young patients. The researchers are also working on developing more advanced AI systems capable of detecting other eye infections and inherited conditions, further transforming ophthalmic care.

This technological breakthrough underscores AI’s potential to improve patient outcomes through precision medicine, making eye care more proactive and efficient.