Researchers have leveraged artificial intelligence (AI) to enhance treatment strategies for patients with keratoconus, a degenerative eye condition primarily affecting young adults. This advancement was discussed at the 43rd Congress of the European Society of Cataract and Refractive Surgeons (ESCRS). Keratoconus can lead to significant visual impairment and, in severe cases, necessitates corneal transplantation. Traditionally, identifying patients who require intervention has relied on prolonged monitoring.
The study, led by Dr. Shafi Balal from Moorfields Eye Hospital NHS Foundation Trust and University College London, involved analyzing 36,673 optical coherence tomography (OCT) images from 6,684 patients. The AI algorithm developed was able to predict the progression of keratoconus based on initial visit data, successfully categorizing approximately two-thirds of patients as low-risk who could continue monitoring without treatment. The accuracy improved to 90% when data from a second visit was included.
Cross-linking treatment, which involves the application of ultraviolet light and riboflavin drops to strengthen the cornea, can halt disease progression in over 95% of cases when applied before scarring occurs. The findings suggest that timely identification of high-risk patients could lead to early intervention, potentially avoiding vision loss and the need for more invasive procedures.
The researchers plan to refine their AI algorithm further, aiming to enhance its applicability to other eye conditions. Dr. José Luis Güell, an expert in the field who was not involved in the study, acknowledged the potential of AI to streamline treatment decisions and reduce the burden of unnecessary monitoring for stable patients, thereby optimizing healthcare resources. Further validation of the algorithm's effectiveness will be essential before its broader clinical implementation.