AS WE LOOK at what is ahead in the myopia space, it’s exciting to see technological advances that will extend myopia management methods across the globe and individualize patient care. How can eyecare practitioners manage the spread of this condition, which has become a public health pandemic? Hopefully, they can do so by implementing these technologies in their offices to expand their effectiveness.
FUNDUS IMAGING
Fundus imaging is becoming the standard of care throughout the eyecare world. It is a required part of every comprehensive assessment that we do on patients in our offices, including those referred by other practitioners. Someday, fundus imaging may serve as a metric for predicting myopia progression.
A recent study looked at 16,211 fundus images from 3,408 children over a period of six years (Kang et al, 2024). Utilizing deep learning, the study group developed a model that achieved 87.9% accuracy in predicting general myopia risk and 99.5% accuracy in predicting high myopia. The authors stated that their model can provide good predictions based on a single measurement (Kang et al, 2024).
Artificial intelligence (AI) has the ability to evaluate huge databases and come up with high statistical predictions. Because fundus imaging is more available than other early myopia diagnostics, such as axial length, practitioners hope that large-scale screenings will become available and will flag patients—particularly those who are at risk of high myopia—for treatment at a very early age.
OCULAR APPEARANCE
An interesting study published by Yang et al (2020) used a deep learning systems (DLS) collected from a variety of cameras, including the iPhone 6 and iPhone 7. The system analyzed many aspects of the images and determined that the temporal sclera is the most significant factor in the detection of myopia (Yang et al, 2020).
Integrating technology and AI to detect myopia, as this model does, is groundbreaking. Although the method is not available on a grand scale, it is easy to imagine the option of activating this feature on a smartphone and comparing images of one’s own child to 200 to 300 images of other children’s eyes. Then, an alert could pop up to tell parents that they should consider an eye exam for their child.
It may seem far-fetched, but something like this may be closer than we think. Additionally, this technology could have substantial implications for the detection and management of myopia across the globe.
REFERENCE
1. M Kang, Y Hu, S Gao, et al. Deep Learning-Based Longitudinal Prediction of Childhood Myopia Progression Using Fundus Image Sequences and Baseline Refraction Data. arXiv. 2024 Jul 31.
2. Yang Y, Li R, Lin D, et al. Automatic identification of myopia based on ocular appearance images using deep learning. Ann Transl Med. 2020 Jun 15;8:705.
