China: Machine learning and big data approaches can improve myopia prediction in Chinese children, according to a recent study published in the journal PLOS Medicine.
Electronic medical records provide large-scale real-world clinical data for use in developing clinical decision systems. However, sophisticated methodology and analytical skills are required to handle the large-scale datasets necessary for the optimization of prediction accuracy.
Myopia is a common cause of vision loss. Current approaches to control myopia progression are effective but have significant side effects. Therefore, identifying those at greatest risk who should undergo targeted therapy is of great clinical importance.
Haotian Lin, Sun Yat-sen University in Guangzhou, China, and colleagues conducted the study with an aim to apply big data and machine learning technology to develop an algorithm that can predict the onset of high myopia, at specific future time points, among Chinese school-aged children.
In eight ophthalmic centers, real-world clinical refraction data were derived from electronic medical record systems. Age, spherical equivalent (SE), and annual progression rate were used to develop the algorithm to predict SE and high myopia onset up to 10 years in the future. For algorithm training and validation, random forest machine learning was used.
- The algorithm accurately predicted high myopia in internal validation (area under the curve [AUC], 0.903 to 0.986 for three years; 0.875 to 0.901 for five years; and 0.852 to 0.888 for eight years), external validation (AUC, 0.874 to 0.976 for three years; 0.847 to 0.921 for five years; and 0.802 to 0.886 for eight years), and multiresource testing (AUC, 0.752 to 0.869 for four years).
- The algorithm provided clinically acceptable accuracy over three years (AUC, 0.94 to 0.985), five years (AUC, 0.856 to 0.901), and eight years (AUC, 0.801 to 0.837) with respect to the prediction of high myopia development by 18 years of age.
- The algorithm achieved clinically acceptable prediction of the actual refraction values at future time points, which is supported by the regressive performance and calibration curves.
“Although the algorithm achieved balanced and robust performance, concerns about the compromised quality of real-world clinical data and over-fitting issues should be cautiously considered,” write the authors.
“Our study demonstrates that machine learning prediction algorithms further translate the benefit of big data research into clinical practice.”
“To our knowledge, this study, for the first time, used large-scale data collected from electronic health records to demonstrate the contribution of big data and machine learning approaches to improved prediction of myopia prognosis in Chinese school-aged children. This work provides evidence for transforming clinical practice, health policy-making, and precise individualized interventions regarding the practical control of school-aged myopia,” they conclude.
For further reference follow the link: https://doi.org/10.1371/journal.pmed.1002674