Carlos S. Hernández, Andrea Gil, Ignacio Casares, Jesús Poderoso, Alec Wehse, Shivang R. Dave, Daryl Lim, Manuel Sánchez-Montañés, Eduardo Lage. Prediction of manifest refraction using machine learning ensemble models on wavefront aberrometry data. Journal of Optometry, 2022. ISSN 1888-4296, https://doi.org/10.1016/j.optom.2022.03.001.
Key findings
Machine learning methods are effective for improving precision in predicting patient’s subjective refraction from objective measurements taken with a low-cost portable device.
Agreement with subjective refraction for M, J0, and J45 was considerably improved compared to the baseline results provided by the autorefractor.
The machine learning approach, implementable via software, may potentially improve upon the accuracy of the handheld autorefractor used in the study in a cost-effective manner.
The technology could be applied to other autorefractors to improve access to vision correction by non-technical ECPs in health disparity populations.