Prediction of manifest refraction using machine learning ensemble models on wavefront aberrometry data

Our last Research paper has been just published in an Special issue in Artificial Intelligence of the Journal of Optometry. This work shows that Machine Learning / Artificial Intelligence techniques can be used on wavefront aberrometry data to enhance the performance of refraction technologies.

The results obtained in this work suggest that a ML approach, implementable via software, may potentially improve upon the accuracy of the handheld autorefractor used in the study in a cost-effective manner. An important point to note is that since the most influential variables for the models were the standard variables in aberrometry, this approach could potentially be extended to other autorefractors, supporting the use of the proposed methodology to improve access to vision correction by non-technical eye care providers in health disparity populations.

Read the Open-access article here