Ocular nutrition counseling has become a routine part of patient education discussions for conditions such as AMD and others. One shortcoming, though, is how hard it is to personalize the recommendations to each individual. However, researchers from China developed an algorithm that can predict the optimal dose of a lutein combination that each patient needs to successfully treat asthenopia.

The machine learning–based model encompassed over 500 features, such as demographics, anthropometrics, eye-related indexes, blood biomarkers and dietary habits. The supplement combination included lutein ester, zeaxanthin and extracts of black currant, chrysanthemum and goji berry.

The randomized control trial enrolled 303 participants—28 were placed in a test set and 30 were included in a validation set. The investigators developed an aggregated score of visual health upon assessing eye fatigue symptoms, visuognosis persistence, macular pigment optical density and Schirmer test results. An algorithm predicted visual health at 45 days after treatment, taking into account three different doses containing 6mg, 10mg and 14mg of lutein.

Among the patients in the test and validation sets, 56 (97%) showed significant eye fatigue improvement, with their visual health elevated by more than 0.1 after 45 days. The researchers noted the other two patients didn’t benefit from the botanical combination, as they already had relatively good visual health at baseline. The algorithm predicted 39 of the participants should take the highest dose, flagged 17 to take a lower dose and suggested two wouldn’t benefit from the combination.

Though the research is preliminary, it could laid the groundwork for a clinically viable approach to personalized eye care counseling in the future.

Kan J, Li A, Zou H, et al. A machine learning based dose prediction of lutein supplements for individuals with eye fatigue. Front Nutr. 2020;7:577923.