Machine Learning based Predictions of Subjective Refractive Errors of the Human Eye

Alexander Leube, Christian Leibig, Arne Ohlendorf, Siegfried Wahl

Abstract

The aim of this research was to demonstrate the suitability of a data-driven approach to identify the subjective refraction. An artificial deep learning network with two hidden layers was trained to predict power vector refraction (M, J0 and J45) from 37 dimensional feature vectors (36 Zernike coefficients + pupil diameter) from a large database of 50,000 eyes. A smaller database of 460 eyes containing subjective and objective refraction from controlled experiment conditions was used to test for prediction power. analysis was performed, calculating the mean difference (eg ΔM) and the 95% confidence interval (CI) between predictions and subjective refraction. Using the machine learning approach, the accuracy (ΔM = +0.08D) and precision (CI for ΔM = ± 0.78D) for the prediction of refractive error corrections was comparable to a conventional metric (ΔM = +0.11D ± 0.89D) as well as the inter-examiner agreement between optometrists (ΔM = -0.05D ± 0.63D). To conclude, the proposed deep learning network for the prediction of refractive error corrections showed its suitability to reliably predict subjective power vectors of refraction from objective wavefront data.

Download


Paper Citation


in Harvard Style

Leube A., Leibig C., Ohlendorf A. and Wahl S. (2019). Machine Learning based Predictions of Subjective Refractive Errors of the Human Eye.In Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 5: HEALTHINF, ISBN 978-989-758-353-7, pages 199-205. DOI: 10.5220/0007254401990205


in Bibtex Style

@conference{healthinf19,
author={Alexander Leube and Christian Leibig and Arne Ohlendorf and Siegfried Wahl},
title={Machine Learning based Predictions of Subjective Refractive Errors of the Human Eye},
booktitle={Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 5: HEALTHINF,},
year={2019},
pages={199-205},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007254401990205},
isbn={978-989-758-353-7},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies - Volume 5: HEALTHINF,
TI - Machine Learning based Predictions of Subjective Refractive Errors of the Human Eye
SN - 978-989-758-353-7
AU - Leube A.
AU - Leibig C.
AU - Ohlendorf A.
AU - Wahl S.
PY - 2019
SP - 199
EP - 205
DO - 10.5220/0007254401990205