Login / Signup

Evaluating intraocular lens power formula constant robustness using bootstrap algorithms.

Achim LangenbucherNóra SzentmáryAlan CaylessJascha A WendelsteinPeter Hoffmann
Published in: Acta ophthalmologica (2022)
We have been able to prove with bootstrapping that nonlinear iterative formula constant optimisation techniques for the Haigis, the Castrop and the SRKT formulae yield consistent results with low uncertainties of the formula constants and low variations in the back-calculated mean, median, mean absolute and root mean squared formula prediction error.
Keyphrases
  • human milk
  • machine learning
  • low birth weight
  • deep learning
  • magnetic resonance imaging
  • magnetic resonance
  • preterm infants
  • image quality