Data generalization for predicting intraocular lens power by means of ANN-models

Abstract

The article studies the possibility of using mathematical models obtained as a result of deep learning of artificial neural networks (ANN models) to predict the optical power of state-of-the-art intraocular lenses (IOLs), widely used in the surgical cataract treatment in ophthalmology. A distinctive feature of such ANN models in comparison with the well-known formulas SRK II, SRK/T, Hoffer-Q, Holladay II, Haigis, Barrett is their ability to take into account a significant number of recorded input quantities, which makes it possible to reduce the mean relative error in calculating the optical power of IOL from 10–12 % to 3,5 %. ANN models have been trained on large-scale samples including depersonalized data for 455 patients. The resulting models, in contrast to the traditionally used formulas, reflect the regional specificity of patients to a much greater extent. They also make it possible to retrain and optimize the structure based on newly received data, which allows taking into account the non-stationarity of the object. The results obtained allow, in principle, to build an intelligent expert system with a continuous flow of new data from a source and a step-by-step retraining of the ANN model. For the construction of ANN-models and their machine learning process, we used the simulator program previously developed by the authors of this article, as well as the Python language tools in the Google Colaboratory. For training models, based on empirical data, the following optimization methods were chosen: the stochastic gradient method, the simple gradient method, and the non-gradient Gauss-Seidel and Monte Carlo coordinate descent methods, which were used interactively.