Evaluation of Orthokeratology Lenses Fitting Using Combination of K-Means and Least Squares Algorithm

摘要

Orthokeratology (OrthoK) is an effective treatment for myopia correction and control. OrthoK lenses decentration usually leads to high-order aberrations. Currently, the evaluation method for lens fitting relies on experienced doctors and changes the normal distribution of tear film. We present an evaluation scheme which combines K-means with weighted least squares to analyze lens fitting. And we assume the treatment zone of OrthoK lenses is an ellipse, of which the parameters can be computed. K-means is used to cluster the data points, and weights of each point are assigned based on the results to increase accuracy. Then, the parameters of ellipse can be calculated by weighted least squares based on the weights. Lens fitting is evaluated by the ratio between the decentration distance and major axis of the ellipse. The experiment results show that 88.6% of the samples using our method have the same results as traditional method.