Machine learning models based on the dimensionality reduction of standard automated perimetry data for glaucoma diagnosis

作者：

Highlights：

• A set of machine learning-based diagnostic models are designed that implement data manipulation, dimensionality reduction, and classification methods.

• Structure-functional and anatomical knowledge are reflected via new input variables derived from visual field clustering schemes.

• Dimensionality reduction is conducted to select important variables so as to alleviate high-dimensionality problems.

• For comparison, we applied various visual field clustering schemes, dimensionality reduction techniques, and classifiers and obtained the best model giving an AUC of 0.912.

摘要

•A set of machine learning-based diagnostic models are designed that implement data manipulation, dimensionality reduction, and classification methods.•Structure-functional and anatomical knowledge are reflected via new input variables derived from visual field clustering schemes.•Dimensionality reduction is conducted to select important variables so as to alleviate high-dimensionality problems.•For comparison, we applied various visual field clustering schemes, dimensionality reduction techniques, and classifiers and obtained the best model giving an AUC of 0.912.

论文关键词：Glaucoma,Machine learning classifier,Dimensionality reduction,Visual field clustering

论文评审过程：Received 27 October 2016, Revised 12 November 2018, Accepted 25 February 2019, Available online 25 February 2019, Version of Record 28 February 2019.

论文官网地址：https://doi.org/10.1016/j.artmed.2019.02.006