Machine learning classifiers for glaucoma diagnosis based on classification of retinal nerve fibre layer thickness parameters measured by Stratus OCT
Machine learning classifiers for glaucoma diagnosis based on classification of retinal nerve fibre layer thickness parameters measured by Stratus OCT
复制标题
DOI：
10.1111/j.1755-3768.2009.01784.x
复制
发表时间：
2010-02-01
影响因子：
3.4
通讯作者：
Bengtsson, Boel
中科院分区：
文献类型：
--
作者：
Bizios, Dimitrios;Heijl, Anders;Bengtsson, Boel
目的：比较基于光学相干断层扫描(OCT)测量视网膜神经纤维层厚度(RNFLT)的两种机器学习分类器(MLCs)--人工神经网络(ANN)和支持向量机(SVMs)--在青光眼诊断中的性能，并评价不同输入参数的影响。使用传统的OCT RNFLT参数加上新的参数，如最小RNFLT值、测量的RNFLT的第10和第90个百分位数，以及A扫描测量的转换，比较了MLCS的性能。结果：神经网络与支持向量机之间无统计学差异。人工神经网络(0.982，95%CI：0.966~0.999)和支持向量机(0.989，95%CI：0.979~1.00)的最佳Arocs均基于转换后的A超测量结果。在此输入上训练的支持向量机比在任何单个RNFLT参数上训练的ANN或支持向量机性能更好(p
Purpose: To compare the performance of two machine learning classifiers (MLCs), artificial neural networks (ANNs) and support vector machines (SVMs), with input based on retinal nerve fibre layer thickness (RNFLT) measurements by optical coherence tomography (OCT), on the diagnosis of glaucoma, and to assess the effects of different input parameters.Methods: We analysed Stratus OCT data from 90 healthy persons and 62 glaucoma patients. Performance of MLCs was compared using conventional OCT RNFLT parameters plus novel parameters such as minimum RNFLT values, 10th and 90th percentiles of measured RNFLT, and transformations of A-scan measurements. For each input parameter and MLC, the area under the receiver operating characteristic curve (AROC) was calculated.Results: There were no statistically significant differences between ANNs and SVMs. The best AROCs for both ANN (0.982, 95%CI: 0.966-0.999) and SVM (0.989, 95% CI: 0.979-1.0) were based on input of transformed A-scan measurements. Our SVM trained on this input performed better than ANNs or SVMs trained on any of the single RNFLT parameters (p