Mardin Christian Y, Hothorn Torsten, Peters Andrea, Jünemann Anselm G, Nguyen Nhung X, Lausen Berthold
Department of Ophthalmology and Eye Hospital, Friedrich-Alexander-University Erlangen-Nürnberg, Erlangen, Germany.
J Glaucoma. 2003 Aug;12(4):340-6. doi: 10.1097/00061198-200308000-00008.
In this article we propose and evaluate nonparametric tree classifiers that can handle non-normal data and a large number of possible predictors using the full set of standard Heidelberg Retina Tomograph measurements for classifying glaucoma.
The classifiers were trained and tested using standard Heidelberg Retina Tomograph parameters from examinations of 98 subjects with glaucoma and 98 normal subjects of the Erlangen Glaucoma Registry. All patients and control subjects were evaluated by 15 degrees -optic disc stereographs, Heidelberg Retina Tomograph measurements, standard computerized white-in-white perimetry, and 24-hour-intraocular pressure profiles. The subjects were matched by age and sex. Standard classification trees as well as bagged classification trees were used. The classification outcome of the trees was compared with the classification by two published linear discriminant functions based on Heidelberg Retina Tomograph variables with respect to their cross-validated misclassification error.
The bagged classification tree had the lowest misclassification error estimate of 14.8% with a sensitivity of 81.6% at a specificity of 88.8%. The cross-validated error rates of the two linear discriminant function procedures were 20.4% (sensitivity 82.6%, specificity 76.7%) and 20.6% (sensitivity 81.4%, specificity 77.3%) for our set of observations. Bagged classification trees were able to reduce the misclassification error of glaucoma classification.
Bagged classification trees promise to be a new and efficient approach for glaucoma classification using morphometric 2- and 3-dimensional data derived from the Heidelberg Retina Tomograph, taking into account all given variables.
在本文中，我们提出并评估了非参数树分类器，该分类器能够处理非正态数据以及大量可能的预测变量，并使用全套标准海德堡视网膜断层扫描测量数据来对青光眼进行分类。
使用来自埃尔朗根青光眼登记处的98例青光眼患者和98例正常受试者的检查中的标准海德堡视网膜断层扫描参数对分类器进行训练和测试。所有患者和对照受试者均通过15度视盘立体照片、海德堡视网膜断层扫描测量、标准计算机化白对白视野检查以及24小时眼压曲线进行评估。受试者按年龄和性别进行匹配。使用了标准分类树以及袋装分类树。将树的分类结果与基于海德堡视网膜断层扫描变量的两个已发表的线性判别函数的分类结果在交叉验证误分类误差方面进行比较。
袋装分类树的误分类误差估计最低，为14.8%，敏感度为81.6%，特异度为88.8%。对于我们的观察组，两种线性判别函数程序的交叉验证错误率分别为20.4%（敏感度82.6%，特异度76.7%）和20.6%（敏感度81.4%，特异度77.3%）。袋装分类树能够降低青光眼分类的误分类误差。
考虑到所有给定变量，袋装分类树有望成为一种使用从海德堡视网膜断层扫描获得的形态学二维和三维数据进行青光眼分类的新的有效方法。
