Glaucoma Diagnostic Accuracy of Machine Learning Classifiers Using Retinal Nerve Fiber Layer and Optic Nerve Data from SD-OCT.
Glaucoma Diagnostic Accuracy of Machine Learning Classifiers Using Retinal Nerve Fiber Layer and Optic Nerve Data from SD-OCT.
复制标题
DOI：
10.1155/2013/789129
复制
发表时间：
2013
影响因子：
1.9
通讯作者：
Gomi ES
中科院分区：
文献类型：
--
作者：
Barella KA;Costa VP;Gonçalves Vidotti V;Silva FR;Dias M;Gomi ES
目的。目的 利用谱域光学相干断层扫描 (SD-OCT) 获得的视网膜神经纤维层 (RNFL) 和视神经 (ON) 参数来研究机器学习分类器 (MLC) 的诊断准确性。方法。招募了 57 名早期至中度原发性开角型青光眼患者和 46 名健康患者。所有 103 名患者均接受了完整的眼科检查、消色差标准自动视野检查以及 SD-OCT 成像。针对 RNFL 和 ON 参数构建了受试者工作特征 (ROC) 曲线。测试了 10 个 MLC。比较每个 SD-OCT 参数和 MLC 获得的 ROC 曲线下面积 (aROC)。结果。健康个体的平均年龄为 56.5 ± 8.9 岁，青光眼患者的平均年龄为 59.9 ± 9.0 岁 (P = 0.054)。健康个体的平均偏差值为-1.4 dB，青光眼患者的平均偏差值为-4.0 dB（P < 0.001）。 aROC 最大的 SD-OCT 参数为杯/椎间盘面积比 (0.846) 和平均杯/椎间盘面积比 (0.843)。使用分类器获得的 aROC 范围为 0.687 (CTREE) 到 0.877 (RAN)。使用 RAN 获得的 aROC (0.877) 与使用最佳单一 SD-OCT 参数 (0.846) 获得的 aROC 没有显着差异 (P = 0.542)。结论。 MLC 显示出良好的准确性，但并未提高 SD-OCT 诊断青光眼的敏感性和特异性。
Purpose. To investigate the diagnostic accuracy of machine learning classifiers (MLCs) using retinal nerve fiber layer (RNFL) and optic nerve (ON) parameters obtained with spectral domain optical coherence tomography (SD-OCT). Methods. Fifty-seven patients with early to moderate primary open angle glaucoma and 46 healthy patients were recruited. All 103 patients underwent a complete ophthalmological examination, achromatic standard automated perimetry, and imaging with SD-OCT. Receiver operating characteristic (ROC) curves were built for RNFL and ON parameters. Ten MLCs were tested. Areas under ROC curves (aROCs) obtained for each SD-OCT parameter and MLC were compared. Results. The mean age was 56.5 ± 8.9 years for healthy individuals and 59.9 ± 9.0 years for glaucoma patients (P = 0.054). Mean deviation values were −1.4 dB for healthy individuals and −4.0 dB for glaucoma patients (P < 0.001). SD-OCT parameters with the greatest aROCs were cup/disc area ratio (0.846) and average cup/disc (0.843). aROCs obtained with classifiers varied from 0.687 (CTREE) to 0.877 (RAN). The aROC obtained with RAN (0.877) was not significantly different from the aROC obtained with the best single SD-OCT parameter (0.846) (P = 0.542). Conclusion. MLCs showed good accuracy but did not improve the sensitivity and specificity of SD-OCT for the diagnosis of glaucoma.
登录
查看更多内容
影响因子：
4.4
作者：
Bowd, Christopher;Hao, Jiucang;Goldbaum, Michael H.
通讯作者：
Goldbaum, Michael H.
影响因子：
2
作者：
Mardin, Christian Y.;Peters, Andrea;Lausen, Berthold
通讯作者：
Lausen, Berthold
影响因子：
7.5
作者：
Tucker, A;Vinciotti, V;Garway-Heath, D
通讯作者：
Garway-Heath, D
影响因子：
4.2
作者：
QUIGLEY, HA;DUNKELBERGER, GR;GREEN, WR
通讯作者：
GREEN, WR
影响因子：
3.4
作者：
Bizios, Dimitrios;Heijl, Anders;Bengtsson, Boel
通讯作者：
Bengtsson, Boel