Glaucoma Diagnostic Accuracy of Machine Learning Classifiers Using Retinal Nerve Fiber Layer and Optic Nerve Data from SD-OCT
Figure 1
Areas under the receiver operating characteristic curve (aROCs) of the best classifier trained with the number of spectral domain optical coherence tomography (SD-OCT) parameters which allowed the best performance (RAN: random forest = 0.877) and aROC of the best SD-OCT parameter (CDA: cup/disc area = 0.846) ().
