Heidelberg retina tomograph measurements of the optic disc and parapapillary retina for detecting glaucoma analyzed by machine learning classifiers

Abstract

Purpose: To determine whether topographical measurements of the parapapillary region analyzed by machine learning classifiers can detect early to moderate glaucoma better than similarly processed measurements obtained within the disc margin and to improve methods for optimization of machine learning classifier feature selection.

Methods: One eye of each of 95 patients with early to moderate glaucomatous visual field damage and of each of 135 normal subjects older than 40 years participating in the longitudinal Diagnostic Innovations in Glaucoma Study (DIGS) were included. Heidelberg Retina Tomograph (HRT; Heidelberg Engineering, Dossenheim, Germany) mean height contour was measured in 36 equal sectors, both along the disc margin and in the parapapillary region (at a mean contour line radius of 1.7 mm). Each sector was evaluated individually and in combination with other sectors. Gaussian support vector machine (SVM) learning classifiers were used to interpret HRT sector measurements along the disc margin and in the parapapillary region, to differentiate between eyes with normal and glaucomatous visual fields and to compare the results with global and regional HRT parameter measurements. The area under the receiver operating characteristic (ROC) curve was used to measure diagnostic performance of the HRT parameters and to evaluate the cross-validation strategies and forward selection and backward elimination optimization techniques that were used to generate the reduced feature sets.

Results: The area under the ROC curve for mean height contour of the 36 sectors along the disc margin was larger than that for the mean height contour in the parapapillary region (0.97 and 0.85, respectively). Of the 36 individual sectors along the disc margin, those in the inferior region between 240 degrees and 300 degrees, had the largest area under the ROC curve (0.85-0.91). With SVM Gaussian techniques, the regional parameters showed the best ability to discriminate between normal eyes and eyes with glaucomatous visual field damage, followed by the global parameters, mean height contour measures along the disc margin, and mean height contour measures in the parapapillary region. The area under the ROC curve was 0.98, 0.94, 0.93, and 0.85, respectively. Cross-validation and optimization techniques demonstrated that good discrimination (99% of peak area under the ROC curve) can be obtained with a reduced number of HRT parameters.

Conclusions: Mean height contour measurements along the disc margin discriminated between normal and glaucomatous eyes better than measurements obtained in the parapapillary region.

Copyright Association for Research in Vision and Ophthalmology

Figures

Figure 1

(A) Thirty-six sectors along the contour line outlining the disc margin. (B) Thirty-six sectors along a contour concentric with the disc margin contour line at a radius of 1.7 mm.

Figure 2

Area under the ROC curve by mean height contour and RNFL sectors along the disc margin and in the parapapillary region (at a radius of 1.7 mm from the center of the disc). Mean height contour (◆) and RNFL (▲) sectors along the disc margin have a larger area under ROC curve than mean height contour (■) and RNFL (✖) in the parapapillary region.

Figure 3

SVM Gaussian optimization using a combination of parameters: global, regional, and mean height contour along disc margin. Optimization using (A) forward selection and (B) backward elimination show that the area under ROC curve is higher for the selection (thick dashed line) ROC curve than the verification (thick solid line) ROC curve. Five cross-validation replications (represented by the five thin dashed lines) surrounding the selection ROC curve and five replications (represented by the five thin dotted lines) were used to create the selection and verification ROC curve, respectively. In addition, forward selection and backward elimination show that less than 10 HRT parameters are required for a model to discriminate at 99% of the peak area under ROC curve (arrows).