Machine Learning Models for Diagnosing Glaucoma from Retinal Nerve Fiber Layer Thickness Maps.

PURPOSE

To assess the diagnostic accuracy of multiple machine learning models using full retinal nerve fiber layer (RNFL) thickness maps in detecting glaucoma.

DESIGN

Case-control study.

PARTICIPANTS

A total of 93 eyes from 69 patients with glaucoma and 128 eyes from 128 age- and sex-matched healthy controls from the Los Angeles Latino Eye Study (LALES), a large population-based, longitudinal cohort study consisting of Latino participants aged ≥40 years residing in El Puente, California.

METHODS

The 6×6-mm RNFL thickness maps centered on the optic nerve head (Cirrus 4000; Zeiss, Dublin, CA) were supplied to 4 different machine learning algorithms. These models included 2 conventional machine learning algorithms, Support Vector Machine (SVM) and K-Nearest Neighbor (KNN), and 2 convolutional neural nets, ResNet-18 and GlaucomaNet, which was a custom-made deep learning network. All models were tested with 5-fold cross validation.

MAIN OUTCOME MEASURES

Area under the curve (AUC) statistics to assess diagnostic accuracy of each model compared with conventional average circumpapillary RNFL thickness.

RESULTS

All 4 models achieved similarly high diagnostic accuracies, with AUC values ranging from 0.91 to 0.92. These values were significantly higher than those for average circumpapillary RNFL thickness, which had an AUC of 0.76 in the same patient population.

CONCLUSIONS

Superior diagnostic performance was achieved with both conventional machine learning and convolutional neural net models compared with circumpapillary RNFL thickness. This supports the importance of the spatial structure of RNFL thickness map data in diagnosing glaucoma and further efforts to optimize our use of this data.