Purpose: To present a new structural biomarker for detecting glaucoma progression based on structural transformation of the optic nerve head (ONH) region.
Methods: A dense ONH deformation was estimated using deep learning methods namely DDCNet-Multires, FlowNet2, and FlowNet-Correlation, and legacy computational methods namely the topographic change analysis (TCA) and proper orthogonal decomposition (POD) methods using longitudinal confocal scans of the ONH for each study eye. A candidate structural biomarker of glaucoma progression in a study eye was estimated as average magnitude of flow velocities within the ONH region. The biomarker was evaluated using longitudinal confocal scans of 12 laser-treated and 12 contralateral normal eyes of 12 primates from the LSU Experimental Glaucoma Study (LEGS); and 36 progressing eyes and 21 longitudinal normal eyes from the UCSD Diagnostic Innovations in Glaucoma Study (DIGS). Area under the ROC curves (AUC) was used to assess the diagnostic accuracy of the candidate biomarker.
Results: AUROC (95\% CI) for LEGS were: 0.83 (0.79, 0.88) for DDCNet-Multires; 0.83 (0.78, 0.88) for FlowNet2; 0.83 (0.78, 0.88) for FlowNet-Correlation; 0.94 (0.91, 0.97) for POD; and 0.86 (0.82, 0.91) for TCA methods. For DIGS: 0.89 (0.80, 0.97) for DDCNet-Multires; 0.82 (0.71, 0.93) for FlowNet2; 0.93 (0.86, 0.99) for FlowNet-Correlation; 0.86 (0.76, 0.96) for POD; and 0.86 (0.77, 0.95) for TCA methods. Lower diagnostic accuracy of the learning-based methods for LEG study eyes were due to image alignment errors in confocal sequences.
Conclusion: Deep learning methods trained to estimate generic deformation were able to detect ONH deformation from confocal images and provided a higher diagnostic accuracy when compared to the classical optical flow and legacy biomarkers of glaucoma progression. Because it is difficult to validate the estimates of dense ONH deformation in clinical population, our validation using ONH sequences under controlled experimental conditions confirms the diagnostic accuracy of the biomarkers observed in the clinical population. Performance of these deep learning methods can be further improved by fine-tuning these networks using longitudinal ONH sequences instead of training the network to be a general-purpose deformation estimator.
