Abstract
Purpose :
(1) To assess whether neural tissue structure and biomechanics could predict functional loss in glaucoma; (2) To evaluate the importance of biomechanics in making such predictions.
Methods :
We recruited 238 glaucoma subjects from the Singapore National Eye Centre. For one eye of each subject, we imaged the optic nerve head (ONH) using spectral-domain OCT under the following conditions: (1) primary gaze and (2) primary gaze with acute IOP elevation (to approximately 33 mmHg) achieved through ophthalmo-dynamometry. Automatic segmentation of neural tissues (retinal nerve fiber layer [RNFL] and pre-lamina) was performed for each OCT volume. Digital volume correlation (DVC) analysis was employed to compute IOP-induced neural tissue strains as detailed in Chuangsuwanich et al. (Ophthalmology, 2023). To predict visual field loss from neural tissue structure and biomechanics, we used a robust geometric deep learning approach known as Point-Net. This method takes 3D point-clouds of neural tissues as inputs (see Figure 1a), together with local tissue strain and thickness, in order to predict the full 24-2 pattern standard deviation (PSD) maps. For each point in each PSD map, we predicted whether it exhibited no defect or a PSD value of less than 5%. Predictive performance was assessed using 5-fold cross-validation and the Dice coefficient, which measures point-wise agreement between the true and predicted PSD maps. The model's performance was compared with and without the inclusion of IOP-induced strains.
Results :
Integrating biomechanical (IOP-induced neural tissue strains) and structural (neural tissues thickness) information yielded a robust predictive model (Dice score: 0.79 ± 0.03) across test subjects (Figure 1b), accurately identifying defective hemispheres. In contrast, relying only on structural information resulted in a significantly lower Dice score of 0.72 ± 0.02 (p < 0.05, Figure 1c).
Conclusions :
Our study has shown that the integration of biomechanical data can significantly improve the accuracy of visual field loss predictions. This highlights the importance of the biomechanics-function relationship in glaucoma, and suggests that biomechanics may serve as a crucial indicator for the development and progression of glaucoma.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.