Individual Structure Function Map using Nerve Fiber Tracing

Content of research project: The general aim is to develop improved methods for early diagnosis of glaucoma. In glaucoma, the retinal nerve fiber bundles (RNFBs), which carry the electrical signals in response to a visual stimulus from the retinal photoreceptors to the brain, get damaged. Therefore, an analysis of the RNFBs can be used for early glaucoma diagnosis. One of the main obstacles for an accurate glaucoma diagnosis the large interindividual variation of the trajectories of the RNFBs within the retinal nerve fiber layer (RNFL). We will tackle this problem by using polarization sensitive optical coherence tomography (PS-OCT) to trace the RNFBs, and use the measured traces to construct an individualized structure-function model (iSFM) describing the topographic association of structural (RNFL thickness and polarizing properties) and functional (visual field) deficits in early glaucoma. Hypotheses: The main hypotheses are that PS-OCT can be used for RNFB tracing in healthy and glaucoma; this will allow for construction of a high-resolution iSFM; which will improve diagnostic accuracy. Additionally we will investigate the suitability of PS-OCT for functional PS-OCT. Methods: We will acquire PS-OCT data, using a prototype instrument, and obtain visual fields (VF) from 200 healthy volunteers and 100 glaucoma patients. Axis information contained in the PS-OCT data, together with local thickness variations will be used to trace the RNFBs. An individualized theoretical model of the RNFB trajectories will be developed describing the association of the RNFB trajectories with retinal and optic nerve head (ONH) biomarkers. This model will then be used as a prior to improve RNFB tracing in areas with thin RNFL. We will finally create an accurate iSFM with high topographic resolution and test its diagnostic accuracy in comparison with the standard parameters of VF and OCT. Explanation indicating what is new and/or special about the project: We will be the first to develop an iSFM with high topographic resolution to enhance glaucoma diagnosis. We build on a pilot study from our group, which demonstrated the feasibility of RNFB tracing with PS-OCT in a few selected healthy volunteers. This project will overcome the difficulties of PS-OCT based RNFB tracing in diseased retinas, by integrating actual measurements with an individualized theoretical model of RNFB trajectories. The resulting high-resolution iSFM will improve the correlation between structural and functional defects and thus support early glaucoma diagnosis. We will be the first to explore functional PS-OCT measurements of the RNFL using the knowledge of the RNFB traces, which allows recording the PS-OCT signals exactly along the measured RNFBs. We will thus examine changes in light polarization by RNFBs as a function of visual stimuli along RNFBs. This might in future allow for objective VF testing.

This research project deals with methods designed to improve the early diagnosis of glaucoma. Glaucoma is a chronic progressive disease of the eye's optic nerve, characterized by the loss of optic nerve fibers and leading to visual disability in about 25% of affected patients. The loss of nerve fibers can be measured indirectly by measuring the thickness of the retinal nerve fiber layer (RNFL) with optical coherence tomography (OCT), while the loss of vision (in the context of glaucoma) is tested by subjective visual field testing. One of the major obstacles to achieve an early and yet accurate diagnosis of glaucoma is the large variation among healthy individuals concerning the appearance of the optic nerve head (ONH) as well as the thickness of the RNFL and its distribution around the ONH. The other major obstacle is the inherent preponderance of subjective visual fields testing to produce both false positive and false negative errors. We have developed an automated retinal nerve fiber bundle (RNFB) tracing method based on polarization-sensitive optical coherence tomography (PS-OCT). This results in individual knowledge of the trajectories of the RNFBs (the way they take from the retinal periphery towards the ONH) which should enable a much more precise joining of the visual field and RNFL data and help with differentiating physiologic variations and errors in visual field testing from true pathology. In this project we have successfully measured 128 healthy eyes and 51 eyes suffering from glaucoma. We achieved in this project to develop a reproducible and automated method of RNFB tracing using PS-OCT data. We could show that this automated tracing outperforms manual tracing in reproducibility. Additionally only an automated method may allow for its routine use, since manual tracing is far too time consuming with 30 to 60 minutes per eye to be routinely used. As an additional outcome we have identified a new parameter, which we assume is linked to transport processes within the RNFL. We could measure that the microstructure of the retinal OCT images is changing much faster in the RNFL as compared to other retinal layers. This is expressed as a much faster speed of signal decorrelation in time series of retinal OCT within the RNFL as compared to the deeper retinal layers. This new parameter may in future serve as a biomarker of the health state of the RNFL, which means it might be used to monitor disease progression and potentially also treatment effects.