Abstract
We describe the hyperspectral Image Mapping Spectrometer (IMS) which is able to acquire spectral images of the retina with no scanning components. We then utilize this technology to obtain absolute oxygen saturation measurements in four patients with retinal diseases. The IMS acquires spatial and spectral information of a scene in a single camera exposure time. The IMS is attached to the port of a Topcon TRC-50EX funduscopic camera in order to acquire images. The IMS has high spatial (350×350) and spectral sampling (40), and avoids motion artifacts associated with eye movements. Oximetry values are obtained by measuring the absorption of retinal blood. Four patients with different retinal diseases were imaged via the IMS in this study. Oxygen saturation is calculated using a least squares fit applied to the whole blood oximetry equation. Oxygen saturation was found to be 88.6% (±1.9) and 34.1% (±7.0) for the artery and vein, respectively, in a patient with non-exudative age-related macular degeneration (AMD) with borderline glaucoma, 100.0% (±0.0) and 50.4% (±8.2) in a patient with exudative AMD, 100.0% (±0.0) and 45.2% (±7.0) in a patient with retinitis pigmentosa, and 94.1% (±2.5) and 44.9% (±5.6) in a patient with chronic iridocyclitis. These values are consistent with previously published data. Based on spatial and spectral data, we created an oxygen saturation map with oxygenation data overlying retinal vasculature. Hyperspectral imaging may be a promising way to measure retinal oxygen saturation. IMS technology is advantageous in its ability to obtain images free of motion artifacts.