Abstract
Abstract: :
Purpose: To evaluate a newly developed Retinal Functional Imager (RFI) designed for non-invasive imaging of functional properties of the retina. Functions examined included (a) retinal blood flow and (b) retinal oximetric state. Methods: Were examined both healthy subjects, diabetic patients and patients with vascular occlusion. (a) Blood flow: Digital images of the retina were recorded using the RFI including a computer controlled stroboscopic flash lamp illumination system and a computerized digital camera attached to a standard fundus camera. The flash lamp system can provide trains of high intensity flashes at up to 50Hz. These images were then re-registered to eliminate eye movements and processed to produce a high-resolution movie of retinal blood flow. Quantitative analysis produced either color-coded maps of flow rate, or numeric flow rates for selected vessels. (b) Oximetric state: Multiple images of the retina were captured at high frequency rate of 20Hz using multiple wavelengths sensitive to the differences in deoxy- and oxy- hemoglobin absorption. These data were processed to reveal the oximetric state in retinal vessels and capillaries. The area of the oximetric map was compared to the fluorescein angiographs (FA). Results: (a) Blood flow: The very high-resolution movies of retinal blood flow showed individual red blood cells or clumps of cells flowing in capillaries and small vessels, in both the macula and optic disc regions. Preliminary results showed that in diabetic patients the flow was slower then in healthy subjects. (b) Oximetric state: Oximetric maps disclosed areas of low oximetric state that correlated to areas of nonperfusion on FA. Conclusion: The high-resolution imaging and quantification of retinal blood flow makes the RFI a powerful tool that may contribute to better understanding and early diagnosis of retinal diseases. The RFI may offers considerable advantages in assessing the oximetric state in several retinal diseases. The non-invasive nature of the recording offers significant advantages.
Keywords: 430 imaging/image analysis: clinical • 388 diabetic retinopathy • 429 image processing