Abstract
Abstract: :
Purpose: Imaging fundus autofluorescence and the mid to late–phase angiograms is difficult in patients with disturbances of the ocular media. To improve the signal–to–noise ratio a series of images can be averaged. In general, for this purpose a series of images is recorded and processed thereafter off–line. We report the first results using a newly developed method for on–line averaging of scanning laser images for the improvement of the signal–to–noise ratio in autofluorescence and angiographic images. Methods: All imaging was performed with a confocal scanning laser ophthalmoscope (HRA2, Heidelberg Engineering, Germany). The new software determines and compensates the eye movements between the live images in realtime by a correlation of automatically extracted landmarks from the different images. A moving average image with an increased signal–to–noise ratio (SNR) is displayed as live image on the screen and can be captured by the user at any time. A total of 100 patients with different retinal and choroidal diseases, mainly age–related macular degeneration and diabetic retinopathy were examined using the new realtime tracking software. In all patients, autofluorescence images and ICG and/or fluorescein angiograms were recorded. Results: The SNR in averaged images increased by a factor of 5.6. Especially for autofluorescence images and in the late phase of ICG angiograms, where there is little dynamic in the dye passage, online–imaging was useful to optimize the quality of the images and for getting higher contrast. The differences in contrast between recordings obtained by online–imaging and those obtained as single images is remarkably high in eyes with opaque optical media, e.g. due to cataract. The real–time tracking software was stable even in eyes with bad fixation. In patients with diabetic retinopathy, we were able to obtain online composite images. As compared with the standard method to store a series of images and compute average or composite images offline, the number of images to be stored is reduced significantly. Conclusions: The newly developed realtime tracking and averaging software allows recording of high quality images with a scanning laser ophthalmoscope even in eyes with opaque media. This new software enables us to record better quality autofluorescence and angiographic images without time consuming post–processing.
Keywords: imaging/image analysis: clinical • age-related macular degeneration • clinical (human) or epidemiologic studies: systems/equipment/techniques