Purpose:
preliminary testing of a newly developed portable, patient- and operator friendly multiwavelength compact scanning laser ophthalmoscope (cSLO), image registration and averaging, to detect exudates and hemorrhages.
Methods:
40 eyes of fifteen normal subjects and five patients with diabetes and known diabetic retinopathy were imaged with the cSLO. Patients with diabetic retinopathy were selected from the retina clinic at the University of Iowa , and had more than minimal background diabetic retinopathy, multiple exudates and multiple hemorraghes in the posterior pole in at least one eye. On the same day, digital stereo fundus photography was performed with a Zeiss 30° camera. The cSLO is a scanning laser ophthalmoscope (size: 70x80x120mm), containing near-infrared (785nm) and visible light (530nm) lasers, that collects 1000x1000 pixel images covering approximately 40° at 5Hz. Subjects' pupils were pharmacologically dilated (normally not required for cSLO alone). Image registration using mutual information and simple frame averaging was applied to increase the signal to noise ratio.
Results:
all subjects could be imaged; image registration and averaging improved subjective contrast and signal to noise ratio. Optic disc, small vessels and pigment epithelial texture could be imaged and depended on the confocality focus depth. At 785nm, comparison with Zeiss images showed that large exudates were visible, but small exudates were not, and hemorrhages were only seen in the focal plane (Figure, gray arrow hemorrhage, white arrow exudate). In the 530nm wavelength images, structures were adequately visualized.
Conclusions:
the results indicate that the cSLO is capable of imaging the posterior pole in normal subjects and patients with diabetes, and can detect lesions characteristic of more than minimal background retinopathy. 530nm visible laser light may be required to detect the subtle lesions typical in a diabetic retinopathy screening setting.
Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • diabetic retinopathy • image processing