Abstract
Purpose :
The visualization of meibomian glands (MG) is an important factor for clinical management of dry eye disease. In this study, we developed a dedicated scan pattern and image processing method for visualizing MG, using a slit-scanning ophthalmoscope.
Methods :
Infrared (IR) images of the meibomian glands were captured using a slit-scanning ophthalmoscope CLARUS™ 500 (ZEISS, Dublin, CA) with prototype software. The patient’s eyelid had to be everted to capture the inner side of the lid. A modified scan pattern enabled alternation of the built-in IR light sources for reducing specular reflections, and a smaller scanning area aided in targeted acquisition of the MG.
For increasing the contrast between glands and background, direct imaging was combined with an offset-aperture technique (figure 1). Image processing algorithms were applied, including filters and contrast enhancement to improve the MG visibility.
Contrast to noise ratio (CNR)
CNR=│μG-μB│/(σG-σB)1/2
was measured from 4 glands (µG,σG) and adjacent background (µB,σB) areas with a size of 10x10-pixels, for each of seven images (example shown in Figure 2b) to determine the effect of the enhancement (a high value indicates that the contrast between gland and background is better than a lower value; figure 2 a).
Results :
The box plot in figure 2a shows the CNR value as a computed average from seven images before and after contrast enhancement. The median CNR value increases from 60.5 to 101.3.
Conclusions :
The offset-aperture technique was not useful in creating a usable image. However, a cropped image in combination with different contrast enhancement algorithm can increase MG visibility (figure 2c-e), potentially extending the capabilities of CLARUS 500 for use in dry eye management.
This is a 2020 ARVO Annual Meeting abstract.