A new image-analysis algorithm was consistent with subjective grading of dry eye syndrome using lissamine green (LG) conjunctival staining, possibly presenting advantages of automation and scalability in clinical trials, according to a research team’s publication in Cornea.

The study was conducted because LG is frequently used with fluorescein for the determination of conjunctival damage in dry eye syndrome but is graded manually, and researchers wanted to describe an algorithm specifically for image analysis of LG conjunctival staining. They examined 20 pictures of dry eye syndrome patients who had visible LG conjunctival staining. Two digital slit lamps were used to take the pictures using a white light source and red filter transmitted over the wavelengths absorbed by LG. The conjunctival staining was black on a red background, and the red channel was extracted from the original image, the report explains. A Laplacian of Gaussian filter was used to detect stained areas while applying a threshold with a value found manually on a subset of images. Algorithm parameters stayed consistent throughout. 

Two experts also drew LG-stained areas manually as reference.


Continue Reading

Their findings: “The delineation obtained by the algorithm closely matched the actual contours of the punctate dots. In 19 cases of 20 (95%), the algorithm found the same Oxford grade as the experts, even for confluent staining that was detected as a multitude of dots by the algorithm but not by the experts, resulting in a high overestimation of the total number of dots (without mismatching the Oxford grade estimated by the experts).” 

And, results were found to be similar for the 2 slit-lamp imaging systems.

“Red-filtered images, acquired with 2 different standard slit-lamp imaging systems, coupled with the image analysis algorithm reported here, yield results consistent with subjective grading and may offer advantages of automation and scalability in clinical trials,” according to the researchers.

One limitation of the study was its ability only to detect small objects, which meant it didn’t detect large aggregates when dots became numerous in Oxford grades IV and V.

Reference

Courrier E, Renault D, Dib E, et al. New freeware for image analysis of lissamine green conjunctival staining. Cornea. 2021;40(3):351-7. doi:10.1097/ICO.0000000000002617