Abstract
Purpose :
To evaluate the potential of a new image processing technique in describing post-natural-blink tear film leveling (TFL) in subjects with healthy eyes (HE) and those with dry eye disease (DED) using Lateral Shearing Interferometry.
Methods :
The new method of interferograms analysis is based on the analysis of image background, rather than the regularity of interference fringes (IF) as tear film surface quality and it was conceived to simplify and speed up the analysis (using a low-pass filter and calculating the zero-order second moment of the Fourier transformed interferogram instead of the first-order moment of an unfiltered image). A sequence of recorded interferograms leads to a time series of TFL indicator that was modeled, in each inter-blink interval, by an exponential function f(t)=a×e−bt+c. This model was previously proposed by Szczesna-Iskander (Ocul Surf. 2018) for describing the regularity of IF in a group of 10 DED and 19 HE subjects. Of particular interest was the decay parameter b, describing the dynamics of TFL. For evaluating the new method, two data sets were considered. One using the retrospective data from the study mentioned above and one that includes recordings from 33 HE subjects aged 21—57 years (avg±SD 27±8) and 33 DED subjects aged 20—78 years (avg±SD 52±15). The Mann-Whitney test was used to test the differences between groups. DED was diagnosed according to DEWS II Report.
Results :
The median decay parameter b was statistically significantly higher for the HE group than for the DED group for both data sets, P=0.008 (previously P=0.026) and P<<0.001, respectively. Additionally, parameter b was significantly correlated with the tear film fluorescein break-up time (FBUT) (Spearman's R=0.404; P=0.033 and R=0.568; P<0.001 for the previous data set and the new data set, respectively.
Conclusions :
Analysis of the time-varying image background is a faster method for TFL assessment than IF estimate proposed before and it shows significantly faster TFL in the DED group than in the HE group. Different TFL dynamics between those groups supports previously stated importance of polar lipids in the mechanism of tear film lipid spreading that might be impaired in subjects with DED. Of particular mention is the moderate correlation between parameter b and FBUT, because often noninvasive evaluation of tear film does not correlate well with invasive methods of its assessment.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.