Abstract
Purpose :
Previous studies have demonstrated that cognitive impairment (CI) is not limited to the brain but also affects the retina. The purpose of this study is to investigate the self-similarity and scaling of the retinal vascular network as potential peripheral biomarkers of CI.
Methods :
Prospective age-matched subjects (n=69) with and without CI and without the presence of any ophthalmic history were recruited (age>55+ years). Retinal images were obtained using the EasyScan (iOptics, The Netherlands). The Montreal Cognitive Assessment was used to measure CI. The multifractal behavior in the skeletonized optic-disc centered images was analyzed using the lacunarity parameter (Λ) and singularity spectrum f(α), both calculated with ImageJ program together with the FracLac plug-in. The Λ was obtained by measuring the gap dispersion inside each retinal image. The multifractal spectra were calculated to obtain the f (α) and its curve asymmetry. Independent sample t tests were used to compare the means of the Λ between the cognitively healthy and the cognitively impaired groups. A p value < 0.05 was considered statistically significant.
Results :
Of the 69 participants, 32 had CI (46%). We found altered microvascular network in individuals with CI. Plots of the spectrum f (α) revealed a parabola with concavity facing down, evidencing the multifractal structure of the retinal vascular network for both groups (Fig.1). The Λ values were not significantly different between the two groups (p=0.48, CI group: 0.35±0.05 versus controls: 0.34±0.03, age 81±6 years). However, the spectrum f(α) followed a trend towards lower alpha values and lower maxima with a narrower spectrum range in the CI group indicating lower singularity values (Fig.1): (α, f(α) [mean±SD]: 1.60±0.06; 1.57±0.06) versus controls (1.63±0.03; 1.61±0.03). Also, drusen-like regions in the peripheral retina along with pigment dispersion were noted in subjects with mild CI.
Conclusions :
Our results showed that the multifractal dimension, as a numeric measure associated with characteristics of the retinal vascular network alteration related to CI, can be integrated into analytical models to facilitate automatic diagnostic and prognostic. However, because of our study sample was small, the full extent of clinical applicability of our approach is provocative and still to be determined.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.