Laser speckle‐based retinal imager as a potential screening tool for mild cognitive impairment

Background As the overall population's life expectancy rises, so does the prevalence of age‐related cognitive impairment (CI). Recent evidence has shown that cognitive impairment (CI) is not limited to the brain and affects the retina. In this study, we characterized dynamic retinal blood flow (RBF) patterns using the XyCAM RI (Vasoptic Medical, Inc.), a non‐invasive, laser speckle‐based retinal imager, in cognitively normal subjects (CN), patients with mild cognitive impairment (MCI), and individuals with diabetes mellitus (DM) including those with diabetic retinopathy (DR). Method A total of thirty‐four individuals (68 eyes) were imaged using the XyCAM RI. Normal cognitive function was determined using neuropsychological tests. High frame‐rate (82 Hz) RBF data was acquired over six seconds from each eye. As described and defined in Fig. 1, various dynamic RBF metrics were computed from the shape of the temporal waveform of blood flow velocity indices (BFVi) averaged within multiple arterial and venous segments intersecting a circular track around the optic disc. One‐way ANOVA was used to compare group means (p<0.05). Results Dynamic RBF metrics were computed using BFVi data from arterial and venous segments of 8 CN, 12 DM, 5DR, and 9 MCI eyes for comparison. Statistical comparisons (Table 1) confirmed significant differences between Dip BFVi in MCI vessels and CN/DR vessels but not DM vessels suggesting similar reduced RBF in both MCI and DM vessels. Compared with the Dip BFVi, no group‐wise differences were observed in Peak BFVi and Mean BFVi, suggesting the usefulness of assessing temporally‐resolved RBF data. A significantly higher arterial VRI (p < 0.002) and smaller venous VFI in MCI subjects demonstrated increased contractility compared to CN and DM subjects. The ANOVA test confirmed a statistically significant difference in the VFI of MCI subjects from each of the CN/DM/DR groups. Conclusion Individuals with MCI have different RBF characteristics compared to CN and DM subjects. These results may be related to systemic changes in arterial stiffness and increased blood pressure, which have been correlated with a decline in cognition. Non‐invasive dynamic RBF metrics using the XyCAM RI should be further investigated as potential biomarkers for classifying individuals with MCI.