One of the most important tools for diagnosing and monitoring the progression of eye disease is optical imaging. In particular, visualization of the retinal vasculature using visible and near-infrared light allows evaluating structural integrity, perfusion and leakage in both clinical and research settings. The recent development of light sources with high radiance and novel spectra, powerful computers, wavefront correctors, high-speed cameras and other optical elements combined with the transparency of the retina, have enabled the invention of numerous retinal vascular imaging modalities, generating image contrast by exploiting absorption, scattering, interference and the Doppler effect, among others.

Bedggood and Metha previously demonstrated in a previous paper that individual blood cells can be beautifully imaged and tracked in the living retina by using epi-illuminating bright field ophthalmoscopy enhanced with adaptive optics. Their landmark study not only showed in vivo imaging of blood cells with unprecedented spatial and temporal resolution, but more importantly, it provided remarkably high contrast non-invasively. In this work, they show how the contrast of plasma and blood cells in this imaging modality changes dramatically with relatively small changes in focus. The quantification by the authors of this reliable change and even reversal of contrast provides food for thought and calls for the re-evaluation of spectroscopic and intrinsic functional signal retinal studies. The former would be strongly affected by the expected wavelength-dependent changes in contrast due to the ocular longitudinal chromatic aberration, while the contradicting findings of the latter across instruments could be potentially explained by focusing differences. The authors’ explanation of the contrast changes in terms of refraction indicates that the irradiance transport equation and techniques that exploit it, such as curvature sensing, could be useful both for both quantifying and further improving the contrast of blood constituents and capillary walls.

Finally, the authors explore the use of temporal and spatial auto-correlation as well as the Radon transform for automated quantification of retinal blood flow at the capillary scale with very encouraging results that transcend the actual imaging modality.

---

---

 Add Comment

You must log in to add comments.