Search

Two major approaches for tracking cellular motion across a range of biological tissues are the manual labelling of cells, and automated analysis of spatiotemporal information represented in a kymograph. Here we compare these two approaches for the measurement of retinal capillary flow, a particularly noisy application due to the low intrinsic contrast of single red blood cells (erythrocytes). Image data were obtained using a flood-illuminated adaptive optics ophthalmoscope at 750 nm, allowing the acquisition of flow information over several cardiac cycles which provided key information in evaluating tracking accuracy. Our results show that in addition to being much faster, the automated method is more accurate in the face of rapid flow and reduced image contrast. This study represents the first validation of commonly used kymograph approaches to capillary flow analysis.

The free diameter of a red blood cell exceeds the lumen diameter of capillaries in the central nervous system, requiring significant deformation of cells. However the deformations undertaken in vivo are not well established due to the difficulty in observing cellular capillary flow in living human tissue. Here, we used high resolution adaptive optics imaging to non-invasively track 17,842 red blood cells in transit through 121 unique capillary segments of diameter 8 µm or less in the retina of 3 healthy human subjects. Within each vessel, a 2D en face profile was generated for the "average cell", whose shape was then inferred in 3D based on the key assumption of a circular capillary cross-section. From this we estimated the average volume, surface area, orientation, and separation between red cells within each capillary tube. Our results showed a network filtration effect, whereby narrower vessels were more likely to contain smaller cells (defined by surface area, which is thought not to vary during a cell's passage through the vascular system). A bivariate linear model showed that for larger cells in narrower vessels: cells re-orient themselves to align with the flow axis, their shape becomes more elongated, there are longer gaps between successive cells, and remarkably, that cell volume is less which implies the ejection of water from cells to facilitate capillary transit. Taken together, these findings suggest that red cells pass through retinal capillaries with some reluctance. A biphasic distribution for cell orientation and separation was evident, indicating a "tipping point" for vessels narrower than approx. 5 µm. This corresponds closely to the typical capillary lumen diameter, and may maximize sensitivity of cellular flow to small changes in diameter. We suggest that the minimization of unnecessary oxygen exchange, and hence of damage via reactive oxygen pathways, may have provided evolutionary pressure to ensure that capillary lumens are generally narrower

Capillary flow is known to be non-homogenous between vessels and variable over time, for reasons that are poorly understood. The local properties of individual vessels have been shown to have limited explanatory power in this regard. This exploratory study investigates the association of network-level properties such as vessel depth, branch order, and distance from the feeding arteriole with capillary flow. Detailed network connectivity analysis was undertaken in 3 healthy young subjects using flood-illuminated adaptive optics retinal imaging, with axial depth of vessels determined via optical coherence tomography angiography. Forty-one out of 70 vessels studied were of terminal capillary type, i.e. fed from an arterial junction and drained by a venous junction. Approximately half of vessel junctions were amenable to fitting with a model of relative branch diameters, with only a few adhering to Murray's Law. A key parameter of the model (the junction exponent) was found to be inversely related to the average velocity (r = -0.59, p = 0.015) and trough velocity (r = -0.67, p = 0.004) in downstream vessels. Aspects of cellular flow, such as the minimum velocity, were also moderately correlated (r = 0.46, p = 0.009) with distance to the upstream feeding arteriole. Overall, this study shows that capillary network topology contributes significantly to the flow variability in retinal capillaries in human eyes. Understanding the heterogeneity in capillary flow is an important first step before pathological flow states can be properly understood. These results show that flow within capillary vessels is not affected by vessel depths but significantly influenced by the upstream feeder distance as well as the downstream vessel junction exponents, but there remains much to be uncovered regarding healthy capillary flow.

The optical density of visual pigment can be measured by imaging the dark-adapted eye while bleaching with visible light. This measurement can be made for individual photoreceptor cells using adaptive optics; however, activation of the phototransduction cascade imparts rapid changes in phase that modulate the signal via optical interference. This limits utility because data must be averaged over many experimental runs. Here we used a "flood" illuminated adaptive optics system at 4000 fps, bright light to achieve rapid bleaching, and broad illumination bandwidth to mitigate interference effects. Data were super-resolved using the natural motion of the eye to overcome the reduced pixel resolution of the ultrafast camera. This approach was applied to classify the trichromatic cone photoreceptor mosaic at a single fixation locus within the foveal region of 3 healthy subjects. Subjects were dark adapted for 6 minutes to replenish cone photopigment. This was followed either directly by imaging at 555 ± 50 nm, or by first pre-adapting the retina to 700 nm light to preferentially deplete "L" cone pigment. A total of 3,252 cones were classified as either "S", "M", or "L" type based on clustering of the intensity data observed under these two conditions. Mean classification probability ranged from 99.3 to 99.8%, with individual cell probabilities exceeding 95% in 97.0 to 99.2% of cones. Accuracy of cone classification peaked when using the first 10-30 ms of data, with significant reductions in accuracy noted with the inclusion of data from later times. Our results show that rapid bleaching and data acquisition significantly improve the robustness of cell-resolved densitometry.

The high power of the eye and optical components used to image it result in "static" distortion, remaining constant across acquired retinal images. In addition, raster-based systems sample points or lines of the image over time, suffering from "dynamic" distortion due to the constant motion of the eye. We recently described an algorithm which corrects for the latter problem but is entirely blind to the former. Here, we describe a new procedure termed "DIOS" (Dewarp Image by Oblique Shift) to remove static distortion of arbitrary type. Much like the dynamic correction method, it relies on locating the same tissue in multiple frames acquired as the eye moves through different gaze positions. Here, the resultant maps of pixel displacement are used to form a sparse system of simultaneous linear equations whose solution gives the common warp seen by all frames. We show that the method successfully handles torsional movement of the eye. We also show that the output of the previously described dynamic correction procedure may be used as input for this new procedure, recovering an image of the tissue that is, in principle, a faithful replica free of any type of distortion. The method could be extended beyond ocular imaging, to any kind of imaging system in which the image can move or be made to move across the detector.

The conventional stimulus for standard automated perimetry is fixed in size, giving elevated contrast thresholds and reduced test reliability in the periphery. Here, we test the hypothesis that appropriate scaling of the size of perimetric stimuli will return fixed thresholds and reduced variability across the visual field. We derived frequency-of-seeing (FOS) curves in five healthy subjects at central (3 degrees) and peripheral (27 degrees) locations with a method of constant stimuli (MOCS) using a desktop LCD display. FOS curves for a Goldmann III (GIII) stimulus were compared with those for size scaled spots. To consider clinical translation, we tested a further five healthy subjects (22-24 years) with the Melbourne Rapid Fields (MRF) tablet perimeter at several locations spanning 1 degree to 25 degrees from fixation, deriving FOS curves (MOCS) and also conducting repeated adaptive clinical thresholding to assess intra- and interobserver variability. We found that GIII contrast thresholds were significantly elevated in the periphery compared with the parafovea, with concomitant reduction of FOS slope. Using appropriately size scaled spots, threshold and slope differences between these locations were significantly reduced. FOS data collected with the tablet perimeter confirmed that size scaling confers broad equivalence of the shape of the FOS curve across the visual field. Repeated adaptive thresholding with size scaled stimuli gave relatively constant intra-observer variability across the visual field, which compares favorably with published normative data obtained with the GIII stimulus. The reduced variability will improve signal-to-noise ratio for correct classification of normal visual field test results, whereas the lower contrast thresholds yield greater dynamic range, which should improve the ability to reliably monitor moderate defects.

Conventional psychophysical methods ignore the degree of confidence associated with each response. We compared the psychometric function for detection with that for "absolute certainty" in a perimetry-style task, to explore how knowledge of response certainty might aid the estimation of detection thresholds. Five healthy subjects performed a temporal 2-AFC detection task, indicating on each trial whether they were "absolutely certain." The method of constant stimuli was used to characterize the shape of the two psychometric functions. Four eccentricities spanning central and peripheral vision were tested. Where possible, conditions approximated those of the Humphrey Field Analyzer (spot size, duration, background luminance, test locations). Based on the empirical data, adaptive runs (ZEST) were simulated to predict the likely improvement in efficiency obtained by collecting certainty information. Compared to detection, threshold for certainty was 0.5 to 1.0 dB worse, and slope was indistinguishable across all eccentricities tested. A simple two-stage model explained the threshold difference; under this model, psychometric functions for detection and for certainty-given-detection are the same. Exploiting this equivalence is predicted to reduce the number of trials required to achieve a given level of accuracy by approximately 30% to 40%. The chances of detecting a spot and the chances of certainty-given-detection were approximately the same in young, healthy subjects. This means, for example, that a spot detected at threshold was labeled as "certainly" detected approximately half the time. The collection of certainty information could be used to improve the efficiency of estimation of detection thresholds.

Capillary flow largely consists of alternating red cells and plasma whose speed oscillates predictably with the cardiac cycle. Superimposed on this regular background are sporadic events potentially disruptive to capillary exchange: the passage of white cells, aggregates of red cells, epochs of sparse haematocrit, or unusually slow flow. Such events are not readily differentiated with velocimetry or perfusion mapping. Here we propose a method to identify these phenomena in retinal capillaries imaged with high frame-rate adaptive optics, by calculating and representing pictorially the autocorrelation of intensity through time at each pixel during short epochs. The phenomena described above manifest as bright regions which transiently appear and propagate across an otherwise dark image. Drawing data from normal subjects and those with Type I diabetes, we demonstrate proof of concept and high sensitivity and specificity of this metric to variations in capillary contents and rate of flow in health and disease. The proposed metric offers a useful adjunct to velocimetry and perfusion mapping in the study of normal and abnormal capillary blood flow.

We present a new method for determining cellular velocity in the smallest retinal vascular networks as visualized with adaptive optics. The method operates by comparing the intensity profile of each movie pixel with that of every other pixel, after shifting in time by one frame. The time-shifted pixel which most resembles the reference pixel is deemed to be a 'source' or 'destination' of flow information for that pixel. Velocity in the transverse direction is then calculated by dividing the spatial displacement between the two pixels by the inter-frame period. We call this method pixel intensity cross-correlation, or "PIX". Here we compare measurements derived from PIX to two other state-of-the-art algorithms (particle image velocimetry and the spatiotemporal kymograph), as well as to manually tracked cell data. The examples chosen highlight the potential of the new algorithm to substantially improve spatial and temporal resolution, resilience to noise and aliasing, and assessment of network flow properties compared with existing methods.

PURPOSE: To determine the normal variation in orientation of the temporal nerve fiber raphe, and the accuracy with which it may be predicted or approximated in lieu of direct measurement. METHODS: We previously described an algorithm for automatic measurement of raphe orientation from optical coherence tomography, using the intensity of vertically oriented macular cubes. Here this method was applied in 49 healthy participants (age 19-81 years) and 51 participants with primary open angle glaucoma (age 51-80 years). RESULTS: Mean fovea-disc-raphe angle was 173.5° ± 3.2° (range = 166°-182°) and 174.2° ± 3.4° (range = 166°-184°) in healthy and glaucoma patients, respectively. Differences between groups were not significant. Fovea-disc-raphe angle was not correlated with age or axial length (P > 0.4), showed some symmetry between eyes in glaucoma (R2 = 0.31, P < 0.001), and little symmetry in the healthy group (P = 0.06). Fovea-disc angle was correlated with fovea-raphe angle (R2 = 0.27, P = 0.0001), but was not a good predictor for raphe orientation (average error = 6.8°). The horizontal axis was a better predictor (average error = 3.2°; maximum error = 9.6°), but still gave approximately twice the error previously reported for direct measurement from macular cubes. CONCLUSIONS: There is substantial natural variation in temporal nerve fiber raphe orientation, which cannot be predicted from age, axial length, relative geometry of the disc and fovea, or the contralateral eye. For applications to which the orientation of the raphe is considered important, it should be measured directly.

We evaluated several approaches for automatic location of the temporal nerve fiber raphe from standard macular cubes acquired on a Heidelberg Spectralis OCT. Macular cubes with B-scan separation of 96-122 µm were acquired from 15 healthy participants, and "high density" cubes with scan separation of 11 µm were acquired from the same eyes. These latter scans were assigned to experienced graders for subjective location of the raphe, providing the ground truth by which to compare methods operating on the lower density data. A variety of OCT scan parameters and image processing strategies were trialed. Vertically oriented scans, purposeful misalignment of the pupil to avoid reflective artifacts, and the use of intensity as opposed to thickness of the nerve fiber layer were all critical to minimize error. The best performing approach "cFan" involved projection of a fan of lines from each of several locations across the foveal pit; in each fan the line of least average intensity was identified. The centroid of the crossing points of these lines provided the raphe orientation with an average error of 1.5° (max = 4.1°) relative to the human graders. The disc-fovea-raphe angle was 172.4 ± 2.3° (range = 168.5-176.2°), which agrees well with other published estimates.

Conventional adaptive optics ophthalmoscopes use wavefront sensing methods to characterize ocular aberrations for real-time correction. However, there are important situations in which the wavefront sensing step is susceptible to difficulties that affect the accuracy of the correction. To circumvent these, wavefront sensorless adaptive optics (or non-wavefront sensing AO; NS-AO) imaging has recently been developed and has been applied to point-scanning based retinal imaging modalities. In this study we show, for the first time, contrast-based NS-AO ophthalmoscopy for full-frame in vivo imaging of human and animal eyes. We suggest a robust image quality metric that could be used for any imaging modality, and test its performance against other metrics using (physical) model eyes.

Retinal image quality from flood illumination adaptive optics (AO) ophthalmoscopes is adversely affected by out-of-focus light scatter due to the lack of confocality. This effect is more pronounced in small eyes, such as that of rodents, because the requisite high optical power confers a large dioptric thickness to the retina. A recently-developed structured illumination microscopy (SIM) technique called HiLo imaging has been shown to reduce the effect of out-of-focus light scatter in flood illumination microscopes and produce pseudo-confocal images with significantly improved image quality. In this work, we adopted the HiLo technique to a flood AO ophthalmoscope and performed AO imaging in both (physical) model and live rat eyes. The improvement in image quality from HiLo imaging is shown both qualitatively and quantitatively by using spatial spectral analysis.

Capture of light in the photoreceptor outer segment initiates a cascade of chemical events that inhibit neurotransmitter release, ultimately resulting in vision. The massed response of the photoreceptor population can be measured non-invasively by electrical recordings, but responses from individual cells cannot be measured without dissecting the retina. Here we used optical imaging to observe individual human cones in the living eye as they underwent bleaching of photopigment and associated phototransduction. The retina was simultaneously stimulated and observed with high intensity visible light at 1 kHz, using adaptive optics. There was marked variability between individual cones in both photosensitivity and pigment optical density, challenging the conventional assumption that photoreceptors act as identical subunits (coefficient of variation in rate of photoisomerization = 23%). There was also a pronounced inverse correlation between these two parameters (p<10(-7)); the temporal evolution of image statistics revealed this to be a dynamic relationship, with cone waveguiding efficiency beginning a dramatic increase within 3 ms of light onset. Beginning as early as 2 ms after light onset and including half of cells by ∼7 ms, cone intensity showed reversals characteristic of interference phenomena, with greater delays in reversal corresponding to cones with more photopigment (p<10(-3)). The timing of these changes is argued to best correspond with either the cessation of dark current, or to related events such as changes in intracellular cGMP. Cone intensity also showed fluctuations of high frequency (332±25 Hz) and low amplitude (3.0±0.85%). Other groups have shown similar fluctuations that were directly evoked by light; if this corresponds to the same phenomenon, we propose that the amplitude of fluctuation may be increased by the use of a bright flash followed by a brief pause, to allow recovery of cone circulating current.

Adaptive optics (AO) retinal image quality of rodent eyes is inferior to that of human eyes, despite the promise of greater numerical aperture. This paradox challenges several assumptions commonly made in AO imaging, assumptions which may be invalidated by the very high power and dioptric thickness of the rodent retina. We used optical modeling to compare the performance of rat and human eyes under conditions that tested the validity of these assumptions. Results showed that AO image quality in the human eye is robust to positioning errors of the AO corrector and to differences in imaging depth and wavelength compared to the wavefront beacon. In contrast, image quality in the rat eye declines sharply with each of these manipulations, especially when imaging off-axis. However, some latitude does exist to offset these manipulations against each other to produce good image quality.