Neurophotonics Laboratory


Current Research

Intrinsic Imaging of the Cerebral Cortex

We have developed an in vivo brain imaging strategy using only intrinsic contrast. Optical Coherence Microscopy (OCM) has the potential to image cortical myelination and neuronal cell bodies at depths of 1.3mm in the cortex. A volumetric, dynamically focused data set (5 micron steps), is used to extract coronal sections (optical slices), without actually cutting the tissue. A color overlay with neuronal cell bodies in green, and myelin in red, corresponded well to the different cortical layers.

OCM imaging of cell viability was performed during spreading depression and anoxic depolarization conditions. OCM shows cell swelling increased scattering during spreading depression while anoxic depolarization caused cells to swell and eventually lose contrast. Direct visualization of cellular changes using OCM could yield a pathway to provide novel optical markers of cell viability. Further details on this paper published in Optics Express can be found here.

Direct 3-D Quantification and Localization of Cell and Vascular Structure in Volumetric Images

Extending our previous work on imaging neurites, we are now using data from focused-tracked Optical Coherence Microscopy (OCM) as well as Optical Coherence Tomography (OCT) experiments to perform quantification of neuronal cells, myelin, and vascular architecture in the brain. Unlike conventional histology, cellular and vascular metrics are obtained directly from three-dimensional data, and do not require stereological extrapolation. These techniques have potential usefulness in the investigation of three-dimensional connectivity mapping in the brain.

Further details on this work can be found in our newly published paper. An image from this paper was chosen as the cover illustration (.gif format)

All-optical quantification of Absolute Flow

We have developed Doppler OCT methods for determining absolute blood flow across many individual vessels in the brain. This technique offers advantages over conventional imaging modalities like fMRI, laser Doppler, etc. In our method, by integrating velocity and area over the en face (xy) plane, absolute blood flow can be determined without explicitly calculating the vessel angle. This is due to the fact that in the en face  plane, the vessel cross-sectional area varies as 1/cos(θ) while the mean velocity axial projection varies as cos(θ). The figure shows a schematic of the method. Complete theoretical details and experimental validation can be found in our Optics Express paper here.

Flow determined by our methods is correlated with flow determined by standard hydrogen clearance methods. By showing conservation of volumetric flow along non-branching and branching vascular segments, the quantitative accuracy of our method was confirmed. This work, published in JCBFM, can be found here. Commentary on this paper is provided here.

OCT methods to determine capillary speed

We developed methods and models of capillary velocity determination based on analysis of the complex autocorrelation function in spectral / Fourier domain OCT. A critical advance was the correct incorporation of static scattering in the model, which otherwise confounds quantification. These advances resulted in the first images showing branch-specific, heterogeneous dynamics in individual brain capillaries. Data were correlated with velocities measured by two-photon microscopy. After further systematic validation, these methods will be used to elucidate capillary heterogeneity in the brain and the implications for oxygen delivery.

Image shows OCT Δf maps with an apparent redistribution of capillary velocity during hypercapnia compared to baseline. On average, the number of high-velocity capillaries is increased during hypercapnia. The paper, published in Biomedical Optics Express, can be found here.

Neurovascular Coupling in the Brain

Historically, it was considered that hemodynamic responses are better correlated with the primary input from the thalamus to layer IV in the cortical column. Recent evidence using non-invasive macroscopic optical imaging methods has suggested that it is more likely the cortico-cortical processing that drives the hemodynamic responses. In our lab, we have developed a method for multi-modal imaging of neuronal and vascular measurements using OCT and electrophysiology.

The image shows positioning of a fire-polished ball electrode to measure surface potentials, a needle electrode to measure neuronal activity in layer IV, and the thick white line shows the OCT scan location. The image next to it shows the cross section Doppler image of the underlying vasculature. Using electrical forepaw stimulation at varying frequency, we determined the SEPs and LFP responses using the 2 electrodes. The flow responses were measured at multiple depths. Our correlation results confirmed previous findings of hemodynamics being better correlated with N1 SEP component (that reflects neuronal activity in layers II/III). Our method also shows subtle differences in amplitude and latency depending on the anesthetic used (Collaboration with MA Franceschini).

We are currently developing advanced optical microscopy techniques for better understanding neurovascular coupling

Imaging of the Neurovascular Unit during Brain Injury

In order to image spatiotemporal evolutions of injury and recovery during acute and chronic stages of ischemic stroke in vivo, we developed and validated high resolution, Doppler OCT, OCT angiography, and OCT signal slope analysis techniques to map hemodynamics and cell viability using intrinsic scattering signatures.

A transient filament middle cerebral artery occlusion (fMCAO) mouse model was used to investigate hemodynamic and cellular scattering signatures. OCT angiograms (A-C), intensity images on a log scale (D-F), and concavity vs depth (G-I) at baseline, during fMCAO, and 60 minutes after reperfusion are shown in the figure. During occlusion, a capillary non-perfused region is apparent, and demarcated with a solid white line. The tissue with anomalous scattering properties (i.e., the log OCT signal is concave down vs. depth) after reperfusion corresponds well to the capillary non-perfused tissue during fMCAO. No comparable changes were observed in the contralateral hemisphere. More results can be found in our PLoS One publication hereThe success of our method could potentially be extended to identify biomarkers for tissue that is at risk but not yet destined for infarction, representing an important therapeutic target.

Angiography and Neurovascular Coupling in the Retina

The relationship between hemodynamics, metabolism, and neuronal activity is routinely studied in the brain. The retina, being an extension of the brain, provides the potential to be used as a as a model system to understand the brain. Extending our angiography techniques developed to visualize the brain to the retina, we show the different vascular layers of the rat eye and the subtle differences in the various vessel types. By using Doppler OCT methods, changes in blood flow rates during a multi-frequency flicker stimulus is  determined. The fundamental understanding of control of hemodynamics and metabolism in the retina may potentially yield promising candidates for non-invasive biomarkers in leading causes of blindness such as glaucoma and diabetic retinopathy.

The figure depicts (A-B) Visualization of the rat retinal vasculature; showing vasculature in the nerve fiber layer (1), inner plexiform layer (2), and outer plexiform layer (3) and the en face color image shows the OCT angiographic technique depicting subtle differences in vessel diameter. (C-D) show averaged en face Doppler OCT image of vessel at baseline and during activation elicited from a visual stimulus. (E) Fractional change time course over 10 trials at a stimulus frequency of 12 Hz. More details on retinal angiography and neurovascular coupling can be found in the recent publications in the Journal of Biomedical Optics.

Label-Free Optical Quantification of Inner Retinal Oxygen Metabolism

Non-invasive measurements of inner retinal oxygen metabolism have the potential to enable better detection of retinal diseases where early tissue hypoxia plays a role. Hemoglobin saturation or blood flow measurements, if considered alone, may be misleading, particularly when metabolism and hemodynamics are uncoupled. We are currently developing a technique using quantitative Doppler OCT flow method in conjunction with a simple conservation principle (Fick’s principle) to measure the inner retinal metabolic rate of oxygen (RMRO2) without the need for exogenous dyes or contrast agents. This is a biomarker that overcomes the mentioned limitations, and may yield more accurate tracking of progression of diseases such as diabetic retinopathy or glaucoma. We employ simultaneous imaging of the retina with visible light multispectral oximetry and infrared light Doppler Optical Coherence Tomography (OCT) to obtain values for saturation and flow, respectively, from the central arteries and veins and thus quantify RMRO2.

Funding for these studies: NIH (R00-NS067050, K99-NS067050, R01-EB001954), The American Heart Association (IRG5440002), and the Glaucoma Research Foundation Catalyst for a Cure 2.