Search

EP-4312743-B1 - SYNTHETIC MULTI-EXPOSURE SPECKLE IMAGING (SYMESI) METHOD

EP4312743B1EP 4312743 B1EP4312743 B1EP 4312743B1EP-4312743-B1

Inventors

  • PARTHASARATHY, Ashwin Bharadwaj
  • SAFI, Abdul Mohaimen

Dates

Publication Date
20260506
Application Date
20220331

Claims (5)

  1. A synthetic Multi-Exposure Imaging method comprising: acquiring one or more raw speckle images (528) by illuminating a sample (120) with divergent light (104) from a laser source of light (514); collecting backscattered light (532) through an optical imaging system (520), and recording speckle intensity as raw speckle images on a camera sensor (524) at a fixed first empirical exposure time; performing data processing by a processor (530), wherein the data processing comprises: converting the raw speckle images to synthetic multi-exposure images (438(j), 448(j)) by spatially averaging a chosen one of the one or more camera sensor recorded images (528(j)) with a corresponding spatial window, thereby synthesizing multi-exposure speckle images, wherein the averaging comprises using multiple binning apertures that have different spatial dimensions to form respectively-corresponding modified synthetic speckle images, wherein each of said modified synthetic speckle images represents a speckle image corresponding to a respectively-corresponding second synthetic exposure time from a plurality of second synthetic exposure times, wherein the plurality of second synthetic exposure times corresponds to a multiplicity of the multiple binning apertures, and wherein each second synthetic exposure time from the plurality of second synthetic exposure times is different from one another and from the fixed first empirical exposure time; and quantifying blood flow information by computing temporal speckle contrast at each image pixel for all exposure times by computing K = σs/(I) at each pixel, where σs is the standard deviation and (I) is the mean of pixel intensities of n sequential modified synthetic speckle images.
  2. A method according to claim 1, further comprising: transforming each of the plurality of modified speckle images into a respectively-corresponding speckle contrast image of a plurality of speckle contrast images corresponding to the same one chosen image of the raw speckle images.
  3. A method according to one of claims 1 and 2, wherein: for each of said modified speckle images from the plurality of modified speckle images, a numerical relationship between the first empirical exposure time and the corresponding second synthetic exposure time depends on a spatial dimension of said pre-determined binning aperture
  4. A method according to one of claims 2 to 3, wherein each pixel of the given speckle contrast image has a value of speckle contrast determined as a ratio of a standard deviation of respectively-corresponding pixel intensities of said modified speckle images to a mean of said intensities.
  5. A method according to one of claims 1 to 4, wherein said averaging includes one of: spatially averaging an irradiance distribution of the chosen raw speckle image with the same binning aperture of the multiple binning apertures that is spatially repositioned across the chosen raw speckle image; and spatially averaging the irradiance distribution of the chosen raw speckle image with the multiple binning apertures of difference sizes and/or shapes, wherein respectively corresponding reference corners of said multiple binning apertures are fixed at the same location of said chosen raw speckle image.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This international patent application claims priority from and benefit of the US Provisional Patent Application No. 63/200,914 filed on April 02, 2021. RELATED ART Blood flow through tissue serves as an important physiological index of health of such tissue, because blood flow directly indicates oxygen delivery to the tissue and is critical for normal tissue functioning. Since even small changes in oxygen supply to the brain, for example, can have a dramatic impact on normal physiological processes, reduction in blood flow to the brain is not without serious consequences. Therefore, imaging of blood flow is important to understand the normal functioning of physiology, monitor disease progression, and to track treatment. Generally, superior spatiotemporal resolution characteristics of optical imaging methods make them more suitable for visualizing cerebral blood flow (CBF) dynamics than, for example, Magnetic Resonance Imaging (MRI), Computed Tomography (CT), or Diffuse Optical Tomography (DOT) - especially for applications that require resolution of individual cerebral blood vessels. Optical imaging methodologies rooted in photon correlation, for example (such as Laser Speckle Contrast Imaging, or LSCI) are particularly well suited for assessing the CBF with the use of intrinsically contrast motion of red blood cells. LSCI, in particular, has been shown useful for imaging cerebral blood flow in small animal models. One of primary advantages of LSCI is its ability to obtain wide-field CBF images with superior spatial and temporal resolution while the employed imaging apparatus is simple and inexpensive. To date, LSCI has been utilized to image CBF dynamics during ischemia in rat and mouse brains, in functional activation studies, and to model ischemic stroke progression. As is recognized in related art, laser speckle is the random interference pattern produced by coherent addition of light fields (for example, of laser light fields) that have backscattered from a sample along trajectories with slightly different path lengths. To this end, FIG. 1C schematically shows a typical LSCI setup. Here, slightly divergent light 104 from a visible or near infrared diode laser light source 110 is directed to impinge on the tissue sample 120; light 104 scatters in the tissue while the optical field components travelling along different paths through the tissue 120 experience different phase shifts. The backscattered light 124 is collected through an imaging lens 128 and recorded, in the form of a speckle, on the camera sensor 130 for data processing at the appropriately configured electronic circuitry 130. The movements of particles (such as, for example, red blood cells) in the sample, cells, impress spatial and temporal fluctuations onto the speckle pattern (see, for example, Ref. 3). Such effect manifests as local blurring (or decorrelation) in the image (FIG. 1A). Quantification of this blurring or decorrelation provides a measure of flow of red blood cells in the vasculature of the tissue 120. Formally, the local speckle contrast, K = σs/〈I〉, i.e., the square root of normalized variance of local intensity, is computed in a small window (usually 7×7 pixels) within the image; here, σs is the standard deviation of the intensities in the 7x7 pixel window, and (I) is the mean of intensities in the 7x7 pixel window. The processed speckle contrast image (FIG. 1B) is then produced by repeating this computation, window-by-window, across the raw speckle image. The higher value of speckle contrast in a given portion of the image of FIG. 1B is a manifestation of the reduced amount of spatial blurring in the corresponding region of the sample (i.e., in FIG. 1A), which is a direct consequence of slow-moving red blood cells. Notably, as recognized in related art, speckle contrast measured with the LSCI approach is appropriate only for acute measurements of relative changes of the CBF changes. In other words, the results of practical use of the conventional implementation of the LSCI system does not allow for quantitative measurement of movement at or in the target tissue, producing instead assessment of the relative (for example, on the scale of arbitrary units from 0 to 1) contrast values across the speckle image that manifests only in and affords only one resulting conclusion about the CBF changes, which can be expressed as follows: a movement of the target particles (say, blood flow) at the first region of the target tissue depicted in the first portion of the image is occurring quicker than a movement of the target particles at the second region of the target tissue depicted in the second portion of the image. The quantitative (that is, numerical) assessment of the speed values of these two movements are simply not possible based on the conventional LSCI, which provides relative - that is, comparative - results. As a result, the application of the conventional LSCI system has been limite