An Adaptive Enhanced Laser Speckle Contrast Blood Flow Imaging Method

Abstract

This invention discloses an adaptively enhanced laser speckle contrast blood flow imaging method. It utilizes a laser speckle contrast imaging system to probe the surface of the tissue under test and acquires a sequence of original speckle images using a CMOS camera. An initial speckle contrast image is calculated based on the original speckle image sequence. The initial speckle contrast image is then segmented to obtain a segmented image. An adaptively enhanced speckle contrast image is calculated based on the original speckle image sequence and the segmented image. The laser speckle contrast blood flow imaging method provided by this invention can improve the imaging effect of blood vessel edges and microvessels, making the contrast image of blood vessels clearer and reflecting more details. In imaging deeper blood vessels, it can improve the visualization effect of blood vessels, with higher noise attenuation and better imaging quality.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to an adaptive enhanced laser speckle contrast blood flow imaging method. Background Technology

[0002] Laser speckle contrast imaging (LSI) is a real-time, non-scanning, full-field hemodynamic imaging technique with advantages such as non-contact, non-invasiveness, no contrast agent required, high spatiotemporal resolution, and simple instrument structure. It is widely used in basic research and clinical applications in ophthalmology, neuroscience, microcirculation, and intraoperative monitoring. LSI can reflect the morphology and flow velocity of different blood vessels at the same time, as well as the changes in blood flow velocity in the same blood vessel at different times, enabling long-term, dynamic monitoring of hemodynamics. Cardiovascular and cerebrovascular diseases, hypertension, and other conditions are closely related to microcirculatory status, and blood flow is one of the key parameters for measuring microcirculatory status. Real-time in vivo measurement of blood flow is of great significance for disease prevention.

[0003] Traditional methods for reconstructing speckle contrast images calculate the standard deviation and mean of light intensity for all pixels within a spatial, temporal, or spatiotemporal window to determine the speckle contrast value at the window's center point, thereby reconstructing a two-dimensional speckle contrast image. However, in deeper vascular imaging, images reconstructed using traditional speckle contrast imaging methods contain high noise levels, resulting in a low contrast-to-noise ratio between dynamic vascular regions and static tissue regions. Furthermore, the imaging quality for vascular edges and microvessels is poor, sometimes even failing to capture them, which is detrimental to monitoring blood flow in deeper microcirculation. Summary of the Invention

[0004] The present invention aims to overcome the shortcomings of the prior art and provide an adaptive enhanced laser speckle contrast blood flow imaging method.

[0005] The technical solution adopted in this invention is: an adaptive enhanced laser speckle contrast blood flow imaging method, the method comprising the following steps:

[0006] S100 uses a laser speckle contrast imaging system to measure the biological tissue under test and obtain the original speckle image sequence.

[0007] S200, calculate the initial speckle contrast image based on the original speckle image sequence;

[0008] S300, the segmented image is calculated based on the initial speckle contrast image;

[0009] S400, based on the original speckle image sequence and the segmented image, an adaptively enhanced speckle contrast image is calculated.

[0010] Furthermore, in S100, the method for obtaining the original speckle image sequence by measuring the biological tissue under test using a laser speckle contrast imaging system is as follows:

[0011] S101, the exposure time of the CMOS camera of the laser speckle contrast imaging system is set to Tms, that is, the number of images acquired per second. frame;

[0012] S102, within the acquisition time t seconds, the laser source of the laser speckle contrast imaging system is turned on and outputs, and after passing through the beam expander and reflector, it illuminates the surface of the biological tissue to be measured.

[0013] S103, the CMOS camera of the laser speckle contrast imaging system acquires the diffuse reflection and scattered light beams on the surface of the biological tissue under test and within the tissue to form the original speckle image.

[0014] S104, setting the total number of records of the CMOS camera within the acquisition time t seconds. The original speckle image described in the frame is preprocessed by a computer to obtain the original speckle image sequence.

[0015] Furthermore, in S200, the method for calculating the initial speckle contrast image based on the original speckle image sequence is as follows:

[0016] S201, the original speckle image sequence is set to I(x, y, z), where I(x, y, z) represents the grayscale value of the original speckle image at pixel coordinates (x, y) in the z-th frame, and the image resolution is set to [M, N], the value range of x is [1, M], the value range of y is [1, N], and the value range of z is [...].

[0017] S202, select a sliding space window of size V×V, and let the sliding space window traverse the pixel coordinates (x, y) within the range of x∈[1, M], y∈[1, N]; based on the original speckle image sequence I(x, y, z), select the corresponding original speckle image subframe data I under the sliding space window. V (x, y, z), where x takes values ​​in the range [1, V], y takes values ​​in the range [1, V], and z takes values ​​in the range [V]. Calculate the spatial speckle contrast image K z (x, y, z):

[0018]

[0019] in

[0020] S203, based on the spatial speckle contrast image K zGiven (x, y, z), calculate the gradient KN(x, y) in each of the four directions at pixel (x, y) in the z-th frame. z , KS(x, y) z , KW(x, y) z , KE(x, y) z The diffusion coefficients dN(x, y) in the four directions are calculated based on the gradients in the four directions. z ds(x, y) z ,dW(x,y) z ,dE(x,y) z The initial speckle contrast image K(x, y, z) was calculated. (n+1) :

[0021] K(x, y, z) (n+1) =K(x, y, z) n +λ×(N+S+WE+E)

[0022] The above equation is used to obtain the initial speckle contrast image K(x, y, z). (n+1) The iterative formula is given by n, where n is the number of iterations and λ is the diffusion coefficient.

[0023] in:

[0024] N = dN(x, y) z ×KN(x, y) z ,

[0025] KN(x, y) z =K z (x, y-1, z)-K z (x, y, z);

[0026] S = dS(x, y) z ×KS(x, y) z ,

[0027] KS(x, y) z =K z (x, y+1, z)-K z (x, y, z);

[0028] WE = dW(x, y) z ×KW(x, y) z ,

[0029] KW(x, y) z =K z (x-1, y, z)-K z (x, y, z);

[0030] E = dE(x, y) z ×KE(x, y) z ),

[0031] KE(x, y) z =K z (x+1, y, z)-K(x, y, z);

[0032] Where Kthr is the edge parameter.

[0033] Furthermore, in S300, the method for calculating the segmented image based on the initial speckle contrast image is as follows:

[0034] S301, based on the initial speckle contrast image K(x, y, z) (n+1) The dynamic and static regions are segmented using a segmentation algorithm to obtain the segmented image Se(x, y):

[0035]

[0036] Se(x,y)=Se(x1,y1)+Se(x2,y2)+Se(x3,y3)

[0037] Where Se(x1, y1) represents the dynamic region, Se(x2, y2) represents the static region, and Se(x3, y3) represents the transition region.

[0038] Furthermore, in S400, the method for calculating the adaptively enhanced speckle contrast image based on the original speckle image sequence and the segmented image is as follows:

[0039] S401, set the size of the selected sliding space window to W×W, and according to the original speckle image sequence I(x, y, z) and the segmented image Se(x, y), the sliding space window traverses the range of values ​​of pixel coordinates (x, y): x∈[1, M], y∈[1, N];

[0040] S402, Select the corresponding original speckle image subframe data I under the sliding window. W (x, y, z) and segmented image subframe data Se W (x, y), where x ranges from [1, W], y ranges from [1, W], and z ranges from [x, W].

[0041] S403, Calculate Se W The number of pixels with a value of 0 in the coordinate (x, y) is C0, the number of pixels with a value of 1 is C1, and the number of pixels with a value of 0.5 is C1. 0.5 If C0 ≠ W 2 Or C1≠W2 Or C 0.5 ≠W 2 This indicates that the original speckle image subframe data I corresponding to the sliding window is... W (x, y, z) is located in the boundary region. The contrast value K at the center point of the window corresponding to (x, y, z) in the original speckle image is calculated using an adaptive window size. a (x, y, z), that is, if Se W If the value of the center point of (x, y) is 1, then use Se. W The pixel I corresponding to 1 in (x, y) W1 The value of (x, y, z) is used to calculate the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z). a1 (x, y, z); similarly, if Se W If the value of the center point of (x, y) is 0, then use Se. W The pixel I corresponding to 0 in (x, y) W0 The value of (x, y, z) is used to calculate the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z). a0 (x, y, z); if Se W If the value of the center point of (x, y) is 0.5, then use Se. W The pixel I corresponding to 0.5 in (x, y) W0.5 The value of (x, y, z) is used to calculate the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z). a0.5 (x, y, z);

[0042] like Then K a (x, y, z) = K a1 (x, y, z)

[0043]

[0044]

[0045] In equations (1) and (2), i and j must satisfy Se W (i, j) = 1;

[0046] like Then K a (x, y, z) = K a0 (x, y, z)

[0047]

[0048]

[0049] In equations (3) and (4), i and j must satisfy Se W (i, j) = 0;

[0050] like Then K a (x, y, z) = K a0.5 (x, y, z)

[0051]

[0052]

[0053] In equations (5)(6), i and j must satisfy Se W (i, j) = 0.5;

[0054] in Indicates to Rounding to the nearest integer, z represents the z-th frame;

[0055] S404, if C0 = W 2 Or C1 = W 2 Or C 0.5 =W 2 This indicates that the original speckle image subframe data I corresponding to the sliding window is... W (x, y, z) is not in the boundary region. The contrast value K at the center point of the window corresponding to (x, y, z) in the original speckle image is calculated using the variance criterion. a (x, y, z);

[0056] In step S404, the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z) is calculated using the variance criterion. a The method for (x, y, z) is:

[0057] S4041, Calculate I W The window variance d of (x, y, z) all and the variance d in four directions (0°, 45°, 90°, 135°). 0° d 45° d 90° d 135° , constitute d n (z):

[0058] d n (z)={d 0° d 45° d 90° d 135° d all}

[0059]

[0060]

[0061] I j (z) represents the pixel light intensity value in the direction corresponding to the z-th frame. This represents the average pixel light intensity in the direction corresponding to the z-th frame. This represents the average pixel light intensity of the entire window;

[0062] S4042, if C1 = W 2 This indicates that the region is a dynamic region, and K is calculated using the pixel value along the direction of minimum variance. a (x, y, z):

[0063] d dynamic (z)=arg d min[var(d n (z))]

[0064]

[0065] d dynamic (z) represents the direction of the dynamic region, that is, the direction of minimum variance, σ(d dynamic (z) represents the standard deviation of the pixel values ​​along the direction of minimum variance. This represents the mean value of the pixels along the direction of minimum variance.

[0066] S4043, if C0 = W 2 This indicates that the region is a static region, and k is calculated using the pixel value along the direction of maximum variance. a (x, y, z):

[0067] d static (z)=arg d max[var(d n (z))]

[0068]

[0069] d static (z) represents the direction of the static region, that is, the direction of maximum variance, σ(d static (z) represents the standard deviation of the pixel values ​​along the direction of maximum variance. This represents the mean value of the pixels along the direction of maximum variance.

[0070] S4044, if C 0.5 =W 2 This indicates that the region belongs to a transition area, and K is calculated using the pixel values ​​along the median variance direction. a (x, y, z):

[0071] d transition (z)=arg d median[var(d n (z))]

[0072]

[0073] d transition (z) represents the direction of the transition region, that is, the direction of the median variance, σ(d transition (z) represents the standard deviation of the pixel values ​​along the median direction of the variance. This represents the mean value of the pixels along the variance median direction;

[0074] S405, the sliding window iterates through the range of pixel coordinates (x, y): x∈[1, M], y∈[1, N]. That is, looping through S402, S403, S404, S4041, S4042, S4043, and S4044 yields a new speckle contrast image K. z (x, y, z).

[0075] As described above, the adaptive enhanced laser speckle contrast blood flow imaging method of the present invention has the following beneficial effects:

[0076] (1) Based on the original speckle image sequence and segmented image, by adaptive window size processing of dynamic and static boundary regions, the imaging effect of blood vessel edges and microvessels can be improved, making the contrast image of blood vessels clearer and reflecting more details.

[0077] (2) Based on the original speckle image sequence and segmented image, dynamic and static regions are segmented. By using different variance criteria to image the dynamic and static regions, the contrast between the dynamic and static regions can be improved. In deeper vascular imaging, the visualization effect of blood vessels can be improved, with higher noise attenuation and better imaging quality. Attached Figure Description

[0078] The above and other features of this disclosure will become more apparent from the detailed description of the embodiments shown in conjunction with the accompanying drawings. In the accompanying drawings, the same reference numerals denote the same or similar elements. Obviously, the drawings described below are merely some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained from these drawings without any inventive effort. In the drawings:

[0079] Figure 1 This is a flowchart of an embodiment of an improved laser speckle contrast blood flow imaging method of the present invention;

[0080] Figure 2This is a schematic diagram of the hardware structure of a laser speckle contrast imaging system in one embodiment of an improved laser speckle contrast blood flow imaging method of the present invention.

[0081] Figure 3 This is a comparison diagram of an improved laser speckle contrast blood flow imaging method of the present invention, based on a disclosed original speckle image, using a spatial speckle contrast imaging method and using the imaging method of the present invention in one embodiment;

[0082] Figure 4 This is a comparison diagram of an improved laser speckle contrast blood flow imaging method of the present invention, based on an embodiment of the original speckle image acquired through a phantom experiment, using a spatial speckle contrast imaging method and the imaging method of the present invention. Detailed Implementation

[0083] The following will provide a clear and complete description of the concept, specific structure, and technical effects of this disclosure in conjunction with the embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of this disclosure. It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other. The illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention and are intended to explain this application, and should not be construed as limiting this application.

[0084] The adaptive enhanced laser speckle contrast blood flow imaging method described in this invention improves the imaging effect of blood vessel edges and microvessels by adaptively processing the window size of dynamic and static boundary regions; it improves the contrast between dynamic and static regions by using different variance criteria for imaging; and it has higher noise attenuation and better imaging quality compared with traditional laser speckle contrast imaging methods in imaging deeper blood vessels.

[0085] like Figure 1 The diagram shown is a flowchart of an adaptive enhanced laser speckle contrast blood flow imaging method according to the present invention. The following is a description of the method in conjunction with... Figure 1 This paper describes an adaptive enhanced laser speckle contrast blood flow imaging method according to an embodiment of the present invention.

[0086] This disclosure proposes an adaptive enhanced laser speckle contrast blood flow imaging method, which specifically includes the following steps:

[0087] S100 uses a laser speckle contrast imaging system to measure the biological tissue under test and obtain the original speckle image sequence.

[0088] S200, calculate the initial speckle contrast image based on the original speckle image sequence;

[0089] S300, the segmented image is calculated based on the initial speckle contrast image;

[0090] S400, based on the original speckle image sequence and the segmented image, an adaptively enhanced speckle contrast image is calculated.

[0091] Furthermore, in S100, the method for obtaining the original speckle image sequence by measuring the biological tissue under test using a laser speckle contrast imaging system is as follows:

[0092] S101, the exposure time of the CMOS camera of the laser speckle contrast imaging system is set to Tms, that is, the number of images acquired per second. frame;

[0093] Preferably, the value of T ranges from 5ms to 10ms;

[0094] S102, within the acquisition time t seconds, the laser source of the laser speckle contrast imaging system is turned on and outputs, and after passing through the beam expander and reflector, it illuminates the surface of the biological tissue to be measured.

[0095] S103, the CMOS camera of the laser speckle contrast imaging system acquires the diffuse reflection and scattered light beams on the surface of the biological tissue under test and within the tissue to form the original speckle image.

[0096] S104, setting the total number of records of the CMOS camera within the acquisition time t seconds. The original speckle image described in the frame is preprocessed by a computer to obtain the original speckle image sequence.

[0097] Optionally, the hardware structure of the laser speckle contrast imaging system in this specific embodiment is configured as follows: Figure 2 As shown, the hardware components of the visible imaging system include a light source module, a lens optical path module, and an imaging acquisition module.

[0098] The light source module includes a laser driver and a laser, preferably with a center wavelength of 785nm;

[0099] The lens optical path module includes a beam expander and a reflector; after the laser beam is expanded by the beam expander, it is then reflected by the reflector to irradiate the biological tissue to be tested.

[0100] The imaging acquisition module includes a camera and a processor. The camera transmits the acquired signals to the processor, which records the original speckle image sequence obtained within a certain acquisition time.

[0101] Furthermore, in S200, the method for calculating the initial speckle contrast image based on the original speckle image sequence is as follows:

[0102] S201, the original speckle image sequence is set to I(x, y, z), where I(x, y, z) represents the grayscale value of the original speckle image at pixel coordinates (x, y) in the z-th frame, and the image resolution is set to [M, N], the value range of x is [1, M], the value range of y is [1, N], and the value range of z is [...].

[0103] S202, select a sliding space window of size V×V, and let the sliding space window traverse the pixel coordinates (x, y) within the range of x∈[1, M], y∈[1, N]; based on the original speckle image sequence I(x, y, z), select the corresponding original speckle image subframe data I under the sliding space window. V (x, y, z), where x takes values ​​in the range [1, V], y takes values ​​in the range [1, V], and z takes values ​​in the range [V]. Calculate the spatial speckle contrast image K z (x, y, z):

[0104]

[0105] in

[0106] Optionally, in this embodiment, V = 5 is selected;

[0107] S203, based on the spatial speckle contrast image K z Given (x, y, z), calculate the gradient KN(x, y) in each of the four directions at pixel (x, y) in the z-th frame. z , KS(x, y) z , KW(x, y) z , KE(x, y) z The diffusion coefficients dN(x, y) in the four directions are calculated based on the gradients in the four directions. z ,dS(x,y) z ,dW(x,y) z ,dE(x,y) z The initial speckle contrast image K(x, y, z) was calculated. (n+1) :

[0108] K(x, y, z) (n+1) =K(x, y, z) n +λ×(N+S+WE+E)

[0109] The above equation is used to obtain the initial speckle contrast image K(x, y, z). (n+1) The iterative formula is given by n, where n is the number of iterations and λ is the diffusion coefficient.

[0110] Optionally, in this embodiment, n = 50 and λ = 0.25 are selected;

[0111] in:

[0112] N = dN(x, y) z ×KN(x, y) z ,

[0113] KN(x, y) z =K z (x, y-1, z)-K z (x, y, z);

[0114] S = dS(x, y) z ×KS(x, y) z ,

[0115] KS(x, y) z =K z (x, y+1, z)-K z (x, y, z);

[0116] WE = dW(x, y) z ×KW(x, y) z ,

[0117] KW(x, y) z =K z (x-1, y, z)-K z (x, y, z);

[0118] E = dE(x, y) z ×KE(x, y) z ),

[0119] KE(x, y) z =K z (x+1, y, z)-K z (x, y, z);

[0120] Where Kthr is the edge parameter.

[0121] Optionally, in this embodiment, select This represents the mean value of the spatial speckle contrast image.

[0122] Furthermore, in S300, the method for calculating the segmented image based on the initial speckle contrast image is as follows:

[0123] S301, based on the initial speckle contrast image K(x, y, z) (n+1)The dynamic and static regions are segmented using a segmentation algorithm to obtain the segmented image Se(x, y):

[0124]

[0125] Se(x,y)=Se(x1,y1)+Se(x2,y2)+Se(x3,y3)

[0126] Where Se(x1, y1) represents the dynamic region, Se(x2, y2) represents the static region, and Se(x3, y3) represents the transition region.

[0127] Optionally, k-means clustering is selected for segmentation in this embodiment.

[0128] Furthermore, in S400, the method for calculating the adaptively enhanced speckle contrast image based on the original speckle image sequence and the segmented image is as follows:

[0129] S401, set the size of the selected sliding space window to W×W, and according to the original speckle image sequence I(x, y, z) and the segmented image Se(x, y), the sliding space window traverses the range of values ​​of pixel coordinates (x, y): x∈[1, M], y∈[1, N];

[0130] Optionally, in this embodiment, W = 5 is selected;

[0131] S402, Select the corresponding original speckle image subframe data I under the sliding window. W (x, y, z) and segmented image subframe data Se W (x, y), where x ranges from [1, W], y ranges from [1, W], and z ranges from [x, W].

[0132] S403, Calculate Se W The number of pixels with a value of 0 in the coordinate (x, y) is C0, the number of pixels with a value of 1 is C1, and the number of pixels with a value of 0.5 is C1. 0.5 If C0 ≠ W 2 Or C1≠W 2 Or C 0.5 ≠W 2 This indicates that the original speckle image subframe data I corresponding to the sliding window is... W (x, y, z) is located in the boundary region. The contrast value K at the center point of the window corresponding to (x, y, z) in the original speckle image is calculated using an adaptive window size. a (x, y, z), that is, if Se W If the value of the center point of (x, y) is 1, then use Se. WThe pixel I corresponding to 1 in (x, y) W1 The value of (x, y, z) is used to calculate the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z). a1 (x, y, z); similarly, if Se W If the value of the center point of (x, y) is 0, then use Se. W The pixel I corresponding to 0 in (x, y) W0 The value of (x, y, z) is used to calculate the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z). a0 (x, y, z); if Se W If the value of the center point of (x, y) is 0.5, then use Se. W The pixel I corresponding to 0.5 in (x, y) W0.5 The value of (x, y, z) is used to calculate the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z). a0.5 (x, y, z);

[0133] like Then K a (x, y, z) = K a1 (x, y, z)

[0134]

[0135]

[0136] In equations (1) and (2), i and j must satisfy Se W (i, j) = 1;

[0137] like Then K a (x, y, z) = K a0 (x, y, z)

[0138]

[0139]

[0140] In equations (3) and (4), i and j must satisfy Se W (i, j) = 0;

[0141] like Then K a (x, y, z) = K a0.5 (x, y, z)

[0142]

[0143]

[0144] In equations (5)(6), i and j must satisfy Se W (i, j) = 0.5;

[0145] in Indicates to Rounding to the nearest integer, z represents the z-th frame;

[0146] S404, if C0 = W 2 Or C1 = W 2 Or C 0.5 =W 2 This indicates that the original speckle image subframe data I corresponding to the sliding window is... W (x, y, z) is not in the boundary region. The contrast value K at the center point of the window corresponding to (x, y, z) in the original speckle image is calculated using the variance criterion. a (x, y, z);

[0147] In step S404, the contrast value K at the center point of the window corresponding to the original speckle image (x, y, z) is calculated using the variance criterion. a The method for (x, y, z) is:

[0148] S4041, Calculate I W The window variance d of (x, y, z) all and the variance d in four directions (0°, 45°, 90°, 135°). 0° d 45° d 90° d 135° , constitute d n (z):

[0149] d n (z)=({ 0° d 45° d 90° d 135° d all}

[0150]

[0151]

[0152] I j (z) represents the pixel light intensity value in the direction corresponding to the z-th frame. This represents the average pixel light intensity in the direction corresponding to the z-th frame. This represents the average pixel light intensity of the entire window;

[0153] S4042, if C1 = W 2This indicates that the region is a dynamic region, and K is calculated using the pixel value along the direction of minimum variance. a (x, y, z)

[0154] d dynamic (z)=arg d min[var(d n (z))]

[0155]

[0156] d dynamic (z) represents the direction of the dynamic region, that is, the direction of minimum variance, σ(d dynamic (z) represents the standard deviation of the pixel values ​​along the direction of minimum variance. This represents the mean value of the pixels along the direction of minimum variance.

[0157] S4043, if C0 = W 2 This indicates that the region is a static region, and K is calculated using the pixel value along the direction of maximum variance. a (x, y, z):

[0158] d static (z)=arg d max[var(d n (z))]

[0159]

[0160] d static (z) represents the direction of the static region, that is, the direction of maximum variance, σ(d static (z) represents the standard deviation of the pixel values ​​along the direction of maximum variance. This represents the mean value of the pixels along the direction of maximum variance.

[0161] S4044, if C 0.5 =W 2 This indicates that the region belongs to a transition area, and K is calculated using the pixel values ​​along the median variance direction. a (x, y, z):

[0162] d transition (z)=arg d median[var(d n (z))]

[0163]

[0164] d transition (z) represents the direction of the transition region, that is, the direction of the median variance, σ(d transition(z) represents the standard deviation of the pixel values ​​along the median direction of the variance. This represents the mean value of the pixels along the direction of the variance median.

[0165] S405, the sliding window iterates through the range of pixel coordinates (x, y): x∈[1, M], y∈[1, N]. That is, looping through S402, S403, S404, S4041, S4042, S4043, and S4044 yields a new speckle contrast image K. a (x, y, z).

[0166] Reference Figure 3 In one embodiment, based on the disclosed original speckle image, the left image is a schematic diagram of a pseudo-color image obtained using the spatial speckle contrast imaging method, and the right image is a schematic diagram of an adaptively enhanced speckle contrast blood flow image pseudo-color obtained using the method of this example. The right image shows more microvascular details than the left image, and the clarity of the blood vessels is effectively improved. This demonstrates that the method of the present invention can make the contrast image of blood vessels clearer, improve the imaging effect of blood vessel edges and microvessels, and reflect more details.

[0167] Reference Figure 4 To conduct the phantom experiments, a phantom model was made using epoxy resin. Different amounts of titanium dioxide and Indian ink were added to the epoxy resin to simulate the dermis and epidermis. A capillary glass tube with an inner diameter of 1 mm and an outer diameter of 2 mm was used to simulate blood vessels. The depth of the capillary glass tube from the surface was 400 μm. A 3% fat emulsion solution was injected into the capillary glass tube at a speed of 10 mm / s using a syringe pump to simulate blood. The original speckle image sequence was obtained by a camera. Figure 4 The left image is a schematic diagram of a pseudo-color image obtained using the spatial speckle contrast imaging method, and the right image is a schematic diagram of an adaptively enhanced speckle contrast blood flow image obtained using the method of this example. Compared with the left image, the right image shows a higher contrast between the blood vessel area and the background area, and the blood vessel area is more obvious. This proves that the method of the present invention can improve the visualization effect of blood vessels in deeper blood vessel imaging, and has higher noise attenuation and better imaging quality.

[0168] Although the embodiments of the present invention have been disclosed above, the present invention is not limited to the above-described embodiments. It can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details.

Claims

1. A method of adaptive enhanced laser speckle contrast blood flow imaging, characterized in that, The method comprises the following steps: S100, measuring a biological tissue to be measured by using a laser speckle contrast imaging system to obtain a raw speckle image sequence I(x, y, z); S200, calculating a spatial speckle contrast image K from the original speckle image sequence I(x,y,z) s S300, calculating an initial speckle contrast image K(x,y,z) from the spatial speckle contrast image K(x,y,z) s S300, calculating an initial speckle contrast image K(x,y,z) from the spatial speckle contrast image K(x,y,z) (n+1) ; S300, segmenting the initial speckle contrast image K(x, y, z) according to the initial speckle contrast image K(x, y, z) (n+1) segmenting the dynamic region and the static region using a segmentation algorithm to obtain a segmented image Se(x, y). S400, calculating an adaptive enhanced speckle contrast image K according to the original speckle image sequence I(x, y, z) and the segmentation image Se(x, y) a (x,y,z), specifically comprising the following steps: S401, setting a selected sliding space window size as WxW, and according to the raw speckle image sequence I(x, y, z) and the segmentation image Se(x, y), the sliding space window traverses the value range of the pixel point coordinates (x, y): x∈[1, M], y∈[1, N]; S402, select the corresponding original speckle image sub-frame data I under the sliding window W (x,y, z) and segmented image sub-frame data Se W (x,y), wherein the value range of x is [1, W], the value range of y is [1, W], and the value range of z is S403, calculate Se W (x,y) is the number of pixels (x, y) that are 0, the number of pixels that are 1, and the number of pixels that are 0.5 0.5 ; if C0≠W 2 or C1≠W 2 or C 0.5 ≠W 2 , it means that the corresponding original speckle image sub-frame data I W (x, y, z) in the sliding window is in the boundary area, and the contrast value K a (x,y,z) corresponding to the center point of the window at the original speckle image (x, y, z) is calculated using an adaptive window size W (x,y) is 1, then Se W (x,y) corresponding to the pixel point I W1 (x,y,z) equal to 1 is used to calculate the contrast value K a1 (x,y,z) corresponding to the center point of the window at the original speckle image (x, y, z) W (x,y) is 0, then Se W (x,y) corresponding to the pixel point I W0 (x,y,z) equal to 0 is used to calculate the contrast value K a0 (x,y,z) corresponding to the center point of the window at the original speckle image (x, y, z) W (x,y) is 0.5, then Se W (x,y) corresponding to the pixel point I W0.5 (x,y,z) equal to 0.5 is used to calculate the contrast value K a0.5 (x,y,z) corresponding to the center point of the window at the original speckle image (x, y, z) S404, if C0=W 2 or C1=W 2 or C 0.5 =W 2 , it indicates that the corresponding original speckle image sub-frame data I W (x, y, z) under the sliding window is not in the boundary region, and the variance criterion is used to calculate the contrast value K a (x, y, z) of the original speckle image (x, y, z) corresponding to the center point of the window. S405, the sliding space window traverses the value range of the pixel point coordinates (x, y): x∈[1, M], y∈[1, N], that is, the new speckle contrast image K can be obtained by circulating S402, S403 and S404 a (x,y, z).

2. The method of claim 1, wherein, In S100, the method for measuring a biological tissue to be measured by using a laser speckle contrast imaging system to obtain a raw speckle image sequence is as follows: In S101, the exposure time of the CMOS camera of the laser speckle contrast imaging system is set to Tms, that is, frames are collected in each second frame; S102, setting that within the collection time t seconds, the laser light source of the laser speckle contrast imaging system is set to be turned on and output, and after passing through a beam expander and a mirror, irradiates the surface of the biological tissue to be measured for measurement; S103, the CMOS camera of the laser speckle contrast imaging system collects the light beams scattered and diffused by the surface of the biological tissue to be measured to form a raw speckle image; S104, set in the acquisition time t seconds, the CMOS camera total record The original speckle image is obtained by gray scale preprocessing of the original speckle image by a computer.

3. The method of claim 1, wherein, In S200, the method for calculating an initial speckle contrast image according to the raw speckle image sequence is as follows: S201, set the original speckle image sequence as I(x, y, z), wherein I(x, y, z) represents the gray value of the original speckle image at the pixel point coordinate (x, y) in the zth frame, wherein the image resolution size is set as [M, N], the value range of x is [1, M], the value range of y is [1, N], and the value range of z is S202, select a sliding spatial window size of VxV, the sliding spatial window traverses the value range of pixel point coordinates (x, y): x e [1, M], y e [1, N]; according to the original speckle image sequence I(x, y, z), select the corresponding original speckle image sub-frame data I V (x, y, z) under the sliding spatial window, wherein the value range of x is [1, V], the value range of y is [1, V], and the value range of z is Calculate the spatial speckle contrast image K s (x, y, z): wherein S203, calculating the spatial speckle contrast image K s (x,y,z), calculating the gradient KN(x,y) in four directions at the pixel point (x,y) in the z-th frame z , KS(x,y) z , KW(x,y) z , KE(x,y) z , calculating the diffusion coefficient dN(x,y) in four directions according to the gradient in four directions z , dS(x,y) z , dW(x,y) z , dE(x,y) z , obtaining the initial speckle contrast image K(x,y,z) (n+1) : K(x, y, z) (n+1) = K(x, y, z) n + λx(N + S + WE + E) The above formula is an iterative formula for obtaining the initial speckle contrast image K(x, y, z) (n+1) n is the number of iterations, and λ is the diffusion coefficient. Wherein: N = dN(x, y) z x KN(x, y) z , S = dS(x, y) z xKS(x, y) z , WE = dW(x, y) z x KW(x, y) z , E = dE(x, y) z x KE(x, y) z ), Wherein Kthr is an edge parameter.

4. The method of claim 1, wherein, In S300, the method for calculating a segmentation image according to the initial speckle contrast image is as follows: S301, segmenting the initial speckle contrast image K(x, y, z) according to the initial speckle contrast image K(x, y, z) (n+1) segmenting the dynamic region and the static region using a segmentation algorithm to obtain a segmentation image Se(x, y): Se(x, y) = Se(x1, y1) + Se(x2, y2) + Se(x3, y3) Wherein Se(x1, y1) represents a dynamic region, Se(x2, y2) represents a static region, and Se(x3, y3) represents a transition region.

5. The adaptive enhanced laser speckle contrast blood flow imaging method according to claim 1, characterized in that, In S403, If then K a (x,y, z) = K a1 (x, y, z) (1) (2) where i, j satisfy Se W (i,j) = 1; If then K a (x, y, z) = K a0 (x, y, z) (3) (4) where i, j satisfy Se W (i, j) = 0; If then K a (x,y, z) = K a0.5 (x, y, z) (5) (6) where i, j must satisfy Se W (i,j) = 0.5; wherein represents a change in rounded to the nearest whole number, and z represents the zth frame.

6. The method of claim 5, wherein, In S404, the contrast ratio K at the original speckle image (x, y, z) corresponding to the center point of the window is calculated by using the variance criterion a The method for calculating the contrast ratio K of the original speckle image (x, y, z) is: S4041, compute I W window variance d of (x, y, z) all and variance d in 4 directions (0°, 45°, 90°, 135°) 0° , d 45° , d 90° , d 135° , make d n (z): d n (z) = {d 0° , d 45° , d 90° , d 135° , d all} I j (z) represents the pixel light intensity value in the direction corresponding to the zth frame, (z) represents the mean value of the pixel light intensity in the direction corresponding to the zth frame, (z) represents the mean value of the pixel light intensity in the direction corresponding to the zth frame, S4042, if C1=W 2 It is explained that the region belongs to a dynamic region, and K is calculated using the value of the pixel point in the direction of minimum variance a (x, y, z): d dynamic (z) = arg d min[var(d n (z))] d dymamic (z) denotes a dynamic region direction, i.e. the direction of minimum variance, σ(d dynamic (z)) denotes the standard deviation of the values of the pixel points in the direction of minimum variance, denotes the mean value of the values of the pixel points in the direction of minimum variance; S4043, if C0=W 2 It is explained that the region belongs to the static region, and the value of the pixel point in the maximum variance direction is used to calculate K a (x, y, z): d static (z) = arg d max[var(d n (z))] d static (z) denotes a static region direction, i.e. the direction of maximum variance, σ(d static (z)) denotes the standard deviation of the values of the pixel points in the direction of maximum variance, denotes the mean value of the values of the pixel points in the direction of maximum variance; S4044, if C 0.5 = W 2 It is explained that the region belongs to the transition region, and K is calculated by using the value of the pixel point in the direction of the median of the variance a (x, y, z): d transition (z) = arg d median[var(d n (z))] d transition (z) denotes a transition region direction, i.e. the variance mean direction, σ(d transition (z)) denotes the standard deviation of the values of the pixel points in the variance mean direction, denotes the mean value of the values of the pixel points in the variance mean direction.

Images