A calculation method for blood flow measurement using diffuse photon autocorrelation

Abstract

The present invention provides a method for measuring and calculating diffuse photon autocorrelation blood flow, comprising the following steps: selecting multiple points on a normalized light field time autocorrelation function curve to obtain data points, segmenting the data points to obtain multiple sequences, multiplying the data points in the sequence by corresponding multiplier values ​​to obtain an amplified sequence, sorting the amplified sequence by numerical value and removing noise data, and restoring the data points in the amplified sequence after noise removal to their original order and values ​​to obtain a first type of denoised sequence, performing a secondary noise data removal operation on the first type of denoised sequence and removing the difference values ​​to obtain a second type of denoised sequence, obtaining characteristic coefficients of the calculated data points through relevant four arithmetic operations, and recursively calculating a five-step iterative blood flow value based on the characteristic coefficients. The present invention has the beneficial effect of overcoming the problems of existing simplex curve fitting methods for solving blood flow, such as difficulty in calculation, long time consumption, and poor stability.

Description

Technical Field

[0001] The invention belongs to the field of blood flow measurement, and in particular relates to a diffuse photon autocorrelation blood flow measurement calculation method. Background Art

[0002] The importance of online monitoring of cerebral blood flow for clinical surgery and critically ill patient care is self-evident. For example, during carotid artery revascularization surgery, localized cerebral hypoperfusion caused by vascular clamping requires online monitoring. The primary cause of stroke is vascular stenosis or obstruction, which leads to decreased cerebral blood flow. Furthermore, patients with Alzheimer's disease experience severely reduced brain function, manifested by inadequate cerebral vascular and blood flow responses. In addition to cerebral blood flow monitoring, skeletal muscle blood flow is also a key physiological parameter for assessing muscle function and conditions such as peripheral vascular disease and senile myofascial pain.

[0003] Currently, the primary cerebral blood flow measurement technologies used in hospitals are ultrasound Doppler (also known as transcranial Doppler) and perfusion magnetic resonance imaging (ASL-MRI). However, both techniques have limitations. Ultrasound Doppler measures the main cerebral vessels, but many brain diseases are related to local microvascular circulation. ASL-MRI can achieve very high spatial resolution, but the measurement cost is high, and dynamic and continuous measurement is not easy to achieve. In recent years, a biotechnology utilizing near-infrared light has been developed, known as diffuse photon autocorrelation spectroscopy (PCS). PCS uses the temporal autocorrelation of photons to rapidly estimate the light field perturbations caused by red blood cell movement, making it a new technology for directly measuring tissue blood flow. This technology has the advantages of being non-invasive, low-cost, and portable, and has developed rapidly in recent years.

[0004] During PCS data acquisition, the light source and detector are placed on the same side of the tissue being measured (such as the forehead), a few millimeters to a few centimeters apart. The near-infrared photons emitted by the light source enter the cerebral cortex and are scattered multiple times by moving red blood cells, causing light field disturbances. A small number of photons escape from the surface and are collected by the detector. The digital correlator performs temporal autocorrelation calculations on the number of photons collected by the detector (i.e., light intensity) to obtain the normalized light intensity autocorrelation function g2(τ), and the normalized light field time autocorrelation function g1(τ) is obtained from the Siegert relationship, where τ is the delay time.

[0005] Because the nonnormalized light field temporal autocorrelation function G1(τ) satisfies the diffusion partial differential equation, the current solution involves fitting the analytical expression of the differential equation to the g1(τ) curve. The basic process of curve fitting involves first substituting an assumed blood flow value into the analytical expression to obtain the theoretical value of the temporal autocorrelation function g1(τ) for different delay times τ. The sum of squared residuals between the theoretical and measured g1(τ) values ​​for all delay times is then calculated. The assumed blood flow value is then iteratively varied to find the minimum sum of squared residuals. This process primarily utilizes the nonlinear simplex method to find the optimal solution. However, this nonlinear fitting method is not a direct arithmetic operation (i.e., addition, subtraction, multiplication, and division) of the data. The iterations are time-consuming, making it difficult to implement online blood flow calculations. Furthermore, this method is susceptible to noise, making the solution unstable. Summary of the Invention

[0006] In view of this, the present invention aims to propose a diffuse photon autocorrelation blood flow measurement and calculation method, in order to solve at least one of the above-mentioned technical problems.

[0007] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0008] In a first aspect, the present invention provides a method for measuring and calculating blood flow using diffuse photon autocorrelation, characterized in that:

[0009] The steps include:

[0010] S1: Select multiple data points on the normalized light field temporal autocorrelation function curve, and perform noise reduction on the multiple data points to obtain calculation data points;

[0011] S2: Perform four related arithmetic operations on the operation data points to obtain the characteristic coefficients of the operation data points;

[0012] S3: A preliminary iterative blood flow value is obtained based on the characteristic coefficient, and recursive calculation is performed to obtain the second to fifth step iterative blood flow values.

[0013] Furthermore, the specific calculation steps of step S1 are as follows:

[0014] S101: selecting a plurality of data points on a normalized light field temporal autocorrelation function curve;

[0015] S102: Segmenting the multiple data points according to the delay time to obtain multiple sequences, and multiplying the data points in the sequences by corresponding multiplier values ​​to obtain amplified sequences;

[0016] S103: sorting the amplified sequence according to the magnitude of the values ​​and removing the noise data, and restoring the data points in the amplified sequence after the noise is removed to their original order and values, to obtain a type of noise-reduced sequence;

[0017] S104: performing a secondary noise point data elimination operation on the first-class denoised sequence and removing the difference values ​​to obtain a second-class denoised sequence;

[0018] S105: Delete the first three and last three points of the second-class denoised sequence, and then take the front part of the amplified sequence as the calculation data points.

[0019] Furthermore, the specific calculation process of step S101 is:

[0020] A normalized light intensity autocorrelation function is obtained from the experiment, and the normalized light intensity autocorrelation function is converted into a normalized light field time autocorrelation curve according to the Siegert relationship. Multiple data points are taken on the normalized light field time autocorrelation curve, and variables are replaced on the data points to generate new data points.

[0021] The number of the plurality of data points is a positive integer not less than 79, and the number of the data points+1 is an integer multiple of 8.

[0022] Furthermore, the specific process of segmenting and multiplying the data points in step S102 is as follows:

[0023] The data points are segmented according to the delay time. The first segmented sequence contains 1 data point, the second segmented sequence contains 14 data points, and the third and subsequent segments each contain 8 data points. Each data point is multiplied by a corresponding multiplier to obtain an amplified sequence.

[0024] Furthermore, the specific process of removing noise data in step S103 is as follows:

[0025] Sort the amplified sequence obtained in step S102 by numerical value to obtain a new sequence. Find the first quarter and the last quarter data points of the new sequence, and make the following judgment:

[0026] If the value of any data point in the first quarter of the data points is less than 0.7 times the value of the last data point in the first quarter of the data points, the first quarter of the new sequence is determined to be a noise point and is removed;

[0027] If the value of any data point in the last quarter of the data points is greater than 1.5 times the value of the first data point in the last quarter of the data points, the last quarter of the new sequence is determined to be a noise point and is removed;

[0028] The data points that are not eliminated are restored to their original data sequence numbers, and the value of each data point is divided by the corresponding multiplication value in step S102 to obtain a type of noise reduction sequence.

[0029] Furthermore, the specific process of secondary noise reduction and removing difference values ​​in step S104 is as follows:

[0030] For the remaining data points in a class of noise reduction sequences, excluding the first three and the last three data points, the following judgment is made:

[0031] If yp(i) < yp(m1), (m1 = i - 1, i - 2, i - 3), then the value of this judgment is 0; otherwise, the value is 1;

[0032] If yp(i) > yp(m2), (m2 = i + 1, i + 2, i + 3), then the value of this judgment is 0; otherwise, the value is 1;

[0033] Add the six values obtained in the above steps. If the result is greater than 2, it is determined as the second type of noise point and is excluded;

[0034] Where yp(i) is the current data point and i is the position index of the current data point.

[0035] Furthermore, in step S2, the following relevant four arithmetic operations are performed on the operation data points (x(i), y(i)):

[0036] Calculate the average value of the product of x(i) and y(i), calculate the average value of the square of x(i), calculate the average value of y(i), calculate the average value of x(i), and obtain the characteristic coefficient of the operation data point through the above four average values.

[0037] In a second aspect, an electronic device includes a processor and a memory communicatively connected to the processor and used to store instructions executable by the processor. The electronic device is characterized in that: the processor is used to execute a diffuse photon autocorrelation blood flow measurement calculation method according to any one of the first aspects above.

[0038] In a third aspect, a server is characterized in that: it includes at least one processor and a memory communicatively connected to the processor. The memory stores instructions executable by the at least one processor, and when the instructions are executed by the processor, the at least one processor is caused to execute a diffuse photon autocorrelation blood flow measurement calculation method according to any one of the first aspects above.

[0039] In a fourth aspect, a computer-readable storage medium stores a computer program, and is characterized in that: when the computer program is executed by a processor, it implements a diffuse photon autocorrelation blood flow measurement calculation method according to any one of the first aspects above.

[0040] Compared with the prior art, a diffuse photon autocorrelation blood flow measurement calculation method according to the present invention has the following beneficial effects:

[0041] The online calculation method for diffuse photon autocorrelation blood flow measurement according to the present invention overcomes the problems of difficult calculation, long time consumption, and poor stability in solving blood flow by curve fitting in the existing simple methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings, which constitute part of the present invention, are provided to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are provided to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0043] Figure 1 This is a flow chart of a diffuse photon autocorrelation blood flow measurement and calculation method according to an embodiment of the present invention;

[0044] Figure 2 Schematic diagram of comparison between original data points and data points in an amplified sequence according to an embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of the first type of noise according to an embodiment of the present invention;

[0046] Figure 4 This is a schematic diagram of the second type of noise according to an embodiment of the present invention;

[0047] Figure 5 This is a schematic diagram of an online measurement curve of cerebral blood flow in a healthy person performing a voluntary breath-holding task according to an embodiment of the present invention;

[0048] Figure 6 The embodiment of the present invention is described Figure 5 Schematic diagram of the online measurement curve of cerebral blood flow obtained with the same PCS data but without using the present invention;

[0049] Figure 7 Schematic diagram of blood flow data points obtained from 200 computer-simulated PCS data according to an embodiment of the present invention;

[0050] Figure 8 The embodiment of the present invention is described Figure 5 The same PCS data, but without the blood flow data points obtained by the present invention. DETAILED DESCRIPTION

[0051] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0052] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0053] Example 1:

[0054] A diffuse photon autocorrelation blood flow measurement and calculation method comprises the following steps:

[0055] S1: Select multiple data points on the normalized light field temporal autocorrelation function curve, and perform noise reduction on the multiple data points to obtain calculation data points;

[0056] S2: Perform four related arithmetic operations on the operation data points to obtain the characteristic coefficients of the operation data points;

[0057] S3: A preliminary iterative blood flow value is obtained based on the characteristic coefficient, and recursive calculation is performed to obtain the second to fifth step iterative blood flow values.

[0058] The specific calculation steps of step S1 are as follows:

[0059] S101: selecting a plurality of data points on a normalized light field temporal autocorrelation function curve;

[0060] S102: Segmenting the multiple data points according to the delay time to obtain multiple sequences, and multiplying the data points in the sequences by corresponding multiplier values ​​to obtain amplified sequences;

[0061] S103: sorting the amplified sequence according to the magnitude of the values ​​and removing the noise data, and restoring the data points in the amplified sequence after the noise is removed to their original order and values, to obtain a type of noise-reduced sequence;

[0062] S104: performing a secondary noise point data elimination operation on the first-class denoised sequence and removing the difference values ​​to obtain a second-class denoised sequence;

[0063] S105: Delete the first three and last three points of the second-class denoised sequence, and then take the front part of the amplified sequence as the calculation data points.

[0064] The specific calculation process of step S101 is:

[0065] The normalized light intensity autocorrelation function g2(τ) was obtained from the experiment and was calculated according to the Siegert relation (g2(τ)=1+β|g1(τ)| 2 ) is converted into a normalized light field time autocorrelation curve g1(τ); here β is taken as 0.5; n0 points are taken on this curve, where n0 is a positive integer not less than 79 and (n0+1) is an integer multiple of 8, to obtain n0 pairs of data (τ(n0), g1(n0)), where n0 is the number of data points, and the following variable substitutions are made:

[0066] x(i)=τ(i),

[0067] y(i)=g1(i)-1, 1≤i≤n0.

[0068] This generates a new data point [xs0(i), ys0(i)].

[0069] The specific process of segmenting and multiplying the data points in step S102 is as follows:

[0070] The data point ys0(i) is segmented according to the delay time. The first segmented sequence contains 1 data point, the second segmented sequence contains 14 data points, and the third and subsequent segments each contain 8 data points. Each data point is multiplied by the corresponding multiplier shown in the following table to obtain the amplified sequence;

[0071]

[0072]

[0073] Among them, the exponential part of the multiplier is: C*(2 (N-1) / [(N-1) 2 +1]); where N is the serial number, the constant C = 2·μs'·ppath·k0·k0·D0·τ0; μs', k0, and ppath are all constants; μs' is the reduced scattering coefficient, k0 = 2π·nf / λ, where nf is the refractive index of the tissue and λ is the wavelength of light; ppath is the average path of photons transmitted in the tissue, and D0 is the Brownian motion coefficient of the particle.

[0074] The specific process of removing noise data in step S103 is as follows:

[0075] Let the sequence ys0(i) be ys1(i), and its position index is [1,2,...n0]

[0076] Sort the amplified sequence obtained in step S102 by numerical value to obtain a new sequence ys2(i), and find the data points in the first quarter and the last quarter of the new sequence;

[0077] Right now

[0078] ys2(1)≤ys2(2)≤...≤ys2(j1)...≤ys2(j2)≤...≤ys2(n0)

[0079] in

[0080] j1=(n0+1) / 4; j2=3(n0+1) / 4

[0081] For the first quarter of the data of ys2(i) (i=1,..,j1-1), if:

[0082] If ys2(i)<0.7*ys2(j1), it is determined to be a type of noise point and will be removed;

[0083] For the second quarter of the data of ys2(i) (i=j2+1,..,n0), if:

[0084] If ys2(i)>1.5*ys2(j2), it is determined to be a type of noise point and will be removed.

[0085] Restore the un-eliminated data points to their original data sequence numbers (i.e., ys1), and divide each data point by the corresponding multiplication value to obtain a new sequence yp(i) (i = 1,.., n00), where the number of data points n00 in the new sequence satisfies n00 ≤ n0.

[0086] The above steps can effectively amplify the noise points (defined as the first type of noise points) at the edges of the g1(τ) data set, making their values fall outside the normal value range, and thus eliminating them; a more stable blood flow change curve can be obtained from the data set after eliminating the first type of noise points.

[0087] The specific process of secondary noise reduction and removal of differential values in step S104 is as follows:

[0088] For the remaining data points in a class of noise reduction sequences except the first three and the last three data points, make the following judgments:

[0089] If yp(i) < yp(m1), (m1 = i - 1, i - 2, i - 3), then the judgment value is 0; otherwise, the value is 1;

[0090] If yp(i) > yp(m2), (m2 = i + 1, i + 2, i + 3), then the judgment value is 0; otherwise, the value is 1;

[0091] Add the six values obtained in the above steps. If the result is greater than 2, it is determined as the second type of noise point and eliminated;

[0092] Where yp(i) is the current data point and i is the position index of the current data point.

[0093] The above steps can select the data points in the g1(τ) data set that do not satisfy the descending order according to the judgment criteria, define them as the second type of noise points, and eliminate them; a more accurate blood flow change curve reflecting the true physiological changes can be obtained from the data set after eliminating the second type of noise points.

[0094] Delete the first three and the last three points of the yp sequence, and then take the first 5 sequences in the amplified sequence as the new data points (x(i), y(i)) (i = 1, 2,... n1); here the number of data points n1 ≤ 36;

[0095] Perform the following relevant four arithmetic operations on the operation data points (x(i), y(i)):

[0096]

[0097] yxm represents the average value of the product of x(i) and y(i);

[0098]

[0099] xxm represents the mean value of the square of x(i);

[0100]

[0101] ym represents the mean value of y(i);

[0102]

[0103] xm represents the mean value of x(i);

[0104] b=(yxm-ym·xm) / (xxm-xm·xm)

[0105] b represents the characteristic coefficient of the x(i) array and the y(i) array.

[0106] Let b1 = b and calculate the initial iterative blood flow value:

[0107] Db1=-b1 / (2·μs'·ppath·k0·k0)

[0108] where μs', k0, and ppath are all constants; μs' is the reduced scattering coefficient, k0 = 2π·nf / λ, where nf is the refractive index of the tissue and λ is the wavelength of light; ppath is the average path of photons transmitted in the tissue;

[0109] The first 6 sequences in the amplified sequence are recorded as new data points (x(i), y(i)) (i = 1, ..., n2); the number of data points n2 ≤ 44; and the method for obtaining the two-step iterative blood flow value Db2 is:

[0110] Let M1 = 2·k0·k0·Db1·μs' and make the following variable substitutions:

[0111] x(i)=τ(i),

[0112] y(i)=g1(i)-1-1 / 2·M1·M1·ppath2·τ(i)·τ(i), 1≤i≤n2;

[0113] Here ppath2 is the mean of the square of the photon transmission path in the tissue, which is called the second-order average path;

[0114] Execute the calculation process of S102 to obtain the updated b;

[0115] Let b2 = b, and calculate the second-step iterative blood flow value Db2:

[0116] Db2=-b2 / (2·μs'·ppath·k0·k0)

[0117] The first 7 sequences in the amplified sequence are recorded as new data points (x(i), y(i)) (i = 1, ..., n3); the number of data points n3 ≤ 52; and the method for obtaining the three-step iterative blood flow value Db3 is:

[0118] M2=2·k0·k0·Db2·μs'

[0119] Make the following variable substitutions: x(i) = τ(i),

[0120] y(i)=g1(i)-1-1 / 2·M2·M2·ppath2·τ(i)·τ(i)+1 / 6·M2·M2·M2·ppath3·τ(i)·τ(i)·τ(i),

[0121] 1≤i≤n3; here ppath3 is the mean of the cube of the photon transmission path in the tissue, called the third-order average path;

[0122] Execute the calculation process of S2 to obtain the updated b;

[0123] Let b3 = b, and calculate the three-step iterative blood flow value Db3:

[0124] Db3=-b3 / (2·μs'·ppath·k0·k0);

[0125] The first 8 sequences in the amplified sequence are recorded as new data points (x(i), y(i)) (i = 1, ..., n4); the number of data points n4 ≤ 60; the number of data points n3 ≤ 52; and the method for obtaining the four-step iterative blood flow value Db4 is:

[0126] Let M3 = 2·k0·k0·Db3·μs' and make the following variable substitutions:

[0127] x(i)=τ(i),

[0128] y(i)=g1(i)-1-1 / 2·M3·M3·ppath2·τ(i)·τ(i)+1 / 6·M3·M3·M3·ppath3·τ(i)·τ(i)·τ(i)-1 / 24·M3·M3·M3·M3·ppath4·τ(i)·τ(i)·τ(i)·τ(i),

[0129] 1≤i≤n4; here ppath4 is the mean of the fourth power of the photon transmission path in the tissue, called the fourth-order average path;

[0130] Execute the calculation process of S2 to obtain the updated b;

[0131] Let b4 = b, and calculate the four-step iterative blood flow value Db4:

[0132] Db4=-b4 / (2·μs'·ppath·k0·k0);

[0133] The first 9 sequences in the amplified sequence are recorded as new data points (x(i), y(i)) (i=1,...,n5); the number of data points n5 ≤ 68; and the method for obtaining the five-step iterative blood flow value Db5 is:

[0134] Let M4 = 2·k0·k0·Db4·μs' and make the following variable substitutions:

[0135] x(i)=τ(i),

[0136] y(i)=g1(i)-1-1 / 2·M4·M4·ppath2·τ(i)·τ(i)+1 / 6·M4·M4·M4·ppath3·τ(i)·τ(i)·τ(i)-1 / 24·M4·M4· M4·M4·ppath4·τ(i)·τ(i)·τ(i)·τ(i)+1 / 120·M4·M4·M4·M4·M4·ppath5·τ(i)·τ(i)·τ(i)·τ(i)·τ(i),

[0137] 1≤i≤n5; here ppath5 is the fifth power mean of the photon transmission path in the tissue, called the fifth-order average path;

[0138] Execute the calculation process of S2 to obtain the updated b;

[0139] Let b5 = b, and calculate the five-step iterative blood flow value Db5:

[0140] Db5=-b5 / (2·μs'·ppath·k0·k0).

[0141] This solution can be used to obtain preliminary iterative blood flow values ​​up to five steps (i.e., Db1, Db2, Db3, Db4, Db5) according to actual needs, and to generalize higher-step iterations based on similar ideas, such as six-step iterative blood flow values ​​(Db6), seven-step iterative blood flow values ​​(Db7), etc.

[0142] This solution can be used for surgical monitoring, critical care monitoring, anesthesia monitoring, sleep monitoring, etc. The present invention uses simple judgment, arrangement, and arithmetic operations, eliminating the need for tedious optimization and iterative calculations, thus significantly shortening calculation time. Furthermore, the present invention does not require the use of external function optimization packages and is suitable for coding in any scientific computing language, allowing for easy embedding in various graphical interactive software (e.g., Visual Basic, Visual C++, Labview, etc.). This enables PCS data acquisition and online blood flow calculation and display, greatly facilitating clinical online blood flow monitoring.

[0143] Working process 1:

[0144] The calculation method of the present invention was quantitatively verified in a liquid phantom, including sequentially performed configuration phantom and liquid phantom experiments.

[0145] The phantom is prepared by mixing black ink, fat emulsion and distilled water in a certain proportion so that the absorption coefficient of the liquid reaches 0.1 cm -1 , the reduced scattering coefficient reaches 8.0cm -1 ; These optical coefficients are similar to those of human tissue, and fat emulsion particles in the liquid simulate red blood cells in microvascular tissue.

[0146] The liquid phantom experiment includes the following process: placing the liquid phantom in a rectangular glass container, and placing the diffuse light probe flat on the surface of the phantom. The diffuse light probe includes a light source fiber (multimode fiber), a detector fiber (single-mode fiber) and a base. The base is placed flat on the surface of the phantom. One end of the light source fiber and one end of the detector fiber are inserted into the base respectively. The distance between the light source fiber and the detector fiber can be adjusted according to actual conditions. The base is bonded to the outer walls of the two optical fibers. The other end of the light source fiber is connected to the infrared light source, and the other end of the detector fiber is connected to the photon detector. The near-infrared light emitted by the infrared light source is incident on the liquid phantom through the light source fiber. A part of the photons that have been scattered multiple times are collected by the detector fiber on the same side and transmitted to the photon detector. The digital correlator reads the detected photons and calculates the light intensity time autocorrelation curve g2(τ).

[0147] The movement of fat emulsion particles is manipulated by changing the temperature, where the liquid phantom measurement data point at the first moment is:

[0148] The data of the first measurement time point is as follows:

[0149] Delay time τ = 1.0e-05*

[0150] [0.02,0.04,0.06,0.08,0.10,0.12,0.14,0.16,0.18,0.20,0.22,0.24,0.26,0.28,0.30,0.34,0.38,0.42,0.46,0.50,0.54,0.58,0.62,0.70,0.78,0.86,0.94,1.02,1.10,1.18,1.26,1.42,1.58,1.74,1.90,2.06,2.22,2.38,2.54

[0151] 2.86,3.18,3.50,3.82,4.14,4.46,4.78,5.10,5.74,6.38,7.02,7.66,8.30,8.94,9.58,10.22,11.50,12.78,14.06,15.34,16.62,17.90,19.18,20.46,23.02,25.58,28.14,30.70,33.26,35.82,38.38,40.94,46.06,51.18,56.30,61.42,66.54,71.66,76.78,81.90], a total of 79 data points, unit is second (s);

[0152] The measured light intensity time autocorrelation function g2(τ) = [2.0344, 1.6790, 1.4939, 1.5471, 1.4446, 1.4382, 1.4414, 1.3877, 1.4095,

[0153] 1.4390,1.4499,1.4736,1.4499,1.4140,1.3864,1.4240,1.3853,1.3491,

[0154] 1.3491,1.4246,1.3523,1.3725,1.3594,1.3532,1.3141,1.3083,1.3146,

[0155] 1.3325,1.3026,1.3023,1.2978,1.2632,1.2591,1.2505,1.2527,1.2170,

[0156] 1.1959,1.2241,1.2032,1.2017,1.1735,1.1581,1.1400,1.1594,1.1370,

[0157] 1.1194,1.1254,1.1125,1.1045,1.0838,1.0781,1.0649,1.0802,1.0615,

[0158] 1.0708,1.0630,1.0525,1.0419,1.0394,1.0464,1.0313,1.0321,1.0266,

[0159] 1.0257,1.0231,1.0173,1.0201,1.0185,1.0131,1.0187,1.0129,1.0141,

[0160] 1.0103, 1.0109, 1.0073, 1.0087, 1.0082, 1.0110, 1.0069] 79 data points in total, no units

[0161] Among them, constants: μs' = 8, k0 = 1.0966e+05, ppath = 14.35, ppath2 = 260.04, ppath3 = 6007, ppath4 = 1.7519e+05, ppath5 = 6.3311e+06, D0 = 1.0e-08;

[0162] By calculation, we get:

[0163] Db1=0.3043e-08; Db2=0.3936e-08; Db3=0.4074e-08; Db4=0.4289e-08; Db5=0.4349e-08.

[0164] The blood flow values ​​of the five-step iterations at all 40 moments are continuously plotted. It can be seen that the blood flow of the phantom is in a stable state before or after the temperature change. The blood flow curve obtained by the method of the present invention is very smooth and has little noise.

[0165] Working process 2:

[0166] The present invention's online blood flow calculation method was also validated in a human brain voluntary breath-holding experiment. The specific steps are as follows: the subject sat flat on a chair, keeping their head still, and began voluntary breath-holding for approximately 100 seconds. Before the task began, one end of the light source fiber 1 and one end of the detector fiber 2 were attached to the subject's forehead. The other end of the light source fiber 1 was connected to an infrared light source, and the other end of the detector fiber 2 was connected to a photon detector and a digital microscope. A baseline measurement took 100 seconds before the start of the experiment. The subject controlled the breath-holding time independently, which was 42 seconds, after which they inhaled gently. After the task ended, an optical measurement was performed for 100 seconds, completing the entire experiment.

[0167] The human brain data points measured at the first moment are as follows:

[0168] Delay time τ = 1.0e-05*

[0169] [0.02,0.04,0.06,0.08,0.10,0.12,0.14,0.16,0.18,0.20,0.22,0.24,0.26,0.28,0.30,0.34,0.38,0.42,0.46,0.50,0.54,0.58,0.62,0.70,0.78,0.86,0.94,1.02,1.10,1.18,1.26,1.42,1.58,1.74,1.90,2.06,2.22,2.38,2.54

[0170] 2.86,3.18,3.50,3.82,4.14,4.46,4.78,5.10,5.74,6.38,7.02,7.66,8.30,8.94,9.58,10.22,11.50,12.78,14.06,15.34,16.62,17.90,19.18,20.46,23.02,25.58,28.14,30.70,33.26,35.82,38.38,40.94,46.06,51.18,56.30,61.42,66.54,71.66,76.78,81.90], a total of 79 data points, unit is second (s);

[0171] The measured light intensity time autocorrelation function g2(τ) = [2.7342, 2.1996, 1.9190, 1.6410, 1.5769, 1.4833, 1.3524, 1.5103, 1.4729,

[0172] 1.4943,1.5370,1.3686,1.5103,1.4194,1.3686,1.4213,1.4467,1.3799,

[0173] 1.3024,1.3919,1.3411,1.2770,1.1969,1.3281,1.2941,1.2059,1.2720,

[0174] 1.2012,1.1638,1.1518,1.1972,1.1317,1.1183,1.1216,1.1087,1.1290,

[0175] 1.0960,1.0872,1.0909,1.0967,1.0825,1.0200,1.0538,1.0309,1.0340,

[0176] 1.0202,1.0210,1.0306,1.0252,1.0029,1.0244,1.0131,0.9973,1.0065,

[0177] 0.9943,0.9928,0.9977,1.0063,0.9884,0.9858,0.9944,1.0140,1.0144,

[0178] 0.9952,1.0030,0.9954,1.0129,0.9927,0.9928,1.0066,0.9875,0.9997,

[0179] 0.9995, 1.0066, 0.9971, 0.9975, 0.9923, 0.9998, 0.9982], 79 data points in total, no units;

[0180] Where: μs' = 8, k0 = 1.0966e+05, ppath = 14.35, ppath2 = 260.04, ppath3 = 6007, ppath4 = 1.7519e+05, ppath5 = 6.3311e+06; D0 = 1.0e-08

[0181] like Figure 2 As shown in the figure: Compared with the original g1 data point (hollow circle o), by multiplying the corresponding value (equilateral triangle Δ), the noise point can be effectively amplified, so that the first type of noise point (i.e. Figure 3 solid dots) and removed them.

[0182] like Figure 4 As shown: the points marked with asterisks (*) are the second type of noise points, which are removed; thereby making the subsequent blood flow value calculation more stable.

[0183] By calculation, we get:

[0184] Db1=0.5169e-08; Db2=0.7743e-08; Db3=0.7432e-08; Db4=0.9147e-08; Db5=0.8604e-08

[0185] The calculated blood flow values ​​Db for the first to fifth iterations are 2.6273e-08; 3.5464e-08; 3.7216e-08; 4.4612e-08; 4.3425e-08;

[0186] The blood flow values ​​of the five-step iterations at all 241 moments are continuously depicted, and the cerebral blood flow response obtained is as follows Figure 5As shown in the figure, cerebral blood flow experiences small fluctuations during the first 100 seconds of baseline, but remains generally stable. After the start of voluntary breath-holding, cerebral blood flow decreases significantly, reflecting microvascular contraction. During breath-holding, cerebral blood flow fluctuates significantly, reflecting the brain's self-regulatory function. After the breath-hold ends, cerebral blood flow decreases slightly due to inertia before rapidly increasing back to baseline, reflecting microvascular dilation. These trends reflect the brain's hemodynamic response to transient hypoxia and recovery.

[0187] It can be seen from the above data that the steps of the present invention can more realistically reflect the physiological state of cerebral blood flow.

[0188] like Figure 6 As shown in the figure, without noise reduction processing, the blood flow curve obtained in the first 100 seconds of the baseline period fluctuates significantly, and the blood flow decreases little after voluntary breath holding, which means that it cannot accurately reflect the physiological changes of cerebral blood flow.

[0189] The present invention also achieves online calculation of cerebral blood flow and synchronization of data acquisition, which has clear clinical significance for monitoring brain status. In other words, the calculation method provided by the present invention has practical application value.

[0190] Working process three:

[0191] Through computer simulation calculation, the following data are obtained:

[0192] Blood flow setting value: Db = 1.0e-08, unit: cm 2 / s

[0193] Delay time τ = 1.0e-05*

[0194] [0.02,0.04,0.06,0.08,0.10,0.12,0.14,0.16,0.18,0.20,0.22,0.24,0.26,0.28,0.30,0.34,0.38,0.42,0.46,0.50,0.54,0.58,0.62,0.70,0.78,0.86,0.94,1.02,1.10,1.18,1.26,1.42,1.58,1.74,1.90,2.06,2.22,2.38,2.54,2.86,3.18,3.50,3.82,4. 14,4.46,4.78,5.10,5.74,6.38,7.02,7.66,8.30,8.94,9.58,10.22,11.50,12.78,14.06,15.34,16.62,17.90,19.18,20.46,23.02,25.58,28.14,30.70,33.26,35.82,38.38,40.94,46.06,51.18,56.30,61.42,66.54,71.66,76.78,81.90], a total of 79 data points, unit is second (s);

[0195] By adding random noise and generating a light intensity time autocorrelation function g2(τ) = [2.0231, 1.6689, 1.6432, 1.6236, 1.5417, 1.5253, 1.4608, 1.4817, 1.4175

[0196] 1.4418,1.3959,1.4385,1.3780,1.5034,1.4321,1.4147,1.4567,1.3977

[0197] 1.3787,1.3751,1.3662,1.3419,1.3568,1.3503,1.3433,1.3415,1.3450

[0198] 1.3308,1.3191,1.3049,1.2769,1.2517,1.2471,1.2064,1.2251,1.21531.1944,1.1161,1.1532,1.1329,1.1050,1.1003,1.0934,1.0078,1.0959

[0199] 1.0349,1.0462,1.0542,1.0487,1.0305,1.0313,1.0126,1.0037,1.0207

[0200] 1.0115,1.0293,1.0012,1.0239,1.0115,1.0410,1.0108,1.0067,1.0068

[0201] 0.9849,1.0122,1.0069,0.9932,1.0001,0.9880,1.0044,1.0085,0.9938

[0202] 1.0050, 0.9865, 1.0028, 1.0055, 0.9879, 1.0073, 0.9948]; 79 data points in total, no units;

[0203] Constants: μs' = 8, k0 = 1.0966e+05, ppath = 14.7625, ppath2 = 291

[0204] .2429, ppath3=8.1391e+03, ppath4=3.0790e+05, ppath5=1.4382e+07; D0=1.0e-08;

[0205] Obtained by calculation

[0206] Db1=0.6747e-08; Db2=0.8859e-08; Db3=0.9721e-08; Db4=1.021e-08; Db5=0.9940e-08;

[0207] Computer simulations show that for 200 cases of noisy PCS data, the accuracy of blood flow values ​​obtained using this method improves with increasing iterations. The average errors achieved by the online algorithm from the initial iteration to the fifth iteration were 32.0%, 10.6%, 5.4%, 2.1%, and 0.7%, respectively. It can be seen that when the number of iterations exceeds two, blood flow calculations achieve an accuracy exceeding 90%, and the accuracy of the five-step online algorithm reaches over 99%.

[0208] like Figure 7 As shown in Figure 2, the blood flow data points obtained from 200 PCS data points are stably concentrated near the target value.

[0209] like Figure 8 As shown in the figure: without noise reduction processing, the blood flow error and instability of the 200 data points obtained were significantly increased; the average errors obtained by the online algorithm from the initial iteration to the five-step iteration increased to: 41.6%, 20.8%, 9.7%, 3.7% and 2.4% respectively.

[0210] This solution can reduce the amount of calculation and shorten the calculation time by using the four arithmetic operations, and realize the online calculation of diffuse light blood flow with high accuracy, and can be widely used in the field of measurement technology of diffuse photon autocorrelation.

[0211] Example 2:

[0212] An electronic device includes a processor and a memory connected to the processor for storing instructions executable by the processor, wherein the processor is configured to execute a diffuse photon autocorrelation blood flow measurement and calculation method as described in any one of the above-mentioned embodiments.

[0213] Example 3:

[0214] A server, characterized in that it includes at least one processor and a memory communicatively connected to the processor, the memory storing instructions executable by the at least one processor, the instructions being executed by the processor so that the at least one processor performs a diffuse photon autocorrelation blood flow measurement calculation method as described in any one of the first embodiments.

[0215] Example 4:

[0216] A computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the method for calculating blood flow measurement based on diffuse photon autocorrelation is implemented.

[0217] Those skilled in the art will appreciate that the units and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0218] In the several embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the division of the units described above is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The above-mentioned units may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiment of the present invention.

[0219] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

[0220] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A diffuse photon autocorrelation blood flow measurement calculation method, characterized in that: It includes the following steps: S1: Select multiple data points on the normalized optical field time autocorrelation function curve, and perform noise reduction processing on the multiple data points to obtain operation data points; S2: Perform relevant four arithmetic operations on the operation data points to obtain the characteristic coefficients of the operation data points; S3: Calculate the preliminary iterative blood flow value based on the characteristic coefficients, and perform recursive calculation to obtain the second to fifth iterative blood flow values; The specific calculation steps of step S1 are as follows: S101: Select multiple data points on the normalized optical field time autocorrelation function curve; S102: Segment the multiple data points according to the delay time to obtain multiple sequences, and multiply the data points in the sequences by the corresponding multiplication values to obtain amplified sequences; S103: Sort the amplified sequences obtained in step S102 according to the numerical size to obtain a new sequence, find the first quarter data points and the last quarter data points of the new sequence, and make the following judgments: If any data point value in the first quarter data points is less than 0.7 times the value of the last data point in the first quarter data points, then it is determined that the first quarter data of the new sequence is a type of noise point and is excluded; If any data point value in the last quarter data points is greater than 1.5 times the value of the first data point in the last quarter data points, then it is determined that the last quarter data of the new sequence is a type of noise point and is excluded; Restore the unexcluded data points to their original data serial numbers, and divide the value of each data point by the corresponding multiplication value in step S102 to obtain a type of noise reduction sequence; S104: Perform a secondary noise point data exclusion operation on the type of noise reduction sequence and remove the difference values to obtain a type two noise reduction sequence; S105: Delete the first three and the last three points of the type two noise reduction sequence, and then take the front part sequence in the amplified sequence as the operation data points; The specific process of secondary noise reduction and removal of difference values in step S104 is: For the remaining data points in the type of noise reduction sequence except the first three and the last three data points, make the following judgments: If yp(i) < yp(m1), m1 = i - 1, i - 2, i - 3, then the judgment value is 0; otherwise the value is 1; If yp(i) > yp(m2), m2 = i + 1, i + 2, i + 3, then the judgment value is 0; otherwise the value is 1; Add the six values obtained in the above steps. If the result is greater than 2, then it is determined as a type two noise point and is excluded; Where yp(i) is the current data point and i is the position index of the current data point.

2. The method for measuring and calculating diffuse photon autocorrelation blood flow according to claim 1, wherein: The specific calculation process of step S101 is: Obtain the normalized light intensity autocorrelation function from the experiment, convert the normalized light intensity autocorrelation function into a normalized optical field time autocorrelation curve according to the Siegert relationship, take multiple data points on the normalized optical field time autocorrelation curve, and perform variable substitution on the data points to generate new data points; Where the number of multiple data points is a positive integer not less than 79, and the number of data points + 1 is an integer multiple of 8.

3. The method for measuring and calculating blood flow using diffuse photon autocorrelation according to claim 1, wherein: The specific process of segmenting and multiplying the data points in step S102 is as follows: the data points are segmented according to the delay time, the first sequence after segmentation contains 1 data point, the second sequence contains 14 data points, and the third and subsequent sequences each contain 8 data points. Each data point is multiplied by the corresponding multiplier to obtain an amplified sequence.

4. The method for measuring and calculating blood flow using diffuse photon autocorrelation according to claim 1, wherein: In step S2, the following four arithmetic operations are performed on the operation data point (x(i), y(i)): Calculate the average value of the product of x(i) and y(i), calculate the average value of the square of x(i), calculate the average value of y(i), and calculate the average value of x(i). The characteristic coefficient of the operation data point is calculated through the above four average calculations.

5. An electronic device comprising a processor and a memory in communication with the processor and configured to store instructions executable by the processor, wherein: The processor is used to execute the diffuse photon autocorrelation blood flow measurement calculation method described in any one of claims 1 to 4.

6. A server, characterized in that: The device comprises at least one processor and a memory in communication with the processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor so that the at least one processor performs a diffuse photon autocorrelation blood flow measurement calculation method as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for calculating blood flow measurement based on diffuse photon autocorrelation is implemented.

Images