Blood flow parameter calculation method and device

Abstract

The invention provides a blood flow parameter calculation method and device, and the method comprises the steps: carrying out the rapid scanning of a target blood vessel through a collection device cooperating with a galvanometer, carrying out the splicing preprocessing of a plurality of scanned images obtained after scanning, obtaining a spliced image, capturing a cell stripe angle representing the motion track of cells in the target blood vessel according to the spliced image, and calculating the blood flow parameter. And the blood flow velocity parameter of the target blood vessel is calculated by combining the cell stripe angle, the blood vessel angle, the optical resolution of the acquisition equipment and the scanning frequency of the galvanometer, so that the accurate calculation of the blood flow parameter is realized. According to the method, the blood flow parameters are calculated in a non-invasive mode, a fluorescent contrast agent does not need to be injected into the blood vessel, and the operation difficulty is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and more specifically, to a method and device for calculating blood flow parameters. Background Art

[0002] The blood flow parameters of biological tissues are one of the important indicators for disease diagnosis. Currently, the blood vessels are generally imaged by injecting fluorescent contrast agents into the blood vessels, and the blood flow parameters are calculated based on the blood vessel imaging.

[0003] However, this method of calculating blood flow parameters is invasive and requires the injection of fluorescent contrast agents into blood vessels, which increases the difficulty of the operation. Summary of the invention

[0004] In view of this, the present invention provides a method and device for calculating blood flow parameters. The method for calculating blood flow parameters is non-invasive and does not require the injection of fluorescent contrast agents into blood vessels, thereby reducing the difficulty of operation.

[0005] In order to achieve the above-mentioned invention object, the specific technical solution provided by the present invention is as follows:

[0006] In a first aspect, an embodiment of the present invention provides a method for calculating blood flow parameters, comprising:

[0007] Acquire a target blood vessel image, where the target blood vessel image is a two-dimensional image of the target blood vessel;

[0008] Acquire a stitched image, wherein the stitched image is obtained by stitching and preprocessing a plurality of scanned images obtained after the acquisition device cooperates with the galvanometer to scan the target blood vessel;

[0009] acquiring a blood vessel angle of the target blood vessel in the target blood vessel image;

[0010] Calculating the cell stripe angle in the target blood vessel according to the stitched image;

[0011] The blood flow velocity parameter of the target blood vessel is calculated according to the blood vessel angle, the cell stripe angle, the optical resolution of the acquisition device and the scanning frequency of the galvanometer.

[0012] In some embodiments, before calculating the cell stripe angle in the target blood vessel according to the stitched image, the method further includes:

[0013] Selecting a detection area in the stitched image;

[0014] Determining whether a blood flow signal is collected in the detection area;

[0015] The detection area where no blood flow signal is collected is removed from the stitched image.

[0016] In some embodiments, determining whether a blood flow signal is collected in the detection area includes:

[0017] Selecting at least one sample area in the detection area;

[0018] Determine whether the preset conditions are met based on the brightness statistics of each sample area;

[0019] If it meets the requirements, it is determined whether it is background based on the brightness difference of each row in the detection area;

[0020] If it is not background, it is determined that a blood flow signal is collected in the detection area.

[0021] In some embodiments, the method further comprises:

[0022] selecting an evaluation area in the stitched image;

[0023] The quadrants where the cell stripe angles are distributed in the statistical evaluation area are obtained to obtain statistical results;

[0024] If the statistical result does not meet the preset condition, the target blood vessel is scanned again.

[0025] In some embodiments, the method further comprises optimizing the stitched image:

[0026] Slice the stitched image in a time dimension to obtain a plurality of slice data;

[0027] The slice data are registered to obtain the registered slice data.

[0028] In some embodiments, registering the slice data includes:

[0029] Determine the reference stripe in the shard data;

[0030] Perform registration based on the static structure of blood vessels in the sliced ​​data to obtain the displacement of other strips except the reference strip;

[0031] The slice data is displacement-adjusted according to the displacement amount.

[0032] In some embodiments, after obtaining the registered slice data, the process further includes performing denoising processing on the registered slice data:

[0033] Performing filtering processing on the registered slice data;

[0034] The registered slice data is subtracted from the filtered slice data, and the obtained result is normalized.

[0035] In some embodiments, acquiring the blood vessel angle of the target blood vessel in the target blood vessel image includes:

[0036] Cutting out a vascular image of a preset area around a pre-marked acquisition point in the target vascular image;

[0037] Obtain K first straight lines of different angles through the center point of the blood vessel image, where K is a positive integer;

[0038] For each of the first straight lines, obtain N second straight lines that are perpendicular to the first straight line and overlap with the blood vessel map, and calculate the line integral of the pixel value of each point on the Nth second straight line in the blood vessel map, where N is a positive integer;

[0039] The target blood vessel centerline in the Kth first straight line is determined according to the line integral result, and the angle corresponding to the target blood vessel centerline is used as the blood vessel angle.

[0040] In some embodiments, the step of calculating the cell stripe angle in the target blood vessel according to the stitched image comprises:

[0041] Extracting a valid blood flow signal area in the stitched image according to the blood vessel angle and blood vessel width;

[0042] Acquire multiple regions of interest ROIs of preset sizes in the effective blood flow signal area;

[0043] Calculate the cell stripe angle in each ROI respectively;

[0044] The preset statistical values ​​of the cell stripe angles in the plurality of ROIs are selected according to the preset rules and determined as the cell stripe angles in the target blood vessel.

[0045] In some embodiments, after calculating the blood flow velocity parameter of the target blood vessel, the method further includes:

[0046] On a perpendicular line to the center line of the target blood vessel, searching from the point of the maximum value of the line integral to both ends of the perpendicular line, obtaining a line segment whose line integral result is greater than a threshold, determining the line segment as a blood vessel portion, and determining the length of the line segment as the blood vessel width;

[0047] The blood flow rate of the target blood vessel is calculated based on the blood flow velocity parameter and the blood vessel width.

[0048] In some embodiments, after calculating the cell stripe angle in the target blood vessel according to the stitched image, the method further includes:

[0049] Determine whether the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°;

[0050] If the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°, the blood vessel angle is determined as the blood flow direction of the target blood vessel;

[0051] If the blood vessel angle and the cell stripe angle are not both greater than 90° and not both less than 90°, the sum of the blood vessel angle and 180° is determined as the blood flow direction of the target blood vessel.

[0052] In a second aspect, an embodiment of the present invention provides a blood flow parameter calculation device, comprising:

[0053] A first image acquisition unit, configured to acquire a target blood vessel image, wherein the target blood vessel image is a two-dimensional image of the target blood vessel;

[0054] A second image acquisition unit is used to acquire a spliced ​​image, where the spliced ​​image is obtained by splicing and pre-processing a plurality of scanned images obtained after the acquisition device cooperates with the galvanometer to scan the target blood vessel;

[0055] a blood vessel angle acquisition unit, configured to acquire a blood vessel angle of the target blood vessel in the target blood vessel image;

[0056] A cell stripe angle calculation unit, used to calculate the cell stripe angle in the target blood vessel according to the spliced ​​image;

[0057] The blood flow velocity parameter calculation unit is used to calculate the blood flow velocity parameter of the target blood vessel according to the blood vessel angle, the cell stripe angle, the optical resolution of the acquisition device and the scanning frequency of the galvanometer.

[0058] In some embodiments, the apparatus further comprises:

[0059] A detection area selection unit, used for selecting a detection area in the stitched image;

[0060] A blood flow signal detection unit, used to determine whether a blood flow signal is collected in the detection area;

[0061] The detection area deletion unit is used to remove the detection area where no blood flow signal is collected from the spliced ​​image.

[0062] In some embodiments, the blood flow signal detection unit is specifically used to select at least one sample area in the detection area; determine whether it meets the preset conditions based on the brightness statistics of each sample area; if it does, determine whether it is the background based on the brightness difference of each row in the detection area; if it is not the background, determine that a blood flow signal is collected in the detection area.

[0063] In some embodiments, the apparatus further comprises:

[0064] The scanning image quality evaluation unit is used to select an evaluation area in the spliced ​​image; count the quadrants in which the cell stripe angles in the evaluation area are distributed to obtain statistical results; if the statistical results do not meet the preset conditions, re-scan the target blood vessel.

[0065] In some embodiments, the apparatus further comprises:

[0066] The data registration unit is used to slice the spliced ​​image in the time dimension to obtain a plurality of slice data; and to register the slice data to obtain registered slice data.

[0067] In some embodiments, the data registration unit is specifically used to determine a reference strip in the sliced ​​data; perform registration based on the static structure of the blood vessels in the sliced ​​data to obtain the displacement of other strips except the reference strip; and perform displacement adjustment on the sliced ​​data according to the displacement.

[0068] In some embodiments, the apparatus further comprises:

[0069] The denoising processing unit is used to filter the registered slice data; subtract the filtered slice data from the registered slice data, and normalize the obtained result.

[0070] In some embodiments, the blood vessel angle acquisition unit is specifically used to:

[0071] Cutting out a vascular image of a preset area around a pre-marked acquisition point in the target vascular image;

[0072] Obtain K first straight lines of different angles through the center point of the blood vessel image, where K is a positive integer;

[0073] For each of the first straight lines, obtain N second straight lines that are perpendicular to the first straight line and overlap with the blood vessel map, and calculate the line integral of the pixel values ​​of each point on the N second straight lines in the blood vessel map, where N is a positive integer;

[0074] The target blood vessel center lines among the K first straight lines are determined according to the line integral results, and the angles corresponding to the target blood vessel center lines are used as the blood vessel angles.

[0075] In some embodiments, the cell stripe angle calculation unit is specifically used to:

[0076] Extracting a valid blood flow signal area in the stitched image according to the blood vessel angle and blood vessel width;

[0077] Acquire multiple regions of interest ROIs of preset sizes in the effective blood flow signal area;

[0078] Calculate the cell stripe angle in each ROI respectively;

[0079] The preset statistical values ​​of the cell stripe angles in the plurality of ROIs are selected according to the preset rules and determined as the cell stripe angles in the target blood vessel.

[0080] In some embodiments, the apparatus further comprises:

[0081] A blood flow calculation unit is used to search from the point of the maximum value of the line integral to both ends of the perpendicular line on the perpendicular line of the center line of the target blood vessel, obtain a line segment whose line integral result is greater than a threshold, determine the line segment as a blood vessel portion, and determine the length of the line segment as the blood vessel width; and calculate the blood flow of the target blood vessel based on the blood flow velocity parameter and the blood vessel width.

[0082] In some embodiments, the apparatus further comprises:

[0083] A blood flow direction determination unit is used to determine whether the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°; if the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°, the blood vessel angle is determined as the blood flow direction of the target blood vessel; if the blood vessel angle and the cell stripe angle are not both greater than 90° and not both less than 90°, the sum of the blood vessel angle and 180° is determined as the blood flow direction of the target blood vessel.

[0084] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0085] The present invention discloses a blood flow parameter calculation method and device, which uses an acquisition device in conjunction with a galvanometer to quickly scan a target blood vessel, obtains a spliced ​​image by splicing preprocessing multiple scanned images obtained after scanning, and captures the cell stripe angle representing the movement trajectory of cells in the target blood vessel according to the spliced ​​image, thereby combining the cell stripe angle, the blood vessel angle, the optical resolution of the acquisition device, and the scanning frequency of the galvanometer to calculate the blood flow velocity parameters of the target blood vessel, thereby achieving accurate calculation of the blood flow parameters. The method of calculating blood flow parameters in the present invention is non-invasive, does not require the injection of fluorescent contrast agents into the blood vessels, and reduces the difficulty of operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0087] Figure 1A schematic diagram of a flow chart of a blood flow parameter calculation method disclosed in an embodiment of the present invention;

[0088] Figure 2 A schematic diagram of a blood vessel diagram disclosed in an embodiment of the present invention;

[0089] Figure 3 A schematic diagram of a blood vessel angle disclosed in an embodiment of the present invention;

[0090] Figure 4 A schematic diagram of a line integral principle for calculating a blood vessel angle disclosed in an embodiment of the present invention;

[0091] Figure 5 A schematic diagram of a blood vessel diagram disclosed in an embodiment of the present invention;

[0092] Figure 6 A schematic diagram of a blood vessel line integration result disclosed in an embodiment of the present invention;

[0093] Figure 7 A schematic diagram of a spliced ​​image for representing a continuously changing blood flow signal disclosed in an embodiment of the present invention;

[0094] Figure 8 The embodiment of the present invention discloses Figure 7 The schematic diagram of the stitched image obtained by rotating the stitched image 90° counterclockwise;

[0095] Fig. 9 A schematic diagram showing the comparison before and after the local registration disclosed in an embodiment of the present invention;

[0096] Fig.10 A schematic diagram of sliced ​​data after removing the static background according to an embodiment of the present invention;

[0097] Fig.11 A schematic diagram of an effective blood flow signal region disclosed in an embodiment of the present invention;

[0098] Fig.12 This is a schematic diagram of the cell stripe angle β disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0099] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0100] The present invention provides a blood flow parameter calculation method and device, which are applied to an electronic device with an image processing function, and the electronic device is communicatively connected with an acquisition device and a galvanometer, and the acquisition device is controlled to cooperate with the galvanometer to quickly scan a target blood vessel. A spliced ​​image is obtained by splicing and preprocessing multiple scanned images obtained after scanning, and the cell stripe angle representing the movement trajectory of cells in the target blood vessel is captured according to the spliced ​​image, so as to calculate the blood flow velocity parameters of the target blood vessel in combination with the cell stripe angle, the blood vessel angle, the optical resolution of the acquisition device and the scanning frequency of the galvanometer, so as to achieve accurate calculation of the blood flow parameters to assist in disease diagnosis.

[0101] See also Figure 1 This embodiment discloses a blood flow parameter calculation method, which specifically includes the following steps:

[0102] S101: Acquire a target blood vessel image, where the target blood vessel image is a two-dimensional image of the target blood vessel.

[0103] The target blood vessel is a fundus blood vessel of a human or animal, and there are many ways to obtain an image of the target blood vessel.

[0104] Exemplarily, a linear array camera is used in conjunction with a dual galvanometer to capture a target blood vessel image, wherein one galvanometer of the dual galvanometer is arranged in a vertical direction, and the other galvanometer is arranged in a horizontal direction.

[0105] Exemplarily, an area array camera is used to capture a target blood vessel image. The area array camera can be pre-adjusted or calibrated so that the target blood vessel image captured by the area array camera is consistent in direction with the stitched image or their conversion relationship is known, so that the angle of the target blood vessel in the target blood vessel image is consistent with that of the target blood vessel in the stitched image or they can be converted to each other.

[0106] S102: Acquire a stitched image. The stitched image is obtained by stitching and preprocessing a plurality of scanned images obtained after the acquisition device cooperates with the galvanometer to scan the target blood vessel.

[0107] Exemplarily, the acquisition device may be a photodetector, which includes but is not limited to a photomultiplier tube, an avalanche photodiode, etc. The acquisition device may also be a linear array camera.

[0108] There can be two scanning galvanometers in the system to facilitate the acquisition of two-dimensional scanning images. However, when executing this method, the movement of the scanning galvanometer in one direction of the system (such as a vertical scanning galvanometer) can be stopped at the target blood vessel position, and only the galvanometer in another direction (such as a horizontal scanning galvanometer) is retained for scanning. In this way, the line scanning image information of high-speed horizontal scanning can be obtained based on the reflection signal detected by the light detector, and the splicing preprocessing can be performed according to the time sequence of the acquisition to obtain the spliced ​​image. The "horizontal" and "vertical" are only used to indicate the two directions of the two-dimensional rotation of the galvanometer, and do not limit its association with the actual horizontal plane and vertical plane.

[0109] The galvanometer moves along a straight line. Every time the galvanometer moves to a point, the acquisition device acquires the image data of the point, and this process is repeated until a line of data is scanned. In this embodiment, a line of data includes 1024 points. Then the galvanometer returns, and data is also acquired on the return journey. The working method is consistent with the above scanning method until a preset number of lines, such as 1024 lines, are acquired to form a scanned image. Multiple scanned images are spliced ​​and pre-processed to obtain a spliced ​​image. Alternatively, one line of data can be used as a scanned image, and the scanned images of the preset line are spliced ​​and pre-processed to obtain a spliced ​​image. The number of lines of the spliced ​​image is set as needed to observe the cell stripes that characterize the blood flow velocity. Different devices have different resolutions and acquisition frequencies, so the number of lines is set differently. Alternatively, there is no limit on the number of lines, and splicing is continued for real-time calculation to monitor the blood flow parameters in real time.

[0110] The sampling frequency of the galvanometer needs to meet the accuracy requirements of the actual application scenario, such as 8 kHz. The higher the accuracy requirement, the higher the sampling frequency of the galvanometer. The present invention does not make any specific limitation.

[0111] S103: Acquire the blood vessel angle of the target blood vessel in the target blood vessel image.

[0112] In this embodiment, the vessel angle of the target vessel is determined as the angle formed by taking the intersection of the center line of the target vessel and the horizontal direction as the origin, the horizontal line where the origin is located as the X-axis, and the positive direction of the X-axis as the starting line along the counterclockwise direction to the center line of the target vessel. Other angles can also be defined by oneself.

[0113] The vascular angle of the target blood vessel can be obtained based on input data, which is obtained by measuring the target blood vessel image manually or manually in combination with auxiliary measurement software. The vascular angle of the target blood vessel can also be obtained by line integral calculation, which is introduced below through a specific embodiment.

[0114] S104: calculating the cell stripe angle in the target blood vessel according to the stitched image;

[0115] Exemplarily, the stitched image includes multiple time slices in the scanning time dimension (a time slice is a scanning image composed of a preset number of rows of data), and a time slice includes multiple cells. The cell stripe angle in the target blood vessel here can be a preset statistical value of the multiple cell stripe angles in the time slice, such as the cell stripe angle that appears most frequently among the multiple cell stripes in the time slice, the average of the multiple cell stripe angles, the mode among the multiple cell stripe angles, etc.

[0116] Exemplarily, the cell stripe angle in the target blood vessel here may also be the stripe angle of a certain cell in the stitched image.

[0117] S105: Calculate the blood flow velocity parameters of the target blood vessel according to the blood vessel angle, the cell stripe angle, the optical resolution of the acquisition device, and the scanning frequency of the galvanometer.

[0118] An optional method of determining the blood flow velocity of a target vessel is as follows:

[0119]

[0120] Among them, V is the blood flow velocity, theta is the cell stripe angle, α is the blood vessel angle, k is the optical resolution of the acquisition device, that is, the resolution of the acquisition device in the direction of the galvanometer movement divided by the micrometer size of the physical image in the area that the galvanometer can scan, and frequence is the scanning frequency of the galvanometer.

[0121] It can be understood that if the above-mentioned cell stripe angle theta is a preset statistical value of multiple cell stripe angles in a time slice, the calculated blood flow velocity is the blood flow velocity of the time slice, and on this basis, the blood flow velocities of all time slices in the stitched image can be counted; if the above-mentioned cell stripe angle theta is the stripe angle of a certain cell in the stitched image, the calculated blood flow velocity is the blood flow velocity of the cell, and on this basis, the blood flow velocity of each cell in the time slice can be counted, and the blood flow velocity of all time slices in the stitched image can be further counted.

[0122] After obtaining the blood flow velocity of all time slices in the stitched image, the blood flow velocity of all time slices in the stitched image can be statistically calculated, such as maximum value, minimum value, mean value, standard deviation, confidence level, etc. These data are of great reference value for assisting disease diagnosis. Moreover, when the sampling rate is high, the blood flow velocity change curve can also be drawn to provide more comprehensive data for assisting disease diagnosis and real-time status monitoring.

[0123] It can be seen that the blood flow parameter calculation method disclosed in this embodiment uses an acquisition device in conjunction with a galvanometer to quickly scan a target blood vessel, obtains a spliced ​​image by splicing preprocessing multiple scanned images obtained after scanning, and captures the cell stripe angle representing the movement trajectory of cells in the target blood vessel according to the spliced ​​image, thereby combining the cell stripe angle, the blood vessel angle, the optical resolution of the acquisition device, and the scanning frequency of the galvanometer to calculate the blood flow velocity parameters of the target blood vessel, thereby achieving accurate calculation of the blood flow parameters. Since the method of calculating blood flow parameters is non-invasive, it is not necessary to inject fluorescent contrast agents into the blood vessels, which reduces the difficulty of operation.

[0124] In the above embodiment S103: obtaining the blood vessel angle of the target blood vessel in the target blood vessel image, an optional implementation method includes the following steps:

[0125] A1: cropping the vascular image of the preset area around the pre-marked acquisition point in the target vascular image;

[0126] like Figure 2 As shown, Figure 2 The blood vessel in is the target blood vessel, and P_vessel(x,y) is a pre-marked acquisition point located in the blood vessel lumen. In the process of acquiring the blood vessel angle of the target blood vessel, in order to reduce the amount of data processing, the blood vessel image Img_vessel of the preset area around P_vessel(x,y) is cropped in the target blood vessel image for subsequent processing.

[0127] The size of the preset area can be set according to the actual application scenario, and the present invention does not make any specific limitation.

[0128] like Figure 3 As shown in FIG. 1 , the vascular angle of the target vessel is the angle VesselAngle of the vessel at the P_vessel point, denoted as α, that is, the intersection point P_vessel of the center line of the target vessel and the horizontal line is taken as the origin, the horizontal line is taken as the X-axis, the positive direction of the X-axis is taken as the starting line, and the angle is formed in the counterclockwise direction to the center line of the target vessel.

[0129] A2: Obtain K first straight lines of different angles through the center point of the blood vessel image, where K is a positive integer;

[0130] A3: for each of the first straight lines, obtain N second straight lines that are perpendicular to the first straight line and overlap with the vascular image, and calculate the line integral of the pixel values ​​of each point on the N second straight lines in the vascular image, where N is a positive integer;

[0131] A4: Determine the target blood vessel center lines among the K first straight lines according to the line integral result, and use the angle corresponding to the target blood vessel center lines as the blood vessel angle.

[0132] Among them, the maximum value of K is 180, which corresponds to the central point of the blood vessel map being in the range of 0 to 180°, and selecting a first straight line at an interval of 1°. In addition, the angle step can be set, and a first straight line is selected at an interval of 1 angle step, and the value of K is reduced accordingly; the maximum value of N can be the number of pixels on the corresponding Kth first straight line. For example, if the angle corresponding to the Kth first straight line is 90°, and there are 1024 pixels on the first straight line, then N can be a maximum of 1024, that is, the line integral of the pixel values ​​of each point on each second straight line in the blood vessel map on these 1024 second straight lines is obtained respectively. In addition, the pixel step can also be set, for example, selecting a second straight line at an interval of 2 pixels to calculate the line integral, then at this 90° angle, 512 line integral results can be obtained in sequence.

[0133] See also Figure 4, K first straight lines pass through the center point of the vascular map. The angle between the Kth first straight line Lk shown in the figure and the horizontal line is ρ, and the value range of ρ is [0°, 179°]. For the first straight line Lk, N second straight lines perpendicular to Lk and overlapping with the vascular map are obtained. The straight lines L1, Lm, and Ln in the figure are all for illustration. It can be seen that there are many straight lines parallel to L1 and intersecting with the vascular map that are not shown.

[0134] Calculate the line integral of the pixel value of each point on the N second straight lines in the vascular map respectively. For example, calculate the line integral of the pixel value of each point on the overlapping part of L1 and the vascular map, that is, calculate the sum of the pixel values ​​of each pixel point on the gray line segment shown in the dotted box 401 in the figure, and so on, calculate the line integral of the N second straight lines respectively. In this embodiment, the line integral value of each second straight line is displayed in the image in sequence to form a curve 402. The curve 402 reflects the integral size of the pixel value of the vascular map on the N second straight lines perpendicular to Lk. The figure is only for illustration and is not an intermediate product that must be obtained by this method.

[0135] According to all the above line integral results, the center line of the blood vessel in the K first straight lines is determined, and the angle corresponding to the center line of the blood vessel is used as the blood vessel angle. If each of the K first straight lines corresponds to N perpendicular lines, there are K*N line integral results in total. If the number of perpendicular lines taken in each of the K first straight lines is different, the total number of line integral results obtained is the sum of the number of all perpendicular lines.

[0136] like Figure 5 As shown, in this embodiment, for the convenience of calculation, a MASK mask operation is performed on the vascular image, and the vascular image is cropped into a circle. In this way, no matter how the angle of the center point of the vascular image is changed, the number of pixel points on the straight line is consistent, so that the same number of perpendicular lines N can be taken for the K first straight lines, and the obtained line integral data is convenient for statistics.

[0137] According to all the above line integral results, the blood vessel center lines among the K first straight lines are determined. The specific method adopted in this embodiment is: Figure 5 The line integral result at each angle in the vascular image shown is taken as a column of data. All the columns of data are combined in the order of the angle ρ, which is another form of the vascular image, such as Figure 6 shown. Figure 6 The horizontal axis represents the angle, the vertical axis represents the straight lines arranged in order of position at the angle, and the gray value of each point in the figure represents the line integral value.

[0138] Afterwards, you can Figure 6Calculate the variance of each column in , determine the angle corresponding to the column with the largest variance, take the angle perpendicular to this angle as the blood vessel angle, and take the first straight line corresponding to the blood vessel angle as the blood vessel centerline; this utilizes the prior knowledge that the integral difference of pixel values ​​on each line parallel to the blood vessel direction is the largest.

[0139] You can also get the point with the highest gray value in the image, that is Figure 6 The brightest point in the image is then determined, and the angle corresponding to the column where the point is located is determined. The angle perpendicular to the angle is taken as the blood vessel angle, and the first straight line corresponding to the blood vessel angle is taken as the blood vessel centerline. This utilizes the prior knowledge that the blood vessel centerline has the highest brightness (i.e., pixel value).

[0140] For example Figure 4 The line integral variance of the N second straight lines corresponding to Lk is the largest. When Lk is 60°, the second straight line perpendicular to Lk, that is, the second straight line at 150°, can be determined as the blood vessel centerline, and 150° is the blood vessel angle.

[0141] Those skilled in the art can also use other prior knowledge of the vascular map to obtain other methods for determining the vascular centerline and vascular angle based on the line integral results, such as calculating the mean and standard deviation of all line integral results, etc., not limited to the above two methods. Figure 6 The manner of characterizing these line integrals by using rows, columns and grayscale values ​​is only an example and is not an intermediate product that must be obtained by the present method.

[0142] According to the line integral result, not only the center line and the angle of the target blood vessel can be determined, but also the perpendicular line to the center line of the blood vessel can be searched from the point of the maximum value of the line integral to both ends of the perpendicular line to obtain the line segment whose line integral result is greater than the threshold value, and the line segment is determined as the blood vessel part, and the length of the line segment is the blood vessel width d. Based on this, the blood vessel part and the non-blood vessel part can be determined, and the blood vessel width d provides a reference for the subsequent clipping of blood flow signals and the calculation of blood flow parameters.

[0143] After the vascular angle of the target blood vessel in the target blood vessel image is obtained, the above-mentioned embodiment S104 is executed: the cell stripe angle in the target blood vessel is calculated according to the spliced ​​image. If the target blood vessel is a human fundus blood vessel, blinking, eye movement, etc. may occur when the fundus blood vessel image is collected. The scanned image collected during blinking has no blood flow signal, and the scanned image collected during eye movement has poor quality, which will affect the accuracy of the subsequent calculation of the cell stripe angle in the target blood vessel.

[0144] In order to solve this technical problem, two optimization methods are provided below based on the above embodiments. Among them, optimization method 1 is used to solve the problem that the scanned image collected has no blood flow signal due to blinking, and optimization method 2 is used to solve the problem that the scanned image collected has poor quality due to eye movement. It should be noted that the following two optimization methods can be implemented separately or in combination.

[0145] Optimization method 1

[0146] Before the above embodiment S104, the following steps are also included:

[0147] B1: Select the detection area in the stitched image;

[0148] The detection area may be a scanned image of a preset number of lines, such as 1024 lines of data.

[0149] B2: Determine whether blood flow signals are collected in the detection area;

[0150] B3: Remove the detection area where no blood flow signal is collected from the stitched image.

[0151] Among them, B2 can be implemented in multiple ways, and an optional implementation is as follows:

[0152] Selecting at least one sample area in the detection area;

[0153] Determine whether the preset conditions are met based on the brightness statistics of each sample area;

[0154] If it meets the requirements, it is determined whether it is background based on the brightness difference of each row in the detection area;

[0155] If it is not background, it is determined that blood flow signals are collected in the detection area.

[0156] Exemplarily, the brightness statistics of each sample area is the sum of the brightness of each sample area, and the preset condition is that the sum of the brightness of each sample area is greater than a first threshold.

[0157] Exemplarily, the brightness statistics of each sample area are the average brightness of each sample area, and the preset condition is that the average brightness of each sample area is greater than the second threshold.

[0158] Exemplarily, the method for determining whether it is the background based on the brightness difference of each row in the detection area is: first, smooth the detection area, and then subtract the image after the smoothing operation from the image before the smoothing operation to remove the static noise of the background; then, statistics are performed on each row of data in the image after the smoothing operation to find out the proportion of blood flow signals that meet the requirements. If the proportion and the signal intensity (i.e., brightness) both meet the settings, it is determined that the frame has a vascular signal and the quality is good.

[0159] By removing the detection area where no blood flow signal is collected from the stitched image, interference with the overall calculation of blood flow parameters is avoided and the accuracy of subsequent image processing is improved.

[0160] Optimization method 2

[0161] The second optimization method is used to evaluate the quality of the scanned image, and further includes the following steps before and / or after the above embodiment S104:

[0162] C1: Select the evaluation area in the stitched image;

[0163] If the optimization method 2 is performed before the above-mentioned embodiment S104, due to the influence of eye movement, the initial image quality of the acquired scanned image is often the best. If the image quality in the initial evaluation area does not meet the requirements, it means that the overall image quality is not good. It can be directly rescanned without evaluating or calculating the entire image, thus saving processing time.

[0164] If the second optimization method is performed after S104 in the above embodiment, it is preferred to evaluate the entire stitched image and select an evaluation area from the entire stitched image.

[0165] Exemplarily, the evaluation area is selected from the stitched image using a sliding window in the order from top to bottom and from left to right. Adjacent evaluation areas may or may not overlap. If there is overlap, the overlap ratio can be set to [0,1]. It can be understood that if adjacent evaluation areas overlap, the number of evaluation areas obtained is relatively large, the processing speed is relatively slow, but the processing accuracy is relatively high; if adjacent evaluation areas do not overlap, the number of evaluation areas obtained is relatively small, the processing speed is relatively fast, but the processing accuracy is relatively low. Whether adjacent evaluation areas overlap in the sliding window and the size of the overlapping area are flexibly set according to the requirements of the processing speed and processing accuracy in the actual application scenario.

[0166] C2: Statistically evaluate the quadrants in which the cell stripe angles are distributed and obtain statistical results;

[0167] The blood flow situation determines that the angle of the cell stripes must be tilted, not horizontal or vertical. The tilt angle represents the speed of blood flow. And because the blood flow in the blood vessels is unidirectional and there is no reverse flow, the collected cell stripe angles should be distributed in the same quadrant under ideal conditions. Therefore, to evaluate the image quality, it is only necessary to count the quadrants in which the cell stripe angles in the evaluation area are distributed. If the cell stripe angles are all in [0°, 90°) or (90°, 180°], it means that the blood flow signal quality in the blood vessel is good and the stripes are clear and consistent.

[0168] C3: If the statistical result does not meet the preset conditions, the target blood vessel is scanned again.

[0169] Considering the influence of factors such as the environment and the error of the acquisition equipment, the preset conditions can be set according to the actual application scenario. For example, whether the proportion of cell stripe angles in the same quadrant exceeds the threshold, or whether there are a preset number of consecutive cell stripe angles in the same quadrant, or whether there are a preset number of cumulative cell stripe angles in the same quadrant.

[0170] Exemplarily, the preset condition is that a preset proportion of cell stripe angles in the evaluation area are in the same quadrant. For example, if 1000 cell stripes are obtained in the evaluation area, if 900 of the cell stripe angles are in the same quadrant, that is, 90% of the cell stripe angles are in the same quadrant, then the preset condition is met; if 500 cell stripe angles are distributed in different quadrants from the other 500 cell stripe angles, then the preset condition is not met.

[0171] Exemplarily, the preset condition is that a preset number of cell stripe angles are in the same quadrant in a row, such as 700 consecutive cell stripe angles are in the same quadrant.

[0172] In order to increase visibility, a cell stripe angle distribution diagram can be drawn. If the statistical results do not meet the preset conditions, that is, there is no high consistency in the cell stripe angles, the data quality is considered to be poor and the user needs to be fed back to rescan.

[0173] It should be noted that if the above optimization method 2 is evaluated before and after the above embodiment S104, the quadrant in which the cell stripe angles in the evaluation area are statistically distributed during the evaluation before the above embodiment S104 is used as a benchmark to evaluate the degree to which all stripe angles after the above embodiment S104 are consistent with the benchmark. In this way, the quadrant benchmark can be determined when the initial image is optimal, the evaluation rate can be improved, and the evaluation error in the later evaluation area can be reduced.

[0174] Furthermore, this embodiment also discloses a method for optimizing the processing of the stitched images, registering the stitched images and eliminating the position deviation in the stitched images, which specifically includes the following steps:

[0175] D1: Slice the spliced ​​image in the time dimension to obtain multiple slice data;

[0176] In order to make the stitched image reflect the movement trajectory of cells in the target blood vessel, the stitched image is obtained by stitching multiple scanned images in the order of scanning time. After stitching, the image is rotated 90° counterclockwise. After rotation, the horizontal direction represents the time dimension, which is convenient for observation and processing.

[0177] Taking the scanning frequency of the galvanometer frequency = 8 khz, and collecting 8000 lines of data every 1 second as an example, the scanning images collected in 1 second are spliced ​​according to the order of scanning time to obtain the continuously changing blood flow signal as shown in Figure 7 As shown, the spliced ​​image is rotated 90° counterclockwise to obtain Figure 8 Stitched image shown.

[0178] Exemplarily, the slice window size is set to frequency / 100, and the spliced ​​image is sliced ​​in the time dimension to obtain a plurality of slice data dataOneStrip.

[0179] D2: Register the slice data to obtain the registered slice data.

[0180] Since there is a certain time difference between two adjacent line scan signals, such as Fig. 9 The schematic diagram of the comparison before and after local registration is shown. The horizontal direction represents the time dimension. Fig. 9 The image on the left is the image before local registration. There is a position deviation in the horizontal direction, so registration is needed to eliminate the position deviation. Specifically, by registering the static structure of the fundus, we can obtain Fig. 9 The image on the right shows the image after local registration. It can be seen that the cell vascular area after registration eliminates jitter and is smoother, which can improve the accuracy of cell vascular parameter calculation. In addition, compared with the solution of directly registering all slices without segmentation, this solution can take into account the reasonable differences caused by vascular jitter between each time slice. The registration result can not only meet the calculation requirements of cell stripe angles, but also be more suitable for actual application scenarios, further improving the calculation accuracy of blood flow parameters.

[0181] Exemplarily, for each slice data dataOneStrip, the NCC (normalized crosscorrelation, normalized cross correlation matching method) algorithm is used for registration, including the following steps:

[0182] E1: Determine the reference stripe in the shard data;

[0183] Preferably, the middle column of the sliced ​​data dataOneStrip is used as the reference strip.

[0184] Of course, other columns except the middle column in the sliced ​​data dataOneStrip may also be selected as reference frames, such as the first column as the reference strip.

[0185] E2: Perform registration based on the static structure of blood vessels in the sliced ​​data to obtain the displacement of other strips except the reference strip;

[0186] The static structures located in the upper and lower parts of the blood vessels in the sliced ​​data dataOneStrip are respectively calculated by NCC to obtain the displacement of other strips in the sliced ​​data dataOneStrip except the reference strip. Registration based on the static structure of the blood vessels in the sliced ​​data can reduce the error caused by the registration of the dynamic part of the blood flow and improve the accuracy of the registration.

[0187] E3: Adjust the displacement of the slice data according to the displacement amount.

[0188] According to the displacement of other strips except the reference strip, a circular displacement operation is performed on the blood flow data in the column dimension to obtain the segmented data after registration, which is used to represent the blood flow signal.

[0189] Furthermore, the registered slice data also includes the static background except the target blood vessel. In order to improve the accuracy of the subsequent calculation of the cell stripe angle in the target blood vessel based on the stitched image and improve the calculation efficiency, the static background needs to be removed.

[0190] Exemplarily, after obtaining the registered slice data, the process further includes performing denoising on the registered slice data to remove the static background, which specifically includes the following steps:

[0191] F1: Filter the registered slice data;

[0192] A one-dimensional mean filtering operation is performed on the slice data dataOneStrip in a direction perpendicular to the scanning direction, and the filter kernel size is 1*N, where N is a positive integer, to obtain a filtered image f(x, y).

[0193]

[0194] F2: Subtract the filtered slice data from the registered slice data and normalize the result.

[0195] The blood flow signal S is obtained by subtracting the image before filtering from the image after filtering and normalizing it, as shown in Fig.10 shown.

[0196] S = Norm(data-f(x,y))

[0197] Among them, data represents the image before filtering.

[0198] After the above optimization processing, S104 can be executed: calculating the cell stripe angle in the target blood vessel according to the optimized spliced ​​image. An optional implementation method specifically includes the following steps:

[0199] G1: Extract the effective blood flow signal area in the spliced ​​image according to the blood vessel angle and blood vessel width;

[0200] For the analysis of most capillaries, it is necessary to crop the blood flow source data to delete the interference data and extract the effective blood flow signal region. Exemplarily, extend 0.5*d / sin(γ) lengths upward and downward at the x coordinate of the acquisition point P_vessel on the stitched image composed of the sliced data after the above background denoising process, where γ is related to VesselAngle, as Figure 3 shown. If 0 < VesselAngle < 90°, then γ = VesselAngle; if 90° < VesselAngle < 180°, γ = 180° - VesselAngle.

[0201] Obtain the blood flow signal distribution width W = d / sin(γ) in the target blood vessel, where d is the width of the target blood vessel. Delete the data outside this range and perform non-maximum suppression in the direction perpendicular to the scanning direction. The specific formula is as follows:

[0202]

[0203] where I(x, y) represents the coordinates of the points in the effective blood flow signal region after non-maximum suppression. The blood flow data after non-maximum suppression is as Fig.11 shown.

[0204] G2: Obtain multiple regions of interest (ROIs) with preset sizes in the effective blood flow signal region;

[0205] Select multiple regions of interest (ROIs) with preset sizes in the effective blood flow signal region. Exemplarily, the ROI can outline the area to be processed in the form of a square, circle, ellipse, irregular polygon, etc. In this embodiment, there is only one cell and only one cell stripe in each selected ROI.

[0206] Exemplarily, if the stitched image where the above effective blood flow signal region is located undergoes the scanning image quality evaluation described in C1 to C3, the method for obtaining multiple regions of interest (ROIs) with preset sizes can be the same as the method for selecting the evaluation region in C1, that is, select the ROIs by means of a sliding window. Adjacent ROIs may overlap or may not overlap. If there is overlap, the overlap ratio can be set to [0, 1].

[0207] Exemplarily, obtain m*n ROIs, where m columns of ROIs are obtained in the time direction and n rows of ROIs are obtained in the position direction.

[0208] G3: Calculate the cell stripe angles in each ROI respectively;

[0209] The method for calculating the cell stripe angle is similar to the method for calculating the blood vessel angle of the target blood vessel, and can be calculated by line integral. In this embodiment, the cell stripe angle β is defined as Fig.12 As shown, no further description is given here.

[0210] The cell stripe angle in each ROI is calculated separately to obtain a set of cell stripe angles in each ROI {β1, β2,…, βm*n}.

[0211] G4: According to a preset rule, a preset statistical value of the cell stripe angles in multiple ROIs is selected to be determined as the cell stripe angle in the target blood vessel.

[0212] Exemplarily, the cell stripe angle that appears most frequently among the cell stripe angles in multiple ROIs is used as a preset statistical value.

[0213] Exemplarily, the average value of the cell stripe angles in multiple ROIs is used as the preset statistical value.

[0214] In the above embodiment S104: after calculating the cell stripe angle in the target blood vessel according to the stitched image, the blood flow direction may be determined based on the blood vessel angle α and the cell stripe angle β.

[0215] Specifically, determine whether the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°; if the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°, determine the blood vessel angle as the blood flow direction of the target blood vessel; if the blood vessel angle and the cell stripe angle are not both greater than 90° and not both less than 90°, determine the sum of the blood vessel angle and 180° as the blood flow direction of the target blood vessel.

[0216] Blood flow directionbloodOren:

[0217]

[0218] According to the direction of blood flow, it can assist in determining whether the target blood vessel is an artery or a vein, further assisting in observation, status monitoring and disease diagnosis.

[0219] In the above embodiment S105: after calculating the blood flow velocity parameter of the target blood vessel, the blood flow rate bloodflowrate may also be calculated according to the blood flow velocity V and the blood vessel width d.

[0220]

[0221] The present invention discloses a blood flow parameter calculation method, which can calculate the blood flow direction and blood volume after calculating the blood flow velocity parameters of the target blood vessel, thereby achieving accurate calculation of the blood flow parameters to assist in disease diagnosis.

[0222] Based on the blood flow parameter calculation method disclosed in the above embodiment, this embodiment correspondingly discloses a blood flow parameter calculation device, which specifically includes:

[0223] A first image acquisition unit, configured to acquire a target blood vessel image, wherein the target blood vessel image is a two-dimensional image of the target blood vessel;

[0224] A second image acquisition unit is used to acquire a spliced ​​image, where the spliced ​​image is obtained by splicing and pre-processing a plurality of scanned images obtained after the acquisition device cooperates with the galvanometer to scan the target blood vessel;

[0225] a blood vessel angle acquisition unit, configured to acquire a blood vessel angle of the target blood vessel in the target blood vessel image;

[0226] A cell stripe angle calculation unit, used to calculate the cell stripe angle in the target blood vessel according to the spliced ​​image;

[0227] The blood flow velocity parameter calculation unit is used to calculate the blood flow velocity parameter of the target blood vessel according to the blood vessel angle, the cell stripe angle, the optical resolution of the acquisition device and the scanning frequency of the galvanometer.

[0228] In some embodiments, the apparatus further comprises:

[0229] A detection area selection unit, used for selecting a detection area in the stitched image;

[0230] A blood flow signal detection unit, used to determine whether a blood flow signal is collected in the detection area;

[0231] The detection area deletion unit is used to remove the detection area where no blood flow signal is collected from the spliced ​​image.

[0232] In some embodiments, the blood flow signal detection unit is specifically used to select at least one sample area in the detection area; determine whether it meets the preset conditions based on the brightness statistics of each sample area; if it does, determine whether it is the background based on the brightness difference of each row in the detection area; if it is not the background, determine that a blood flow signal is collected in the detection area.

[0233] In some embodiments, the apparatus further comprises:

[0234] The scanning image quality evaluation unit is used to select an evaluation area in the spliced ​​image; count the quadrants in which the cell stripe angles in the evaluation area are distributed to obtain statistical results; if the statistical results do not meet the preset conditions, re-scan the target blood vessel.

[0235] In some embodiments, the apparatus further comprises:

[0236] The data registration unit is used to slice the spliced ​​image in the time dimension to obtain a plurality of slice data; and to register the slice data to obtain registered slice data.

[0237] In some embodiments, the data registration unit is specifically used to determine a reference strip in the sliced ​​data; perform registration based on the static structure of the blood vessels in the sliced ​​data to obtain the displacement of other strips except the reference strip; and perform displacement adjustment on the sliced ​​data according to the displacement.

[0238] In some embodiments, the apparatus further comprises:

[0239] The denoising processing unit is used to filter the registered slice data; subtract the filtered slice data from the registered slice data, and normalize the obtained result.

[0240] In some embodiments, the blood vessel angle acquisition unit is specifically used to:

[0241] Cutting out a vascular image of a preset area around a pre-marked acquisition point in the target vascular image;

[0242] Obtain K first straight lines of different angles through the center point of the blood vessel image, where K is a positive integer;

[0243] For each of the first straight lines, obtain N second straight lines that are perpendicular to the first straight line and overlap with the blood vessel map, and calculate the line integral of the pixel values ​​of each point on the N second straight lines in the blood vessel map, where N is a positive integer;

[0244] The target blood vessel center lines among the K first straight lines are determined according to the line integral results, and the angles corresponding to the target blood vessel center lines are used as the blood vessel angles.

[0245] In some embodiments, the cell stripe angle calculation unit is specifically used to:

[0246] Extracting a valid blood flow signal area in the stitched image according to the blood vessel angle and blood vessel width;

[0247] Acquire multiple regions of interest ROIs of preset sizes in the effective blood flow signal area;

[0248] Calculate the cell stripe angle in each ROI respectively;

[0249] The preset statistical values ​​of the cell stripe angles in the plurality of ROIs are selected according to the preset rules and determined as the cell stripe angles in the target blood vessel.

[0250] In some embodiments, the apparatus further comprises:

[0251] A blood flow calculation unit is used to search from the point of the maximum value of the line integral to both ends of the perpendicular line on the perpendicular line of the center line of the target blood vessel, obtain a line segment whose line integral result is greater than a threshold, determine the line segment as a blood vessel portion, and determine the length of the line segment as the blood vessel width; and calculate the blood flow of the target blood vessel based on the blood flow velocity parameter and the blood vessel width.

[0252] In some embodiments, the apparatus further comprises:

[0253] A blood flow direction determination unit is used to determine whether the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°; if the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°, the blood vessel angle is determined as the blood flow direction of the target blood vessel; if the blood vessel angle and the cell stripe angle are not both greater than 90° and not both less than 90°, the sum of the blood vessel angle and 180° is determined as the blood flow direction of the target blood vessel.

[0254] The present embodiment discloses a blood flow parameter calculation device, which uses an acquisition device in conjunction with a galvanometer to quickly scan a target blood vessel, obtains a spliced ​​image by splicing and preprocessing multiple scanned images obtained after scanning, and captures the cell stripe angle representing the movement trajectory of cells in the target blood vessel according to the spliced ​​image, thereby combining the cell stripe angle, the blood vessel angle, the optical resolution of the acquisition device, and the scanning frequency of the galvanometer to calculate the blood flow velocity parameters of the target blood vessel, thereby achieving accurate calculation of the blood flow parameters. The method of calculating blood flow parameters in the present invention is non-invasive, does not require the injection of fluorescent contrast agents into the blood vessels, and reduces the difficulty of operation.

[0255] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0256] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0257] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0258] The above-mentioned embodiments may be combined arbitrarily. For the above description of the disclosed embodiments, the features recorded in the embodiments in this specification may be replaced or combined with each other, so that professional and technical personnel in this field can implement or use the present application.

[0259] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calculating blood flow parameters, characterized in that: include: Acquire a target blood vessel image, where the target blood vessel image is a two-dimensional image of the target blood vessel; Acquire a stitched image, wherein the stitched image is obtained by stitching and preprocessing a plurality of scanned images obtained after the acquisition device cooperates with the galvanometer to scan the target blood vessel; acquiring a blood vessel angle of the target blood vessel in the target blood vessel image; Calculating the cell stripe angle in the target blood vessel according to the stitched image; The blood flow velocity parameter of the target blood vessel is calculated according to the blood vessel angle, the cell stripe angle, the optical resolution of the acquisition device and the scanning frequency of the galvanometer.

2. The method according to claim 1, characterized in that Before calculating the cell stripe angle in the target blood vessel according to the spliced ​​image, the method further includes: Selecting a detection area in the stitched image; Determining whether a blood flow signal is collected in the detection area; The detection area where no blood flow signal is collected is removed from the stitched image.

3. The method according to claim 2, characterized in that The determining whether a blood flow signal is collected in the detection area includes: Selecting at least one sample area in the detection area; Determine whether the preset conditions are met based on the brightness statistics of each sample area; If it meets the requirements, it is determined whether it is background based on the brightness difference of each row in the detection area; If it is not background, it is determined that a blood flow signal is collected in the detection area.

4. The method according to claim 1, characterized in that The method further comprises: selecting an evaluation area in the stitched image; The quadrants where the cell stripe angles are distributed in the statistical evaluation area are obtained to obtain statistical results; If the statistical result does not meet the preset condition, the target blood vessel is scanned again.

5. The method according to claim 1, characterized in that The method further comprises optimizing the stitched image: Slice the stitched image in a time dimension to obtain a plurality of slice data; The slice data are registered to obtain the registered slice data.

6. The method according to claim 5, characterized in that The registering the slice data comprises: Determine the reference stripe in the shard data; Perform registration based on the static structure of blood vessels in the sliced ​​data to obtain the displacement of other strips except the reference strip; The slice data is displacement-adjusted according to the displacement amount.

7. The method according to claim 5, characterized in that After obtaining the registered slice data, the process further includes performing denoising processing on the registered slice data: Performing filtering processing on the registered slice data; The registered slice data is subtracted from the filtered slice data, and the obtained result is normalized.

8. The method according to claim 1, characterized in that The acquiring the blood vessel angle of the target blood vessel in the target blood vessel image includes: Cutting out a vascular image of a preset area around a pre-marked acquisition point in the target vascular image; Obtain K first straight lines of different angles through the center point of the blood vessel image, where K is a positive integer; For each of the first straight lines, obtain N second straight lines that are perpendicular to the first straight line and overlap with the blood vessel map, and calculate the line integral of the pixel values ​​of each point on the N second straight lines in the blood vessel map, where N is a positive integer; The target blood vessel center lines among the K first straight lines are determined according to the line integral results, and the angles corresponding to the target blood vessel center lines are used as the blood vessel angles.

9. The method according to claim 1, characterized in that: The step of calculating the cell stripe angle in the target blood vessel according to the spliced ​​image comprises: Extracting a valid blood flow signal area in the stitched image according to the blood vessel angle and the blood vessel width; Acquire multiple regions of interest ROIs of preset sizes in the effective blood flow signal area; Calculate the cell stripe angle in each ROI respectively; The preset statistical values ​​of the cell stripe angles in the plurality of ROIs are selected according to the preset rules and determined as the cell stripe angles in the target blood vessel.

10. The method according to claim 8, characterized in that After calculating the blood flow velocity parameter of the target blood vessel, the method further includes: On a perpendicular line to the center line of the target blood vessel, searching from the point of the maximum value of the line integral to both ends of the perpendicular line, obtaining a line segment whose line integral result is greater than a threshold, determining the line segment as a blood vessel portion, and determining the length of the line segment as the blood vessel width; The blood flow rate of the target blood vessel is calculated based on the blood flow velocity parameter and the blood vessel width.

11. The method according to claim 1, characterized in that: After calculating the cell stripe angle in the target blood vessel according to the stitched image, the method further includes: Determine whether the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°; If the blood vessel angle and the cell stripe angle are both greater than 90° or both less than 90°, the blood vessel angle is determined as the blood flow direction of the target blood vessel; If the blood vessel angle and the cell stripe angle are not both greater than 90° and not both less than 90°, the sum of the blood vessel angle and 180° is determined as the blood flow direction of the target blood vessel.

12. A blood flow parameter calculation device, characterized in that: include: A first image acquisition unit, configured to acquire a target blood vessel image, wherein the target blood vessel image is a two-dimensional image of the target blood vessel; A second image acquisition unit is used to acquire a spliced ​​image, where the spliced ​​image is obtained by splicing and pre-processing a plurality of scanned images obtained after the acquisition device cooperates with the galvanometer to scan the target blood vessel; a blood vessel angle acquisition unit, configured to acquire a blood vessel angle of the target blood vessel in the target blood vessel image; A cell stripe angle calculation unit, used to calculate the cell stripe angle in the target blood vessel according to the spliced ​​image; The blood flow velocity parameter calculation unit is used to calculate the blood flow velocity parameter of the target blood vessel according to the blood vessel angle, the cell stripe angle, the optical resolution of the acquisition device and the scanning frequency of the galvanometer.

Images