A method and device for calculating the maximum pressure gradient of a coronary artery based on an X-ray contrast image

Abstract

The application provides a coronary artery maximum pressure gradient calculation method and device based on an X-ray contrast image, and applies to the technical field of data processing. The application performs automatic screening on functional contrast frames based on preprocessed contrast images, obtains a difference image by subtracting a first frame of background frames from each preprocessed image, calculates a gray scale variance of the difference image, and selects a frame with the maximum variance as a target contrast image of each projection angle. The target contrast image is subjected to three-dimensional blood vessel model reconstruction and coordinate conversion to construct a coronary artery three-dimensional network model. Hemodynamic parameters are calculated based on the coronary artery three-dimensional network model. A PPG index is constructed based on the hemodynamic parameters, and lesions are classified. A PPG index formula is derived by integrating a 20mm maximum pressure gradient and a diffuse lesion ratio, a single PPG index value is made to correspond to a treatment strategy through an inverse proportional weighting algorithm, and classification information of coronary artery lesions is generated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a coronary artery maximum pressure gradient calculation method and device based on X-ray contrast images. BACKGROUND

[0002] In the clinical evaluation of coronary atherosclerotic heart disease, the maximum pressure gradient (PPG) is a core index for judging the influence of coronary stenosis on hemodynamics, which is defined as the pressure difference between the two ends of the stenosis section. The current mainstream method relies on invasive pressure guide wire manual withdrawal to collect the fractional flow reserve (FFR), combined with artificial analysis of the stenosis site based on coronary angiography two-dimensional images. The specific process is as follows: the physician manually checks the X-ray contrast image sequence frame by frame, and subjectively selects the "optimal frame" as the evaluation basis; the pressure guide wire is sent to the distal end of the coronary artery through the guide catheter, and the FFR value is recorded by manually withdrawing at a uniform speed and spacing of 10-20mm, and the curve of FFR changing with the length of the blood vessel is generated; according to the stenosis position identified by the "optimal frame" of the angiography, the pressure difference of the corresponding FFR curve is extracted, and the PPG is estimated combined with the stenosis rate of the vessel diameter.

[0003] The existing technology relies on digital subtraction angiography (DSA), pressure guide wire and hemodynamic detector, but has the following significant defects: manual screening of the optimal frame is low in efficiency (5-10 minutes for a single blood vessel) and strong in subjectivity (20%-30% of cases are misjudged); the manual withdrawal speed of the pressure guide wire is uneven, resulting in insufficient data accuracy, incomplete capture of diffuse lesions (missed diagnosis rate of 15%-20%), and underestimation of the PPG of serial lesions (deviation of 30%-40%); two-dimensional angiography cannot reflect the three-dimensional characteristics of the blood vessel, resulting in 25%-30% of "mild stenosis" being missed, and ignoring the progressive pressure loss of diffuse plaques.

[0004] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The purpose of the present application is to provide a coronary artery maximum pressure gradient calculation method and device based on X-ray contrast images, which at least to some extent overcomes the problems existing in the prior art, and realizes automatic PPG calculation through X-ray contrast sequence: obtain multi-view DICOM sequence and parameters, and pre-process through Gaussian filtering and down-sampling; select the optimal frame according to the gray variance, and reconstruct the three-dimensional model through coordinate conversion, angle correction and least squares method; calculate the blood flow parameters using the Navier-Stokes equation, and construct the PPG index corresponding to the treatment strategy, to solve the problems of low efficiency and inaccurate acquisition, and improve the evaluation efficiency.

[0006] Other features and advantages of the present application will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, or some aspects of the application.

[0007] According to an aspect of the present application, a method for calculating maximum pressure gradient of coronary artery based on X-ray contrast image is provided, comprising: acquiring multi-view coronary artery contrast images and pre-processing to generate pre-processed contrast images; performing automatic screening of functional contrast frames based on the pre-processed contrast images, obtaining difference images by subtracting each pre-processed image from the first background frame, calculating the gray variance of the difference images, and selecting the frame with the maximum variance for each projection angle as the target contrast image of that view; performing three-dimensional vascular model reconstruction and coordinate conversion on the target contrast image to construct a three-dimensional coronary artery network model; calculating hemodynamic parameters based on the three-dimensional coronary artery network model; constructing PPG index and classifying lesions based on the hemodynamic parameters, integrating 20mm maximum pressure gradient and diffuse lesion ratio to derive PPG index formula, and using inverse proportional weighting algorithm to make a single PPG index value correspond to a treatment strategy to generate classification information of coronary artery lesions.

[0008] According to another aspect of the present application, a device for calculating maximum pressure gradient of coronary artery based on X-ray contrast image is provided, characterized in that it comprises: an acquisition module for acquiring multi-view coronary artery contrast images and pre-processing to generate pre-processed contrast images; a processing module for performing automatic screening of functional contrast frames based on the pre-processed contrast images, obtaining difference images by subtracting each pre-processed image from the first background frame, calculating the gray variance of the difference images, and selecting the frame with the maximum variance for each projection angle as the target contrast image of that view; performing three-dimensional vascular model reconstruction and coordinate conversion on the target contrast image to construct a three-dimensional coronary artery network model; calculating hemodynamic parameters based on the three-dimensional coronary artery network model; constructing PPG index and classifying lesions based on the hemodynamic parameters, integrating 20mm maximum pressure gradient and diffuse lesion ratio to derive PPG index formula, and using inverse proportional weighting algorithm to make a single PPG index value correspond to a treatment strategy to generate classification information of coronary artery lesions.

[0009] According to still another aspect of the present application, an electronic device is provided, characterized in that it comprises: a first processor; and a memory for storing executable instructions of the first processor; wherein the first processor is configured to execute the executable instructions to implement the above-mentioned method for calculating maximum pressure gradient of coronary artery based on X-ray contrast image.

[0010] According to still another aspect of the present application, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a second processor, implements the above-described method for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image.

[0011] The method and device for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image provided by the present application realize automatic and accurate calculation of the maximum pressure gradient (PPG) of a coronary artery through an X-ray contrast image sequence. The process is as follows: a multi-view DICOM contrast sequence and device parameters are acquired, and 3x3 kernel Gaussian filtering and 2x2 mean downsampling preprocessing are performed; through gray variance analysis of a difference image, the most optimal frame with the largest variance is automatically selected; through coordinate conversion, angle correction and least square fitting, a three-dimensional blood vessel model is reconstructed; blood flow parameters are calculated based on the Navier-Stokes equation, and the maximum pressure gradient and the length of a functional lesion in a 20mm window are extracted; a PPG index is constructed, and a corresponding treatment strategy is weighted in inverse proportion. The method and device effectively solve the problems of low efficiency of manual screening and inaccuracy of pressure acquisition, and improve the evaluation efficiency and accuracy.

[0012] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 A flowchart of a method for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image according to an embodiment of the present application is shown;

[0014] Figure 2 A structural schematic diagram of a device for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0015] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, and it should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application.

[0016] The method for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image according to the exemplary embodiments of the present application is described below in conjunction with Figure 1 It should be noted that the following application scenarios are only shown for the purpose of facilitating understanding of the spirit and principles of the present application, and the embodiments of the present application are not limited in this respect. On the contrary, the embodiments of the present application are applicable to any applicable scenario.

[0017] In one embodiment, the present application further provides a method and device for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image. Figure 1A flowchart of a method for calculating maximum pressure gradient of coronary artery based on X-ray contrast image is shown schematically.

[0018] In S101, multi-view coronary angiography images are acquired and preprocessed to generate preprocessed contrast images.

[0019] In an embodiment, a DICOM format X-ray contrast sequence is acquired, and device parameters are recorded synchronously. A clinician uses a digital subtraction angiography (DSA) machine to perform coronary angiography on a patient suspected of having coronary heart disease, and scans the left anterior descending branch (LAD) of the left coronary artery from two projection angles, i.e., a left anterior oblique angle of 30° and a right anterior oblique angle of 20°. The device automatically generates a DICOM format image sequence. Thirty images are acquired at each angle, with a resolution of 512x512 pixels, and a single image grayscale value range of 0-4095 (12-bit depth), serving as contrast sequence information.

[0020] The device parameters recorded synchronously are as follows: a source-to-detector distance (SID) of 100 cm; a left anterior oblique angle of β=30° and α=-15°; a right anterior oblique angle of β=20° and α=10°; a pixel physical pitch of Δx=0.2 mm / pixel and Δy=0.2 mm / pixel; and an image size of width W=102.4 mm (512x0.2) and height H=102.4 mm.

[0021] A 3x3 kernel and a Gaussian filter with a standard deviation σ=1.0 are used to denoise the original image, and 2x2 mean down-sampling is performed on the 512x512 pixel image to generate a 256x256 pixel image, thereby generating a preprocessed contrast image. The original image is an image in an X-ray contrast sequence in DICOM format that is directly acquired without processing after multi-angle scanning of the coronary artery by a digital subtraction angiography machine. The denoising formula is where I(x,y) is the pixel value of the original image at coordinates (x,y), G(x,y) is the pixel value of the filtered image, the 3x3 kernel indicates that the filter has an action range of a 3x3 neighborhood centered at (x,y) on the image, and σ=1.0 controls the decay rate of the Gaussian function (the smaller σ is, the weaker the filtering effect is, and the more image details are retained; the larger σ is, the stronger the denoising effect is, but the edges may be blurred).

[0022] Before filtering, the edges of coronary artery branches in the original image have obvious particle noise, the gray contrast between blood vessels and surrounding tissues is low, and the outlines of small blood vessels (diameter <1 mm) are blurred. After filtering, through Gaussian filtering, the particle noise in the image is significantly reduced, and the smoothness of the blood vessel edge is improved. For example, in the image of a coronary artery segment with a diameter of 2 mm, the standard deviation of pixel gray scale after filtering is reduced from 25 to 12, the contrast between the blood vessel and the background is improved by 15%, and a clearer input is provided for subsequent blood vessel segmentation.

[0023] The filtered image of 512x512 pixels is down-sampled by 2x2 mean value to generate a pre-processed image of 256x256 pixels. During the down-sampling process, the original image is divided into 2x2 non-overlapping pixel blocks, and the output pixel value of each block is the mean value of the 4 pixels in the block.

[0024] Before down-sampling, the filtered 512x512 image contains a large amount of redundant pixel information, and the data volume is large (about 512KB per image), which is not conducive to the rapid inference of subsequent deep learning models. After down-sampling, an image of 256x256 pixels is generated, and the data volume is reduced to 128KB, reducing the consumption of computing resources. At the same time, the mean value down-sampling preserves the overall morphological features of the blood vessels, such as the key information of the main branch direction of the coronary artery and the location of the stenosis site, which is not lost. In the down-sampled image, the blood vessel segment with an original diameter of 3 pixels can still be clearly identified, and there is no breakage or distortion.

[0025] In S102, based on the pre-processed contrast image, functional contrast frame automatic screening is performed, the difference value image is obtained by subtracting each pre-processed image from the first background frame, the gray variance of the difference value image is calculated, and the frame with the maximum variance is selected as the target contrast image for each projection angle.

[0026] In one embodiment, the pre-processed image of each frame is subtracted from the first frame of background frame without contrast agent filling, eliminating the interference of extravascular tissue and device noise, generating a difference image. Subtracting the pre-processed contrast image of each frame from the first frame of background frame without contrast agent filling eliminates the interference of extravascular tissue (such as bone, myocardium) and device noise, and highlights the blood vessel region containing contrast agent. The first frame of background frame has no contrast agent, and the gray difference between the blood vessel region and the surrounding tissue is small; in the contrast frame, the blood vessels show high gray due to contrast agent filling, and the significant difference between the blood vessels and the background can be preserved after subtraction. The background frame is the image at the initial stage of coronary angiography examination (without contrast agent injection), with a size of 256x256 pixels, and the gray value of the blood vessel region (such as left anterior descending branch LAD) and myocardial tissue is close (about 150-200), and the noise standard deviation is about 30. The contrast frame is the image 10 seconds after the contrast agent is injected, and the gray value of the LAD blood vessel region is increased to 350-400 due to contrast agent filling, and the gray value of the surrounding tissue remains 150-200. Subtracting the corresponding pixel gray values, the difference value of the blood vessel region is 150-250 (350-150=200), and the difference value of the surrounding tissue is close to 0 (200-190=10), and the generated difference image has clear blood vessel contours without tissue interference.

[0027] The gray variance of the difference image is calculated, and the variance formula is where D n (x,y) is the pixel gray value of the nth frame of difference image at coordinate (x,y), μ n is the gray mean value of the difference image, and WxH is the image size, such as 256x256. The larger the variance, the more significant the gray difference between the blood vessel region and the background in the difference image, and the clearer the lesion information (such as contrast agent retention at the stenosis).

[0028] Taking the difference image of 256x256 pixels as an example, the gray mean value μ n is calculated, the average gray value of the blood vessel region (about 10% of the pixels) is 200, and the average gray value of the background region (90% of the pixels) is 10, then μ n = 0.1x200 + 0.9x10 = 29. The variance Variance n is calculated, the contribution of the blood vessel region is 0.1x(200-29) 2 = 2924.1, the contribution of the background region is 0.9x(10-29) 2 = 324.9, and the total variance Variance n = 2924.1 + 234.9 = 3249. The variance of this frame is 3249, and if the mean value of the same sequence image is 2000, then 3249÷2000 = 1.62 (>1.5), which meets the condition of "prior screening as the target contrast image", because the high variance indicates that the blood vessel lesions (such as stenosis) are clearly displayed.

[0029] For each projection angle image sequence, the frame with the largest variance is selected as the target angiographic image for that viewpoint. When the variance of a frame is greater than 1.5 times the sequence mean, it is prioritized as the target angiographic image to ensure that it contains the lesion of the target vessel. For each projection angle image sequence (e.g., 30 frames of left anterior oblique 30°), the frame with the largest variance is selected as the target angiographic image; if the variance of a frame is ≥ 1.5 times the sequence mean, it is prioritized. A patient's right coronary artery (RCA) angiography sequence contains 25 frames with a variance range of 1800-4500 and a sequence mean of 2500. The selection process is as follows: the 15th frame has a variance of 4200 (4200 ÷ 2500 = 1.68 > 1.5), which is the maximum value in the sequence; in this frame, the contrast filling defect at the mid-segment stenosis of the RCA is obvious (grayscale difference reaches 300), and the surrounding vessel branches are clear; it is determined to be the target angiographic image for this viewpoint and used for subsequent 3D reconstruction and pressure gradient calculation.

[0030] By eliminating non-vascular interference through interpolation and combining it with variance quantification of image information, objective screening of target angiographic frames can be achieved, avoiding the subjectivity of manual judgment. For example, in cases of calcified stenosis, high-variance frames can clearly show the grayscale difference between the stenotic segment and normal vessels, providing accurate anatomical basis for subsequent pressure gradient calculation, reducing the stenosis localization error from 5-10 mm in manual screening to 1-2 mm.

[0031] S103, the target angiography image is reconstructed into a three-dimensional vascular model and its coordinates are transformed to construct a three-dimensional network model of the coronary arteries.

[0032] In one implementation, pixel coordinates (u,v) are converted to physical coordinates (x,y) using the formulas x = (uW / 2) × Δx and y = (vH / 2) × Δy, where the origin is located at the image center. Converting image pixel coordinates (u,v) to actual physical coordinates (x,y) establishes a scale relationship between the two-dimensional image and three-dimensional space. (u,v) represents the image pixel coordinates (u is the horizontal pixel index, v is the vertical pixel index); W and H represent the image width and height (unit: pixels); Δx and Δy represent the pixel physical spacing (unit: mm / pixel), provided by the DSA device parameters; the origin (0,0) is located at the image center, i.e., the (W / 2,H / 2) pixel position.

[0033] Take the left anterior oblique projection image as an example: image parameters: W=256 pixels, H=256 pixels, Δx=Δy=0.2 mm / pixel. A blood vessel point is located at (u=150, v=100) in the image; x=(150-256 / 2)×0.2=(150-128)×0.2=22×0.2=4.4 mm; y=(100-256 / 2)×0.2=(100-128)×0.2=(-28)×0.2=-5.6 mm. The physical coordinates of the blood vessel point are (4.4 mm, -5.6 mm), indicating that it is located 4.4 mm to the right and 5.6 mm below the center of the image.

[0034] The Z-axis coordinate is initialized as the distance d from the human body to the detector, generating the initial three-dimensional coordinates (x, y, d). The two-dimensional physical coordinates (x, y) are combined with the depth information to generate the initial three-dimensional coordinates (x, y, d), where d is the distance from the human body to the detector (unit: mm) and is directly provided by the DSA device (usually 1000-1200 mm). Specifically, the device parameter is the detector distance d=1000 mm, and the initial three-dimensional coordinates of the blood vessel point are (4.4 mm, -5.6 mm, 1000 mm).

[0035] The intermediate coordinates (x', y', z') are generated by rotating around the Y-axis, and the rotation matrix is The conversion formula is Specifically, the projection angle is left anterior oblique β=30°, so cos30°≈0.866, sin30°=0.5. The rotation calculation process is x'=x×cosβ+z×sinβ=4.4×0.866+100×0.5=503.81 mm; y'=y=-5.6 mm; z'=-x×sinβ+z×cosβ=-4.4×0.5+1000×0.866≈863.8 mm. The intermediate coordinates are (503.81 mm, -5.6 mm, 863.8 mm).

[0036] The three-dimensional coordinates (X, Y, Z) in the world coordinate system are generated by rotating around the X-axis, and the rotation matrix is The conversion formula is Specifically, the projection angle is left anterior oblique α=-15°, so cos(-15°)≈0.966, sin(-15°)≈-0.259. The rotation calculation process is X=x'=503.81 mm; Y=y'×cosα-z'×sinα=-5.6×0.966-863.8×(-0.259)≈218.29 mm; Z=y'×sinα+z'×cosα=-5.6×(-0.259)+863.8×0.966≈835.85 mm. The world coordinate system coordinates are (503.81 mm, 218.29 mm, 835.85 mm).

[0037] The three-dimensional coordinates of multi-view reconstruction are grouped and fitted by least squares method to determine the three-dimensional coordinates of blood vessels. For three-dimensional coordinates belonging to the same blood vessel position under different views, a set of spatial straight lines is constructed by connecting lines to provide a geometric basis for subsequent target point fitting. Coordinates of the same blood vessel position (such as the midpoint of a certain stenosis segment of the coronary artery) are obtained from ≥2 views such as left anterior oblique and right anterior oblique. Each connecting line represents the spatial mapping of the position under different views.

[0038] The multi-view coordinates are as follows: the blood vessel point coordinate A (500, 200, 800) mm is measured by left anterior oblique (LAO 30°); the corresponding blood vessel point coordinate B (490, 190, 820) mm is measured by right anterior oblique (RAO 20°); a spatial straight line L1 is formed by connecting A and B, and the direction vector is The radiation source coordinate S1 is the left anterior oblique radiation source position (0, 0, 0) mm.

[0039] The coordinates of the same blood vessel position under different views are connected by lines, and the target point P is solved so that the sum of the distances from all connecting lines is minimized, and the formula is where P is the three-dimensional coordinate of the target point to be solved, S i is the radiation source coordinate of the i-th view (determined by the DSA device parameters), is the direction vector of the i-th connecting line, the numerator is the modulus of the vector cross product, representing the perpendicular distance from p to the connecting line; and the denominator is the modulus of the direction vector, used to standardize the distance.

[0040] Taking two views as an example: the radiation source coordinate S1 (0, 0, 0) is known, and the connecting line L1 direction vector The radiation source coordinate S2 (1000, 0, 0) (right anterior oblique radiation source), and the connecting line L2 direction vector Let the target point P (x, y, z), then Cross product Distance Similarly, d2 is calculated, and the total distance sum is d1+d2. The optimal solution P (495, 195, 810) mm is obtained by minimizing d1+d2 through numerical iteration (such as gradient descent); the sum of the distances from this point to L1 and L2 is minimized, and the error is ≤0.5 mm, which meets the clinical accuracy requirements.

[0041] All the fitted points (e.g. one point every 1mm) are connected to form a continuous line segment, forming a three-dimensional mesh model, completely restoring the spatial morphology of the blood vessel shape, branching and stenosis. The left anterior descending branch (LAD) is fitted, and 200 target points are obtained, and the three-dimensional model formed after connection: accurately restores the change of blood vessel diameter (the diameter of the stenosis segment is narrowed from 3mm to 1.5mm); accurately captures the branching angle (the angle between LAD and D1 branch is 75°); the spatial position error is ≤1mm, which provides an accurate anatomical basis for subsequent hemodynamic calculation (such as pressure gradient, flow velocity distribution).

[0042] By least square fitting to eliminate multi-view imaging errors, the two-dimensional contrast image is converted into a high-precision three-dimensional blood vessel model, which solves the problems of blood vessel overlap and spatial distortion in traditional two-dimensional evaluation. For example, for tandem stenosis lesions, the three-dimensional model can clearly distinguish the relative position (distance 15mm) of the two stenoses, provide reliable geometric parameters for calculating the pressure superposition effect caused by "lesion crosstalk", and reduce the MaxPPG20mm calculation error from 30%-40% of the traditional method to within 5%, completing the construction of the coronary artery three-dimensional mesh model.

[0043] S104, calculating the hemodynamic parameters based on the three-dimensional network model of the coronary artery.

[0044] In one embodiment, for each cross section of the three-dimensional network model, the blood vessel contour is extracted by an edge detection algorithm, and the cross-sectional area is calculated. For each cross section of the three-dimensional blood vessel model, the blood vessel contour is extracted by an edge detection algorithm (Canny operator), and the cross-sectional area is calculated based on the contour, and the formula is where D is the diameter of the blood vessel cross section (directly obtained from the three-dimensional reconstruction result), A is the cross-sectional area (unit: mm 2 ).

[0045] The cross-sectional diameter D of a certain stenosis segment in the three-dimensional model is 2.4mm, and the cross-sectional area is This parameter is used for subsequent blood flow and flow velocity calculation, reflecting the physical restriction of blood vessel stenosis on blood flow (e.g. the smaller the stenosis area, the greater the blood flow resistance).

[0046] According to the filling time of the contrast agent and the blood vessel volume, the blood flow is calculated, and the flow velocity distribution under laminar flow state is calculated, and under laminar flow state, the flow velocity distribution in the blood vessel follows the parabolic law, wherein the flow velocity formula is v(r)=v max (1-r 2 / R 2 ), wherein, V(r) is the flow velocity at radial coordinate r (unit: mm / s); V max is the maximum flow velocity; R is the radius of the blood vessel, and r is the distance from a point to the center of the blood vessel (0≤r≤R). The blood flow Q=100mm3 s, vessel radius R = 1.2 mm, cross-sectional area A ≈ 4.52 mm 2 ; maximum flow rate At the center of the vessel (r = 0), the flow rate V(0) = 22.1 x (1 - 0) = 22.1 mm / s; at the wall of the vessel (r = R = 1.2 mm), the flow rate V(1.2) = 22.1 x (1 - 1) = 0 mm / s (consistent with the no-slip boundary condition).

[0047] The N-S equation is discretized using the finite element method and the boundary conditions are set, and the whole segment pressure distribution is obtained by iterative solution, wherein the N-S equation is discretized using the finite element method wherein, is the continuity equation, which indicates that the blood is incompressible, u is the flow rate vector, p is the blood density, m is the dynamic viscosity coefficient, and f is the volume force; the boundary conditions are set as follows: at the inlet, the pressure p in = 100 mmHg; at the outlet, the flow rate At the wall of the vessel: no-slip boundary condition (u = 0). A coronary artery with a length of 50 mm is simulated, and the stenosis segment is located in the middle 10 mm region (the diameter is narrowed from 3 mm to 1.5 mm); after iterative solution, the pressure at the proximal end of the stenosis segment is 95 mmHg, and the pressure at the distal end is 70 mmHg, and the pressure gradient is The results reflect the significant pressure loss caused by stenosis, providing a basis for subsequent PPG calculation.

[0048] Sliding a 20 mm window along the axis of the vessel, the pressure gradient in the window is calculated, and the maximum value is taken as the pressure loss intensity of the focal stenosis. Sliding a 20 mm window along the axis of the vessel, the maximum value of the pressure gradient in the window is calculated, and if the proximal pressure in a window is 80 mmHg and the distal pressure is 55 mmHg, then the pressure gradient is This value is used as an index of the pressure loss intensity of the focal stenosis.

[0049] The total length of the continuous vessel segment with a pressure gradient slope of ≥0.0015 mmHg / mm is taken as the functional lesion length. The total length of the continuous vessel segment with a pressure gradient slope of ≥0.0015 mmHg / mm is calculated, and the functional lesion length is 30 mm for a vessel with a total length of 100 mm, of which 30 mm segments meet the slope condition.

[0050] The fluid mechanics modeling accurately calculates the pressure distribution of the coronary artery, solving the error problem of traditional empirical formulas. For example, for serial stenosis lesions, the pressure superposition effect caused by "lesion crosstalk" can be captured, making the MaxPPG20mm calculation value 0.08 FFRunits higher than the traditional method, which is closer to the clinical measurement results, providing a quantitative basis for accurate diagnosis of coronary heart disease.

[0051] S105, constructing PPG index based on hemodynamic parameters and classifying lesions, integrating 20mm maximum pressure gradient and diffuse lesion ratio to derive PPG index formula, making single PPG index value correspond to treatment strategy through inverse proportion weighting algorithm, and generating classification information of coronary artery lesions.

[0052] In an embodiment, PPG index is constructed based on hemodynamic parameters and lesion classification, PPG index formula is derived by integrating 20mm maximum pressure gradient and diffuse lesion ratio, and the formula is Wherein, MaxPPG 20mm is the maximum pressure gradient (unit: FFR units) calculated by sliding a 20mm window along the blood vessel axis, reflecting the pressure loss intensity of focal stenosis; DiffuseRatio is the diffuse lesion ratio (range 0-1), that is, the proportion of the total length of the blood vessel that is occupied by the length of the lesion with a pressure gradient slope ≥0.0015 FFR units / mm; through the inverse proportion structure of “focal intensity × (1-diffuse ratio)” and “diffuse ratio + correction term (0.1)”, the nonlinear coupling of focal and diffuse lesions is realized.

[0053] Taking a patient with left anterior descending coronary artery (LAD) lesion as an example, the total length of the blood vessel is 120mm, and after three-dimensional reconstruction and hemodynamic calculation, the following data are obtained: the maximum pressure gradient MaxPPG 20mm of the 20mm sliding window is 0.25 FFR units (corresponding to the pressure difference of the most severe stenosis region); the continuous lesion length with a pressure gradient slope ≥0.0015 FFR units / mm is 48mm (i.e. Length diffuse = 48mm).

[0054] Calculate MaxPPG 20mm , slide along the blood vessel axis with a 20mm window, calculate the pressure gradient in each window, Length window = 20mm.

[0055] The pressure P proximal of a certain 20mm window near the proximal end is 0.90 FFR units, and the pressure P distal of the distal end is 0.65 FFR units, so the pressure gradient of this window is 0.90-0.65=0.25 FFR units, which is the maximum value among all windows. Calculate DiffuseRatio and substitute it into the formula Wherein, DiffuseRatio is the diffuse lesion ratio, Length Diffuse is the total length of the continuous lesion that meets the slope condition (pressure gradient changes with length ≥0.0015 FFR units / mm), reflecting the range of diffuse lesions.

[0056] Calculate PPG index , into the formula

[0057] When PPG index <0.4, diffuse lesions are determined, and drug treatment or coronary artery bypass grafting (CABG) is preferentially recommended.

[0058] The patient's lesions are mainly diffuse (Diffuse Ratio = 0.4), and the PPG index is low, suggesting that unnecessary PCI (percutaneous coronary intervention) should be avoided, which can reduce the length of stent implantation and reduce the risk of perioperative myocardial infarction.

[0059] The present application realizes the automatic and accurate calculation of the maximum pressure gradient (PPG) through the X-ray contrast image sequence. The multi-view DICOM format contrast sequence is improved to obtain the device parameters, and the Gaussian filter is used for noise reduction and 2x2 mean downsampling preprocessing; The functional optimal frame is automatically screened, and the difference image is generated by subtracting the first frame background frame, the gray variance is calculated, and the frame with the maximum variance is selected; The target frame is reconstructed into a three-dimensional vascular model, which is converted and corrected in coordinates, combined with the least square method fitting; Based on the three-dimensional model, the Navier-Stokes equation is used to calculate the hemodynamic parameters, and the maximum pressure gradient and functional lesion length of the 20mm window are extracted; The PPG index is constructed, the above two parameters are fused, and the corresponding treatment strategy is obtained through the inverse proportional weighting algorithm. It solves the problems of low efficiency of manual screening, inaccurate pressure data acquisition and the like, improves the efficiency and accuracy, provides support for clinical decision-making, and is suitable for coronary heart disease functional lesion evaluation.

[0060] In one embodiment, as shown in Figure 2 The present application also provides a coronary artery maximum pressure gradient calculation device based on X-ray contrast images, comprising:

[0061] The acquisition module 201 is used for acquiring multi-view coronary angiography images and preprocessing to generate preprocessed contrast images;

[0062] The processing module 202 is configured to perform functional contrast frame automatic screening based on the preprocessed contrast image, obtain a difference image by subtracting a first frame of background frame from each preprocessed image, calculate a gray scale variance of the difference image, select a frame with the maximum variance as a target contrast image of each projection angle for an image sequence of each projection angle, perform three-dimensional blood vessel model reconstruction and coordinate conversion on the target contrast image, and construct a three-dimensional coronary artery network model; calculate a hemodynamic parameter based on the three-dimensional coronary artery network model; construct a PPG index and perform lesion classification based on the hemodynamic parameter, derive a PPG index formula by integrating a 20mm maximum pressure gradient and a diffuse lesion ratio, make a single PPG index value correspond to a treatment strategy through an inverse proportional weighting algorithm, and generate classification information of the coronary artery lesion.

[0063] The computer readable storage medium provided by the above embodiments of the present application has the same beneficial effects as the method adopted, run or implemented by the application program stored in the computer readable storage medium.

[0064] Each of the embodiments in the present application is described in a related manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the difference from other embodiments. In particular, the evaluation of the method for calculating the maximum pressure gradient of the coronary artery based on the X-ray contrast image, the electronic device, the electronic equipment, and the readable storage medium are basically similar to the above-mentioned embodiments of the method for calculating the maximum pressure gradient of the coronary artery based on the X-ray contrast image, and therefore the description is relatively simple, and the relevant parts can be referred to the above-mentioned embodiments of the method for calculating the maximum pressure gradient of the coronary artery based on the X-ray contrast image.

Claims

1. A method for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image, characterized by, The method comprises the following steps: obtaining multi-view coronary angiography images and preprocessing to generate preprocessed angiography images; based on the preprocessed angiography images, functional contrast frame automatic screening is performed, each frame of the preprocessed images is subtracted from the first frame of the background frame to obtain a difference image, the gray variance of the difference image is calculated, and the frame with the maximum variance of each projection angle image sequence is selected as the target angiography image of the view; three-dimensional vascular model reconstruction and coordinate conversion are performed on the target angiography image to construct a coronary three-dimensional network model; based on the coronary three-dimensional network model, hemodynamic parameters are calculated; Based on the hemodynamic parameters, a PPG index is constructed and the lesion is classified, the PPG index formula is derived by integrating the 20mm maximum pressure gradient and the diffuse lesion ratio, and a single PPG index value is obtained by inverse proportional weighting algorithm to correspond to the treatment strategy, and the classification information of the coronary artery lesion is generated, including based on the hemodynamic parameters, a PPG index is constructed and the lesion is classified, the PPG index formula is derived by integrating the 20mm maximum pressure gradient and the diffuse lesion ratio, and the formula is ; , wherein, is the 20 mm maximum pressure gradient, and is the proximal and distal pressure values within the window, is the length of the window used for the calculation, is the final calculated PPG index; where DiffuseRatio is the diffuse lesion ratio, is the total length of lesions continuously meeting the slope condition, reflecting the range of diffuse lesions; By inverse proportional weighting algorithm of the single PPG index value corresponding to the treatment strategy, the classification information of coronary artery lesions is generated.

2. The method of claim 1, wherein, obtaining multi-view coronary angiography images and preprocessing to generate preprocessed angiography images, comprising: obtaining X-ray contrast sequences in DICOM format, and synchronously recording device parameters; using a 3*3 kernel and a Gaussian filter with a standard deviation σ=1.0 to denoise the original image, performing 2*2 mean downsampling on the 512*512 pixel image to generate a 256*256 pixel image, and generating the preprocessed angiography image, wherein the original image is an image in the X-ray contrast sequence in DICOM format obtained directly without processing after multi-angle scanning of the coronary artery by a digital subtraction angiography machine; The noise reduction formula is wherein, is the original image pixel gray value, is the filtered image, is the standard deviation of the Gaussian filter.

3. The method of claim 2, wherein, based on the preprocessed angiography images, functional contrast frame automatic screening is performed, each frame of the preprocessed images is subtracted from the first frame of the background frame to obtain a difference image, the gray variance of the difference image is calculated, and the frame with the maximum variance of each projection angle image sequence is selected as the target angiography image of the view, comprising: subtracting each frame of the preprocessed images from the first frame of the contrast agent non-filled background frame to eliminate extravascular tissue and device noise interference and generate a difference image; The gray scale variance of the difference image is calculated, and a variance formula is wherein, The gray scale variance of the nth frame difference image is used to quantify the gray scale difference between the blood vessel region and the background in the difference image. The size of the image, that is, the product of the image width W and the height H; The pixel gray scale value of the nth frame difference image at the coordinate (x, y) is represented by The gray scale mean value of the difference image is represented by selecting the frame with the maximum variance of each projection angle image sequence as the target angiography image of the view, and when the variance of a certain frame is 1.5 times or more than the average value of the sequence, the frame is preferentially selected as the target angiography image to make it contain the lesion display of the target blood vessel.

4. The method of claim 1, wherein, three-dimensional vascular model reconstruction and coordinate conversion are performed on the target angiography image to construct a coronary three-dimensional network model, comprising: By the formula Convert pixel coordinates (u, v) to physical coordinates (x, y) with the origin at the image center; the Z-axis coordinate is initialized as the distance d from the human body to the detector to generate an initial three-dimensional coordinate (x, y, d); The intermediate coordinates (x', y', z') are generated by rotating around the Y axis, and the rotation matrix is , and the conversion formula is ; The three-dimensional coordinates (X, Y, Z) in the world coordinate system are generated by rotating around the X axis, and the rotation matrix is , and the conversion formula is ; the least square method is used for grouping fitting to determine the three-dimensional coordinates of the blood vessels: The coordinates of the same blood vessel position under different perspectives are connected to solve the target point P, so that the sum of the distances from all the lines is minimized, and the formula is wherein P is the target point, is the radiation source coordinate, is the line direction vector, and a coronary artery three-dimensional grid model is constructed.

5. The method of claim 4, wherein, based on the coronary three-dimensional network model, hemodynamic parameters are calculated, comprising: for each cross section of the three-dimensional network model, the blood vessel contour is extracted by an edge detection algorithm, and the cross-sectional area is calculated; Blood flow is calculated from the contrast filling time and the vessel volume, and the flow velocity distribution is calculated in the case of laminar flow, where the flow velocity is given by where R is the vessel radius and r is the radial coordinate. The finite element method is used to discretize the N-S equation and set boundary conditions, and the whole section pressure distribution is obtained by iterative solution, wherein the finite element method is used to discretize the N-S equation , wherein u is a flow velocity vector, is the blood density, is a dynamic viscosity coefficient, and f is a volume force; the boundary conditions are set as follows: inlet: pressure ; outlet: flow rate ; cross-section average flow rate ; and the blood vessel wall: no-slip boundary condition (u=0); a 20mm window is slid along the blood vessel axis, the pressure gradient in the window is calculated, and the maximum value is taken as the pressure loss intensity of the focal stenosis; the total length of the continuous blood vessel segments with a pressure gradient slope greater than or equal to 0.0015mmHg / mm is counted as the functional lesion length.

6. An apparatus for calculating a maximum pressure gradient of a coronary artery based on an X-ray contrast image, characterized by, The device for implementing the method of claim 1 comprises: an acquisition module for obtaining multi-view coronary angiography images and preprocessing to generate preprocessed angiography images; The processing module is used for automatic screening of functional contrast frames based on the preprocessed contrast images, obtaining a difference image by subtracting each preprocessed image from a first background frame, calculating a gray variance of the difference image, selecting a frame with the maximum variance as a target contrast image for each image sequence of a projection angle, performing three-dimensional vascular model reconstruction and coordinate conversion on the target contrast image, constructing a three-dimensional network model of the coronary artery, calculating hemodynamic parameters based on the three-dimensional network model of the coronary artery, constructing a PPG index based on the hemodynamic parameters and classifying lesions, integrating a 20mm maximum pressure gradient and a diffuse lesion ratio to derive a PPG index formula, using an inverse proportional weighting algorithm to make a single PPG index value correspond to a treatment strategy, and generating classification information of the coronary artery lesions.

7. An electronic device, comprising: Comprise: a first processor; and a memory for storing executable instructions of the first processor; wherein the first processor is configured to implement the method for calculating the maximum pressure gradient of the coronary artery based on the X-ray contrast image according to any one of claims 1-5 by executing the executable instructions.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the second processor to implement the method for calculating the maximum pressure gradient of the coronary artery based on the X-ray contrast image according to any one of claims 1-5.

Images