Aortic valve automatic segmentation method based on multi-temporal CTA image

Abstract

The invention relates to the technical field of medical image processing, in particular to an aortic valve automatic segmentation method based on a multi-temporal CTA image. The method comprises the following steps: identifying a laminar flow dark area and a vortex bright area on a cross section of an aortic sinus area; according to the first feature size of the laminar flow dark area and the second feature size of the vortex bright area, the valve leaflet-sinus wall spacing is determined; determining a first deviation index according to the change trend of the valve leaflet-sinus wall spacing in the cardiac cycle; determining a second deviation index according to a difference characteristic between the CT value of the laminar flow dark area and the CT value of the eddy current bright area; if the first deviation index and the second deviation index both exceed the corresponding threshold values, judging that the data is in a to-be-corrected time phase; correcting the spacing by using the correction weight calculated by the second deviation index; and extracting key parameters of each time phase on the longitudinal section, and associating the corrected valve leaflet-sinus wall spacing to construct a valve three-dimensional dynamic model, thereby facilitating the three-dimensional dynamic model to output more accurate valve motion data.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to an automatic aortic valve segmentation method based on multi-phase CTA images. Background Art

[0002] With the development of medical imaging technology, CTA (CT angiography) has become the key to accurate segmentation of the aortic valve due to its high resolution and three-dimensional imaging advantages.

[0003] Currently, existing technologies mainly rely on single-phase CTA images for valve segmentation. However, this method cannot capture the dynamic changes in the structural morphology and boundary position of the aortic valve leaflets during the cardiac cycle, making it difficult to accurately identify the valve's trajectory parameters. On the other hand, CT imaging is prone to data noise due to factors such as partial volume effect and uneven distribution of contrast agents, which leads to errors in the calculation results of key parameters and affects the accuracy of the quantified segmented aortic valve motion trajectory parameters. Summary of the Invention

[0004] In order to solve the technical problem that it is difficult to accurately identify the trajectory parameters of the valve when segmenting the valve based on single-phase CTA images, the present invention provides an automatic aortic valve segmentation method based on multi-phase CTA images. The technical solution adopted is as follows: The present invention proposes an automatic aortic valve segmentation method based on multi-phase CTA images, the method comprising: Acquire multi-phase CTA images of the aortic root covering the entire cardiac cycle; locate the aortic sinus region in each phase of the CTA images; and identify laminar dark areas and eddy bright areas in cross-sectional images of the aortic sinus region. The leaflet-sinus wall spacing of each phase is determined based on the first characteristic size of the laminar dark area and the second characteristic size of the vortex bright area in each phase; the first deviation index of each phase is determined based on the change trend of the leaflet-sinus wall spacing of each phase during the cardiac cycle; the second deviation index of each phase is determined based on the difference characteristics between the CT value of the laminar dark area and the CT value of the vortex bright area in each phase; If the first deviation index exceeds a preset first index threshold and the second deviation index exceeds a preset second index threshold, determining the corresponding phase as a phase of data to be corrected; determining a correction weight based on the second deviation index of the phase of data to be corrected; and correcting the leaflet-sinus wall distance of the phase of data to be corrected using the correction weight; The morphological and functional dynamic parameters of the aortic valve at each time phase are extracted from the longitudinal cross-sectional images of the aortic sinus region, and the leaflet-sinus wall spacing at each time phase is associated to construct a three-dimensional dynamic model of the valve. The model outputs trajectory parameters that characterize the valve motion, wherein the spacing of the phase to be corrected is the corrected value.

[0005] Furthermore, the process of determining the first characteristic size of the laminar dark area and the second characteristic size of the eddy bright area includes: On the cross-sectional image of the aortic sinus region, the inner edge boundary outline of the laminar dark area and the outer edge boundary outline of the eddy bright area are extracted; A polar coordinate system was established with the center of the aortic root as the origin. Multiple preset sampling angles were set along the circumference. At each sampling angle, the radial distance from the origin to the inner edge of the laminar dark zone was calculated as the first characteristic distance for the corresponding angle. The radial distance from the origin to the outer edge of the eddy bright zone was calculated as the second characteristic distance for the corresponding angle. Calculate the arithmetic mean of the first characteristic distances corresponding to all sampling angles in each phase as the first characteristic size of the laminar dark area in each phase; The arithmetic mean of the second characteristic distances corresponding to all sampling angles in each time phase is calculated as the second characteristic size of the vortex bright area in each time phase.

[0006] Furthermore, after extracting the inner edge boundary contour of the laminar dark area and the outer edge boundary contour of the eddy bright area, the contours are smoothed. The smoothing process includes removing isolated noise points on the contours.

[0007] Furthermore, the laminar dark area and the eddy bright area are identified on the cross-sectional image of the aortic sinus region, including: Grayscale analysis of cross-sectional images; Based on a preset CT value threshold range, identifying a high CT value region and a low CT value region in the image; wherein the CT value of the high CT value region is within a first preset range, the CT value of the low CT value region is within a second preset range, and the lower limit of the first preset range is higher than the upper limit of the second preset range; The identified low CT value area is used as the laminar dark area; the identified high CT value area is used as the eddy bright area.

[0008] Furthermore, the automatic segmentation method of the aortic valve based on multi-phase CTA images also includes: For multiple phases within the cardiac cycle, the first area ratio of the laminar dark area and the second area ratio of the eddy bright area are calculated; the first area ratio is the ratio of the laminar dark area to the total cross-sectional area of ​​the aortic sinus region, and the second area ratio is the ratio of the eddy bright area to the total cross-sectional area of ​​the aortic sinus region; Based on the temporal variation characteristics of the leaflet-sinus wall distance, the first area ratio, and the second area ratio during the cardiac cycle, each phase is classified as a first blood flow state period or a second blood flow state period.

[0009] Furthermore, the first deviation index determination process includes: Performing polynomial fitting on the leaflet-sinus wall spacing of all phases within the same period to generate a spacing change trend curve, and obtaining the fitting spacing value of each phase within the same period, wherein the same period represents the first blood flow state period or the second blood flow state period obtained by classification; The difference between the leaflet-sinus wall distance at each time phase and the fitted distance at the same time phase was calculated as the original residual; the standard deviation of all original residuals in the same period was calculated; The absolute value of the ratio of the original residual to the standard deviation of each phase in the same period is taken as the first deviation index of the corresponding phase.

[0010] Furthermore, the second deviation index determination process includes: Calculating a first statistical characteristic value of the laminar dark area based on the CT value distribution of the pixels in the laminar dark area, and calculating a second statistical characteristic value of the eddy bright area based on the CT value distribution of the pixels in the eddy bright area; Calculate the difference between the first statistical characteristic value of the laminar dark area and the second statistical characteristic value of the eddy bright area in each time phase to obtain the CT difference value of each time phase; Performing polynomial fitting on the CT difference values ​​of all time phases within the same period to generate a CT difference value change trend curve, and obtaining the fitted CT difference value of each time phase, wherein the same period represents the first blood flow state period or the second blood flow state period obtained by classification; The difference between the CT difference value of each phase and the corresponding fitted CT difference value was calculated as the CT original residual of each phase; the standard deviation of all CT original residuals in the same period was calculated; The absolute value of the ratio of the CT raw residual to the standard deviation of each phase in the same period is used as the second deviation index of the corresponding phase.

[0011] Furthermore, the calculation process of the first statistical eigenvalue and the second statistical eigenvalue includes: Calculate the arithmetic mean of the CT values ​​of all pixels in the laminar dark area at each time phase as the first statistical characteristic value of the laminar dark area; The arithmetic mean of the CT values ​​of all pixels in the eddy current bright area of ​​each time phase is calculated as the second statistical characteristic value of the eddy current bright area.

[0012] Furthermore, the correction weight determination process includes: The correction weight of each phase is obtained by performing exponential operation with the natural constant as the base and the opposite number of the second deviation index of each phase as the exponent.

[0013] Furthermore, the leaflet-sinus wall distance of the data at the time phase to be corrected is corrected using the correction weight, including: Get the corresponding fitting spacing value when the data is to be corrected; Calculating the product of the correction weight and the leaflet-sinus wall distance at the phase of the data to be corrected as the first product; calculating the difference between the value 1 and the correction weight as the modified weight, and calculating the product of the modified weight and the fitting distance value as the second product; The sum of the first product and the second product was calculated as the corrected leaflet-sinus wall distance.

[0014] The present invention has the following beneficial effects: The present invention determines a first deviation index by analyzing the changing trend of the leaflet-sinus wall spacing in each phase during the cardiac cycle; determines a second deviation index by analyzing the difference characteristics between the CT values ​​of the laminar dark area and the CT values ​​of the eddy bright area in each phase, and combines the first deviation index and the second deviation index to accurately identify the phase with abnormal data in the cardiac cycle to eliminate accidental errors caused by single-phase segmentation; quantifies the spatial scales of the laminar dark area and the eddy bright area through the first characteristic size and the second characteristic size to accommodate their possible irregular shapes, thereby reducing the error in the determined leaflet-sinus wall spacing; corrects the problematic leaflet-sinus wall spacing using the correction weight calculated by the second deviation index, and dynamically balances the leaflet-sinus wall spacing through the correction weight so that the changing trend of the leaflet-sinus wall spacing conforms to the hemodynamic characteristics of blood flow through the aortic valve, thereby avoiding distortion of key data subsequently acquired; finally, the corrected spacing, leaflet attachment points, opening and closing angles and other key parameters are synchronously integrated to construct a three-dimensional dynamic model, so that more accurate valve motion data can be quickly output based on the three-dimensional dynamic model. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flowchart of a method for automatic segmentation of the aortic valve based on multi-phase CTA images provided by one embodiment of the present invention; Figure 2 An example diagram of the changing process of the laminar dark area and the eddy bright area provided by one embodiment of the present invention; Figure 3 An example diagram of a first deviation index determination process provided by one embodiment of the present invention; Figure 4 This is an example diagram of the second deviation index determination process provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of an automatic aortic valve segmentation method based on multi-phase CTA images proposed by the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following describes in detail a specific solution of the method for automatic segmentation of the aortic valve based on multi-phase CTA images provided by the present invention with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flow chart of a method for automatic segmentation of the aortic valve based on multi-phase CTA images provided by one embodiment of the present invention, the method comprising: S101: Acquire multi-phase CTA images of the aortic root covering the entire cardiac cycle; locate the aortic sinus region in the CTA images of each phase; and identify laminar dark areas and eddy bright areas in the cross-sectional images of the aortic sinus region.

[0021] Computed Tomography Angiography (CTA) is a medical imaging method that combines computed tomography (CT) technology and angiography principles.

[0022] To accurately segment the aortic valve, clear CTA images of the aortic root need to be obtained. Therefore, before acquiring multi-phase CTA images, some preliminary preparations are required. For example, a higher concentration of iodine contrast agent is used while ensuring that it does not cause significant adverse effects on the patient.

[0023] A complete cardiac cycle refers to the entire process that the heart goes through from the beginning of one contraction to the beginning of the next contraction, including two stages: systole (ventricular ejection) and diastole (ventricular filling). During systole: the ventricles contract, the aortic valve opens, and blood is ejected from the left ventricle into the aorta. During this period, the distance between the valve leaflets and the sinus wall gradually decreases to the minimum value; during diastole: the ventricles relax, the aortic valve closes, and blood flows back to form a vortex. During this period, the distance between the valve leaflets and the sinus wall gradually increases to the maximum value.

[0024] It should be noted that the specific methods for locating the aortic sinus region and identifying the laminar dark area and eddy bright area in CTA images are well known to those skilled in the art and will not be further described in this embodiment. For example, the existing nnU-Net model is used to locate the aortic sinus region and segment a cross-sectional image of the aortic sinus region. The cross-sectional image of the aortic sinus region is analyzed using the Otsu annular profile analysis algorithm to determine the cross-sectional image grayscale histogram, and then the cross-sectional image is divided into two parts: the laminar dark area and the eddy bright area.

[0025] It is important to understand that during cardiac systole, due to the opening of the aortic valve leaflets, blood is rapidly ejected from the left ventricle through the aortic valve orifice into the mainstream area of ​​the aorta. The blood flow presents a regular parallel flow (laminar state), with a fast flow rate and rapid flushing of the contrast agent. In addition, due to the dilation of the aortic sinus structure, the blood flow forms a rotating, slow vortex after entering, and the contrast agent is retained in the sinus for a longer time, resulting in a significantly higher contrast agent concentration than in the laminar flow area in the center of the valve orifice. It can be seen that the CT value directly reflects the degree of X-ray absorption by the blood. The contrast agent is a strong X-ray absorbing substance. The higher the contrast agent concentration, the higher the corresponding CT value. Therefore, in the CTA images of the aortic root during cardiac systole, the blood flow channel in the center of the valve orifice, that is, the laminar flow area, appears as a "dark area" with low CT value due to the low contrast agent concentration. Correspondingly, the vortex area appears as a "bright ring" with high CT value in the CTA image, surrounding the laminar flow area.

[0026] For example, the changing process of laminar dark area and eddy bright area in a complete cardiac cycle is shown in the figure below. Figure 2 As shown, the change in CT value difference is synchronized with the opening and closing of the aortic valve and its hemodynamic characteristics. For example, from early systole to mid-systole, left ventricular pressure typically rises, the valve leaflets gradually open, blood flow accelerates into the aorta, the laminar dark zone velocity continues to rise, contrast agent is rapidly washed away, and the corresponding CT value gradually decreases. In contrast, the eddy zone, due to contrast agent retention within the sinusoids, has a consistently high CT value. Therefore, the CT difference between the CT value of the eddy bright zone and the CT value of the laminar dark zone gradually increases as the CT value of the laminar dark zone decreases, and the change curve is smooth (without sharp fluctuations). From mid-systole to early diastole, ventricular contraction typically weakens, the valve leaflets gradually close, the laminar dark zone velocity slows, contrast agent washout weakens, and the CT value of the laminar dark zone decreases, slowing the decline or even slightly recovering. Contrast agent concentrations in the eddy zone remain high, and the CT value of the eddy bright zone remains stable. Therefore, the increase in CT value difference slows until it stabilizes after the valve leaflets fully close.

[0027] Among them, it should be noted that based on the change law from contraction to diastole, the change law from early diastole to mid-diastole and then to systole can be reversely analyzed. The change from early diastole to mid-diastole is manifested as a gradual decrease in CT difference, and the change curve is smooth; the change from mid-diastole to systole is manifested as a slowdown in the decrease trend of CT difference.

[0028] In this embodiment, grayscale analysis is performed on the cross-sectional image; based on a preset CT value threshold range, high CT value areas and low CT value areas in the image are identified; wherein the CT value of the high CT value area is within a first preset range, the CT value of the low CT value area is within a second preset range, and the lower limit value of the first preset range is higher than the upper limit value of the second preset range; the identified low CT value area is regarded as a laminar dark area, and the identified high CT value area is regarded as an eddy bright area.

[0029] A cross-sectional image refers to a two-dimensional image slice obtained by cutting perpendicular to the long axis of the aorta. The cross-sectional image can clearly show the annular structure of the aortic sinus, the attachment position of the valve leaflets and the blood flow distribution status. The cross-sectional image can usually observe an approximately circular cross-sectional outline of the aortic sinus.

[0030] It should be noted that the specific value ranges of the first preset range and the second preset range are adaptively adjusted according to actual conditions and are not specifically limited in this embodiment. For example, to avoid regional overlap and ensure clear division between high and low CT value regions, there is a clear interval between the first preset range and the second preset range. The first preset range (high CT value region) can be set to 150HU~350HU, corresponding to the eddy current bright area; the second preset range (low CT value region) can be set to 30HU~120HU, corresponding to the laminar dark area.

[0031] S102: Determine the leaflet-sinus wall spacing of each time phase based on the first characteristic size of the laminar dark area and the second characteristic size of the vortex bright area in each time phase; determine the first deviation index of each time phase based on the change trend of the leaflet-sinus wall spacing of each time phase during the cardiac cycle; determine the second deviation index of each time phase based on the difference characteristics between the CT value of the laminar dark area and the CT value of the vortex bright area in each time phase.

[0032] The leaflet-sinus wall distance is the average distance between the leaflet and the sinus wall calculated based on the first characteristic size of the laminar dark area and the second characteristic size of the vortex bright area, which may be referred to as the distance in the subsequent description.

[0033] It should be noted that, according to step S101, it can be understood that in the cardiac cycle, when transitioning from the systole to the diastole, the range of the laminar dark area gradually shrinks, the range of the vortex bright area gradually expands, and the leaflet-sinus wall distance tends to increase; conversely, when transitioning from the diastole to the systole, the range of the laminar dark area and the vortex bright area and the change trend of the leaflet-sinus wall distance are inversely corresponding.

[0034] In this embodiment, for multiple phases within the cardiac cycle, the first area ratio of the laminar dark area and the second area ratio of the vortex bright area are calculated; based on the temporal change characteristics of the leaflet-sinus wall distance, the first area ratio and the second area ratio during the cardiac cycle, each phase is classified as a first blood flow state period or a second blood flow state period.

[0035] The first area proportion reflects the proportion of laminar flow in the local blood flow, where the first area proportion is the ratio of the laminar dark area to the total cross-sectional area of ​​the aortic sinus region.

[0036] The second area ratio reflects the proportion of vortex in local blood flow, where the second area ratio is the ratio of the vortex bright area to the total cross-sectional area of ​​the aortic sinus region.

[0037] It should be noted that, under non-extreme circumstances, the aortic sinus region cannot be non-existent, and thus the total cross-sectional area of ​​the aortic sinus region cannot be zero.

[0038] It should be understood that, combined with the physiological movement and hemodynamic characteristics of the aortic valve, it can be analyzed that when the aortic valve opens (systole): the ventricular ejection pressure pushes the free edge of the valve leaflet to expand outward toward the aortic sinus wall, and the valve leaflet gradually approaches the sinus wall. Therefore, the valve leaflet-sinus wall distance gradually decreases until the valve leaflet is fully expanded and attached to the sinus wall (the valve leaflet-sinus wall distance reaches the minimum value). At this time, the laminar dark area expands due to the opening of the valve leaflet (the first area accounts for a high proportion), and the vortex area is washed away by the high-speed blood flow and shrinks (the second area accounts for a low proportion); when the aortic valve closes (diastole): the ventricle stops ejecting blood, and the blood flow in the aorta reverses, pushing the free edge of the valve leaflet inward to close, and the valve leaflet gradually moves away from the sinus wall. Therefore, the valve leaflet-sinus wall distance gradually increases until the free edge of the valve leaflet fits together (the valve leaflet-sinus wall distance reaches the maximum value). At this time, the vortex area (the reflow area between the valve leaflet and the sinus wall) expands (the second area accounts for a high proportion), and the laminar area disappears or shrinks (the first area accounts for a low proportion).

[0039] For example, if the leaflet-sinus wall distance in a certain phase is smaller than the leaflet-sinus wall distance in the previous phase and the first area ratio is greater than that in the previous phase, it is determined to be the first blood flow state period, that is, the cardiac systole; if the leaflet-sinus wall distance in a certain phase is larger than the leaflet-sinus wall distance in the previous phase and the first area ratio is less than that in the previous phase, it is determined to be the second blood flow state period, that is, the cardiac diastole.

[0040] The first blood flow state corresponds to the aortic valve systolic phase. During this phase, the ventricles contract, the valve leaflets open, and blood flows from the left ventricle into the aorta in a high-speed laminar flow. Laminar flow dominates the blood flow (large dark areas), while eddy currents are weak (small bright areas). The distance between the valve leaflets and the sinus wall decreases.

[0041] The second blood flow state corresponds to the aortic valve diastole phase. During this phase, the ventricles relax, the valve leaflets close, blood flow slows down, and vortices (large bright areas) form within the aortic sinus. Laminar flow disappears or weakens (small dark areas). These vortices help cushion the impact of blood flow on the sinus wall and maintain vascular stability.

[0042] It should be understood that in the cross-sectional images of the aortic sinus, the laminar dark area and the vortex bright area are usually approximately circular or quasi-circular in shape (because the anatomical structure of the aortic sinus is a symmetrical sac-like structure, it is easy to form a surrounding vortex in hemodynamics). However, during cardiac systole, the laminar dark area expands and extends axially, and the cross-section may be elliptical; during cardiac diastole, when the leaflets are closed, the vortex bright area is squeezed by the leaflets and may appear as an irregular fan-shaped "three-leaf embrace". Therefore, when determining the leaflet-sinus wall spacing, the impact of the irregular morphology of the laminar dark area or vortex bright area on the spacing must also be considered.

[0043] The first characteristic dimension is a quantitative value used to characterize the spatial scale of the laminar dark zone, reflecting the average radial position of the laminar dark zone relative to the center point of the aortic root, and is a quantitative description of the overall spatial distribution of the valve leaflet.

[0044] The second characteristic size is a quantitative value used to characterize the spatial scale of the vortex bright area, reflecting the average radial position of the vortex bright area relative to the center point of the aortic root, and is a quantitative description of the overall spatial distribution of the sinus wall.

[0045] In this embodiment, on the cross-sectional image of the aortic sinus region, the inner edge boundary contour of the laminar dark area and the outer edge boundary contour of the eddy bright area are extracted; a polar coordinate system is established with the center point of the aortic root as the origin; a plurality of preset sampling angles are set along the circumferential direction; in the direction of each sampling angle, the radial distance from the origin to the inner edge boundary contour of the laminar dark area is calculated as the first characteristic distance of the corresponding angle; the radial distance from the origin to the outer edge boundary contour of the eddy bright area is calculated as the second characteristic distance of the corresponding angle; the arithmetic mean of the first characteristic distances corresponding to all sampling angles in each time phase is calculated as the first characteristic size of the laminar dark area in each time phase; the arithmetic mean of the second characteristic distances corresponding to all sampling angles in each time phase is calculated as the second characteristic size of the eddy bright area in each time phase.

[0046] It should be noted that before calculating the arithmetic mean of the first or second characteristic distance, outlier detection (for example, using the Z-score method) is performed on the first or second characteristic distance of all sampling angles to eliminate extreme values ​​that deviate from the mean by more than 3 times the standard deviation to ensure the robustness of the arithmetic mean.

[0047] The inner edge boundary outline of the laminar dark area is the continuous contour line of the inner edge of the laminar dark area (low CT value area) in the cross-sectional image of the aortic sinus. In the heart, this inner edge boundary outline usually corresponds to the position of the free edge of the aortic valve leaflet. The laminar dark area is formed by high-speed laminar blood, and its inner side is close to the leaflet. Therefore, the inner edge boundary outline can be regarded as the projection of the free edge of the leaflet on the cross-sectional image.

[0048] The outer edge boundary contour of the vortex bright area. In the cross-sectional image of the aortic sinus, the continuous contour line of the outer edge of the vortex bright area (high CT value area) is the boundary that distinguishes the vortex area from the surrounding tissues (such as the aortic sinus wall and myocardium). In the heart, this outer edge boundary contour usually corresponds to the inner surface of the aortic sinus wall. The vortex bright area is formed by the retention of contrast agent, and its outer side is close to the sinus wall. Therefore, the outer edge boundary contour can be regarded as the projection of the inner surface of the sinus wall on the cross-sectional image.

[0049] It should be noted that after extracting the inner edge boundary contour of the laminar dark area and the outer edge boundary contour of the eddy bright area, the contours are smoothed. The smoothing process includes removing isolated noise points on the contours, for example, using Gaussian filtering to smooth the boundary contours.

[0050] It should be noted that in order to ensure data consistency, the sampling angles can be set evenly along the circumferential direction, and the differences between adjacent sampling angles are equal. For example, a sampling angle is set every 30 degrees, and a total of 12 sampling angles are set. The specific method of setting multiple sampling angles along the circumferential direction is a technical means well known to those skilled in the art and will not be repeated in this embodiment.

[0051] In this embodiment, the difference between the second characteristic size of the vortex bright area and the first characteristic size of the laminar dark area in each time phase is calculated as the leaflet-sinus wall distance in each time phase.

[0052] It can be understood that the distance between the leaflet and the sinus wall = the second characteristic size - the first characteristic size.

[0053] The first deviation index determination process is as follows Figure 3 As shown, including: S102-1: Perform polynomial fitting on the leaflet-sinus wall distances of all phases within the same period to generate a distance variation trend curve, and obtain fitting distance values ​​of each phase within the same period.

[0054] It should be noted that the same period refers to periods classified as the same blood flow state. For example, all phases classified as the first blood flow state period are referred to as phases of the same period.

[0055] Among them, it should be noted that the specific method of obtaining the fitting spacing value through polynomial fitting is a technical means well known to those skilled in the art, and will not be repeated in this embodiment. For example, a polynomial function is used to fit the leaflet-sinus wall spacing of all phases in the same period. During the fitting process, a suitable order (such as 2nd order) is selected for the polynomial function, and then the coefficient is solved by the least squares method to obtain the final fitting function, which is the "spacing change trend curve". For each phase in the same period, the fitting function is substituted into the calculation to obtain the corresponding fitting spacing value.

[0056] It should be noted that, in the polynomial fitting process, polynomials of different orders are selected for fitting according to the data trend. For example, during the cardiac systole, the polynomial fitting is performed according to the trend of "gradually decreasing valve leaflet-sinus wall spacing", and the second order is used when the blood flow is stable; during the cardiac diastole, the polynomial fitting is performed according to the trend of "gradually increasing spacing" (such as the rising trend of a quadratic function), and the third order is used when the eddy current changes are complex.

[0057] S102-2: Calculate the difference between the leaflet-sinus wall distance at each time phase and the fitted distance value at the same time phase as the original residual; calculate the standard deviation of all original residuals in the same period.

[0058] It should be noted that the method of calculating the standard deviation is an existing technical means and will not be described in detail in this embodiment.

[0059] S102-3: The absolute value of the ratio of the original residual to the standard deviation of each phase in the same period is used as the first deviation indicator of the corresponding phase.

[0060] The first deviation index is used to quantify the overall deviation between the leaflet-sinus wall distance at each phase in the same blood flow state period and the fitted distance value of the corresponding phase.

[0061] It should be understood that if the absolute value of the ratio of the original residual to the standard deviation of a certain phase is larger, it reflects that the leaflet-sinus wall spacing obtained from this phase is an outlier on the spacing change trend curve, that is, the degree of deviation is greater.

[0062] It should be noted that, even in the same period, in actual situations, the leaflet-sinus wall spacing is constantly changing with time. Due to some objective factors such as voxel mixing, the leaflet-sinus wall spacing and the corresponding fitted spacing value are usually not exactly the same. Therefore, the original residuals cannot all be zero, and the standard deviation cannot be zero either.

[0063] The second deviation index determination process is as follows Figure 4 As shown, including: S102-4: Calculate a first statistical characteristic value of the laminar dark area based on the CT value distribution of the pixels in the laminar dark area, and calculate a second statistical characteristic value of the eddy bright area based on the CT value distribution of the pixels in the eddy bright area.

[0064] For example, when the CT values ​​of the laminar dark area or the eddy bright area meet the following conditions, the arithmetic mean is used to calculate the statistical characteristic value: 1. The CT value distribution in the area is unimodal and approximately normal; 2. The CT values ​​of the pixels in the area are spatially uniformly distributed (such as homogeneous blood flow and no artifact soft tissue); 3. The signal-to-noise ratio of the laminar dark area or the eddy bright area is > .

[0065] It should be noted that the specific methods of determining the CT value distribution of the laminar dark area or the eddy bright area and calculating the signal-to-noise ratio are technical means well known to those skilled in the art and will not be described in detail in this embodiment.

[0066] S102-5: Calculate the difference between the first statistical characteristic value of the laminar dark area and the second statistical characteristic value of the eddy bright area in each time phase to obtain the CT difference value of each time phase.

[0067] In this embodiment, the arithmetic mean of the CT values ​​of all pixels in the laminar dark area of ​​each time phase is calculated as the first statistical characteristic value of the laminar dark area; the arithmetic mean of the CT values ​​of all pixels in the eddy bright area of ​​each time phase is calculated as the second statistical characteristic value of the eddy bright area.

[0068] S102-6: Perform polynomial fitting on the CT difference values ​​of all time phases in the same period to generate a CT difference value change trend curve, and obtain the fitted CT difference value of each time phase.

[0069] S102-7: Calculate the difference between the CT difference value of each time phase and the corresponding fitted CT difference value as the CT raw residual of each time phase; calculate the standard deviation of all CT raw residuals in the same period.

[0070] S102-8: The absolute value of the ratio of the CT raw residual to the standard deviation of each phase in the same period is used as the second deviation indicator of the corresponding phase.

[0071] The second deviation index is used to quantify the overall deviation between the CT difference value of each phase in the same blood flow state period and the fitted CT difference value of the corresponding phase in the period.

[0072] It should be understood that if the absolute value of the ratio of the CT original residual to the standard deviation of a certain phase is larger, it reflects that the CT difference value obtained from the phase is an outlier on the CT difference value change trend curve, that is, the degree of deviation is greater.

[0073] S103: If the first deviation index exceeds the preset first index threshold and the second deviation index exceeds the preset second index threshold, the corresponding phase is determined to be the phase of data to be corrected; based on the second deviation index of the phase of data to be corrected, a correction weight is determined; and the leaflet-sinus wall distance of the phase of data to be corrected is corrected using the correction weight.

[0074] It should be understood that in order to avoid a single indicator exceeding the standard, which may be an accidental error (such as instantaneous imaging noise), when two indicators exceed the standard at the same time, the probability of accidental error is extremely low, and it is more likely to be a systematic problem (such as equipment parameter errors, changes in patient position during data collection, abnormal valve movement causing measurement reference disorder, etc.). Therefore, the first deviation indicator and the second deviation indicator are combined to determine whether the data extracted in the time phase needs to be corrected.

[0075] Among them, it should be noted that the specific values ​​of the preset first indicator threshold and the preset second indicator threshold are determined according to actual conditions, and this embodiment does not make specific limitations. For example, when the preset first indicator threshold is 3, it can be clearly found that the leaflet-sinus wall distance measurement value deviates too much from the normal trend of the distance in the same period; when the preset second indicator threshold is 2, it can be clearly found that the CT difference value deviates too much from the normal change trend of the CT difference value in the same period.

[0076] In this embodiment, an exponential operation is performed with a natural constant as a base and the opposite number of the second deviation index of each time phase as an exponent to obtain the correction weight of each time phase.

[0077] It should be noted that when using the correction weight to correct the leaflet-sinus wall distance of the phase to be corrected, only the phases that simultaneously meet the first and second deviation index exceeding the threshold value within the same blood flow state period are corrected, and correction across periods (such as systole and diastole) is not performed.

[0078] Since, if the second deviation index of a certain phase is larger, it reflects that the overall deviation between the leaflet-sinus wall spacing of the phase and the fitted spacing value of the phase is larger. At this time, the leaflet-sinus wall spacing of the phase should be corrected to be closer to the fitted spacing value. Then, during the correction process, the smaller the weight of the leaflet-sinus wall spacing determined at the phase, that is, the smaller the correction weight, the greater the weight of the fitted spacing value. Therefore, the correction weight of each phase can be expressed by the following formula: in, represents the correction weight; represents an exponential function with a natural constant as its base; Represents the second deviation indicator.

[0079] In this embodiment, the corresponding fitting spacing value when the data is to be corrected is obtained; the product of the correction weight and the leaflet-sinus wall spacing at the time when the data is to be corrected is calculated as the first product; the difference between the value 1 and the correction weight is calculated as the correction weight, and the product of the correction weight and the fitting spacing value is calculated as the second product; the sum of the first product and the second product is calculated as the corrected leaflet-sinus wall spacing.

[0080] S104: Extract the morphological and functional dynamic parameters of the aortic valve at each time phase from the longitudinal cross-sectional image of the aortic sinus region, associate the leaflet-sinus wall spacing at each time phase to construct a three-dimensional dynamic model of the valve, and output the trajectory parameters representing the valve movement, wherein the spacing of the phase to be corrected is the corrected value.

[0081] The longitudinal cross-sectional image of the aortic sinus region refers to the cross-sectional imaging along the long axis of the aorta (i.e., the mainstream direction of blood flowing from the left ventricle into the aorta), which can intuitively present the upper and lower diameters of the aortic sinus, the longitudinal movement trajectory of the valve leaflets, and the longitudinal adjacent relationship with the surrounding structures.

[0082] Morphological and functional dynamic parameters of the aortic valve: The morphological parameters of the aortic valve mainly describe the anatomical structural characteristics of the aortic valve; the functional dynamic parameters of the aortic valve mainly reflect the hemodynamic state of the aortic valve during the cardiac cycle; for example, the morphological and functional dynamic parameters of the aortic valve at least include the coordinates of the leaflet attachment point, the leaflet apex position and the leaflet opening and closing angle.

[0083] The leaflet attachment point coordinates represent the position of the fixed point where the leaflet is connected to the aortic sinus wall in the coordinate system. For example, in a three-dimensional coordinate system with the center of the aortic root as the origin, the coordinates of the fixed point where the leaflet is connected to the aortic sinus wall.

[0084] The leaflet apex position refers to the coordinates of the highest point of the free edge of the leaflet (during cardiac systole) or the closing point (during cardiac diastole). For example, in a three-dimensional coordinate system with the center of the aortic root as the origin, the coordinates of the highest point of the free edge of the leaflet during cardiac systole.

[0085] The leaflet opening and closing angle is the angle formed by the line segment connecting the leaflet attachment point and the leaflet vertex with the long axis of the aorta. During the systole of the heart, the leaflet opening and closing angle increases (for example, from 30° to 90°, reflecting the degree of leaflet opening); during the diastole of the heart, the leaflet opening and closing angle decreases (for example, from 90° to 10°, reflecting the degree of leaflet closure).

[0086] It should be noted that the trajectory parameters of the valve movement include at least the maximum opening and closing angle, the spacing change rate, and the attachment point-apex distance change rate.

[0087] The maximum opening and closing angle is the maximum opening and closing angle of the leaflet during cardiac contraction; the spacing change rate is the rate of change of the corrected leaflet-sinus wall distance over time, which is used to reflect the speed of leaflet opening; the attachment point-apex distance change rate is the fluctuation range of the distance from the leaflet attachment point to the apex in the same period, which is used to reflect the expansion and contraction of the leaflet itself.

[0088] It should be noted that the specific method of constructing a three-dimensional dynamic model of the valve is a technical means well known to those skilled in the art and will not be described in detail in this embodiment. For example, the coordinates of the leaflet attachment points, vertex positions, and opening and closing angles extracted from the longitudinal section image are first time-space matched with the leaflet-sinus wall spacing corrected from the cross-sectional image. The parameters of the same phase correspond to the same moment in the cardiac cycle, forming a "static snapshot" of the valve at that moment. Then, the "static snapshots" of all phases in the cardiac cycle are connected in chronological order, and the transition states between adjacent phases are supplemented by an interpolation algorithm to form a continuous three-dimensional dynamic model. The model can intuitively display the complete movement process of the leaflet from opening to closing and then opening again.

[0089] It should be noted that the specific method of obtaining the maximum opening and closing angle, the spacing change rate, and the attachment point-vertex distance change rate is an existing technical means and will not be repeated in this embodiment.

[0090] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for automatic segmentation of the aortic valve based on multi-phase CTA images, characterized in that: The method comprises: Acquire multi-phase CTA images of the aortic root covering the entire cardiac cycle; locate the aortic sinus region in each phase of the CTA images; and identify laminar dark areas and eddy bright areas in cross-sectional images of the aortic sinus region. The leaflet-sinus wall spacing of each phase is determined based on the first characteristic size of the laminar dark area and the second characteristic size of the vortex bright area in each phase; the first deviation index of each phase is determined based on the change trend of the leaflet-sinus wall spacing of each phase during the cardiac cycle; the second deviation index of each phase is determined based on the difference characteristics between the CT value of the laminar dark area and the CT value of the vortex bright area in each phase; If the first deviation index exceeds a preset first index threshold and the second deviation index exceeds a preset second index threshold, determining the corresponding phase as a phase of data to be corrected; determining a correction weight based on the second deviation index of the phase of data to be corrected; and correcting the leaflet-sinus wall distance of the phase of data to be corrected using the correction weight; The morphological and functional dynamic parameters of the aortic valve at each time phase are extracted from the longitudinal cross-sectional images of the aortic sinus region, and the leaflet-sinus wall spacing at each time phase is associated to construct a three-dimensional dynamic model of the valve. The model outputs trajectory parameters that characterize the valve motion, wherein the spacing of the phase to be corrected is the corrected value.

2. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 1, characterized in that: The process of determining the first characteristic size of the laminar dark area and the second characteristic size of the eddy bright area includes: On the cross-sectional image of the aortic sinus region, the inner edge boundary outline of the laminar dark area and the outer edge boundary outline of the eddy bright area are extracted; A polar coordinate system was established with the center of the aortic root as the origin. Multiple preset sampling angles were set along the circumference. At each sampling angle, the radial distance from the origin to the inner edge of the laminar dark zone was calculated as the first characteristic distance for the corresponding angle. The radial distance from the origin to the outer edge of the eddy bright zone was calculated as the second characteristic distance for the corresponding angle. Calculate the arithmetic mean of the first characteristic distances corresponding to all sampling angles in each phase as the first characteristic size of the laminar dark area in each phase; The arithmetic mean of the second characteristic distances corresponding to all sampling angles in each time phase is calculated as the second characteristic size of the vortex bright area in each time phase.

3. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 2, characterized in that: After extracting the inner edge boundary contour of the laminar dark area and the outer edge boundary contour of the eddy bright area, the contours are smoothed. The smoothing process includes removing isolated noise points on the contours.

4. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 1, characterized in that: The identifying of the laminar dark area and the eddy bright area on the cross-sectional image of the aortic sinus region includes: Grayscale analysis of cross-sectional images; Based on a preset CT value threshold range, identifying a high CT value region and a low CT value region in the image; wherein the CT value of the high CT value region is within a first preset range, the CT value of the low CT value region is within a second preset range, and the lower limit of the first preset range is higher than the upper limit of the second preset range; The identified low CT value area is used as the laminar dark area; the identified high CT value area is used as the eddy bright area.

5. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 1, characterized in that: The method further comprises: For multiple phases within the cardiac cycle, the first area ratio of the laminar dark area and the second area ratio of the eddy bright area are calculated; the first area ratio is the ratio of the laminar dark area to the total cross-sectional area of ​​the aortic sinus region, and the second area ratio is the ratio of the eddy bright area to the total cross-sectional area of ​​the aortic sinus region; Based on the temporal variation characteristics of the leaflet-sinus wall distance, the first area ratio, and the second area ratio during the cardiac cycle, each phase is classified as a first blood flow state period or a second blood flow state period.

6. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 5, characterized in that: The first deviation indicator determination process includes: Performing polynomial fitting on the leaflet-sinus wall spacing of all phases within the same period to generate a spacing change trend curve, and obtaining the fitting spacing value of each phase within the same period, wherein the same period represents the first blood flow state period or the second blood flow state period obtained by classification; The difference between the leaflet-sinus wall distance at each time phase and the fitted distance at the same time phase was calculated as the original residual; the standard deviation of all original residuals in the same period was calculated; The absolute value of the ratio of the original residual to the standard deviation of each phase in the same period is taken as the first deviation index of the corresponding phase.

7. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 5, characterized in that: The second deviation indicator determination process includes: Calculating a first statistical characteristic value of the laminar dark area based on the CT value distribution of the pixels in the laminar dark area, and calculating a second statistical characteristic value of the eddy bright area based on the CT value distribution of the pixels in the eddy bright area; Calculate the difference between the first statistical characteristic value of the laminar dark area and the second statistical characteristic value of the eddy bright area in each time phase to obtain the CT difference value of each time phase; Performing polynomial fitting on the CT difference values ​​of all time phases within the same period to generate a CT difference value change trend curve and obtain the fitted CT difference value of each time phase, wherein the same period represents the first blood flow state period or the second blood flow state period obtained by classification; The difference between the CT difference value of each phase and the corresponding fitted CT difference value was calculated as the CT original residual of each phase; the standard deviation of all CT original residuals in the same period was calculated; The absolute value of the ratio of the CT raw residual to the standard deviation of each phase in the same period is used as the second deviation index of the corresponding phase.

8. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 7, characterized in that: The calculation process of the first statistical eigenvalue and the second statistical eigenvalue includes: Calculate the arithmetic mean of the CT values ​​of all pixels in the laminar dark area at each time phase as the first statistical characteristic value of the laminar dark area; The arithmetic mean of the CT values ​​of all pixels in the eddy current bright area of ​​each time phase is calculated as the second statistical characteristic value of the eddy current bright area.

9. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 6, characterized in that: The correction weight determination process includes: The correction weight of each phase is obtained by performing exponential operation with the natural constant as the base and the opposite number of the second deviation index of each phase as the exponent.

10. The method for automatic segmentation of the aortic valve based on multi-phase CTA images according to claim 9, characterized in that: The method of correcting the leaflet-sinus wall distance at the phase of the data to be corrected by using the correction weights includes: Get the corresponding fitting spacing value when the data is to be corrected; Calculating the product of the correction weight and the leaflet-sinus wall distance at the phase of the data to be corrected as the first product; calculating the difference between the value 1 and the correction weight as the modified weight, and calculating the product of the modified weight and the fitting distance value as the second product; The sum of the first product and the second product was calculated as the corrected leaflet-sinus wall distance.

Images