Coronary artery blood vessel hemodynamic quantification method and system based on IVUS image

Abstract

The invention relates to the technical field of medical image processing, in particular to a coronary artery blood vessel hemodynamic quantification method and system based on an IVUS image, and the method comprises the following steps: in the IVUS image, respectively enhancing a target, and obtaining a snowflake point target enhanced image and an interference target enhanced image; performing target detection on the enhanced image by using a Mask R-CNN model to obtain a snowflake point target mask image and an interference target mask image; processing the IVUS image by using a fused mask of the snowflake point target mask image and the interference target mask image to obtain a snowflake point segmentation result image; and the hemodynamics is quantified on the image based on the snowflake point segmentation result. According to the method, the snowflake point region in the IVUS image is accurately segmented by using the neural network model, accurate extraction of low-echo region information caused by blood flow in the intima embodied by the IVUS image is realized, and then accurate quantitative evaluation of hemodynamic parameters of coronary artery blood vessels is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and in particular to a method and system for quantifying coronary artery hemodynamics based on IVUS images. Background Art

[0002] In the cardiovascular field, IVUS (Intravascular Ultrasound) technology has, after decades of development, become increasingly the mainstream of precision treatment, especially for the treatment of coronary arteries with complex lesions, such as left main trunk lesions, bifurcation lesions, calcified lesions, CTO (Chronic Total Occlusion), etc. It can fully evaluate the status of coronary artery lesions and has strong guiding significance and optimization strategies for PCI (Percutaneous Coronary Intervention).

[0003] At present, the clinical significance of IVUS images is limited to the evaluation of vascular structure, analysis of plaque characteristics, and post-operative evaluation of stent conditions. There has been no characteristic analysis of the low-echo area, namely the "snowflake point", caused by blood flow in the endothelium reflected by IVUS images. Since the image in this area is mainly caused by the front end of the ultrasound catheter emitting and receiving blood flow, if this part of the data can be extracted and further analyzed, the hemodynamic parameters of the coronary artery vessels can be evaluated, thereby providing important information for evaluating cardiovascular health, diagnosing coronary artery lesions, and individualized treatment strategies.

[0004] Therefore, the existing technology lacks characteristic analysis of the low-echo areas in the endothelium caused by blood flow in IVUS images, namely "snowflakes", making it difficult to quantitatively evaluate the hemodynamic parameters of the coronary arteries. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for quantifying coronary artery hemodynamics based on IVUS images, so as to solve the technical problem that the prior art lacks characteristic analysis of low-echo areas, namely "snowflake points", caused by blood flow in the endothelium reflected in IVUS images, making it difficult to quantitatively evaluate the hemodynamic parameters of coronary arteries.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions: A method for quantifying coronary artery hemodynamics based on IVUS images comprises the following steps: IVUS images of the coronary arteries are obtained while the patient undergoes coronary angiography; In the IVUS image, the snowflake point target and the interference target are enhanced respectively to obtain the snowflake point target enhanced image and the interference target enhanced image; The Mask R-CNN model is used to detect snowflake point targets and interference targets on the snowflake point target enhanced image and the interference target enhanced image respectively, and the snowflake point target mask image and the interference target mask image are obtained; The IVUS image is processed by a fusion mask of the snowflake point target mask image and the interference target mask image to obtain a snowflake point segmentation result image; The hemodynamics is quantified based on the snowflake point segmentation result image to obtain the coronary artery hemodynamic parameters.

[0007] As a preferred embodiment of the present invention, the interference targets include the lumen of the IVUS catheter, the caliber of the IVUS catheter, and plaques.

[0008] As a preferred solution of the present invention, the method for acquiring the snowflake point target enhanced image comprises: The Sobel operator is used to perform transverse edge detection and longitudinal edge detection on the IVUS image to obtain an IVUS transverse edge detection image and an IVUS longitudinal edge detection image; The combined image of IVUS transverse edge detection image and IVUS longitudinal edge detection image is added and fused with the IVUS image to obtain the snowflake point target enhanced image; The enhancement operation process of the snowflake point target enhanced image is as follows: ; In the formula, Enhance the image for snowflake targets, For IVUS images, is the horizontal detection template of the Sobel operator, is the vertical detection template of the Sobel operator, is the enhancement factor, and All are absolute values; in, ; In the formula, is the L2 norm.

[0009] As a preferred solution of the present invention, the method for acquiring the interference target enhanced image includes: Subtract and fuse the combined image of IVUS transverse edge detection image and IVUS longitudinal edge detection image with the IVUS image to obtain an interference target enhanced image; The enhancement operation process of the interference target enhanced image is as follows: ; In the formula, Enhance the image for the interference target, For IVUS images, is the horizontal detection template of the Sobel operator, is the vertical detection template of the Sobel operator, is the enhancement factor, and All are absolute values; in, ; In the formula, is the L2 norm.

[0010] As a preferred solution of the present invention, the method for acquiring the snowflake point target mask image includes: Selecting multiple snowflake point target enhanced images, and marking snowflake point area mask images on the snowflake point target enhanced images; A data set is formed by combining multiple snowflake target enhanced images and snowflake region mask images, and a Mask R-CNN model is trained based on the data set to pre-establish a snowflake target detection model for detecting snowflake targets in snowflake target enhanced images; Using the snowflake point target detection model, the snowflake point target is detected on the snowflake point target enhanced image to obtain the snowflake point target mask image; The snowflake target detection model is: ; In the formula, is the snowflake target mask image, The image is enhanced for snowflake targets, and Mask R-CNN is the Mask R-CNN model.

[0011] As a preferred solution of the present invention, the method for acquiring the interference target mask image includes: Selecting multiple interference target enhanced images, and marking interference area mask images on the interference target enhanced images; A plurality of interference target enhanced images and interference area mask images are combined into a data set, and a Mask R-CNN model is trained based on the data set to pre-establish an interference target detection model for detecting interference targets in interference target enhanced images; Using the interference target detection model, the interference target is detected on the interference target enhanced image to obtain the interference target mask image; The interference target detection model is: ; In the formula, To interfere with the target mask image, To enhance the image for distracting targets, Mask R-CNN is the Mask R-CNN model.

[0012] As a preferred solution of the present invention, the method for obtaining the snowflake point segmentation result image includes: In the snowflake point target mask image, the overlapping parts of the snowflake point target and the interference target mask image are removed to obtain the fused mask; Perform pixel extraction on the IVUS image based on the fusion mask to obtain the snowflake point segmentation result image; The calculation process of the fusion mask is: ; In the formula, is the fusion mask, is the snowflake target mask image, is the interference target mask image, mix is ​​the coincidence operator, It is the overlapping part of the snowflake point target and the interference target in the snowflake point target mask image and the interference target mask image.

[0013] As a preferred embodiment of the present invention, the method for quantifying the coronary artery hemodynamic parameters includes: In the snowflake point segmentation result image of the IVUS image of each coronary artery cross-section, the number of blood flow snowflake points is counted; The number of pixels of each blood flow snowflake point is counted, and the sum is calculated to obtain the blood flow snowflake point area of ​​each coronary artery cross-section IVUS image; The blood flow snowflake volume is obtained by multiplying the blood flow snowflake area of ​​the IVUS image of each coronary artery cross section by the thickness of each coronary artery cross section. In the snowflake point segmentation result image of the IVUS image of each coronary artery cross-section, the lumen cross-sectional area of ​​the coronary artery is calculated; The difference between the lumen cross-sectional area of ​​the coronary artery in the IVUS image of each coronary artery cross-section and the normal lumen cross-sectional area of ​​the coronary artery is calculated by ratio with the normal lumen cross-sectional area of ​​the coronary artery to obtain the degree of coronary artery stenosis; The blood flow velocity at the coronary artery is obtained by multiplying the ratio of the lumen cross-sectional area of ​​the coronary artery in the IVUS image to the normal lumen cross-sectional area of ​​the coronary artery by the normal blood flow velocity; The square of the blood flow velocity in the coronary artery minus the square of the normal blood flow velocity multiplied by a coefficient of 4 gives the pressure gradient in the coronary artery. The blood flow velocity in the coronary arteries is multiplied by the blood flow snowflake area to obtain the blood flow in the coronary arteries.

[0014] As a preferred solution of the present invention, the loss function for training the Mask R-CNN model is the segmentation mask loss, and the segmentation mask loss is: ; In the formula, is the segmentation mask loss, is the fusion mask, is the true value of the snowflake target mask image, for The total number of pixels, is the overlapped part of the snowflake point target mask image and the interference target mask image. is the snowflake target mask image, is to interfere with the target mask image, and mix is ​​the coincidence operator.

[0015] As a preferred embodiment of the present invention, the present invention provides a coronary artery hemodynamic quantification system based on IVUS images, which is applied to a coronary artery hemodynamic quantification method based on IVUS images. The system includes: A data acquisition unit, used for simultaneously acquiring IVUS images of coronary arteries based on coronary angiography of the patient; An image enhancement unit is used to enhance the snowflake point target and the interference target in the IVUS image, respectively, to obtain a snowflake point target enhanced image and an interference target enhanced image; A mask making unit is used to detect snowflake point targets and interference targets on the snowflake point target enhanced image and the interference target enhanced image respectively by using the Mask R-CNN model to obtain a snowflake point target mask image and an interference target mask image; The target segmentation unit is used to process the IVUS image with a fusion mask of the snowflake point target mask image and the interference target mask image to obtain a snowflake point segmentation result image; The hemodynamic quantification unit is used to quantify the hemodynamics based on the snowflake point segmentation result image to obtain the coronary artery vascular hemodynamic parameters.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention utilizes a neural network model to accurately segment the snowflake area in the IVUS image, thereby realizing the accurate extraction of information on the low-echo area in the endothelium caused by blood flow in the IVUS image, and then performing characteristic analysis on the low-echo area in the endothelium caused by blood flow, namely the "snowflake", to realize accurate quantitative evaluation of the hemodynamic parameters of the coronary artery vessels. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0018] Figure 1 A flow chart of a method for quantifying coronary artery hemodynamics based on IVUS images provided in an embodiment of the present invention; Figure 2 A coronary artery hemodynamic quantification system frame based on IVUS images provided in an embodiment of the present invention; Figure 3 A schematic diagram of interference target marking provided by an embodiment of the present invention; Figure 4 The interference target mask image provided by the embodiment of the present invention; Figure 5 A snowflake point target mask image provided by an embodiment of the present invention; Figure 6 A fused mask image provided by an embodiment of the present invention; Figure 7 This is a snowflake point segmentation result image provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, the present invention provides a method for quantifying coronary artery hemodynamics based on IVUS images, comprising the following steps: IVUS images of the coronary arteries are obtained while the patient undergoes coronary angiography; In the IVUS image, the snowflake point target and the interference target are enhanced respectively to obtain the snowflake point target enhanced image and the interference target enhanced image; The Mask R-CNN model is used to detect snowflake point targets and interference targets on the snowflake point target enhanced image and the interference target enhanced image respectively, and the snowflake point target mask image and the interference target mask image are obtained; The IVUS image is processed by a fusion mask of the snowflake point target mask image and the interference target mask image to obtain a snowflake point segmentation result image; The hemodynamics is quantified based on the snowflake point segmentation result image to obtain the coronary artery hemodynamic parameters.

[0021] In order to quantify the precise hemodynamic parameters of the coronary arteries, the present invention uses a neural network model to perform precise target detection on IVUS images, thereby accurately identifying the low-echo area caused by blood flow in the endothelium, i.e., the snowflake target, in the IVUS images of the coronary arteries.

[0022] Specifically, the present invention adopts a combination of direct snowflake point target detection and indirect snowflake point target detection to achieve accurate image detection of snowflake point targets, wherein direct snowflake point target detection includes enhancing the snowflake point targets in the IVUS image and segmenting the enhanced snowflake point targets in the IVUS image (i.e., in the IVUS image, the snowflake point targets are enhanced to obtain a snowflake point target enhanced image; and the Mask R-CNN model is used to detect snowflake point targets and interference targets on the snowflake point target enhanced image to obtain a snowflake point target mask image), and a mask image for extracting snowflake point targets in the IVUS image is obtained. Since the mask image can be directly mapped to the original IVUS image to segment the snowflake point targets therefrom, the snowflake point target detection can segment the snowflake point targets on the IVUS image and directly complete the detection of the snowflake point targets.

[0023] Indirect snowflake point detection includes enhancing the interference targets (i.e., other targets other than snowflake point targets, which will interfere with the recognition of snowflake point targets) in the IVUS image and segmenting the enhanced interference targets in the IVUS image (in the IVUS image, the interference targets are enhanced to obtain an interference target enhanced image; the Mask R-CNN model is used to detect snowflake point targets and interference targets on the interference target enhanced image to obtain an interference target mask image), and a mask image is obtained for extracting interference targets in the IVUS image. Since the mask image can be directly mapped to the original IVUS image to segment the interference targets therefrom, the accurate segmentation of the interference targets will leave the snowflake point targets, which can avoid the snowflake point targets being mistakenly segmented into interference targets. Therefore, interference target detection can segment the interference targets on the IVUS image, eliminate the interference noise during the segmentation of the snowflake point targets, and indirectly complete the detection of the snowflake point targets.

[0024] Therefore, the overlapping part of the snowflake point target mask image obtained by direct detection of the snowflake point target and the interference target mask image is eliminated to eliminate the interference noise in the snowflake point target mask image, thereby obtaining an accurate segmentation result with only snowflake point targets, providing a guarantee for the quantification of accurate hemodynamic parameters of the coronary artery vessels.

[0025] like Figure 3 As shown, interference targets include the lumen of the IVUS catheter, the caliber of the IVUS catheter, and plaques.

[0026] The method for acquiring the snowflake point target enhanced image includes: The Sobel operator is used to perform transverse edge detection and longitudinal edge detection on the IVUS image to obtain an IVUS transverse edge detection image and an IVUS longitudinal edge detection image; The combined image of IVUS transverse edge detection image and IVUS longitudinal edge detection image is added and fused with the IVUS image to obtain the snowflake point target enhanced image; The enhancement operation process of the snowflake target enhanced image is as follows: ; In the formula, Enhance the image for snowflake targets, For IVUS images, is the horizontal detection template of the Sobel operator, is the vertical detection template of the Sobel operator, is the enhancement factor, and All are absolute values; in, ; In the formula, is the L2 norm.

[0027] In the present invention, the Sobel operator for edge detection is used to detect the edge information of the IVUS image. Since the snowflake point targets are discrete and numerous, the number of edges possessed by the snowflake point targets is also large. The edges of the snowflake point targets are detected by using the Sobel operator to form an edge detection image, and the edge detection image is superimposed on the original IVUS image, so that the edge information of the snowflake point targets in the original IVUS image can be superimposed and enhanced.

[0028] In this edge enhancement process, not only the edge information of the snowflake point target is superimposed and enhanced, but the edge information of the interference target will also be enhanced to a certain extent. However, the interference target generally has fewer edges and is relatively different from the snowflake point target. The difference will also be magnified after enhancement. Therefore, when the edge information of the snowflake point target and the edge information of the interference target are enhanced, the edge information of all discrete snowflake point targets can be obtained, and the difference information between the snowflake point target and the interference target is magnified. In comparison, the expectation of snowflake point target enhancement is achieved.

[0029] In addition, an enhancement coefficient is added. The enhancement coefficient is a measure of the edge detection effect between the edge detection image and the original IVUS image. The greater the difference in edge information between the edge detection image and the original IVUS image, the better the edge detection effect. The edge detection credibility is high, and a high enhancement coefficient is assigned so that the edge detection image is highly superimposed on the original IVUS image to improve the snowflake point target enhancement effect. Correspondingly, the edge detection credibility is low, and a low enhancement coefficient is assigned so that the edge detection image is lowly superimposed on the original IVUS image to avoid erroneous edge detection that reduces the snowflake point target enhancement.

[0030] The method for acquiring the interference target enhanced image includes: Subtract and fuse the combined image of IVUS transverse edge detection image and IVUS longitudinal edge detection image with the IVUS image to obtain an interference target enhanced image; The enhancement operation process of the interference target enhanced image is: ; In the formula, Enhance the image for the jammer, For IVUS images, is the horizontal detection template of the Sobel operator, is the vertical detection template of the Sobel operator, is the enhancement factor, and All are absolute values; in, ; In the formula, is the L2 norm.

[0031] In the present invention, the edge information of the IVUS image is detected by using the Sobel operator for edge detection. Since the snowflake point targets are discrete and numerous, the number of edges possessed by the snowflake point targets is also large. The edges of the snowflake point targets are detected by using the Sobel operator to form an edge detection image, and the edge information of the edge detection image is subtracted from the original IVUS image, so that the edge information of the snowflake point targets in the original IVUS image can be superimposed and weakened, and the edge information intensity of the interference target will also be superimposed and weakened during the subtraction process.

[0032] Since the enhancement coefficient is set, the edge information intensity in the edge detection image is attenuated. This subtraction avoids subtracting the edge information of the original IVUS image from the edge detection image, so that the edge information of the snowflake point target and the edge information of the infected target are subtracted from themselves to 0. In addition, since the edge information intensity of the snowflake point target is relatively small and the edge information intensity of the interference target is relatively large, the edge information of the snowflake point target almost disappears after superposition and attenuation, while the edge information of the existing interference target is still amplified compared to the almost disappeared edge information of the snowflake point target. Therefore, the edge information of all discrete snowflake point targets can be made to disappear, while the difference information between the snowflake point targets and the interference targets is amplified, which achieves the expectation of enhancing the interference targets.

[0033] The method for obtaining the snowflake point target mask image includes: Selecting multiple snowflake point target enhanced images, and marking snowflake point area mask images on the snowflake point target enhanced images; A data set is formed by combining multiple snowflake target enhanced images and snowflake region mask images, and a Mask R-CNN model is trained based on the data set to pre-establish a snowflake target detection model for detecting snowflake targets in snowflake target enhanced images; Using the snowflake target detection model, the snowflake target is detected on the snowflake target enhanced image to obtain the snowflake target mask image, such as Figure 5 As shown; The snowflake target detection model is: ; In the formula, is the snowflake target mask image, The image is enhanced for snowflake targets, and Mask R-CNN is the Mask R-CNN model.

[0034] The method for obtaining the interference target mask image includes: Selecting multiple interference target enhanced images, and marking interference area mask images on the interference target enhanced images; A plurality of interference target enhanced images and interference area mask images are combined into a data set, and a Mask R-CNN model is trained based on the data set to pre-establish an interference target detection model for detecting interference targets in interference target enhanced images; Using the interference target detection model, the interference target is detected on the interference target enhanced image to obtain the interference target mask image, such as Figure 4 As shown; The interference target detection model is: ; In the formula, To interfere with the target mask image, To enhance the image for distracting targets, Mask R-CNN is the Mask R-CNN model.

[0035] The method for obtaining the snowflake point segmentation result image includes: In the snowflake point target mask image, the overlapping parts of the snowflake point target and the interference target mask image are removed to obtain a fused mask, such as Figure 6 As shown; Pixel extraction is performed on the IVUS image based on the fusion mask to obtain the snowflake point segmentation result image, such as Figure 7 As shown; The calculation process of the fusion mask is: ; In the formula, is the fusion mask, is the snowflake target mask image, is the interference target mask image, mix is ​​the coincidence operator, It is the overlapping part of the snowflake point target and the interference target in the snowflake point target mask image and the interference target mask image.

[0036] Methods for quantifying coronary vascular hemodynamic parameters include: The snowflake segmentation result images of the IVUS images of each coronary artery cross-section (i.e. Figure 6 In the snowflake target), count the number of blood flow snowflakes; The number of pixels of each blood flow snowflake point is counted, and the sum is calculated to obtain the blood flow snowflake point area of ​​each coronary artery cross-section IVUS image; The blood flow snowflake volume is obtained by multiplying the blood flow snowflake area of ​​the IVUS image of each coronary artery cross section by the thickness of each coronary artery cross section. In the snowflake point segmentation result image of the IVUS image of each coronary artery cross-section, the lumen cross-sectional area of ​​the coronary artery is calculated; The difference between the lumen cross-sectional area of ​​the coronary artery in the IVUS image of each coronary artery cross-section and the normal lumen cross-sectional area of ​​the coronary artery is calculated by ratio with the normal lumen cross-sectional area of ​​the coronary artery to obtain the degree of coronary artery stenosis; The blood flow velocity at the coronary artery is obtained by multiplying the ratio of the lumen cross-sectional area of ​​the coronary artery in the IVUS image to the normal lumen cross-sectional area of ​​the coronary artery by the normal blood flow velocity; The square of the blood velocity in the coronary arteries minus the square of the normal blood velocity multiplied by a factor of 4 gives the pressure gradient in the coronary arteries, which can be estimated by the Bernoulli equation, especially in cases of significant stenosis.

[0037] Among them, the normal lumen cross-sectional area and normal blood flow velocity of the coronary artery are within a rough range due to individual differences. The left main coronary artery (LMCA): usually has a diameter of about 3.5-4.5 mm, which means a cross-sectional area of ​​about 10-16 square millimeters. The left anterior descending branch (LAD) and the left circumflex branch (LCx): usually have a diameter of about 2.5-4 mm, and a cross-sectional area of ​​5-12 square millimeters. The right coronary artery (RCA): usually has a diameter of about 2.5-3.5 mm, and a cross-sectional area of ​​5-10 square millimeters. In this case, the final calculation may lead to deviations. In order to eliminate this deviation as much as possible, the snowflake point values ​​of multiple coronary arteries in one or more cardiac cycles displayed by the electrocardiogram are calculated, so as to obtain the mean value of the coronary artery hemodynamic parameters, which is more accurate.

[0038] The blood flow velocity in the coronary arteries is multiplied by the blood flow snowflake area to obtain the blood flow in the coronary arteries.

[0039] The loss function for training the Mask R-CNN model is the segmentation mask loss, which is: ; In the formula, is the segmentation mask loss, is the fusion mask, is the true value of the snowflake target mask image, for The total number of pixels, is the overlapped part of the snowflake point target and the interference target in the snowflake point target mask image and the interference target mask image. is the snowflake target mask image, is to interfere with the target mask image, and mix is ​​the coincidence operator.

[0040] The present invention uses segmentation mask loss as the loss function for training the Mask R-CNN model. It measures the fusion mask between the snowflake target mask image and the interference target mask image, that is, the difference between the target segmentation result of the combination of direct detection and indirect detection and the real result. Taking this as the loss function, the snowflake target mask image and the interference target mask image detected by the trained Mask R-CNN model can achieve the best segmentation result and the most accurate target detection result. It measures the overlap between the snowflake point target and the interference target in the snowflake point target mask image and the interference target mask image. Taking this as the loss function, the snowflake point target mask image detected by the trained Mask R-CNN model can achieve the minimum overlap with the interference target mask image, ensuring the minimum overlap between the segmented snowflake point target and the interference target, and achieving accurate results in both snowflake point target detection and interference target detection.

[0041] like Figure 2 As shown, the present invention provides a coronary artery hemodynamic quantification system based on IVUS images, which is applied to a coronary artery hemodynamic quantification method based on IVUS images. The system includes: A data acquisition unit, used for simultaneously acquiring IVUS images of coronary arteries based on coronary angiography of the patient; An image enhancement unit is used to enhance the snowflake point target and the interference target in the IVUS image, respectively, to obtain a snowflake point target enhanced image and an interference target enhanced image; A mask making unit is used to detect snowflake point targets and interference targets on the snowflake point target enhanced image and the interference target enhanced image respectively by using the Mask R-CNN model to obtain a snowflake point target mask image and an interference target mask image; The target segmentation unit is used to process the IVUS image with a fusion mask of the snowflake point target mask image and the interference target mask image to obtain a snowflake point segmentation result image; The hemodynamic quantification unit is used to quantify the hemodynamics based on the snowflake point segmentation result image to obtain the coronary artery vascular hemodynamic parameters.

[0042] The present invention is different from the FFR (Fractional Flow Reserve) and QFR (Quantitative Flow Ratio) in the coronary artery function assessment method. It can quantitatively quantify the hemodynamics of the coronary artery without adding an additional catheter guidewire, and further evaluate the impact of coronary artery stenosis on hemodynamics. In actual operation, reading IVUS data requires professional interventional technicians, and there is a great threshold for reading images. The automated detection system of the present invention can not only quickly read images and detect the location and characteristics of low-echo areas and other related current technologies, but also detect the patient's coronary artery hemodynamic parameters, quantify the blood flow before and after stent implantation, reduce the threshold for reading images, and save various costs and time; secondly, it avoids the side effects of injecting the vasodilator drug to be tested, which will cause the patient to have an uncomfortable reaction, and the guidewire is easy to cause damage to the patient's blood vessels to be tested during the intervention process. This method provides more detailed and quantitative data for clinical assessment of the severity of coronary artery disease, especially for preoperative hemodynamic assessment of patients, intraoperative plaque characteristics and stability, personalized treatment, and comprehensive postoperative assessment, and fills the gaps in some traditional assessment methods.

[0043] The present invention utilizes a neural network model to accurately segment the snowflake area in the IVUS image, thereby realizing the accurate extraction of information on the low-echo area in the endothelium caused by blood flow in the IVUS image, and then performing characteristic analysis on the low-echo area in the endothelium caused by blood flow, namely the "snowflake", to realize accurate quantitative evaluation of the hemodynamic parameters of the coronary artery vessels.

[0044] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A method for quantifying coronary artery hemodynamics based on IVUS images, characterized in that: The following steps are involved: IVUS images of the coronary arteries are obtained while the patient undergoes coronary angiography; In the IVUS image, the snowflake point target and the interference target are enhanced respectively to obtain the snowflake point target enhanced image and the interference target enhanced image; The Mask R-CNN model is used to detect snowflake point targets and interference targets on the snowflake point target enhanced image and the interference target enhanced image respectively, and the snowflake point target mask image and the interference target mask image are obtained; The IVUS image is processed by a fusion mask of the snowflake point target mask image and the interference target mask image to obtain a snowflake point segmentation result image; The hemodynamics is quantified based on the snowflake point segmentation result image to obtain the coronary artery hemodynamic parameters.

2. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 1, characterized in that: The interference targets include the lumen of the IVUS catheter, the caliber of the IVUS catheter, and plaques.

3. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 2, characterized in that: The method for acquiring the snowflake point target enhanced image comprises: The Sobel operator is used to perform transverse edge detection and longitudinal edge detection on the IVUS image to obtain an IVUS transverse edge detection image and an IVUS longitudinal edge detection image; The combined image of IVUS transverse edge detection image and IVUS longitudinal edge detection image is added and fused with the IVUS image to obtain the snowflake point target enhanced image; The enhancement operation process of the snowflake point target enhanced image is as follows: ； In the formula, Enhance the image for snowflake targets, For IVUS images, is the horizontal detection template of the Sobel operator, is the vertical detection template of the Sobel operator, is the enhancement factor, and All are absolute values; in, ; In the formula, is the L2 norm.

4. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 3, characterized in that: The method for acquiring the interference target enhanced image comprises: Subtract and fuse the combined image of IVUS transverse edge detection image and IVUS longitudinal edge detection image from the IVUS image to obtain an interference target enhanced image; The enhancement operation process of the interference target enhanced image is as follows: ； In the formula, Enhance the image for the jammer, For IVUS images, is the horizontal detection template of the Sobel operator, is the vertical detection template of the Sobel operator, is the enhancement factor, and All are absolute values; in, ; In the formula, is the L2 norm.

5. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 4, characterized in that: The method for acquiring the snowflake point target mask image comprises: Selecting multiple snowflake point target enhanced images, and marking snowflake point area mask images on the snowflake point target enhanced images; A data set is formed by combining multiple snowflake target enhanced images and snowflake region mask images, and a Mask R-CNN model is trained based on the data set to pre-establish a snowflake target detection model for performing snowflake target detection on snowflake target enhanced images; Using the snowflake point target detection model, the snowflake point target is detected on the snowflake point target enhanced image to obtain the snowflake point target mask image; The snowflake target detection model is: ； In the formula, is the snowflake target mask image, The image is enhanced for snowflake targets, and Mask R-CNN is the Mask R-CNN model.

6. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 5, characterized in that: The method for acquiring the interference target mask image includes: Selecting multiple interference target enhanced images, and marking interference area mask images on the interference target enhanced images; A plurality of interference target enhanced images and interference area mask images are combined into a data set, and a Mask R-CNN model is trained based on the data set to pre-establish an interference target detection model for detecting interference targets in interference target enhanced images; Using the interference target detection model, the interference target is detected on the interference target enhanced image to obtain the interference target mask image; The interference target detection model is: ； In the formula, To interfere with the target mask image, To enhance the image for distracting targets, MaskR-CNN is the Mask R-CNN model.

7. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 6, characterized in that: The method for obtaining the snowflake point segmentation result image comprises: In the snowflake point target mask image, the overlapping parts of the snowflake point target and the interference target mask image are removed to obtain the fused mask; Perform pixel extraction on the IVUS image based on the fusion mask to obtain the snowflake point segmentation result image; The calculation process of the fusion mask is as follows: ； In the formula, is the fusion mask, is the snowflake target mask image, is the interference target mask image, mix is ​​the coincidence operator, It is the overlapping part of the snowflake point target and the interference target in the snowflake point target mask image and the interference target mask image.

8. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 7, characterized in that: The method for quantifying the coronary artery hemodynamic parameters comprises: In the snowflake point segmentation result image of the IVUS image of each coronary artery cross-section, the number of blood flow snowflake points is counted; The number of pixels of each blood flow snowflake point is counted, and the sum is calculated to obtain the blood flow snowflake point area of ​​each coronary artery cross-section IVUS image; The blood flow snowflake volume is obtained by multiplying the blood flow snowflake area of ​​the IVUS image of each coronary artery cross section by the thickness of each coronary artery cross section. In the snowflake point segmentation result image of the IVUS image of each coronary artery cross-section, the lumen cross-sectional area of ​​the coronary artery is calculated; The difference between the lumen cross-sectional area of ​​the coronary artery in the IVUS image of each coronary artery cross-section and the normal lumen cross-sectional area of ​​the coronary artery is calculated by ratio with the normal lumen cross-sectional area of ​​the coronary artery to obtain the degree of coronary artery stenosis; The blood flow velocity at the coronary artery is obtained by multiplying the ratio of the lumen cross-sectional area of ​​the coronary artery in the IVUS image to the normal lumen cross-sectional area of ​​the coronary artery by the normal blood flow velocity; The square of the blood flow velocity in the coronary artery minus the square of the normal blood flow velocity multiplied by a coefficient of 4 gives the pressure gradient in the coronary artery. The blood flow velocity in the coronary arteries is multiplied by the blood flow snowflake area to obtain the blood flow in the coronary arteries.

9. The method for quantifying coronary artery hemodynamics based on IVUS images according to claim 7, characterized in that: The loss function for training the Mask R-CNN model is the segmentation mask loss, which is: ； In the formula, is the segmentation mask loss, is the fusion mask, is the true value of the snowflake target mask image, for The total number of pixels, is the overlapped part of the snowflake point target mask image and the interference target mask image. is the snowflake target mask image, is to interfere with the target mask image, and mix is ​​the coincidence operator.

10. A coronary artery hemodynamic quantification system based on IVUS images, characterized in that: A method for quantifying coronary artery hemodynamics based on IVUS images as described in any one of claims 1 to 9, the system comprising: A data acquisition unit, used for simultaneously acquiring IVUS images of coronary arteries based on coronary angiography of the patient; An image enhancement unit is used to enhance the snowflake point target and the interference target in the IVUS image, respectively, to obtain a snowflake point target enhanced image and an interference target enhanced image; A mask making unit is used to detect snowflake point targets and interference targets on the snowflake point target enhanced image and the interference target enhanced image respectively by using the Mask R-CNN model to obtain a snowflake point target mask image and an interference target mask image; The target segmentation unit is used to process the IVUS image with a fusion mask of the snowflake point target mask image and the interference target mask image to obtain a snowflake point segmentation result image; The hemodynamic quantification unit is used to quantify the hemodynamics based on the snowflake point segmentation result image to obtain the coronary artery vascular hemodynamic parameters.

Images