Image dynamic processing method and imaging device

Abstract

The present invention belongs to the field of ultrasonic imaging technology and provides a method for dynamic image processing, including: acquiring multiple ultrasonic images of the area to be imaged, dividing the total grayscale range of the image into multiple step grayscale ranges based on the multiple ultrasonic images; extracting the ultrasonic image to be processed based on the step grayscale ranges to obtain multiple ultrasonic feature maps; processing the multiple ultrasonic feature maps to obtain multiple processed ultrasonic feature maps; merging the multiple processed ultrasonic feature maps to obtain a target ultrasonic image. The present invention also provides an imaging device for implementing the above-mentioned dynamic image processing method, which is particularly suitable for ultrasonic imaging of body tissues with periodic physiological characteristics. The present invention automatically divides the grayscale of the ultrasonic image according to the actual situation of the area to be imaged, without the need for manual intervention or presetting a fixed grayscale range. It can selectively enhance the ultrasonic image and suppress noise and speckle, thereby improving the imaging quality of the target ultrasonic image.

Description

Technical Field

[0001] The present invention belongs to the technical field of ultrasonic imaging, and in particular relates to an image dynamic processing method and imaging equipment. Background Art

[0002] In vivo ultrasound imaging technology is a new diagnostic method that has been clinically used in recent years to diagnose vascular and cardiac diseases. Intravascular ultrasound imaging can display the histomorphological characteristics of the vascular lumen, wall and atherosclerotic plaques, as well as the pathological composition of plaques, such as calcification, fibrous tissue, lipid core and plaque rupture, in vitro. Intracardiac ultrasound imaging (echocardiography) has been routinely used in the diagnosis, treatment and follow-up of cardiac diseases.

[0003] Clear ultrasound images are essential for doctors to effectively extract features, identify lesions, and make accurate diagnoses. However, actual ultrasound images are often affected by factors such as pixel noise, speckle, and low contrast. Furthermore, the different locations, shapes, and types of physiological structures have varying effects on ultrasound waves. These factors can degrade image quality, making tissue identification difficult, imaging physiological structures unclear, and making image post-processing difficult, thus affecting ultrasound diagnostic results.

[0004] Conventional imaging methods for reducing ultrasound image noise and suppressing speckle currently include filtering algorithms and spatial compounding algorithms. Filtering algorithms typically smooth the original image while preserving image features and edges. The principle of spatial compounding algorithms is to compound multiple images extracted from the same imaging object at different locations. This method enhances image features that appear in all images and suppresses random features such as noise and speckle that appear in individual images. However, none of the above imaging methods enhance image contrast. The histogram-based threshold method, a commonly used method for enhancing ultrasound image contrast in the prior art, works by changing the image grayscale value within the threshold range through grayscale transformation to achieve the purpose of enhancing image contrast. However, since the thresholds in the above methods are typically fixed preset values ​​or based on specific assumptions, when the input image conditions change, it is easy for image pixels to be misclassified. Summary of the Invention

[0005] Based on this, an object of the present invention is to provide an image processing method and imaging device that effectively reduce ultrasonic image noise and speckle while enhancing ultrasonic image contrast.

[0006] An object of the present invention is to provide a method for dynamic image processing, comprising the following steps:

[0007] Acquire multiple ultrasonic images of the area to be imaged, and divide the total grayscale range of the image into multiple step grayscale ranges according to the multiple ultrasonic images;

[0008] Extracting the ultrasonic image to be processed according to the graded grayscale range to obtain a plurality of ultrasonic feature maps;

[0009] Processing the multiple ultrasonic characteristic images to obtain multiple processed ultrasonic characteristic images;

[0010] The multiple processed ultrasonic feature maps are merged to obtain a target ultrasonic image.

[0011] Furthermore, the image dynamic processing method is used to process ultrasonic images of body tissues with periodic physiological characteristics, and the specific method for acquiring multiple ultrasonic images of the area to be imaged is: acquiring multiple ultrasonic images of the area to be imaged within the periodic time T of at least one physiological characteristic.

[0012] Furthermore, the method for decomposing image grayscale according to multiple ultrasound images includes:

[0013] S11, respectively counting the total number of accumulated pixels H of the kth grayscale of each ultrasound image in the plurality of ultrasound images k , calculate the cumulative total number of pixels H of the multiple ultrasound images k The standard deviation σ(H k ), determine its standard deviation σ(H k ) is the maximum value, and the corresponding grayscale λ1 is the first grayscale threshold. The total grayscale range is divided into the first grayscale range [0, λ1] and the second grayscale range [λ1, K] according to the first grayscale threshold λ1. The total number of pixels H k and its standard deviation σ(H k ) are respectively expressed as:

[0014]

[0015]

[0016] Wherein, k is the grayscale count in each ultrasound image, K is the total grayscale count of the image, and h i is the number of grayscale pixels of the i-th grayscale, t is the t-th ultrasound image in time T, N is the total number of ultrasound images, H k (t) is the total number of pixels with the kth gray level in the tth ultrasound image; μ(H k ) is the average value of the total number of k-th grayscale cumulative pixels in N ultrasonic images within the time period;

[0017] S12, within the grayscale range obtained in the previous step, take the coordinate point As the starting point, coordinate point Get the line for the end point Determine the distance from the straight line The coordinate point P when the distance is the maximum x(x, σ(H x )) The grayscale x corresponding to it is the grayscale threshold λ x , with each gray threshold λ obtained in this step x The grayscale range obtained in the previous step is further divided to obtain multiple grayscale ranges [λ a ,λ x ] and [λ x ,λ b ];

[0018] Among them, the λ a and λ b are the endpoint values ​​of each grayscale range obtained in the previous step, and the grayscale x∈[λ a ,λ b ];

[0019] S13, the grayscale ranges [λ a ,λ x ] and [λ x ,λ b ], calculate the distance from the straight line The distance is the minimum and the distance from λ x The grayscale y1 and grayscale y2 corresponding to the coordinate point with the smallest absolute value of the difference, where grayscale y1∈[λ a ,λ x ], grayscale y2∈[λ x ,λ b ];

[0020] S14, when the grayscale y1 and the grayscale y2 are the boundary values ​​of the grayscale range, further segmentation is stopped within the grayscale range;

[0021] When the grayscale y1 and grayscale y2 are not within the grayscale range boundary value, the grayscale threshold λ is determined. y1 and grayscale threshold λ y2 , with gray threshold λ y1 For grayscale range [λ a ,λ x ] is further segmented with gray threshold λ y2 For grayscale range [λ x ,λ b ] to further segment and obtain multiple grayscale ranges [λ a ,λ y1 ],[λ y1 ,λ x ],[λ x ,λ y2 ] and [λ y2 ,λ b ] and then return to step S13;

[0022] S15 , when segmentation is stopped in each grayscale range, determining each grayscale range as a plurality of graded grayscale ranges obtained by segmenting the total grayscale range of the image.

[0023] Furthermore, the steps of acquiring multiple ultrasonic images of the area to be imaged and dividing the total grayscale range of the image into multiple step grayscale ranges according to the multiple ultrasonic images are specifically as follows:

[0024] S11, collecting multiple ultrasound images to be processed within time T;

[0025] S12, determining effective grayscale ranges of multiple ultrasound images to be processed, and mapping the effective grayscale ranges to the total grayscale range of the images to obtain a mapped grayscale range;

[0026] S13. Divide the mapped grayscale range into multiple stepped grayscale ranges according to the multiple ultrasound images.

[0027] Furthermore, the method for determining the effective grayscale range of the multiple ultrasound images is:

[0028] Calculate the probability p of the kth grayscale occurrence in the multiple ultrasound images respectively k and the standard deviation of the grayscale occurrence probability in multiple ultrasound images σ(p k ), which can be expressed as:

[0029] p k =h k / M;

[0030]

[0031] Among them, h k is the number of pixels of the kth grayscale, M is the total number of pixels in the ultrasound image; p k (t) is the probability of the kth grayscale appearing in the tth ultrasound image; μ(p k ) is the average probability of occurrence of the kth grayscale in N ultrasound images within the time period;

[0032] When σ(p k ) value is less than the preset value σ l and σ h When , the corresponding grayscale k is the boundary threshold λ of the effective grayscale range l and λ h , which can be expressed as:

[0033] λ l ={max(i),i∈[1,K]:[σ(p0),σ(p i )]<σ l};

[0034] λh ={min(j),j∈[1,K]:[σ(p j ),σ(p K )]<σ h}.

[0035] Furthermore, the step of processing the plurality of ultrasonic characteristic images to obtain the plurality of processed ultrasonic characteristic images includes a region processing step and / or a time processing step.

[0036] The regional processing step includes:

[0037] Region division: dividing the multiple ultrasonic characteristic images into multiple non-overlapping sub-regions;

[0038] Region transformation: performing image transformation on the multiple sub-regions to obtain multiple transformed sub-regions, wherein the image transformation method is S-curve transformation or gamma transformation;

[0039] Regional compounding: Merge multiple transformed sub-regions of each ultrasonic feature map to obtain multiple regional processed ultrasonic feature maps;

[0040] The time processing step includes:

[0041] Time point division: dividing the plurality of regional processed ultrasound feature maps into a plurality of time images according to the ultrasound image acquisition time range;

[0042] Time point transformation: Select one of the time images as a reference image, and perform image registration on other time images in different time ranges with the reference image as the target to obtain multiple registered images. The image registration method is rigid transformation, affine transformation or elastic transformation, and multiple transformed time images are determined to be the reference image and multiple registered images;

[0043] Time point compounding: Multiple transformed time images of each region processed ultrasound image are merged to obtain multiple processed ultrasound feature maps.

[0044] Furthermore, the steps of merging the multiple processed ultrasonic feature maps to obtain a target ultrasonic image are as follows:

[0045] Performing grayscale transformation on the multiple processed ultrasonic characteristic images to obtain multiple transformed ultrasonic characteristic images, wherein the transformation method includes linear transformation, S-curve transformation, logarithmic transformation or gamma transformation;

[0046] The multiple transformed ultrasonic feature maps are merged to obtain a target ultrasonic image.

[0047] Furthermore, the step of obtaining the target ultrasonic image also includes the steps of filtering the multiple transformed ultrasonic feature maps, retaining feature edges, smoothing the image, and enhancing contrast.

[0048] A second object of the present invention is to provide an imaging device comprising:

[0049] An ultrasonic transducer, the ultrasonic transducer being used to collect ultrasonic image signals of a patient's body tissue;

[0050] a processor, configured to receive the ultrasonic image signal and dynamically process the ultrasonic image signal;

[0051] A storage medium, wherein a plurality of instructions are stored in the storage medium, and when the instructions are executed by a processor, any one of the above-mentioned methods for dynamic image processing is performed.

[0052] Furthermore, the imaging device also includes a catheter body, the distal end of which can be inserted into the area to be imaged in the patient's body through a human lumen; and the ultrasonic transducer is arranged at the distal end of the catheter body.

[0053] The dynamic image processing method provided by the present invention can selectively enhance the ultrasound image and suppress noise and speckle by grayscale-dividing multiple dynamic ultrasound images of the area to be imaged. A multiple composite algorithm that combines regional processing and time point processing features can further suppress noise and speckle, thereby improving the imaging quality of the target ultrasound image. The image processing method disclosed by the present invention automatically obtains each grayscale range based on the actual situation of the area to be imaged, without the need for manual intervention or presetting a fixed grayscale range, thereby making the image processing results more accurate. The dynamic image processing method and imaging device provided by the present invention are particularly suitable for imaging body tissues with periodic physiological characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flowchart of the image dynamic processing method in the first embodiment of the present invention. DETAILED DESCRIPTION

[0055] To make the content of the present invention clearer, the present invention is further described below in conjunction with the accompanying drawings and embodiments. It should be understood that the present invention is not limited to the specific embodiments described below, and the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

[0056] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

[0057] The present invention provides a dynamic image processing method for processing an ultrasound image to be processed in an imaging area to reduce the influence of factors such as noise, speckle and low contrast on the ultrasound image, comprising the following steps:

[0058] Acquire multiple ultrasonic images of the area to be imaged, and divide the total grayscale range of the image into multiple step grayscale ranges according to the multiple ultrasonic images;

[0059] Extracting the ultrasonic image to be processed according to the graded grayscale range to obtain a plurality of ultrasonic feature maps;

[0060] Processing the multiple ultrasonic characteristic images to obtain multiple processed ultrasonic characteristic images;

[0061] The multiple processed ultrasonic feature maps are merged to obtain a target ultrasonic image.

[0062] The dynamic ultrasonic imaging method provided by the present invention can automatically calculate multiple thresholds that produce obvious grayscale changes in the image, and then divide the grayscale range of the ultrasonic image to be processed in the same area according to the multiple thresholds, which is beneficial to enhancing the image contrast of ultrasonic diagnosis and suppressing noise and speckle, thereby improving the quality of the processed ultrasonic image; and by collecting multiple ultrasonic images of the area to be imaged and performing grayscale segmentation based on the collected ultrasonic images, the grayscale segmentation can have better adaptability, making the image processing results more accurate; segmenting the image to be processed according to the grayscale range not only retains the complete information in the original image as a whole, but also each ultrasonic feature map shows independent feature information, so that beneficial features can be targetedly enhanced and interference features can be suppressed; after processing multiple ultrasonic feature maps separately and then compounding them, they can be selectively processed at the feature level, thereby achieving the purpose of improving ultrasonic imaging.

[0063] It should be noted that, in the step of acquiring multiple ultrasonic images of the area to be imaged, the acquisition may be continuous video images within a certain period of time, or may be discontinuous multiple images.

[0064] The number of ultrasound image acquisition frames is at least two, preferably, 30 to 150.

[0065] The present invention places no particular restrictions on image acquisition time. The image processing method provided by the present invention is particularly suitable for imaging body tissues with periodic physiological characteristics. Multiple ultrasound images can be acquired within at least one period of the physiological characteristics of the body tissue. By performing grayscale segmentation on ultrasound images within the period of the physiological characteristics of the body tissue, the grayscale segmentation range can be made more applicable.

[0066] In addition, for body tissues with periodic physiological characteristics, their image grayscale and corresponding pixel numbers will also show periodic changes due to their periodic changes. Therefore, analyzing the grayscale changes within a certain period of time and dividing the grayscale range according to the obvious changes in the number of grayscale pixels is more applicable to the extraction of the processed image.

[0067] In the present invention, the ultrasound image to be processed is a captured ultrasound image of an imaging area. There is no order in which to capture the ultrasound image to be processed and capture multiple ultrasound images of the imaging area. Multiple ultrasound images of the imaging area can be captured first, followed by the ultrasound image to be processed; or multiple ultrasound images of the imaging area can be captured first, followed by multiple ultrasound images of the imaging area; or multiple ultrasound images can be captured within a certain period of time, with one or more of them serving as the ultrasound image to be processed.

[0068] refer to Figure 1 A first embodiment of the present invention provides a dynamic image processing method for performing ultrasonic imaging and image processing on cardiac tissue. The processing method includes the following steps:

[0069] S1. Acquire multiple ultrasound images of the area to be imaged within a time T, and divide the total grayscale range of the image into multiple step grayscale ranges according to the multiple ultrasound images;

[0070] S2. Acquire an ultrasonic image to be processed, extract the ultrasonic image to be processed according to the graded grayscale range, and obtain multiple ultrasonic feature maps;

[0071] S3. Processing the plurality of ultrasonic characteristic images respectively to obtain a plurality of processed ultrasonic characteristic images;

[0072] S4. Merge the multiple processed ultrasound feature maps to obtain a target ultrasound image.

[0073] Since heartbeat has a certain periodicity, its ultrasound image will also show periodic grayscale changes over time. For example, for the ultrasound image to be processed, a dark grayscale feature map, a medium grayscale feature map, and a bright grayscale feature map can be extracted based on the grayscale range.

[0074] For heart tissue, the time T is at least one heartbeat cycle. In this embodiment, the time T is one heartbeat cycle.

[0075] The plurality of ultrasound images refers to the number of ultrasound images acquired within one cardiac cycle, which is generally related to the acquisition conditions of the imaging device and preferably ranges from 30 to 150. In this embodiment, the number of ultrasound images is 60, meaning that 60 ultrasound images of cardiac tissue are acquired within one cardiac cycle.

[0076] Furthermore, the step S1 includes the following steps:

[0077] S11, acquiring multiple ultrasound images of the area to be imaged within a time T;

[0078] S12, respectively counting the total number of accumulated pixels H of the kth grayscale of each ultrasound image in the plurality of ultrasound images k , calculate the cumulative total number of pixels H of the kth grayscale of the multiple ultrasound images k The standard deviation σ(H k ), determine the cumulative standard deviation of the total number of pixels σ(H k ) is the maximum value and the corresponding grayscale λ1 is the first grayscale threshold. The total grayscale range is divided into the first grayscale range [0, λ1] and the second grayscale range [λ1, K] according to the first grayscale threshold λ1;

[0079] The total number of accumulated pixels of the kth gray level H k and the cumulative standard deviation of the total number of pixels σ(H k ) are respectively expressed as:

[0080]

[0081]

[0082] Wherein, k is the grayscale count in each ultrasound image, K is the total grayscale count of the image, and h i is the number of grayscale pixels of the i-th grayscale, t is the t-th ultrasound image in time T, N is the total number of ultrasound images, H k (t) is the total number of pixels with the kth gray level in the tth ultrasound image; μ(H k ) is the average value of the total number of k-th grayscale cumulative pixels in N ultrasonic images within the time period;

[0083] S13, within the grayscale range obtained in the previous step, take the coordinate point As the starting point, coordinate point Get the line for the end point Determine the distance from the straight line The coordinate point P when the distance is the maximum x (x, σ(H x )) The grayscale x corresponding to it is the grayscale threshold λ x , with each gray threshold λ obtained in this step x The grayscale range obtained in the previous step is further divided to obtain multiple grayscale ranges [λ a ,λ x ] and [λ x ,λ b ];

[0084] Among them, the λa and λ b are the endpoint values ​​of each grayscale range obtained in the previous step, and the grayscale x∈[λ a ,λ b ];

[0085] S14, the grayscale ranges [λ a ,λ x ] and [λ x ,λ b ], calculate the distance from the straight line The distance is the minimum and the distance from λ x The grayscale y1 and grayscale y2 corresponding to the coordinate point with the smallest absolute value of the difference, where grayscale y1∈[λ a ,λ x ], grayscale y2∈[λ x ,λ b ];

[0086] S15, when the grayscale y1 and the grayscale y2 are at the boundary values ​​of the grayscale range, further segmentation is stopped within the grayscale range;

[0087] When the grayscale y1 and grayscale y2 are not within the grayscale range boundary value, the grayscale threshold λ is determined. y1 And gray threshold λy2, gray threshold λ y1 For grayscale range [λ a ,λ x ] is further segmented with gray threshold λ y2 For grayscale range [λ x ,λ b ] to further segment and obtain multiple grayscale ranges [λ a ,λ y1 ],[λ y1 ,λ x ],[λ x ,λ y2 ] and [λ y2 ,λ b ] and then return to step S14

[0088] S16 , when segmentation is stopped in each grayscale range, determining each grayscale range as a plurality of graded grayscale ranges obtained by segmenting the total grayscale range of the image.

[0089] It should be noted that when step S12 is completed and step S13 is entered, within the first grayscale range [0, λ1], λ in step S13 is a is 0, λ b is λ1. That is, the coordinate point As the starting point, the coordinate point P0(0,σ(H o ))Get the straight line for the end point Define the distance line Coordinate point P when the distance is maximum x (x, σ(H x ))The grayscale x corresponding to the second grayscale threshold λ2 is further divided by the second grayscale threshold λ2 to obtain the grayscale range [0, λ2] and the grayscale range [λ2, λ1], where the grayscale x∈[0, λ1].

[0090] Likewise, within the second grayscale range [λ1, K], λ in step S13 a is λ1,λ b is K. That is, the coordinate point As the starting point, the coordinate point PK (K, σ (HK)) is the end point to obtain a straight line Determine the distance from the straight line Coordinate point P when the distance is maximum x (x, σ(H x ))The grayscale x corresponding to the third grayscale threshold λ3 is used to further divide the grayscale range [λ1, K] to obtain the grayscale range [λ1, λ2] and the grayscale range [λ2, K] where x∈[λ1, K].

[0091] Specifically, in step S13, the specific calculation method of the second grayscale threshold λ2 and the third grayscale threshold λ3 and the method of segmenting the first grayscale range [O, λ1] and the second grayscale range [λ1, K] are as follows:

[0092] straight line It can be expressed as:

[0093]

[0094] Coordinate point P x (x, σ(H x ))Distance from straight line The distance is When the distance The coordinate point P is the maximum value x (x, σ(H x )) The grayscale x corresponding to the second grayscale threshold λ2 can be expressed as:

[0095]

[0096] Similarly, the straight line It can be expressed as:

[0097]

[0098] Coordinate point P x (x, σ(H x ))Distance from straight line The distance is When the distance The coordinate point P is the maximum value x (x, σ(H x )) The grayscale x corresponding to the third grayscale threshold λ3 can be expressed as:

[0099]

[0100] After determining the second grayscale threshold λ2 and the third grayscale threshold λ3, the first grayscale range [0, λ1] can be further divided into [0, λ2] and [λ2, λ1]; the second grayscale range [λ1, K] can be further divided into [λ1, λ3] and [λ3, K].

[0101] In step S14, the grayscale range [0, λ2] obtained in step S13 is used as an example for specific description, and the grayscale range [0, λ2] is defined to determine the distance from the straight line. The grayscale y corresponding to the coordinate point with the smallest distance and the smallest absolute difference from the second grayscale threshold λ2 is the fourth grayscale threshold λ4. The grayscale y can be expressed as:

[0102]

[0103] Wherein, abs(y-λ2) is the absolute value of the difference between the grayscale y and the second grayscale threshold λ2.

[0104] It should be noted that, in step S15, when the grayscale y reaches the boundary value of the grayscale range, further segmentation within the grayscale range is stopped.

[0105] The grayscale range [0, λ2] is taken as an example for specific description; if the grayscale y=0 or the grayscale y=λ2, λ4 is not included in the grayscale threshold, that is, [0, λ2] is the first segmentation grayscale range determined.

[0106] If grayscale y≠0 and grayscale y≠λ2, λ4 is used as the grayscale threshold, and the grayscale range [0, λ2] can be further divided into [0, λ4] and [λ4, λ2], and the process returns to step S15.

[0107] It should be noted that in step S16, when the segmentation is stopped in each grayscale range, the multiple grayscale thresholds obtained within the total grayscale range of the image are defined as λ c1 ,λ c2 ,λ c3 ...λ cn , the multiple grayscale ranges are [0, λ c1 ]、[λ c1 ,λ c2 ]、[λ c2 ,λ c3 ]……[λ c(n-1) ,λcn ].

[0108] In this embodiment, the step of acquiring the ultrasound image to be processed in step S2 is performed simultaneously with the step of acquiring multiple ultrasound images of the area to be imaged within time T in step S11. Specifically, the ultrasound image to be processed is one or more of the multiple ultrasound images acquired in step S11.

[0109] In step S2 of this embodiment, according to the multiple grayscale ranges [0, λ c1 ]、[λ c1 ,λ c2 ]、[λ c2 ,λ c3 ]……[λ c(n-1) ,λ cn ]The ultrasonic image to be processed is extracted to obtain multiple ultrasonic feature maps.

[0110] In the present invention, the step S3 includes a region processing step and / or a time point processing step, and the order of the region processing and the time processing is irrelevant.

[0111] In this embodiment, step S3 includes a regional processing step and a time processing step. First, the multiple ultrasonic feature images are regionally processed to obtain multiple regional processed ultrasonic feature images; then, the multiple regional processed ultrasonic feature images are time-processed to obtain multiple processed ultrasonic feature images.

[0112] The regional processing step includes:

[0113] S311, region division: dividing the plurality of ultrasonic characteristic images into a plurality of non-overlapping sub-regions;

[0114] S312, region transformation: performing image transformation on the multiple sub-regions to obtain multiple transformed sub-regions, wherein the image transformation method is S-curve transformation, gamma transformation or other nonlinear transformation;

[0115] S313, regional compounding: merging the multiple transformed sub-regions of each ultrasonic feature map to obtain multiple regional processed ultrasonic feature maps.

[0116] It should be noted that the area division can be in various forms, specifically according to the shape of the ultrasound image of the area to be imaged, for example, it can be divided at equal intervals along the length and width directions, or it can be divided at equal intervals along the radial direction of a sector or circle.

[0117] Preferably, the regions can be divided according to the distance range from the ultrasonic sound source. In this embodiment, the ultrasonic image obtained by the cardiac tissue ultrasonic imaging is a sector-shaped region. Starting from the center of the sector-shaped region, each ultrasonic characteristic map can be divided into three non-overlapping sub-regions at equal intervals along the radial direction.

[0118] Through regional division, image areas with different distances from the ultrasound source can be classified and processed, thereby adjusting the contrast of images in different ultrasound imaging areas, solving the problem of different responses of body tissues to ultrasound propagation due to different distances from the ultrasound source; through regional changes, the grayscale contrast of body tissues in different areas can be adjusted, thereby obtaining better imaging effects.

[0119] The time processing step includes:

[0120] S321, time point division: selecting at least three regions according to the ultrasound image acquisition time points to process the ultrasound feature maps;

[0121] S322, time point transformation: selecting one of the time images as a reference image, performing image registration on other time images within different time ranges with the reference image as the target to obtain multiple registered images, wherein the image registration method is a rigid body transformation, an affine transformation, or other nonlinear transformation, and determining multiple transformed time images as the reference image and multiple registered images;

[0122] S323, time point compounding: merging the multiple transformed time images of each regional processed ultrasound image to obtain multiple processed ultrasound feature maps.

[0123] It should be noted that the spatial division refers to the division of the ultrasound image acquisition time T. This can be done by evenly dividing the acquisition time T and selecting regions to process the ultrasound feature map; it can also be done by unevenly dividing the acquisition time T and processing the ultrasound feature map; or it can be done by selecting multiple regions from different time periods to process the ultrasound feature map. In this embodiment, the time T is evenly divided into 10 equal time periods, and the multiple ultrasound feature maps are divided into multiple time images.

[0124] The method for combining multiple time-shifted images is to perform weighted summation according to image weights, wherein when a pixel is 0, the weight of the pixel is 0.

[0125] Since noise and speckle in images usually appear randomly, the time processing steps can effectively suppress noise and speckle in images.

[0126] It should be noted that, in an alternative embodiment of the present invention, step S3 only includes a regional processing step. Compared with the first embodiment, the difference of this embodiment is only that the multiple regional processed ultrasonic feature maps obtained after regional compounding are the multiple processed ultrasonic feature maps in step S3.

[0127] In another alternative embodiment of the present invention, step S3 only includes a time processing step. Compared with the first embodiment, the difference of this embodiment is that the time division is to divide the multiple ultrasonic feature maps into multiple time images according to the ultrasonic image acquisition time range; the time compounding step is to respectively aggregate the multiple transformed time images of each ultrasonic feature map to obtain multiple processed ultrasonic feature maps.

[0128] In this embodiment, step S4 includes:

[0129] S41, performing grayscale transformation on the multiple processed ultrasonic feature maps to obtain multiple transformed ultrasonic feature maps, wherein the transformation method includes linear transformation, S-curve transformation, logarithmic transformation, gamma transformation or other nonlinear transformation;

[0130] S42: Merge the multiple transformed ultrasonic feature maps to obtain a target ultrasonic image.

[0131] The step S41 also includes the steps of filtering the multiple transformed ultrasonic feature maps, retaining feature edges, smoothing the images, and enhancing contrast.

[0132] The second embodiment of the present invention discloses another method for dynamic image processing, which differs from the first embodiment in that step S1 includes:

[0133] S11, collecting multiple ultrasound images to be processed within time T;

[0134] S12, determining effective grayscale ranges of multiple ultrasound images to be processed, and mapping the effective grayscale ranges to the total grayscale range of the images to obtain a mapped grayscale range;

[0135] S13. Divide the mapped grayscale range into multiple stepped grayscale ranges according to the multiple ultrasound images.

[0136] It should be noted that the effective grayscale range refers to the range between the grayscale of the darkest part and the grayscale of the brightest part in multiple ultrasound images that change with time within the ultrasound image acquisition time T; and in the total grayscale range, the grayscale range outside the effective grayscale range is the invalid grayscale range.

[0137] By mapping the effective grayscale range to the total grayscale range, invalid grayscale can be reduced, and the dynamic tolerance of the effective grayscale or feature map in the ultrasound image can be improved, which is conducive to enhancing the feature details and contrast of the ultrasound image.

[0138] In this embodiment,

[0139] The specific method of mapping the effective grayscale range to the total grayscale range is as follows:

[0140] Calculate the probability p of the kth grayscale occurrence in multiple ultrasound images within time T respectively k and the standard deviation of the grayscale occurrence probability in multiple ultrasound images σ(p k ), which can be expressed as:

[0141] p k =h k / M;

[0142]

[0143] Among them, h k is the number of pixels of the kth grayscale, M is the total number of pixels in the ultrasound image; p k (t) is the probability of the kth grayscale appearing in the tth ultrasound image; μ(p k ) is the average probability of the kth grayscale appearance in N ultrasound images within the time period.

[0144] The standard deviation σ(p k ) changes in the range of k∈[0,K], reflecting the change in the probability of the kth grayscale appearing within this time T. The smaller the standard deviation σ(p k ) reflects that the change in the probability of occurrence of this grayscale is small.

[0145] In this embodiment, when σ(p k ) value is less than a preset standard deviation, it can be considered that the probability of occurrence of the grayscale k has no significant change within the time period, that is, the effective grayscale range and invalid grayscale range in the ultrasound image within time T can be determined by the grayscale k.

[0146] In this embodiment, by σ(p k ) can calculate the boundary threshold λ of the effective grayscale range l and λ h , which can be expressed as:

[0147] λ l ={max(i),i∈[1,K]:[σ(p0),σ(p i )]<σ l};

[0148] λ h ={min(j),j∈[1,K]:[σ(p j ),σ(p K )]<σh},

[0149] Among them, σ l and σ h is the preset standard deviation.

[0150] That is, the effective grayscale range is [λ l ,λ h ]; the invalid grayscale range is [0, λ l ) and (λ h , K].

[0151] The effective grayscale range [λ l ,λ h ] is mapped to the total grayscale range of the image [0, K] to obtain the mapping grayscale range, and then the mapping grayscale ranges of multiple ultrasonic feature maps are segmented to obtain the segmented grayscale range.

[0152] A linear grayscale transformation can be expressed as:

[0153]

[0154] in,

[0155] A third embodiment of the present invention further provides an imaging device, comprising:

[0156] An ultrasonic transducer, the ultrasonic transducer being used to collect ultrasonic image signals of a patient's body tissue;

[0157] a processor, configured to receive the ultrasonic image signal and dynamically process the ultrasonic image signal;

[0158] A storage medium, wherein a plurality of instructions are stored in the storage medium, and when the instructions are executed by the processor, the image dynamic processing method described in any one of the above embodiments is implemented.

[0159] Furthermore, the imaging device also includes a catheter body, the distal end of which can be inserted into the area to be imaged in the patient's body through a human lumen; and the ultrasonic transducer is arranged at the distal end of the catheter body.

[0160] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any appropriate manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

Claims

1. A dynamic image processing method for processing ultrasonic images of body tissues with periodic physiological characteristics, characterized in that: The following steps are involved: Acquire multiple ultrasound images of the area to be imaged within a period T of at least one physiological characteristic, and divide the total grayscale range of the image into multiple step grayscale ranges according to the multiple ultrasound images; Extracting the ultrasonic image to be processed according to the graded grayscale range to obtain a plurality of ultrasonic feature maps; Processing the multiple ultrasonic characteristic images to obtain multiple processed ultrasonic characteristic images; Merging the multiple processed ultrasonic feature maps to obtain a target ultrasonic image; Methods for decomposing image grayscale based on multiple ultrasound images include: S11, respectively counting the total number of accumulated pixels H of the kth grayscale of each ultrasound image in the plurality of ultrasound images k , calculate the cumulative total number of pixels H of the multiple ultrasound images k The standard deviation σ(H k ), determine the cumulative standard deviation of the total number of pixels σ(H k ) is the maximum value, and the corresponding grayscale λ1 is the first grayscale threshold. The total grayscale range is divided into the first grayscale range [0, λ1] and the second grayscale range [λ1, K] according to the first grayscale threshold λ1. The total number of accumulated pixels H k and its standard deviation σ(H k ) are respectively expressed as: Wherein, k is the grayscale count in each ultrasound image, K is the total grayscale count of the image, and h i is the number of grayscale pixels of the i-th grayscale, t is the t-th ultrasound image in time T, N is the total number of ultrasound images, H k (t) is the total number of pixels with the kth gray level in the tth ultrasound image; μ(H k ) is the average value of the total number of k-th grayscale cumulative pixels in N ultrasonic images within the time period; S12, within the grayscale range obtained in the previous step, take the coordinate point As the starting point, coordinate point Get the line for the end point Determine the distance from the straight line The coordinate point P when the distance is the maximum x (x,σ(H x )) The grayscale x corresponding to it is the grayscale threshold λ x , with each gray threshold λ obtained in this step x The grayscale range obtained in the previous step is further divided to obtain multiple grayscale ranges [λ a ,λ x ] and [λ x ,λ b ]; Among them, the λ a and λ b are the endpoint values ​​of each grayscale range obtained in the previous step, and the grayscale x∈[λ a ,λ b ]; S13, the grayscale ranges [λ a ,λ x ] and [λ x ,λ b ], calculate the distance from the straight line The distance is the minimum and the distance from λ x The grayscale y1 and grayscale y2 corresponding to the coordinate point with the smallest absolute value of the difference, where grayscale y1∈[λ a ,λ x ], grayscale y2∈[λ x ,λ b ]; S14, when the grayscale y1 and the grayscale y2 are the boundary values ​​of the grayscale range, further segmentation is stopped within the grayscale range; When the grayscale y1 and grayscale y2 are not within the grayscale range boundary value, the grayscale threshold λ is determined. y1 and grayscale threshold λ y2 , with gray threshold λ y1 For grayscale range [λ a ,λ x ] is further segmented with gray threshold λ y2 For grayscale range [λ x ,λ b ] to further segment and obtain multiple grayscale ranges [λ a ,λ y1 ],[λ y1 ,λ x ],[λ x ,λ y2 ] and [λ y2 ,λ b ] and then return to step S13; S15 , when segmentation is stopped in each grayscale range, determining each grayscale range as a plurality of graded grayscale ranges obtained by segmenting the total grayscale range of the image.

2. The image dynamic processing method according to claim 1, characterized in that: The steps of acquiring multiple ultrasonic images of the area to be imaged and dividing the total grayscale range of the image into multiple step grayscale ranges according to the multiple ultrasonic images are specifically as follows: S11, collecting multiple ultrasonic images to be processed within a time T; S12, determining effective grayscale ranges of multiple ultrasound images to be processed, and mapping the effective grayscale ranges to the total grayscale range of the images to obtain a mapped grayscale range; S13. Divide the mapped grayscale range into multiple stepped grayscale ranges according to the multiple ultrasound images.

3. The image dynamic processing method according to claim 2, characterized in that: The method for determining the effective grayscale range of the multiple ultrasound images is: Calculate the probability p of the kth grayscale occurrence in the multiple ultrasound images respectively k and the standard deviation of the grayscale occurrence probability in multiple ultrasound images σ(p k ), which can be expressed as: p k =h k / M; Among them, h k is the number of pixels of the kth grayscale, M is the total number of pixels in the ultrasound image; p k (t) is the probability of the kth grayscale appearing in the tth ultrasound image; μ(p k ) is the average probability of occurrence of the kth grayscale in N ultrasound images within the time period; When σ(p k ) value is less than the preset value σ l and σ h When , the corresponding grayscale k is the boundary threshold λ of the effective grayscale range l and λ h , which can be expressed as: l l ={max(i),i∈[1,K]:[σ(p0),σ(p i )]<σ l }; l h ={min(j),j∈[1,K]:[σ(p j ),σ(p K )]<σ h }。 4. The image dynamic processing method according to claim 1 or 2, characterized in that: The step of processing the plurality of ultrasonic characteristic images to obtain the plurality of processed ultrasonic characteristic images comprises a region processing step and / or a time processing step. The regional processing step includes: Region division: dividing the multiple ultrasonic characteristic images into multiple non-overlapping sub-regions; Region transformation: performing image transformation on the multiple sub-regions to obtain multiple transformed sub-regions, wherein the image transformation method is S-curve transformation or gamma transformation; Regional compounding: Merge multiple transformed sub-regions of each ultrasonic feature map to obtain multiple regional processed ultrasonic feature maps; The time processing step includes: Time point division: dividing the plurality of regional processed ultrasound feature maps into a plurality of time images according to the ultrasound image acquisition time range; Time point transformation: Select one of the time images as a reference image, and perform image registration on other time images in different time ranges with the reference image as the target to obtain multiple registered images. The image registration method is rigid transformation, affine transformation or elastic transformation, and multiple transformed time images are determined to be the reference image and multiple registered images; Time point compounding: Multiple transformed time images of each region processed ultrasound image are merged to obtain multiple processed ultrasound feature maps.

5. The image dynamic processing method according to claim 4, characterized in that: The steps of merging the multiple processed ultrasonic feature maps to obtain a target ultrasonic image are as follows: Performing grayscale transformation on the multiple processed ultrasonic characteristic images to obtain multiple transformed ultrasonic characteristic images, wherein the transformation method includes linear transformation, S-curve transformation, logarithmic transformation or gamma transformation; The multiple transformed ultrasonic feature maps are merged to obtain a target ultrasonic image.

6. The image dynamic processing method according to claim 5, characterized in that: The step of obtaining the target ultrasonic image also includes the steps of filtering the multiple transformed ultrasonic feature maps, retaining feature edges, smoothing the image, and enhancing contrast.

7. A dynamic imaging device, characterized in that: include: An ultrasonic transducer, wherein the ultrasonic transducer is used to collect ultrasonic image signals of a patient's body tissue; a processor, configured to receive the ultrasonic image signal and dynamically process the ultrasonic image signal; A storage medium, wherein a plurality of instructions are stored in the storage medium, and when the instructions are executed by a processor, the image dynamic processing method according to any one of claims 1 to 6 is implemented.

8. The dynamic imaging device according to claim 7, characterized in that The imaging device further comprises a catheter body, the distal end of which can be inserted into a region to be imaged in a patient's body through a human lumen; and the ultrasonic transducer is arranged at the distal end of the catheter body.

Images