WO2025050534A1 - Method and apparatus for acquiring bifurcated blood vessel on the basis of frequency domain signals, medium and device - Google Patents

Abstract

A method and apparatus for acquiring a bifurcated blood vessel on the basis of frequency domain signals, a medium and an electronic device. The method comprises: acquiring a plurality of radio frequency signals; performing frequency domain analysis on part of the radio frequency signals among the plurality of radio frequency signals, so as to obtain a blood flow region corresponding to each frame of an intravascular ultrasound image; and at least removing a main blood vessel from the blood flow region to obtain a bifurcated blood vessel connected to the main blood vessel.

Description

Method, device, medium and equipment for acquiring bifurcated blood vessels based on frequency domain signals

This application claims priority to the Chinese patent application filed with the China Patent Office on September 7, 2023, with application number 2023111545148, the entire contents of which are incorporated by reference into this application.

Technical Field

The present application relates to the field of signal processing, for example, to a method, device, medium and electronic device for acquiring bifurcated blood vessels based on frequency domain signals.

Background Art

Imaging technology plays a certain role in assisting doctors in evaluating vascular anatomy and lesion characteristics. Intravascular ultrasound (IVUS) inserts an ultrasound probe into a guidewire and places an ultrasound imaging device inside the coronary artery. It can provide cross-sectional and longitudinal images, allowing doctors to make detailed assessments of lesions inside the coronary artery. However, due to the resolution limitations of IVUS images themselves, side branch vessels may not be clear enough or difficult to distinguish in the image.

At present, the commonly used strategies and methods for the problem of side branch identification in IVUS images are:

Use of contrast agents: The use of contrast agents to enhance the contrast of IVUS images can improve the visualization of side branch vessels; however, the need for additional contrast agents increases the complexity of the technical solution.

Auxiliary analysis tools: such as those based on deep learning and other technologies to help identify and locate side branch vessels, which requires the construction of complex network models and is a large workload.

Combining with other imaging techniques: In addition to IVUS, combining with other imaging techniques such as coronary angiography or optical coherence tomography (OCT) can provide a more comprehensive vascular assessment and help identify side branch vessels more accurately. However, it needs to rely on other imaging techniques, and the implementation of the solution is more complex.

Summary of the invention

The embodiments of the present application provide a method, device, medium and electronic device for acquiring bifurcated blood vessels based on frequency domain signals. The embodiments of the present application propose an analysis technology combined with frequency domain signals to achieve fast and accurate side branch identification without the need for contrast agents or other imaging assistance.

In a first aspect, an embodiment of the present application provides a method for acquiring bifurcated blood vessels based on frequency domain signals, the method comprising: acquiring multiple radio frequency signals; performing frequency domain analysis on some of the multiple radio frequency signals to obtain a blood flow area corresponding to each frame of an intravascular ultrasound image; and removing at least a main branch blood vessel from the blood flow area to obtain a bifurcated blood vessel connected to the main branch blood vessel.

In a second aspect, some embodiments of the present application provide a device for acquiring bifurcated blood vessels based on frequency domain signals, the device comprising: a radio frequency signal acquisition module, configured to acquire multiple radio frequency signals; a blood flow area acquisition module, configured to perform frequency domain analysis on some of the multiple radio frequency signals to obtain a blood flow area corresponding to each frame of an intravascular ultrasound image; and a bifurcated blood vessel acquisition module, configured to remove at least the main branch blood vessel from the blood flow area to obtain a bifurcated blood vessel connected to the main branch blood vessel.

In a third aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the method described in any embodiment of the first aspect can be implemented.

In a fourth aspect, some embodiments of the present application provide an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can implement the method described in any embodiment of the first aspect.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of the composition of an IVUS imaging system provided in an embodiment of the present application;

FIG2 is a schematic diagram of a probe provided in an embodiment of the present application rotating in a blood vessel;

FIG3 is a flow chart of a method for acquiring bifurcated blood vessels based on frequency domain signals provided in an embodiment of the present application;

FIG4 is a schematic diagram of multiple radio frequency signals corresponding to three frames of intravascular ultrasound images provided by an embodiment of the present application;

FIG5 is a schematic diagram of three matching radio frequency signals provided in an embodiment of the present application;

FIG6 is a block diagram of a device for acquiring bifurcated blood vessels based on frequency domain signals provided in an embodiment of the present application;

FIG. 7 is a schematic diagram of the composition of an electronic device provided in an embodiment of the present application.

DETAILED DESCRIPTION

The embodiments of the present application propose an analysis technology that combines frequency domain signals to solve the problem of detecting side branch vessels, thereby achieving rapid and accurate side branch vessel identification and positioning without the need for contrast agents or other imaging assistance.

That is to say, the embodiments of the present application aim to propose a set of solutions for automatically determining the position and structure of side branch vessels in IVUS images. For example, the method of acquiring bifurcated vessels based on frequency domain signals in some embodiments of the present application exemplarily includes: acquiring radio frequency signals; performing frequency domain analysis based on radio frequency signals to acquire all blood flow areas; and extracting side branch areas through morphological processing and continuity analysis.

Please refer to Figure 1, which shows an IVUS imaging system provided by some embodiments of the present application. The IVUS imaging system includes: a computer 110, an ultrasound imaging system 120, an ultrasound signal line 124, a proximal drive module 130 and a probe 180, wherein the ultrasound imaging system exemplarily includes: an echo processing module 121, a coordinate transformation module 122 and a coding transmission module 123.

IVUS imaging systems are divided into two typical types, namely mechanical rotation type and phased array type, and the two types use different ultrasonic probes. The construction principle of the former is that the motor maintains a certain speed to guide the single-element transducer in the conductor to rotate accordingly, thereby realizing the sending and receiving of signals and completing the drawing of images. The latter uses an electronic phased array system. The system uses multiple ultrasonic sensors to arrange the array elements in a regular pattern, and uses a timing control method to form the required image. The embodiments of the present application can adopt either of these two types.

The proximal drive module 130 is also called a retraction device, a motor drive unit, etc., but they are similar. Its main functions are: controlling the rotation speed of the motor, feeding back a synchronous control signal to the entire system, and controlling the IVUS catheter (or guide wire 140) to carry the probe 180 to retract in the blood vessel wall 150 to capture different frame images. For example, the retraction direction 160 is shown in FIG. 1. In FIG. 1, the first cross section 111 is obtained. (the corresponding probe rotates one circle in this direction and stops at multiple rotation angles to transmit signals to obtain corresponding radio frequency signals, and a frame of intravascular ultrasound image on the first cross-section is obtained based on the multiple radio frequency signals) and the second cross-section 112 (the corresponding probe rotates one circle in this direction and stops at multiple rotation angles to transmit signals to obtain corresponding radio frequency signals, and a frame of intravascular ultrasound image on the second cross-section is obtained based on the multiple radio frequency signals).

For example, FIG2 shows that the probe continuously rotates the corresponding h-th rotation angle 171 and the h+m-th rotation angle 172 when it stays in any retraction direction, wherein h and m are both integers greater than or equal to 1. It can be understood that the probe needs to rotate one circle in different retraction directions to obtain the radio frequency signal of each corresponding rotation angle in that direction, and then obtain the intravascular ultrasound image of the corresponding frame based on these signals.

The following is an illustrative description of a method for acquiring bifurcated blood vessels based on frequency domain signals provided by some embodiments of the present application and executed by the computer of FIG. 1 , in conjunction with FIG. 3 .

As shown in FIG3 , some embodiments of the present application provide a method for acquiring a bifurcated blood vessel based on a frequency domain signal, the method comprising:

S101: Acquire multiple radio frequency signals.

It should be noted that the multiple radio frequency signals recorded in S101 are acquired by the probe 180 that penetrates deep into the main branch blood vessel, and the multiple radio frequency signals recorded in S101 are time domain signals corresponding to multiple frames of intravascular ultrasound images (for example, multiple consecutive frames of intravascular ultrasound images) acquired by the probe 180.

S102, performing frequency domain analysis on some of the multiple radio frequency signals to obtain a blood flow area corresponding to each frame of the intravascular ultrasound image.

S103, removing at least the main branch blood vessel from the blood flow area to obtain the bifurcated blood vessel connected to the main branch blood vessel.

It is not difficult to understand that some embodiments of the present application identify the blood flow area through the frequency domain characteristics of the radio frequency signal, and then obtain the bifurcated blood vessels based on the blood flow area analysis, thereby improving the accuracy of the bifurcated blood vessel structure and position discrimination.

The implementation process of each step in FIG. 3 is exemplarily described below.

The following is an exemplary description of the process of obtaining multiple radio frequency signals in conjunction with FIG. 1. In some embodiments, the plurality of radio frequency signals are radio frequency signals corresponding to one frame or at least two consecutive frames of intravascular ultrasound images.

When the probe 180 of FIG. 1 reaches the first cross section inside the blood vessel through the guide wire 140, the ultrasonic transmitting element inside the probe 180 will emit a sound wave pulse and receive the echo signal returned from the blood vessel wall 150 at the same time. Through the data acquisition card, the first radio frequency signal can be obtained. When the probe rotates a certain angle at the position corresponding to the first cross section to reach a new rotation angle, the sound wave pulse will be emitted again and the echo signal returned from the blood vessel wall 150 will be received to obtain a second radio frequency signal (two different rotation angles as shown in FIG. 2). When the probe rotates 360 degrees at the position corresponding to the first cross section in the blood vessel, multiple radio frequency signals are obtained, and a frame of intravascular ultrasound image can be obtained through these radio frequency signals. Next, the probe 180 moves along the withdrawal direction of FIG. 1 to reach the position corresponding to the second cross section, repeats the above process, obtains multiple radio frequency signals, and another frame of intravascular ultrasound image can be obtained according to the obtained multiple radio frequency signals.

It should be noted that the multiple radio frequency signals obtained in S101 may be multiple radio frequency signals corresponding to one frame of intravascular ultrasound image or may be multiple radio frequency signals corresponding to at least two frames of intravascular ultrasound image.

In some embodiments of the present application, if the multiple radio frequency signals obtained in S101 are multiple radio frequency signals corresponding to a frame of intravascular ultrasound image, when determining the vascular region corresponding to the i-th frame of intravascular ultrasound image, S102 exemplarily includes: performing high-frequency filtering on the multiple radio frequency signals corresponding to the i-th frame of intravascular ultrasound image through a high-frequency filter to obtain the vascular region corresponding to the frame of intravascular ultrasound image.

In some embodiments of the present application, if the multiple radio frequency signals obtained in S101 are all radio frequency signals corresponding to at least two consecutive frames of intravascular ultrasound images, when determining the blood vessel region corresponding to the i-th frame of intravascular ultrasound image, S102 exemplarily includes:

In the first step, a matching radio frequency signal that matches each radio frequency signal to be matched is determined.

It should be noted that the RF signal to be matched in this step belongs to any RF signal corresponding to the i-th frame of intravascular ultrasound image, and the matching RF signal belongs to the RF signal corresponding to any adjacent frame image adjacent to the i-th frame of intravascular ultrasound image, and one RF signal corresponds to one rotation angle.

As shown in FIG. 4 , the figure shows the probe moving along the main branch vessel in the process of withdrawing three times 440. Three RF signal sequences should be collected, which are a first RF signal sequence 410, a second RF signal sequence 420 and a third RF signal sequence 430, and each RF signal sequence corresponds to generating a frame of intravascular ultrasound image (these RF signals are referred to as multiple RF signals corresponding to the frame of intravascular ultrasound image). In Figure 4, the first RF signal sequence 410 is three RF signals corresponding to the i-1th frame of intravascular ultrasound image (corresponding to three different rotation angles or simply referred to as three different angles), the second RF signal sequence 420 is three RF signals corresponding to the i-th frame of intravascular ultrasound image, and the third RF signal sequence 430 is three RF signals corresponding to the i+1th frame of intravascular ultrasound image. In each RF signal sequence in Figure 4, three consecutive rotation angles corresponding to each retraction are given as an example to obtain RF signals, and each rotation angle corresponds to an RF signal. The three RF signal lines indicated by the RF sequence 450 in Figure 4 are three RF signals located on the third RF signal sequence 430. It can be understood that each RF signal sequence in Figure 4 is an RF signal obtained by three consecutive rotation angles corresponding to each retraction of the probe.

The motion correction process of the embodiment of the present application is to find the matching signal of each RF signal corresponding to the i-th frame from all RF signals corresponding to the adjacent frames of the i-th frame, so as to subsequently complete the angle alignment of signals of different frames.

In the second step, all matching RF signals that match the kth RF signal to be matched in the i-th frame of the intravascular ultrasound image are summed with the kth RF signal to be matched to obtain the kth target RF signal, wherein the kth RF signal to be matched is any RF signal corresponding to the i-th frame of the intravascular ultrasound image. As shown in FIG5 , the figure shows three matching RF signals 510 of three adjacent frames, wherein the three matching RF signals 510 of the three adjacent frames respectively include the i-th frame RF signal to be matched 522 (i.e., any RF signal corresponding to the i-th frame of the intravascular ultrasound image in FIG4 ), the matching RF signal 521 of the i-1th frame (i.e., the RF signal matching the i-th frame RF signal to be matched found from all RF signals corresponding to the i-1th frame of the intravascular ultrasound image in FIG4 ), and the matching RF signal 523 of the i+1th frame (i.e., the RF signal matching the i-th frame RF signal to be matched found from all RF signals corresponding to the i+1th frame of the intravascular ultrasound image in FIG4 ).

Repeat the first and second steps until the image corresponding to the i-th frame of intravascular ultrasound image is obtained. A target radio frequency signal of all radio frequency signals to be matched is obtained, and a frequency domain analysis is performed on the target radio frequency signal to obtain a blood flow area of the i-th frame of the intravascular ultrasound image.

By repeating the related processes of the first and second steps mentioned above, the technical effect of motion correction of the acquired related data can be achieved. The reason why motion correction is needed is that it can effectively improve the problem that radio frequency signals with the same rotation angle cannot be aligned due to the jitter of the guide wire during IVUS signal acquisition and the movement of the observed object itself.

The multiple RF signal lines of each RF signal sequence in Figure 4 represent the scanning results at different angles in the same frame image, and different frames are obtained along the retreat direction, where each curve represents the corresponding RF signal. In Figure 5, it is assumed that a signal at an angle in a frame has been selected for analysis, then according to the result of angle alignment, the RF signal of the corresponding angle in the adjacent frame can be taken out.

It should be noted that the motion correction process of some embodiments of the present application includes: a process of searching for an image block that matches a local image block on the current frame of intravascular ultrasound image from an adjacent frame of intravascular ultrasound image corresponding to the current frame of intravascular ultrasound image.

That is to say, in some embodiments of the present application, the process of determining the matching RF signal that matches each RF signal to be matched in the first step exemplarily includes: selecting an image block from the i-th frame of intravascular ultrasound image as the image block to be matched; acquiring image blocks that respectively match the image block to be matched from the multiple adjacent frame images according to the search area, to obtain candidate matching RF signals; repeating the above process until all candidate matching RF signals corresponding to the i-th frame of intravascular ultrasound image are obtained; and obtaining the matching RF signal based on all the candidate matching RF signals.

It should be noted that in some embodiments of the present application, the process of obtaining a matching image block exemplarily includes: taking the pixel value matrix of the image block corresponding to the image block to be matched as a kernel, and performing convolution operations on the kernel with the pixel value matrices of the image blocks corresponding to multiple search areas on multiple adjacent frame images, respectively, to obtain the RF signal corresponding to the image block on the adjacent frame image that matches the image block to be matched. For example, in some embodiments of the present application, each two-dimensional image block to be matched uses a relatively large kernel (for example, the kernel is 6.8mm x 10angles) and a 7mm x 20angle search area to determine the offset with the maximum cross-correlation between adjacent frames (the offset with the maximum cross-correlation is the offset that matches the image block to be matched). The RF signal number of the matching image block in the adjacent frame or adjacent frame is recorded, and the offset value is recorded to facilitate the subsequent superposition operation of the matching RF signal. Taking three consecutive frames as an example, it is explained how to use the offset with the largest cross-correlation. For example, the 200-degree RF signal line corresponding to the 10th frame of intravascular ultrasound image, the 201-degree RF signal line corresponding to the 11th frame of intravascular ultrasound image, and the 204-degree RF signal line corresponding to the 12th frame of intravascular ultrasound image are matched. After recording the corresponding information, the RF signals at the corresponding angles of other adjacent frames can be comprehensively considered when determining the blood vessel area of the 11th frame.

It should be noted that the meaning of the description information 10angles of the kernel in the above example is, for example, if the current frame of the intravascular ultrasound image is composed of 360 RF signal lines (for example, the three RF signal lines corresponding to each frame of the intravascular ultrasound image shown in FIG4 ), and each RF signal line represents a different rotation angle, then 10angles corresponds to ten different continuous rotation angles. Similarly, the angles corresponding to the search area have the same meaning as above.

It is not difficult to understand that some embodiments of the present application complete the motion correction process by obtaining matching image blocks. Specifically, an image block on the current frame is used as a convolution kernel, and a convolution operation is performed with the search area of the RF signal corresponding to the adjacent frame image. The matching RF signal of the image block on the current frame, that is, the corresponding RF signal to be matched, can be obtained.

It should be noted that the matching results obtained by the above-mentioned image block convolution method in some embodiments of the present application may be false positive results. Therefore, in some embodiments of the present application, the matching results (i.e., candidate matching RF signals) obtained by image block convolution are not directly used as the matching RF signals of the image blocks to be matched, but it is necessary to reselect an RF signal as the final matching RF signal based on the characteristics of the candidate matching RF signals. Two examples of obtaining false positive matching results are described below.

In some embodiments of the present application, speed limitation is used to remove false positive matches and maintain spatial consistency. The reason why some embodiments of the present application can use speed limitation to identify and correct false positive matching results is that due to the physical characteristics of probe rotation, the maximum interpolation of different angle matches between adjacent frames is limited by the rotation speed of the probe, i.e., speed limitation.

That is, in some embodiments of the present application, the process of obtaining the matching radio frequency signal according to all candidate matching radio frequency signals exemplarily includes: in response to confirming that the matching radio frequency signal is consistent with the image block to be matched If the difference between the corresponding rotation angle serial number and the rotation angle serial number corresponding to any candidate matching RF signal is less than or equal to the first threshold, then any candidate matching RF signal is used as the matching RF signal corresponding to the image block to be matched, otherwise a RF signal that satisfies the first threshold is selected from the RF signals corresponding to the corresponding adjacent frame images as the matching RF signal that matches the image block to be matched. Wherein, the rotation angle of the image block to be matched is the rotation angle of the RF signal located in the middle of the image block to be matched. For example, an example of determining a false positive matching result by speed limitation is: the candidate matching RF signal corresponding to the 100th RF signal line corresponding to the first frame of the intravascular ultrasound image includes: 102 RF signals corresponding to the second frame of the intravascular ultrasound image, 110 RF signals corresponding to the third frame of the intravascular ultrasound image, and 105 RF signals corresponding to the fourth frame of the intravascular ultrasound image. According to the speed limitation principle, it can be known that the 110 RF signals corresponding to the third frame of the intravascular ultrasound image belong to the false positive matching result. It is not difficult to understand that the embodiments of the present application can correct the problem of some false matching results by limiting the maximum interpolation of different angle matching between adjacent frames, thereby improving the accuracy of the obtained matching RF signal.

In some embodiments of the present application, false positive matches are removed and spatial consistency is maintained by angle matching monotonicity (or curve fitting). This is because according to the properties of the physical system, it can be considered that the angle matching between adjacent frames should be monotonic, so its curve satisfies monotonicity.

That is to say, in some embodiments of the present application, the above process of obtaining the matching RF signal based on the candidate matching RF signal exemplarily includes: obtaining a curve corresponding to all candidate matching RF signals corresponding to a frame of intravascular ultrasound image, wherein a point on the curve corresponds to a candidate matching RF signal; filtering out points that do not satisfy monotonicity from the curve; and using an updated RF signal as the matching RF signal corresponding to the point (for example, the updated RF signal is a RF signal that satisfies monotonicity selected from RF signals corresponding to corresponding adjacent frame images). Some embodiments of the present application select another suitable matching RF signal for the false matching result obtained by image block matching according to the monotonicity requirement. For example, an example of determining a false positive matching result through angle matching monotonicity is: the candidate matching RF signal corresponding to the 100th RF signal line corresponding to the 1st frame of intravascular ultrasound image is the 102nd RF signal corresponding to the second frame of intravascular ultrasound image, and the 101st RF signal corresponding to the 1st frame of intravascular ultrasound image is the candidate matching RF signal corresponding to the 100th RF signal line. The candidate matching RF signal corresponding to the RF signal line is the 103rd RF signal corresponding to the second frame of intravascular ultrasound image, the candidate matching RF signal corresponding to the 102nd RF signal line corresponding to the first frame of intravascular ultrasound image is the 106th RF signal corresponding to the second frame of intravascular ultrasound image, and the candidate matching RF signal corresponding to the 101st RF signal line corresponding to the first frame of intravascular ultrasound image is the 105th RF signal corresponding to the second frame of intravascular ultrasound image. Then, a curve corresponding to the candidate matching RF signals of the multiple RF signals corresponding to the first frame is drawn (the values on the curve are 102, 103, 106 and 105 respectively). It can be seen that the 106th RF signal corresponding to the second frame of intravascular ultrasound image is a false positive matching result. Subsequently, the 104th RF signal corresponding to the second frame of intravascular ultrasound image can be used as the matching RF signal that matches the 102nd RF signal line corresponding to the first frame of intravascular ultrasound image.

It can be understood that some embodiments of the present application believe that the angle matching between adjacent frames should be monotonic based on the properties of the physical system, so its curve satisfies monotonicity. Based on this monotonicity, false positive matching results obtained by image block convolution matching can be removed.

In some embodiments of the present application, the above-mentioned obtaining the matching RF signal based on all the candidate matching RF signals includes: if it is confirmed that the difference between the rotation angle number corresponding to the image block to be matched and the rotation angle number corresponding to any one of the candidate matching RF signals is less than or equal to a first threshold, then using any one of the candidate matching RF signals as the matching RF signal corresponding to the image block to be matched; otherwise, selecting an RF signal that meets the first threshold from the RF signals corresponding to the corresponding adjacent frame images as the matching RF signal that matches the image block to be matched; and obtaining a curve corresponding to all the candidate matching RF signals corresponding to a frame of intravascular ultrasound image (the frame of intravascular ultrasound image belongs to any one of the adjacent frame images), wherein a point on the curve corresponds to a candidate matching RF signal; filtering out points that do not meet monotonicity from the curve; and using an updated RF signal as the matching RF signal corresponding to the point.

It is not difficult to understand that some embodiments of the present application complete the matching of the same angle RF signal on different frame images through motion correction. The matching RF signals can be summed in the integration stage, and then the RF signals in the adjacent frames are comprehensively considered when obtaining the blood flow area of the current frame (i.e., the frame corresponding to the i-th frame of the intravascular ultrasound image mentioned above), thereby improving the accuracy of the judgment of the blood flow area of the current frame.

It should be noted that after the above-mentioned motion correction, it can be considered that the RF signals of different frames in some embodiments of the present application have completed the angle alignment. Therefore, by integrating the RF signal information between different frames at the same angle and using a high-pass filter, it is possible to extract the rapidly changing components, remove tissues with slower signal changes such as plaques, and find the envelope of the signal. For example, the high-pass filtering method can be implemented by a Butterworth high-pass filter or a singular value decomposition (SVD). That is, the embodiments of the present application filter out the rapidly changing areas as blood flow areas (relative to the blood flow areas in the tissue, which belong to the rapidly changing areas) through a high-pass filter, thereby improving the accuracy of blood flow area identification. The following exemplifies the implementation process of performing frequency domain analysis of the target RF signal in S102 to obtain the blood flow area of the i-th frame of the intravascular ultrasound image.

For example, in some embodiments of the present application, the process of performing frequency domain analysis on the target radio frequency signal to obtain the blood flow area of the i-th frame of intravascular ultrasound image in S102 exemplarily includes:

In the first step, Fourier transform is performed on the target RF signal to obtain the frequency domain signal.

As described above, the target RF signal of this step is illustrated by taking the kth target RF signal corresponding to the i-th frame of intravascular ultrasound image as an example. The process of acquiring the kth target RF signal is as follows: summing all matching RF signals that match the kth RF signal to be matched in the i-th frame of intravascular ultrasound image (including matching RF signals that are reselected after removing false positive matching results) with the kth RF signal to be matched to obtain the kth target RF signal, wherein the kth RF signal to be matched is any RF signal corresponding to the i-th frame of intravascular ultrasound image.

For example, a target RF signal can be obtained by superimposing the three matching RF signals shown in FIG. 5 .

In a second step, the frequency domain signal is filtered at least by a high-pass filter to obtain the blood flow area of the i-th frame of the intravascular ultrasound image (as an example of the current frame of the intravascular ultrasound image).

In some embodiments of the present application, the second step exemplarily includes: directly filtering the frequency domain signal using a high-pass filter to obtain the blood flow area of the i-th frame of the intravascular ultrasound image.

It should be noted that, in order to obtain a more robust blood flow area, in some embodiments of the present application, the area corresponding to the signal of the high-pass filter is not directly used as the blood flow area, but the RF signal of the adjacent rotation angle adjacent to the RF signal of the angle to be processed on the current frame is compared with the RF signal of the angle to be processed. The result of the superposition or delayed superposition of the signals is used as the final blood flow area. That is, in some embodiments of the present application, the process of obtaining the blood flow area is: high-pass filtering the target radio frequency signal to obtain the initial blood flow area corresponding to the i-th frame, and then superimposing the signals based on the similarity between the radio frequency signals of adjacent rotation angles in the initial blood flow area to obtain the final blood flow area.

For example, in some embodiments of the present application, the process of filtering the frequency domain signal at least through a high-pass filter to obtain the blood flow area of the i-th frame of the intravascular ultrasound image exemplarily includes: filtering the frequency domain signal (i.e., the signal after Fourier transforming the target radio frequency signal) through the high-pass filter to obtain the i-th frame of high-frequency time domain signal corresponding to the i-th frame of the intravascular ultrasound image; repeating the following process to obtain the i-th frame of composite time domain signal: determining that the similarity value of the n-th high-frequency radio frequency signal corresponding to the i-th frame of the high-frequency time domain signal and the adjacent angle high-frequency radio frequency signal is greater than a set threshold (for example, the value of the set threshold is greater than or equal to 0.5, which is the value of the set threshold). An optional numerical value), the nth high-frequency radio frequency signal and the adjacent-angle high-frequency radio frequency signal are composited (for example, the composite operation includes summation or delayed summation) to obtain an nth composite radio frequency signal, wherein the nth high-frequency radio frequency signal is the high-frequency radio frequency signal of the nth rotation angle corresponding to the i-frame high-frequency time domain signal, and the adjacent-angle high-frequency radio frequency signal is the high-frequency radio frequency signal of the n+uth or n-uth rotation angle corresponding to the i-frame high-frequency time domain signal, and u is an integer greater than 0 and less than a set integer value; the frequency domain signal corresponding to the nth composite radio frequency signal is filtered by the high-pass filter to obtain the blood flow area of the i-frame intravascular ultrasound image.

That is to say, some embodiments of the present application have preliminarily obtained the fast-changing area representing blood and other tissues after high-pass filtering. In order to obtain a more stable response, some embodiments of the present application also perform signal compounding based on the rotation angle, that is, each high-frequency time domain signal passing through the high-pass filter is divided into multiple small segments along the time axis, and the following formula is used for each segment to obtain the angle range (i.e., similarity value) required for signal compounding. The specific calculation formula is as follows:

Among them, N represents the total number of points on the signal segment, n represents the coordinate of the axis, Sθ(n) represents the signal value of the nth point, μθ represents the mean value on the signal segment, and the calculation method is: sθ+Δθ(n) represents the signal value at adjacent angles, μθ±Δθ represents the mean value at adjacent angles, and the last symbol in the numerator of the above formula is a multiplication sign and the multiplication sign is set to represent the multiplication of the two values preceding the multiplication sign.

For example, in some embodiments of the present application, if ρ≥0.5, it is considered that signal compounding is required, and the compounding method can be addition. According to the example formula of compounding two values, it is:
Sθ(n)＝Sθ(n)+sθ+Δθ(n)

Alternatively, the compounding method can be delayed addition. According to the example formula of compounding two values, it is:
Sθ(n)=Sθ(n)+sθ+Δθ(n+δn)

After signal combination, a relatively robust blood flow area response is obtained.

After the blood flow region is obtained through the above-mentioned embodiments, in some embodiments of the present application, the side branch region can be extracted from the blood flow region through morphological processing and continuity analysis. It is not difficult to understand that the blood flow region obtained through S102 should include the main branch blood flow, the side branch blood flow and the collateral blood flow. Through connectivity analysis, the independent collateral blood flow region is excluded. For example, in some embodiments of the present application, the blood flow region boundary is obtained through multiple longitudinal sections, and the boundary is curve fitted to obtain a continuous main branch boundary.

That is, in some embodiments of the present application, S103 exemplarily includes: according to the non-connectivity between the side branch vessel and the main branch vessel, removing the corresponding side branch vessel from the blood flow area to obtain a corrected vessel area; acquiring a three-dimensional area corresponding to the main branch vessel; performing a difference operation between the three-dimensional area and the corrected vessel area to obtain a bifurcated vessel connected to the main branch vessel.

Some embodiments of the present application utilize connectivity and morphology to remove the main branch (i.e., main branch vessel) and the side branch (i.e., collateral vessel) to obtain the side branch (i.e., bifurcated vessel).

In some embodiments of the present application, the process of obtaining the three-dimensional region corresponding to the main branch vessel exemplarily includes: obtaining the blood flow region boundary of the modified blood vessel region through multiple longitudinal section data, performing curve fitting on the blood flow region boundary to obtain multiple main branch boundaries; Perform three-dimensional surface fitting to obtain the three-dimensional region.

Some embodiments of the present application provide a process for acquiring a three-dimensional region corresponding to a main branch vessel, thereby improving the accuracy of acquiring the three-dimensional region of the main branch vessel.

Please refer to FIG6 , which shows a device for acquiring bifurcated blood vessels based on frequency domain signals provided in an embodiment of the present application. It should be understood that the device corresponds to the method embodiment of FIG3 above, and can execute multiple steps involved in the method embodiment above. The specific functions of the device can be found in the description above. To avoid repetition, detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system of the device. The device for acquiring bifurcated blood vessels based on frequency domain signals includes: a radio frequency signal acquisition module 501, a blood flow area acquisition module 502, and a bifurcated blood vessel acquisition module 503.

The radio frequency signal acquisition module is configured to acquire multiple radio frequency signals. For example, in some embodiments of the present application, the multiple radio frequency signals are acquired by a probe that penetrates deep into the main branch blood vessel.

The blood flow region acquisition module is configured to perform frequency domain analysis on some of the multiple radio frequency signals to obtain a blood flow region corresponding to each frame of the intravascular ultrasound image.

The bifurcated vessel acquisition module is configured to remove at least the main branch vessel from the blood flow area to obtain a bifurcated vessel connected to the main branch vessel.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method, and will not be described in detail here.

An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method described in any embodiment of the method for acquiring a bifurcated blood vessel based on a frequency domain signal can be implemented.

As shown in FIG. 7 , some embodiments of the present application provide an electronic device 700, which exemplarily includes a memory 710, a processor 720, and a computer program stored in the memory 710 and executable on the processor 720, wherein when the processor 720 reads and executes the program through a bus 730, the method described in any embodiment of the method for acquiring bifurcated blood vessels based on frequency domain signals can be implemented.

Processor 720 can process digital signals and can include various computing structures, such as complex instruction set computer structure, reduced instruction set computer structure, or a structure that implements a combination of multiple instruction sets. In some examples, processor 720 can be a microprocessor.

The memory 710 may be configured to store instructions executed by the processor 720 or data related to the execution of instructions. These instructions or data may include codes, which are configured to implement part or all of the functions of at least one module described in the embodiments of the present application. The processor 720 of the embodiment of the present disclosure may be configured to execute instructions in the memory 710 to implement the method shown in FIG3. The memory 710 includes a dynamic random access memory, a static random access memory, a flash memory, an optical memory, or other memory known to those skilled in the art.

In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, program segment or a part of the code contains at least one executable instruction for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and the flowchart, and the combination of the boxes in the block diagram and the flowchart, can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

In addition, the functional modules in multiple embodiments of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. The essence of the solution or the part that contributes to the relevant technology or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

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

Claims (16)

A method for acquiring bifurcated blood vessels based on frequency domain signals, comprising: Acquire multiple radio frequency signals; Performing frequency domain analysis on some of the multiple radio frequency signals to obtain a blood flow area corresponding to each frame of the intravascular ultrasound image; At least the main branch vessel is removed from the blood flow area to obtain a bifurcated vessel connected to the main branch vessel. According to the method of claim 1, the multiple radio frequency signals are acquired by a probe that penetrates deep into the main branch blood vessel, and the multiple radio frequency signals are time domain signals corresponding to multiple frames of intravascular ultrasound images acquired by the probe. The method according to claim 2, wherein performing frequency domain analysis on some of the multiple radio frequency signals to obtain a blood flow area corresponding to each frame of the intravascular ultrasound image comprises: Determine a matching radio frequency signal that matches each radio frequency signal to be matched, wherein the radio frequency signal to be matched belongs to any radio frequency signal corresponding to the i-th frame of intravascular ultrasound image, the matching radio frequency signal belongs to a radio frequency signal corresponding to any adjacent frame image adjacent to the i-th frame of intravascular ultrasound image, and one radio frequency signal corresponds to one rotation angle; Summing all matching radio frequency signals that match the kth radio frequency signal to be matched in the i-th frame of intravascular ultrasound image with the kth radio frequency signal to be matched, to obtain a kth target radio frequency signal, wherein the kth radio frequency signal to be matched is any radio frequency signal corresponding to the i-th frame of intravascular ultrasound image; Repeat the above process until the target radio frequency signals of all radio frequency signals to be matched corresponding to the i-th frame of intravascular ultrasound image are obtained; The target radio frequency signal is subjected to frequency domain analysis to obtain the blood flow region of the i-th frame of the intravascular ultrasound image. The method of claim 3, wherein determining a matching radio frequency signal that matches each radio frequency signal to be matched comprises: Selecting an image block from the i-th frame of the intravascular ultrasound image as the image block to be matched; According to the search area, the image blocks to be matched are obtained from the plurality of adjacent frame images. image block, obtaining candidate matching RF signals; Repeat the above process until all candidate matching radio frequency signals corresponding to the i-th frame of intravascular ultrasound image are obtained; The matching radio frequency signal is obtained according to all the candidate matching radio frequency signals. The method according to claim 4, wherein obtaining the matching radio frequency signal according to all the candidate matching radio frequency signals comprises: In response to confirming that the difference between the rotation angle sequence number corresponding to the image block to be matched and the rotation angle sequence number corresponding to any one of the candidate matching radio frequency signals is less than or equal to a first threshold, taking any one of the candidate matching radio frequency signals as the matching radio frequency signal corresponding to the image block to be matched; In response to confirming that the difference between the rotation angle corresponding to the image block to be matched and the rotation angles corresponding to all candidate matching radio frequency signals is greater than a first threshold, an radio frequency signal that meets the first threshold is selected from the radio frequency signals corresponding to the corresponding adjacent frame images as the matching radio frequency signal that matches the image block to be matched. The method according to claim 4 or 5, wherein obtaining the matching radio frequency signal according to all the candidate matching radio frequency signals comprises: Acquire a curve corresponding to all candidate matching radio frequency signals corresponding to a frame of intravascular ultrasound image, wherein one point on the curve corresponds to one candidate matching radio frequency signal, and the frame of intravascular ultrasound image belongs to any frame of the adjacent frame images; Screening out points that do not satisfy monotonicity from the curve; An updated radio frequency signal is used as a matching radio frequency signal corresponding to the point that does not satisfy monotonicity. The method according to claim 6, wherein the updated RF signal is a RF signal that satisfies monotonicity and is selected from RF signals corresponding to corresponding adjacent frame images. The method according to claim 3, wherein the step of performing frequency domain analysis on the target radio frequency signal to obtain the blood flow area of the i-th frame of the intravascular ultrasound image comprises: Performing Fourier transform on the target radio frequency signal to obtain a frequency domain signal; The frequency domain signal is filtered at least by a high-pass filter to obtain the i-th frame of intravascular ultrasound Blood flow area of the image. The method according to claim 8, wherein filtering the frequency domain signal at least by a high-pass filter to obtain the blood flow area of the i-th frame of the intravascular ultrasound image comprises: Filtering the frequency domain signal through the high-pass filter to obtain an i-th frame of high-frequency time domain signal corresponding to the i-th frame of intravascular ultrasound image; Repeat the following process to obtain the i-th frame composite time domain signal: determine that the similarity value between the n-th high-frequency radio frequency signal corresponding to the i-th frame high-frequency time domain signal and the adjacent angle high-frequency radio frequency signal is greater than a set threshold, then perform a composite operation on the n-th high-frequency radio frequency signal and the adjacent angle high-frequency radio frequency signal to obtain the n-th composite radio frequency signal, wherein the n-th high-frequency radio frequency signal is the high-frequency radio frequency signal of the n-th rotation angle corresponding to the i-th frame high-frequency time domain signal, and the adjacent angle high-frequency radio frequency signal is the high-frequency radio frequency signal of the n+u-th or n-u-th rotation angle corresponding to the i-th frame high-frequency time domain signal, and u is an integer greater than 0 and less than a set integer value; The frequency domain signal corresponding to the nth composite radio frequency signal is filtered by the high-pass filter to obtain the blood flow area of the i-th frame of the intravascular ultrasound image. The method according to claim 9, wherein the value of the set threshold is greater than or equal to 0.5. The method of claim 9, wherein the compound operation comprises a summation or a delayed summation. The method according to any one of claims 1 to 11, wherein removing at least the main branch vessel from the blood flow area to obtain a bifurcated vessel connected to the main branch vessel comprises: According to the non-connectivity between the side branch vessels and the main branch vessels, the corresponding side branch vessels are removed from the blood flow area to obtain a corrected blood vessel area; Acquire a three-dimensional region corresponding to the main branch blood vessel; A difference operation is performed between the three-dimensional region and the corrected blood vessel region to obtain a bifurcated blood vessel connected to the main branch blood vessel. The method of claim 12, wherein acquiring the three-dimensional region corresponding to the main branch vessel comprises: Acquire the blood flow area boundary of the modified blood vessel area through a plurality of longitudinal section data, and perform curve fitting on the blood flow area boundary to obtain a plurality of main branch boundaries; The three-dimensional region is obtained by performing three-dimensional surface fitting on the multiple main branch boundaries. A device for acquiring bifurcated blood vessels based on frequency domain signals, comprising: A radio frequency signal acquisition module is configured to acquire a plurality of radio frequency signals; A blood flow region acquisition module is configured to perform frequency domain analysis on some of the multiple radio frequency signals to obtain a blood flow region corresponding to each frame of the intravascular ultrasound image; The bifurcated vessel acquisition module is configured to remove at least the main branch vessel from the blood flow area to obtain a bifurcated vessel connected to the main branch vessel. A computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, can implement the method according to any one of claims 1 to 13. An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor can implement the method according to any one of claims 1 to 13 when executing the program.