Method for estimating the angle between an ultrasound beam and the direction of blood flow movement, calibration system and method

Abstract

The application discloses an estimation method of an angle between an ultrasonic beam and a blood flow movement direction, a calibration system and a method. The estimation method of the angle between the ultrasonic beam and the blood flow movement direction comprises the following steps: training a neural network model based on image datasets composed of ultrasonic blood vessel wall images with different deflection angles of different blood vessels in advance; calling the neural network model to segment an ultrasonic blood vessel wall image of a blood vessel to be estimated to obtain a blood vessel cross-section graph; acquiring pixel points forming a blood vessel lumen boundary and calculating a lumen area; analyzing time sequence changes of the lumen area by using a time sequence analysis method; if the time sequence changes of the lumen area are stable, fitting an elliptical geometric graph corresponding to the blood vessel cross-section; otherwise, fitting a convex hull geometric graph; and estimating the angle between the ultrasonic beam and the blood flow movement direction by using a formula, wherein a is a long semi-axis of the geometric graph, and b is a short semi-axis of the geometric graph. The application can accurately estimate the angle between the ultrasonic beam and the blood flow movement direction.

Description

Technical Field

[0001] This invention relates to the field of ultrasound imaging technology, and more particularly to an intravascular ultrasound imaging technique, specifically a method for estimating the angle between the ultrasound beam and the direction of blood flow, and a technique for calibrating blood flow velocity using this angle. Background Technology

[0002] Monitoring blood flow velocity within blood vessels is crucial for understanding vascular changes and is one of the most important parameters for assessing vascular health. However, current ultrasound techniques for measuring blood flow velocity primarily rely on the Doppler effect, meaning that the echo signal received by the ultrasound probe from blood cells exhibits a Doppler frequency shift, the amplitude and direction of which are related to the relative motion between the probe and the blood. An angle exists between the ultrasound beam emitted by the probe and the direction of blood flow (referred to in some literature as the beam angle or Doppler angle), which is the main source of error in blood flow velocity measurement.

[0003] The effectiveness of using external probes such as linear arrays and convex arrays to measure blood flow velocity often depends on the operator's experience in scanning the cross-section, thus hindering quantitative measurement of blood flow velocity. Because ultrasound B-mode imaging only supports observation of limited cross-sectional information of blood vessels, it is difficult to determine the overall orientation and structure of the vessel. Therefore, when measuring blood flow velocity externally, the θ angle is generally set between 15° and 60°. The greater the difference between the actual θ and the set value, the greater the measurement error.

[0004] Traditional intravascular ultrasound imaging technology uses an external motor to drive a flexible drive shaft, enabling the ultrasound catheter to perform rotational scanning within the blood vessel. However, this imaging method has an unavoidable drawback: the angle between the ultrasound catheter and the midline of the blood vessel is constantly changing, making it difficult to determine the value of θ using geometric characteristics.

[0005] To obtain the angle between the ultrasound beam and the blood flow direction, there are currently three main methods: the artificial method, dual-beam and multi-beam Doppler methods, and three-dimensional vascular reconstruction methods.

[0006] The manual method is necessary because commercially available color Doppler ultrasound diagnostic systems cannot automatically determine the angle between the ultrasound beam and the direction of blood flow. Therefore, the operator needs to analyze the color Doppler image and continuously adjust the sampling volume and the direction of the sampling line. In general examinations, the sampling volume is less than one-third of the lumen, and the sampling line should always be kept parallel to the lumen. However, manual calibration is time-consuming and labor-intensive, and the calibrated measurements have drawbacks such as poor repeatability and high subjectivity, which may interfere with the doctor's accurate assessment of changes in vascular hemodynamic function.

[0007] In dual-beam and multi-beam Doppler ultrasound, the ultrasound probe is equipped with piezoelectric transducers arranged at different tilt angles to receive blood flow echo signals from different ultrasound incident angles. Then, through geometric relationships and equations, the angle θ between the ultrasound beam and the blood flow direction, as well as the blood flow velocity v, are derived. Although this technique can theoretically calculate θ accurately, it suffers from several practical limitations. For example, when a physician holds the probe or uses an ultrasound patch, pressure on the patient's tissues is unavoidable. This causes deformation of the probe's contact surface with the skin, and the blood vessels are also compressed to some extent. Consequently, the incident angle of the ultrasound beam is no longer equal to the tilt angle of the piezoelectric transducers, thus violating the theoretical assumptions. Furthermore, measuring blood flow velocity using dual-beam Doppler requires highly sophisticated probe manufacturing processes and extensive, complex calculations to determine θ, making it difficult to scale to commercial color Doppler ultrasound monitoring systems.

[0008] Optical coherence tomography (OCT) for three-dimensional vascular reconstruction is an emerging technology that eliminates the need for intravascular contrast agent injection. For example, existing techniques, such as those described in publication CN105286779A, typically require coronary angiography to acquire cross-sectional images of the blood vessels. Optical scattering signals are obtained through circular or horizontal scanning, and the two-dimensional images are analyzed using decorrelation to perform three-dimensional reconstruction of the retinal and choroidal vascular systems. The three-dimensional vascular structure is analyzed to determine the position of the vessel's central axis and obtain the angle between the incident light ray and the vessel's central axis. Although OCT offers high spatial resolution, its penetration depth is only a few millimeters. Therefore, it can only be used to measure blood flow velocity in superficial skin or retinal vessels, lacking the universality of ultrasound. Furthermore, three-dimensional vascular reconstruction requires high-performance hardware, undoubtedly increasing the cost of instrument system development and hindering its adoption in primary healthcare institutions.

[0009] Therefore, how to provide a method that is easy to promote and can accurately estimate the angle between the ultrasound beam and the direction of blood flow is an urgent technical problem to be solved. Summary of the Invention

[0010] To address the limitations or inaccuracies of existing technologies for accurately estimating the angle between the ultrasound beam and the direction of blood flow, this invention proposes a method, calibration system, and approach for estimating the angle between the ultrasound beam and the direction of blood flow.

[0011] The method for estimating the angle between the ultrasound beam and the direction of blood flow in this invention includes:

[0012] A neural network model was trained using an image dataset composed of ultrasound images of the vessel walls at different deflection angles for different blood vessels.

[0013] The neural network model is invoked to segment the ultrasound image of the blood vessel wall to be estimated, thereby obtaining a cross-sectional image of the blood vessel.

[0014] Obtain the pixels that make up the boundary of the blood vessel lumen and calculate the lumen area;

[0015] The temporal variation of the lumen area was analyzed using time series analysis.

[0016] If the lumen area changes steadily over time, then fit an elliptical geometry corresponding to the blood vessel cross-section; otherwise, fit a convex hull geometry.

[0017] Using formula Estimate the angle between the ultrasound beam and the direction of blood flow, where a is the major semi-axis of the geometric figure and b is the minor semi-axis of the geometric figure.

[0018] Furthermore, by analyzing the binary image of the lumen region of the blood vessel cross-section pattern, the pixels that make up the boundary of the blood vessel lumen are found.

[0019] Furthermore, when the lumen area changes steadily over time, the zero-order moment M of the image grayscale is calculated using the grayscale values ​​f(i,j) of the pixels in the geometric image. 00 =∑ i ∑ j f(i,j) and first-order moment M 10 =∑ i ∑ j i·f(i,j),M 01 =∑ i ∑ j j·f(i,f) yields the coordinates (x0,y0) of the centroid of the geometric figure.

[0020] Based on the orientation angle, major semi-axis, and minor semi-axis of the ellipse, the standard inclined ellipse equation is derived using coordinate transformation to obtain the elliptical geometric shape corresponding to the blood vessel cross-section.

[0021] Furthermore, when the lumen area changes non-stationarily over time, the distance relationship between heel-side pixels is analyzed using a rotating caliper algorithm. The two farthest pixels are connected as the major axis of the geometric figure. Utilizing the relationship that the product of the minor axis slope and the major axis slope is always equal to -1, a point-slope equation is established to solve the equation of the straight line representing the minor axis. The coordinates of the two overlapping pixels in the pixel set of the straight line equation and the pixel set of the geometric figure are taken as the endpoint coordinates of the minor axis. The coordinates of the intersection of the minor axis and the major axis are taken as the coordinates of the centroid of the geometric figure. All vertices in the pixel set of the image are then connected to obtain the convex hull-shaped geometric figure corresponding to the blood vessel cross-section.

[0022] Furthermore, when there are no two overlapping pixels in the pixel set of the straight line equation and the pixel set of the geometric figure, the distance from all pixels in the pixel set of the geometric figure to the minor axis is calculated. The coordinates of the two pixels with the smallest distance to the minor axis are taken as the endpoint coordinates of the minor axis, and the coordinates of the intersection of the minor axis and the major axis are taken as the coordinates of the centroid of the geometric figure.

[0023] The blood flow velocity calibration method proposed in this invention includes:

[0024] Acquire B-mode ultrasound images of the vessel wall and Doppler shift signals of blood cells within the blood vessel;

[0025] The angle between the ultrasound beam and the direction of blood flow is estimated using the above-mentioned technical solution.

[0026] Relationship based on the ultrasonic Doppler effect The blood flow velocity is calculated, where v is the blood flow velocity, c is the speed of sound in blood, and f is the blood flow velocity. d It is the Doppler frequency shift of the first echo signal reflected back from red blood cells in the blood flow, f0 is the center frequency of the device measuring the first echo signal, and θ is the angle between the ultrasound beam and the direction of blood flow.

[0027] The blood flow velocity calibration system proposed in this invention includes:

[0028] Ultrasonic catheters are used to simultaneously acquire B-mode ultrasound images of the vessel wall and Doppler shift signals of blood cells within the blood vessel.

[0029] The calculation module uses the estimation method for the angle between the ultrasound beam and the blood flow direction described in the above technical solution to obtain the angle between the ultrasound beam and the blood flow direction, and is based on the relationship of the ultrasound Doppler effect. The blood flow velocity is calculated, where v is the blood flow velocity, c is the speed of sound in blood, and f is the blood flow velocity. d It is the Doppler frequency shift of the first echo signal reflected back from red blood cells in the blood flow, f0 is the center frequency of the device measuring the first echo signal, and θ is the angle between the ultrasound beam and the direction of blood flow.

[0030] Furthermore, the ultrasound catheter includes:

[0031] tube body;

[0032] The transducer fixing block is located at one end of the tube body;

[0033] The first ultrasonic transducer is disposed at the first end of the transducer fixing block near the end of the tube body, for emitting a first sound beam toward the outside of the tube body and receiving the first echo signal reflected back by red blood cells in the blood flow.

[0034] The second ultrasonic transducer is disposed at the second end opposite to the transducer fixing block and is positioned opposite to the first ultrasonic transducer. It is used to emit a second sound beam perpendicular to the tube wall and receive a second echo signal.

[0035] An acoustic reflector, disposed opposite to the second ultrasonic transducer within the tube, includes a reflector and a drive assembly. The reflector reflects the emitted second sound beam perpendicular to the tube wall, and the drive assembly drives the reflector to rotate around its axis.

[0036] Furthermore, the reflector is made of a cylindrical magnet, and the first end facing the second ultrasonic transducer is an inclined plane at 45° to the axis of the tube body. The driving assembly includes an inner tube that fixes the reflector inside the tube body, a coil wound on the inner tube, and a ball bearing located at the second end of the reflector inside the inner tube.

[0037] Compared with the prior art, the present invention has the following advantages and beneficial effects.

[0038] (1) Compared with the manual method of the prior art, the present invention can automatically estimate and calibrate blood flow velocity. It shortens the time spent on each calibration and improves the accuracy and repeatability of blood flow velocity measurement.

[0039] (2) Compared with dual-beam and multi-beam Doppler methods based on external ultrasound probes, the ultrasound catheter of this invention is placed inside the blood vessel, avoiding compression of tissues and blood vessels and reducing external interference caused by the measurement principle. Therefore, it is beneficial to measure the true blood flow velocity more accurately. Compared with dual-beam Doppler methods based on intravascular ultrasound catheters, the ultrasound catheter of this invention has only two transducers, so the structure is simpler and more compact, and the manufacturing cost is much lower.

[0040] (3) Compared with existing three-dimensional vascular reconstruction methods, the main steps of the three-dimensional vascular reconstruction method include: acquiring vascular volume metadata, three-dimensional segmentation, three-dimensional vascular reconstruction, and analyzing the trend of the vascular midline, etc. However, this invention utilizes deep learning technology to segment the lumen region of a two-dimensional B-mode image, and then fits the lumen boundary to estimate it. Therefore, it reduces the hardware development cost and computation time of the instrument system, and is more conducive to application in low- to mid-range color Doppler ultrasound diagnostic systems.

[0041] (4) Compared with the prior art, the present invention takes into account the turbulence phenomenon caused by the deformation of blood vessels in cardiovascular patients. When blood vessels are diseased and compressed, the shape of the lumen in the resulting B-mode image is irregular. By analyzing the variation law of the lumen area, the lumen boundary is fitted into an inclined ellipse or convex hull according to the actual situation, and the Doppler angle is solved and calibrated using the principal axis parameters. This technical approach is more in line with real clinical problems.

[0042] (5) In order to ensure the relative accuracy of the measurement, the cutting surface of the external ultrasound probe in the prior art cannot be parallel to the skin surface or the tilt angle is too large. Therefore, it is generally only suitable for measuring blood flow velocity in the case of 15°-60°. However, the deflection angle range of the intravascular ultrasound catheter of the present invention in the blood vessel can reach 0°-60°, which is suitable for the case of 0°-60°.

[0043] (6) This invention can be applied to the measurement and calibration of blood flow velocity in both superficial and deep blood vessels. Based on the Doppler effect of ultrasound, the displacement of blood cells is analyzed to directly measure blood flow velocity. Because it does not need to consider the influence of physiological parameters such as patient weight, blood vessel diameter, and blood pressure, it eliminates the systematic error caused by converting other physiological parameters into blood flow velocity in indirect measurement methods. Attached Figure Description

[0044] The present invention will now be described in detail with reference to the embodiments and accompanying drawings, wherein:

[0045] Figure 1 This is the main flowchart of the angle estimation and blood flow velocity calibration method of the present invention.

[0046] Figure 2 A three-dimensional structural diagram of the ultrasonic catheter of the present invention.

[0047] Figure 3 This is a cross-sectional view of an ultrasonic catheter according to an embodiment of the present invention.

[0048] Figure 4 This is a flowchart of the blood flow detection process of the present invention.

[0049] Figure 5 This is a flowchart of the intravascular ultrasound imaging process of the present invention.

[0050] Figure 6(a) is a set of pixels representing the boundary of a cavity according to an embodiment of the present invention.

[0051] Figure 6(b) shows the convex hull fitting result of the lumen boundary in Figure 6(a).

[0052] Figure 7(a) shows the blood flow velocity results before calibration.

[0053] Figure 7(b) shows the blood flow velocity results after calibration using the present invention.

[0054] Figure 8 This is a schematic diagram of the off-axis and torsion phenomena of an ultrasound catheter within a blood vessel.

[0055] Figure 9 This is a schematic diagram illustrating the basic theoretical analysis of the included angle estimation method of this invention.

[0056] Figure 10This is a schematic diagram illustrating how the angle between the ultrasound beam and the blood flow direction is equal to the angle between the axis of the ultrasound catheter and the midline of the blood vessel.

[0057] Figure 11 (a) is an intravascular ultrasound image of a blood vessel obtained according to an embodiment of the present invention.

[0058] Figure 11 (b) is a mask marking the lumen area.

[0059] Figure 11 (c) is the segmentation result of the lumen region.

[0060] Figure 11 (d) is a binary image representing the lumen region.

[0061] Figure 11 (e) is a binary image representing the boundary of the lumen.

[0062] Figure 11 (f) is the ellipse fitting result of the lumen boundary.

[0063] Explanation of reference numerals in the attached figures:

[0064] 1. Transducer fixing block; 2. Tube body; 3. Inner tube; 4. Coil; 5. Reflector; 6. Ball bearing; 7. Piezoelectric crystal of the first ultrasonic transducer; 8. Matching layer of the first ultrasonic transducer; 9. Backing layer of the first ultrasonic transducer; 10. Piezoelectric crystal of the second ultrasonic transducer; 11. Matching layer of the second ultrasonic transducer; 12. Backing layer of the second ultrasonic transducer. Detailed Implementation

[0065] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0066] Therefore, a feature pointed out in this specification is used to describe one feature of one embodiment of the invention, and does not imply that every embodiment of the invention must have the described feature. Furthermore, it should be noted that this specification describes many features. Although certain features may be combined to illustrate possible system designs, these features may also be used in other combinations not explicitly stated. Therefore, unless otherwise stated, the described combinations are not intended to be limiting.

[0067] The method for estimating the angle between the ultrasound beam and the direction of blood flow in this invention mainly includes the following steps.

[0068] Before estimation, a neural network model for image segmentation needs to be prepared in advance. To prepare the corresponding neural network model, an image dataset needs to be acquired. The image dataset needs to contain ultrasound images of blood vessels with different angles and different deflection angles. After acquiring ultrasound images of blood vessels with different angles and different deflection angles, a neural network model is trained based on the image dataset composed of ultrasound images of blood vessels with different angles and different deflection angles. The input of the neural network model is the ultrasound image of the blood vessel wall of the blood vessel whose angle is to be estimated, and the output is the cross-sectional image of the blood vessel whose angle is to be estimated.

[0069] After obtaining the trained neural network model, when it is necessary to estimate the direction of ultrasound beam and blood flow, first obtain the ultrasound image of the blood vessel wall to be estimated, and then call the trained neural network model to segment the ultrasound image of the blood vessel wall to be estimated to obtain the cross-sectional image of the blood vessel.

[0070] Based on the obtained cross-sectional image of the blood vessel, the pixels representing the vascular lumen boundary are acquired to calculate the lumen area. After the ultrasound catheter is assembled, B-mode imaging can be performed using a standard ring-shaped wire target phantom to test axial and lateral resolution. The wire diameter of the nylon or metal wire target, as well as its position and angle embedded in the blood vessel phantom, are generally known and can be found in the instruction manual. The delay of the echo signal reflected by the wire target on the oscilloscope is observed, and the Euclidean distance from the wire target to the centerline of the ultrasound catheter is calculated using the formula d = c * t / 2. A B-mode image consists of a two-dimensional matrix containing grayscale brightness values. Therefore, by analyzing the distance of the set of bright pixels representing the wire target from the image center, and the number of pixels comprising the set, the true size represented by a pixel, i.e., the image scale, can be estimated. Based on the analysis of multiple sets of the above experiments, the scale can be obtained. Once the number of pixels constituting the lumen region is obtained, the area of ​​the lumen region can be estimated.

[0071] The temporal variation of the lumen area was analyzed using time series analysis.

[0072] If the lumen area changes steadily over time, then fit an elliptical geometry corresponding to the blood vessel cross-section; otherwise, fit a convex hull geometry.

[0073] Using formula Estimate the angle between the ultrasound beam and the direction of blood flow. 'a' is the major semi-axis of the geometric figure, which is half of the major axis of the geometric figure, and 'b' is the minor semi-axis of the geometric figure, which is half of the minor axis of the geometric figure.

[0074] This invention analyzes the temporal changes in the lumen area using time series analysis and fits ellipses or convex hulls according to different situations, so that the obtained geometric shape of the blood vessel cross section is closer to the original shape of the blood vessel. Compared with the existing technology that always uses ellipses to represent the blood vessel cross section, this invention can obtain a more accurate angle between the blood flow direction and the ultrasound beam.

[0075] Based on the above technical solution, one embodiment of the present invention finds the pixels that make up the boundary of the blood vessel lumen by analyzing the binary image of the lumen region of the blood vessel cross-section graphic.

[0076] In one embodiment, when the temporal change in lumen area is stable, it indicates that there is no embolism or other abnormalities in the blood vessel, and it is a normal blood vessel. The zero-order moment M of the image grayscale is calculated using the grayscale values ​​f(i,j) of the pixels in the geometric image. 00 =∑ i ∑ j f(i,j) and first-order moment M 10 =∑ i ∑ j i·f(i,j),M 01 =∑ i ∑ j j·f(i,j) yields the coordinates (x0,y0) of the centroid of the geometric figure.

[0077] Then, based on the orientation angle, major semi-axis, and minor semi-axis of the ellipse, the standard inclined ellipse equation is derived using coordinate transformation to obtain the elliptical geometric shape corresponding to the blood vessel cross-section. In other words, after obtaining the centroid of the geometric shape, this invention fits the equation of a general ellipse, then obtains the orientation angle, major semi-axis, and minor semi-axis of the ellipse corresponding to the general ellipse equation, and finally obtains the standard inclined ellipse equation.

[0078] In one embodiment, when the lumen area changes non-stationarily over time, the distance relationship between opposing pixels is analyzed using a rotating caliper algorithm. The two farthest pixels are connected as the major axis of the geometric figure. Utilizing the relationship that the product of the minor axis slope and the major axis slope is always equal to -1, a point-slope equation is established to solve for the linear equation representing the minor axis. The coordinates of the two overlapping pixels in the pixel set of the linear equation and the pixel set of the geometric figure are used as the endpoint coordinates of the minor axis. The coordinates of the intersection of the minor and major axes are used as the coordinates of the centroid of the geometric figure. All vertices in the pixel set of the image are then connected to obtain the convex hull-shaped geometric figure corresponding to the blood vessel cross-section. The reason for calculating the centroid coordinates in this invention is that when fitting the convex hull, the major axis of the convex hull does not necessarily pass through the calculated centroid, but it is certainly near the major axis. This invention assumes that the minor axis passes through the centroid, thus establishing a point-slope equation to solve for the linear equation of the minor axis.

[0079] For blood vessels, thrombi may exist within them, causing the cross-section of the vessel to be neither circular nor elliptical. For example, as atherosclerotic plaques develop, blood vessels become increasingly narrow. Due to the irregular shape of plaques in the coronary and carotid arteries, the velocity gradient of each particle changes randomly when high-speed arterial blood flows through the plaque, often resulting in turbulence. When imaging narrowed vessel segments, the shape of the vessel lumen is often irregular. When measuring blood flow velocity in narrowed vessel segments, the blood flow distribution is not laminar. Therefore, the lumen boundary should not be fitted as an ellipse, but rather as a convex hull to analyze the major and minor axes of the lumen region. When imaging normal blood vessels, the lumen area changes periodically with pulsation. When imaging narrowed vessels containing plaques, the lumen shapes in adjacent image sequences are not identical, thus the lumen area changes randomly. Segmenting the lumen region in B-mode images under linear time series and analyzing the strength of randomness in lumen area changes can determine whether the ultrasound catheter is currently measuring blood flow velocity at a narrowed vessel segment.

[0080] Assuming the presence of small thrombi or plaques within the blood vessel, and that these plaques are randomly distributed along the vessel wall, imaging the narrowed segment reveals a clear B-mode image of the vessel exhibiting both arc-shaped and irregularly curved structures. As shown in Figure 6(a), after segmenting the lumen boundary, the binary image results are analyzed to find the set of pixels S representing the lumen boundary. The arc-shaped structure represents the normal inner wall of the blood vessel, while the irregularly curved structure represents the plaque portion. Clearly, points on the irregular curve are closer to the inner wall of the blood vessel, while points on the arc are farther away. The distance relationship between opposing pixels is analyzed using the rotating caliper algorithm. Connecting the two farthest pixels forms the line segment l, which is the diameter of the convex hull (major axis). The slope of the major axis is solved using a two-point equation. Theoretically, the minor axis is perpendicular to the major axis and passes through the centroid. Therefore, the product of the slope of the minor axis and the slope of the major axis is always equal to -1. Then, the infinitely long straight line t representing the minor axis is solved using the point-slope equation. However, there may be a situation where no two points in set S lie exactly on line t. In this case, the distances from all pixels in S to t should be analyzed, and the two points with the smallest distance should be taken as the two endpoints of the minor axis. As shown in Figure 6(b), the positions of all pixels in set S are analyzed, a convex hull is fitted, and then the vertices of the convex hull are connected to obtain a convex polygon. "*" represents the centroid of the lumen region, the dashed line represents the major axis of the lumen region, and the dotted line represents the minor axis of the lumen region.

[0081] This invention is based on the aforementioned method for estimating the angle between the ultrasound beam and the direction of blood flow, and also protects the blood flow velocity calibration method, such as... Figure 1 As shown, the blood flow velocity calibration method of the present invention mainly includes the following steps.

[0082] Acquire B-mode ultrasound images of the vessel wall and Doppler shift signals of blood cells within the blood vessel;

[0083] The angle between the ultrasound beam and the direction of blood flow is estimated using the above-mentioned technical solution.

[0084] Relationship based on the ultrasonic Doppler effect The blood flow velocity was calculated.

[0085] v is the velocity of blood flow, c is the speed of sound in blood, f d It is the Doppler frequency shift of the first echo signal reflected back from red blood cells in the blood flow, f0 is the center frequency of the device measuring the first echo signal, and θ is the angle between the ultrasound beam and the direction of blood flow.

[0086] The blood flow velocity calibration system of the present invention includes an ultrasound catheter and a calculation module.

[0087] Ultrasonic catheters are used to simultaneously acquire B-mode ultrasound images of the vessel wall and Doppler shift signals of blood cells within the blood vessel.

[0088] The calculation module uses the estimation method for the angle between the ultrasound beam and the blood flow direction in the above-mentioned technical solution to obtain the angle between the ultrasound beam and the blood flow direction, and based on the relationship of the ultrasound Doppler effect. The blood flow velocity is calculated, where v is the blood flow velocity, c is the speed of sound in blood, and f is the blood flow velocity. d It is the Doppler frequency shift of the first echo signal reflected back from red blood cells in the blood flow, f0 is the center frequency of the device measuring the first echo signal, and θ is the angle between the ultrasound beam and the direction of blood flow.

[0089] In one embodiment, a display module may also be included to display the calibrated blood flow velocity results, i.e., pulsed Doppler blood flow imaging.

[0090] Figure 2 , Figure 3 The diagram shows the specific structure of the ultrasonic catheter of the present invention. The ultrasonic catheter includes a tube body 2, a transducer fixing block 1, a first ultrasonic transducer, a second ultrasonic transducer, and an acoustic reflector.

[0091] The tube body is the outer shell of the 2 ultrasonic catheters.

[0092] The transducer fixing block 1 is installed inside one end of the tube body 2.

[0093] The first ultrasonic transducer, located at the first end of the transducer fixing block near the tube body, is used to emit a first sound beam toward the outside of the tube body and receive the first echo signal reflected back from red blood cells in the blood flow; in the relationship of the ultrasonic Doppler effect, f df0 is the center frequency of the first ultrasonic transducer, as measured by the first transducer. The first ultrasonic transducer includes a piezoelectric crystal 7, a matching layer 8, and a backing layer 9.

[0094] The second ultrasonic transducer is disposed at the second end opposite to the transducer fixing block 1 and is positioned back-to-back with the first ultrasonic transducer. It is used to emit a second sound beam perpendicular to the tube wall and to receive a second echo signal. The second ultrasonic transducer includes a piezoelectric crystal 10, a matching layer 11, and a backing layer 12.

[0095] The acoustic reflector is designed to allow the second ultrasonic transducer to emit a second sound beam perpendicular to the tube wall. The acoustic reflector is positioned opposite the second ultrasonic transducer within the tube. The acoustic reflector includes a reflector 5 and a drive assembly. The reflector 5 reflects the emitted second sound beam perpendicular to the tube wall, while the drive assembly drives the reflector to rotate around its axis. The drive assembly includes an inner tube, a coil, and ball bearings.

[0096] The reflector 5 is made of a cylindrical magnet, and its first end facing the second ultrasonic transducer is an inclined plane at a 45° angle to the axis of the tube body. This inclined plane is coated with a metal film layer that can reflect sound waves, thus serving as a reflector and also capable of rotating 360 degrees in conjunction with the coil 4. The driving assembly includes an inner tube 3 that fixes the reflector 5 inside the tube body 2. The coil 4 is wound on the inner tube 3. When three-phase alternating current is applied to the coil 4, the reflector 5 can rotate 360 ​​degrees. Ball bearings 6 are located inside the inner tube at the second end of the reflector. The inner tube is made of a thin pure copper tube, and the ball bearings are made of steel balls.

[0097] Figure 4 The flowchart illustrates blood flow detection using a first transducer. A pulse transceiver excites the first transducer, causing it to emit ultrasound waves into the blood. An A / D converter acquires the echo signals received by the first transducer. The one-dimensional radio frequency (RF) data is arranged according to the directions of fast and slow signals, reconstructing a two-dimensional RF data matrix. A Fourier transform is performed on the signals to analyze their frequency domain characteristics, and suitable low-pass and wall filters are designed. The low-pass filter removes high-frequency components, while the wall filter removes the Doppler frequency shift signal of vascular pulsation. The fast signal is demodulated, and the slow signal at a certain sampling depth is analyzed. Blood flow velocity is estimated using spectral Doppler methods, autocorrelation methods, and autoregression methods, and the results are finally displayed.

[0098] Figure 5The flowchart illustrates intravascular ultrasound imaging using a second transducer. A pulse transceiver excites the second transducer, causing it to emit ultrasonic waves towards an acoustic mirror. As a micromotor rotates, the ultrasonic waves are uniformly reflected into the vascular tissue by the acoustic mirror, achieving a 360° circular scan of the blood vessel. An A / D acquisition card acquires the echo signals received by the second transducer. The one-dimensional radio frequency (RF) data is arranged according to the fast and slow signal directions, reconstructing a two-dimensional RF data matrix. A Fourier transform is performed on the signals to analyze their frequency domain characteristics, and a suitable bandpass filter is designed to demodulate the fast signal. Due to the scanning characteristics of the second transducer, the two-dimensional RF data matrix is ​​arranged in polar coordinates. To conform to human image reading habits, the two-dimensional RF data matrix needs to be transformed from polar coordinates to Cartesian coordinates through digital coordinate conversion. Since the maximum and minimum pixel brightness values ​​often differ significantly, logarithmic compression should be performed to adapt to the grayscale levels of the display. Then, algorithms such as bilinear interpolation and nearest neighbor interpolation are used to fill the gaps between each sound line in the image matrix, and the circular B-mode image result is then displayed.

[0099] This invention places an ultrasound catheter on a deflection angle adjustment platform to simulate varying degrees of catheter off-axis and torsion within a blood vessel. Echo signals are acquired to obtain a series of intravascular ultrasound B-mode images. These images are annotated by a professional physician to construct a dataset. When the dataset has a small number of images, data augmentation processing is performed using elastic transformation and affine transformation to increase the sample size. Two-dimensional Gaussian filtering and adaptive denoising filtering are used to suppress white noise and speckle noise in the images, and histogram equalization is used to improve image contrast. Variant network structures of U-Net and DeepLabV3+ are designed using PyTorch and TensorFlow, and the network is trained until convergence, outputting a segmentation model (input is given). Inference code is deployed to call the model for real-time segmentation of the lumen region in the B-mode images. The binary image of the segmentation result is analyzed to detect the lumen boundary and calculate the lumen area. The temporal variation of the lumen area is analyzed using a differential autoregressive moving average model. If the area size is a stationary time series, the lumen boundary is fitted as a tilted ellipse; otherwise, the lumen boundary is fitted as a convex hull. Analyze the fitted geometry and solve for parameters such as the centroid, major axis, and minor axis. Utilize... To determine the angle between the ultrasound beam and the direction of blood flow, use... The blood flow velocity is calibrated, and the calibrated blood flow velocity result is displayed.

[0100] To conveniently and effectively obtain the angle between the ultrasound beam and the blood flow direction, this invention employs a method of simultaneously acquiring B-mode ultrasound images of the blood vessel wall and Doppler shift signals of blood cells within the vessel. The reliability of the angle calculation between the ultrasound beam and the blood flow direction in this invention will be explained below.

[0101] When a plane intersects a cylindrical surface, three scenarios can occur, and the shape of the intersection line depends on the relative position and angle between the plane and the cylinder's axis. When the plane is parallel to the cylinder's axis, the intersection line is rectangular. When the plane is perpendicular to the cylinder's axis, the intersection line is circular. When the plane is inclined to the cylinder's axis, the intersection line is elliptical, with the minor semi-axis of the ellipse equal to the radius of the cylinder's base. The second ultrasonic transducer of this invention, because it can emit a first sound beam perpendicular to the blood vessel wall, has its emitted sound waves reflected 360 degrees onto the blood vessel by a reflector, which can be considered as the scanning plane, while the blood vessel wall can be considered as a cylindrical surface.

[0102] This invention utilizes a peristaltic pump, a blood phantom, isolated blood vessels, and an ultrasound catheter to construct a hemodynamic circulation loop. Taking spectral Doppler as an example, Figure 7(a) shows the power spectrum results over 0.29 seconds, where the grayscale value of the pixels in the coordinate region represents the amplitude of the Doppler frequency shift signal. At 0.04004 seconds, the blood flow velocity is approximately 0.453125 meters per second. Clearly, the blood phantom flows regularly, with each cycle lasting approximately 0.1 seconds. The spectral envelope is analyzed using power spectral density integration and a three-line fitting method, and plotted as a solid line. This will be further explained through... Substituting the solution θ The calibrated blood flow velocity was obtained. As shown in Figure 7(b), at 0.04004 seconds, the calibrated blood flow velocity was approximately 0.482206 m / s.

[0103] When an intravascular ultrasound catheter is subjected to the impact of high-speed blood flow, off-axis and torsion phenomena occur. Assuming the second ultrasound transducer and acoustic reflector are aligned within the catheter, the extension of the ultrasound beam emitted by the first ultrasound transducer should be the normal vector of the scanning plane of the second transducer. For a finite-length blood vessel, under the influence of blood friction and blood pressure, the cross-sectional shape of the vessel lumen is approximately circular. Figure 8 As shown, assuming the blood vessel wall is cylindrical and the blood flow within the vessel is a periodic, pulsating laminar flow, the angle between the ultrasound beam emitted forward from the first ultrasound transducer and the direction of blood flow is the beam angle θ, and the angle between the axis of the ultrasound catheter and the midline of the blood vessel is δ. Figure 2 The ducts and blood vessels in the image are projected onto the xoz plane, and then... Figure 9 Geometric analysis was performed. Here, the DN line segment represents the projection of the scanning plane of the second transducer onto the xoz plane, and JK represents the projection of the lumen region in the annular B-mode image onto the xoz plane. Rectangle PQP'Q' represents a vessel of finite length, TR lies on the central axis of the vessel, and the straight line L is the extension of the ultrasound catheter, L⊥DN. Figure 10 A detailed argument yields θ = δ:

[0104] For normal blood vessels, the imaging result of the vessel lumen is approximately circular or elliptical. In this invention, a circle can be considered a special ellipse with equal major and minor axes. Therefore, by fitting the lumen boundary to an ellipse, the equation of a tilted ellipse can be derived, and the major and minor axes of the lumen region can be analyzed. Traditional image segmentation methods include edge detection, region growing, adaptive thresholding, etc. Machine learning-based image segmentation methods include simple linear iterative clustering, Naive Bayes, minimum error Bayesian decision method, etc. Deep learning-based image segmentation networks include U-Net and its variants, DeepLab V3+, Faster R-CNN, Mask R-CNN, etc.

[0105] Figure 11 It is the result of segmenting the lumen region and fitting the lumen boundary using deep learning technology. Figure 11 (a) shows the image acquired by the second ultrasonic transducer. First, the images in the dataset are preprocessed, such as through affine transformation, histogram equalization, and two-dimensional median filtering. Then, as... Figure 11 As shown in (b), image annotation software is used to annotate the luminal boundary along the inner wall of the blood vessel using circles or polygons, outputting labels and masks. Then, the convolutional neural network structure is designed, including the selection of activation functions, upsampling convolution kernel settings, downsampling pooling kernel settings, loss function selection, etc. The ratio of training to validation sets in the dataset is differentiated. The neural network is trained using a GPU, the learning rate is adjusted, and multiple iterations are performed until the loss function converges, at which point the segmentation model is saved. Figure 11 As shown in (c), the segmentation model is invoked to segment the lumen region from the intravascular ultrasound image. Figure 11 (d)- Figure 11 As shown in (e), output the binary image of the lumen region, analyze the connected components in the image, and find all the pixels that make up the lumen boundary.

[0106] A rectangular coordinate system is established with the center of the image as the origin. The coordinates of the pixels are analyzed, and the cavity boundary is fitted into an ellipse using an algorithm. The standard ellipse equation is then solved. There are various methods for fitting an ellipse, such as the least squares method, the Hough transform method, and the random sampling consistency method. The centroid coordinates P0(x0,y0) of the image are calculated by normalizing the first-order central moments. As shown in equations (2)-(3), the gray values ​​of the pixels in the two-dimensional image are represented by f(i,j), and the sum of the image gray values ​​is the zero-order moment M. 00 The first moment is M 10 and M 01 As shown in equation (4), the gray-level center of the image is the centroid of the image, which is the ratio of the first moment to the zeroth moment. As shown in equation (5), based on the orientation angle of the ellipse (the angle α between the major axis and the x-axis), the semi-major axis a, and the semi-minor axis b, the equation of the standard inclined ellipse is derived using coordinate transformation. Figure 11As shown in (f), the result of the ellipse fitting is depicted with a solid line near the boundary of the lumen, where α∈(0,π / 2).

[0107]

[0108] M 10 =∑ i ∑ j i·f(i,j),M 01 =∑ i ∑ j j·f(i,j) (2)

[0109]

[0110] Based on the above description, the present invention has the following characteristics.

[0111] This invention relates to an intravascular ultrasound catheter with two transducers at different center frequencies, enabling simultaneous detection of blood flow and intravascular ultrasound imaging. A synchronization signal triggers a pulse transceiver to excite the first and second ultrasound transducers. The echo signals from both transducers are acquired, simultaneously performing blood flow velocity measurement and vascular loop B-mode imaging.

[0112] This invention acquires B-mode images at different deflection angles during the training of a neural network model and uses deep learning technology to segment the lumen region. Variants of convolutional neural network structures such as U-Net and DeepLabV3+ are designed, and real-time segmentation is achieved by calling the model through inference code.

[0113] This invention analyzes connected regions and components in binary images to locate luminal boundaries. By analyzing the variation in luminal area, it determines whether blood flow velocity is measured in a normal or narrow vessel, thus using different methods to fit the luminal boundary. After obtaining the luminal boundary, the invention acquires the coordinates of image particles using the image center distance. Analyzing the fitting results of the luminal boundary, the major and minor axis lengths of the luminal region are calculated. Finally, based on the major and minor axes, the angle θ between the ultrasound beam and the direction of blood flow is estimated, and the blood flow velocity is then calibrated.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for estimating the angle between an ultrasound beam and the direction of blood flow, characterized in that, include: A neural network model was trained using an image dataset composed of ultrasound images of the vessel walls at different deflection angles for different blood vessels. The neural network model is invoked to segment the ultrasound image of the blood vessel wall to be estimated, thereby obtaining a cross-sectional image of the blood vessel. Obtain the pixels that make up the boundary of the blood vessel lumen and calculate the lumen area; The temporal variation of the lumen area was analyzed using time series analysis. If the lumen area changes steadily over time, then the elliptical geometry corresponding to the blood vessel cross-section is fitted. Otherwise, fit a convex hull geometry; Using formula Estimate the angle between the ultrasound beam and the direction of blood flow, where a is the major semi-axis of the geometric figure and b is the minor semi-axis of the geometric figure; When the lumen area changes non-stationarily over time, the distance relationship between heel pixels is analyzed by the rotating caliper algorithm. The two farthest pixels are connected as the major axis of the geometric figure. The product of the minor axis slope and the major axis slope is always equal to -1. A point-slope equation is established to solve the equation of the straight line representing the minor axis. The coordinates of the two overlapping pixels in the pixel set of the straight line equation and the pixel set of the geometric figure are taken as the endpoint coordinates of the minor axis. The coordinates of the intersection of the minor axis and the major axis are taken as the coordinates of the centroid of the geometric figure. All vertices in the pixel set of the image are connected to obtain the convex hull-shaped geometric figure corresponding to the blood vessel cross section. If there are no two overlapping pixels in the pixel set of the straight line equation and the pixel set of the geometric figure, then calculate the distance from all pixels in the pixel set of the geometric figure to the minor axis, take the coordinates of the two pixels with the smallest distance to the minor axis as the endpoint coordinates of the minor axis, and take the coordinates of the intersection of the minor axis and the major axis as the coordinates of the centroid of the geometric figure.

2. The method for estimating the angle between the ultrasound beam and the direction of blood flow as described in claim 1, characterized in that, By analyzing the binary image of the lumen region of the blood vessel cross-section, the pixels that make up the boundary of the blood vessel lumen are found.

3. The method for estimating the angle between the ultrasound beam and the direction of blood flow as described in claim 2, characterized in that, When the lumen area changes steadily over time, the zero-order moment of the image grayscale is calculated using the grayscale value f(i,j) of the pixel in the geometric image. and first moment This yields the coordinates (x0, y0) of the centroid of the geometric figure. ; Based on the orientation angle, major semi-axis, and minor semi-axis of the ellipse, the standard inclined ellipse equation is derived using coordinate transformation to obtain the elliptical geometric shape corresponding to the blood vessel cross-section.

4. A blood flow velocity calibration method, characterized in that, include: Acquire B-mode ultrasound images of the vessel wall and Doppler shift signals of blood cells within the blood vessel; The angle between the ultrasound beam and the direction of blood flow is obtained by using the estimation method described in any one of claims 1 to 3. Relationship based on the ultrasonic Doppler effect The blood flow velocity was calculated. It's the blood flow velocity. It's the speed of sound in blood. It is the Doppler shift of the first echo signal reflected back from red blood cells in the bloodstream. It is the center frequency of the device used to measure the first echo signal. It is the angle between the ultrasound beam and the direction of blood flow.

5. A blood flow velocity calibration system, characterized in that, include: Ultrasonic catheters are used to simultaneously acquire B-mode ultrasound images of the vessel wall and Doppler shift signals of blood cells within the blood vessel. The calculation module uses the estimation method for the angle between the ultrasound beam and the blood flow direction as described in any one of claims 1 to 3 to obtain the angle between the ultrasound beam and the blood flow direction, and is based on the relationship of the ultrasound Doppler effect. The blood flow velocity was calculated. It's the blood flow velocity. It's the speed of sound in blood. It is the Doppler shift of the first echo signal reflected back from red blood cells in the bloodstream. It is the center frequency of the device used to measure the first echo signal. It is the angle between the ultrasound beam and the direction of blood flow.

6. The blood flow velocity calibration system as described in claim 5, characterized in that, The ultrasonic catheter includes: tube body; The transducer fixing block is located at one end of the tube body; The first ultrasonic transducer is disposed at the first end of the transducer fixing block near the end of the tube body, for emitting a first sound beam toward the outside of the tube body and receiving the first echo signal reflected back by red blood cells in the blood flow. The second ultrasonic transducer is disposed at the second end opposite to the transducer fixing block and is positioned opposite to the first ultrasonic transducer. It is used to emit a second sound beam perpendicular to the tube wall and receive a second echo signal. An acoustic reflector, disposed opposite to the second ultrasonic transducer within the tube, includes a reflector and a drive assembly. The reflector reflects the emitted second sound beam perpendicular to the tube wall, and the drive assembly drives the reflector to rotate around its axis.

7. The blood flow velocity calibration system as described in claim 6, characterized in that, The reflector is made of a cylindrical magnet, and the first end facing the second ultrasonic transducer is an inclined plane at 45° to the axis of the tube. The drive assembly includes an inner tube that fixes the reflector inside the tube, a coil wound on the inner tube, and a ball bearing located at the second end of the reflector inside the inner tube.

Images