Ultrasound imaging method with improved transverse and lateral resolution

Abstract

The present invention discloses an ultrasonic imaging method for improving transverse and lateral resolution, comprising: controlling a single-row one-dimensional ultrasonic probe to acquire two-dimensional ultrasonic images, and setting the transverse focal depth to be the same as the transverse focal depth of the sensor; moving the probe to continuously scan and synchronously acquire corresponding posture information until all depth ranges of the target imaging area are overscanned near the transverse focal depth of the ultrasonic probe; rotating the probe or the imaging target approximately 90 degrees laterally and repeating the scanning process; reconstructing a three-dimensional ultrasonic image by combining the two-dimensional ultrasonic image sequence and the corresponding posture information; and extracting cross sections from the three-dimensional volume data. The present invention fully utilizes the acoustic field distribution properties of the ultrasonic sensor to effectively improve the transverse and lateral resolution without increasing cost or system complexity, thereby overcoming the disadvantages of low transverse and lateral resolution of conventional ultrasonic imaging.

Description

Technical Field

[0001] The present invention relates to an ultrasonic imaging method, in particular to an ultrasonic imaging method for improving transverse and lateral resolution. Background Art

[0002] Currently, common ultrasound imaging uses one-dimensional ultrasound probes, which have varying resolutions in the axial, transverse, and lateral directions. The transverse direction of a one-dimensional ultrasound probe refers to the direction in which the transducer array is arranged, while the axial direction refers to the direction perpendicular to the transducer plane in which the ultrasound signal is transmitted. Both the transverse and axial directions lie within the ultrasound scanning plane, while the direction perpendicular to the ultrasound scanning plane is called the lateral direction. The axial resolution of a one-dimensional ultrasound probe is theoretically half the spatial pulse length, offering the best resolution in the three directions. The lateral resolution, primarily related to the width of the ultrasound beam, is approximately four times that of the axial resolution, meaning the resolution is one-quarter that of the axial resolution. Focusing can slightly improve lateral resolution, but it remains inferior to axial resolution. Lateral resolution is the worst. This is because the transducer units have a fixed lateral focal depth, determined by the acoustic lens. Generally, for image uniformity, the properties of each unit in the same ultrasound transducer array are identical, resulting in a fixed lateral focal depth for the probe as a whole. Beyond this depth, there is significant deviation, typically up to several millimeters, with the deviation increasing with distance from the lateral focal depth. While a few one-dimensional ultrasound probes utilize a multi-row transducer array structure, with each row having a different lateral focal depth, and multiple focusing techniques can be used to improve lateral resolution, the manufacturing process for multi-row transducers is more difficult, and the scanner must provide multiple channels for the multi-row transducer array. This significantly increases the cost and complexity of the system, limiting its application. On the other hand, three-dimensional ultrasound imaging is a new type of ultrasound imaging technology. One of the methods is to use a single-line one-dimensional ultrasound probe, perform multiple two-dimensional scans, and use a positioning system to obtain the position of each two-dimensional scan, and reconstruct multiple two-dimensional images into a three-dimensional ultrasound image. The positioning system is the key to reconstruction accuracy. Commonly used positioning systems include acoustic, electromagnetic, and optical positioning systems, but they are generally expensive. Patent application number CN202110564219.4 discloses a method for correcting manual translation stage errors using visual calibration. The positioning method using a calibration plate is highly accurate and low-cost, and with slight modifications, it can also be used for positioning in three-dimensional ultrasound imaging. Summary of the Invention

[0003] Purpose of the invention: The technical problem to be solved by the present invention is to provide an ultrasonic imaging method with improved transverse and lateral resolution in response to the shortcomings of the existing technology.

[0004] In order to solve the above technical problems, the present invention discloses an ultrasonic imaging method for improving transverse and lateral resolution, comprising the following steps:

[0005] Step 1: Use an ultrasonic sensor to perform a single scan of the target imaging area; control the ultrasonic sensor probe to transmit and collect ultrasonic signals through the ultrasonic collector, and set the transverse focusing depth of the ultrasonic sensor probe to be the same as the lateral focusing depth of the ultrasonic sensor probe, reconstruct a two-dimensional ultrasonic image, and complete a single scan of the target imaging area;

[0006] Step 2: Scan layer by layer and move the ultrasonic sensor probe layer by layer, continuously repeating the single scanning process described in Step 1, and using the positioning system to synchronously obtain the spatial position and angle information of each two-dimensional ultrasonic image until all depth ranges of the target imaging area are overscanned near the lateral focus depth of the ultrasonic sensor probe, thereby obtaining a two-dimensional ultrasonic image sequence and its corresponding spatial position and angle information;

[0007] Step 3: Rotate the probe or imaging target 90 degrees laterally and repeat steps 1 to 2.

[0008] Step 4: Reconstructing the three-dimensional volume data by combining the two-dimensional ultrasound image sequence and its corresponding spatial position and angle information;

[0009] Step 5: extracting cross sections from the three-dimensional ultrasound volume data to obtain ultrasound imaging results with high transverse resolution and high lateral resolution.

[0010] The method of moving the ultrasonic sensor probe in step 2 of the present invention is free hand-held.

[0011] The method for synchronously acquiring the spatial position and angle information of each two-dimensional ultrasound image in step 2 includes:

[0012] Step 2-1, fix a calibration plate on the ultrasonic sensor probe;

[0013] Step 2-2, by calibrating the ultrasonic sensor probe, obtaining the relative spatial position and angle information of the calibration plate and the two-dimensional ultrasonic image;

[0014] Step 2-3: Fix a camera in space and use image detection algorithm to identify the spatial position and angle information of the calibration plate in the camera coordinate system;

[0015] Step 2-4, combining the above two sets of spatial position and angle information, transforming the two-dimensional ultrasound image into the camera coordinate system.

[0016] The ultrasonic sensor described in step 1 of the present invention is a single-row one-dimensional array ultrasonic sensor.

[0017] The image detection algorithm used in step 1 to identify the spatial position and angle information of the calibration plate in the camera coordinate system includes:

[0018] In step 2-3-1, use the ArUco (full name: Augmented Reality Marker, a square feature marker block that can be used for camera calibration and pose estimation, with the advantages of simplicity, speed and robustness) calibration block detection algorithm (reference: S. Garrido-Jurado, R. Mu noz Salinas, FJ Madrid-Cuevas, and MJ Marín-Jiménez. Automatic generation and detection of highly reliable fiducial markers under occlusion. Pattern Recognition, 47(6): 2280–2292, 2014.) to obtain the image coordinates of the four ArUco calibration blocks, and use the rectangle formed by them to crop the original image A taken by the camera to obtain the image B containing the calibration plate;

[0019] Step 2-3-2, use the contour detection algorithm (reference: Satoshi Suzuki, Keiichi Abe, Topological structural analysis of digitized binary images by border following, Computer Vision, Graphics, and Image Processing, Volume 30, Issue 1, 1985, Pages 32-46,) to extract the contours of all circular spots in image B and calculate their respective center coordinates;

[0020] Step 2-3-3, according to the center coordinates of all circular spot contours in step 2-3-2 and the positional relationship between the center coordinates of each circular spot known on the calibration plate, solve the PnP (Perspective N Points) equation (the PnP problem is the process of solving the target posture, i.e., the external parameters, when the coordinates of N points of the target object in three-dimensional space and their corresponding coordinates on the camera image, the camera intrinsic parameter matrix and the distortion coefficient are known. The solution can be solved by the Levenberg-Marquardt iteration method, reference: Marquardt, Donald (1963). "An Algorithm for Least-Squares Estimation of Nonlinear Parameters". SIAM Journal on Applied Mathematics. 11(2): 431–441.), and obtain the rotation matrix and translation vector of the calibration plate relative to the camera.

[0021] The method for reconstructing three-dimensional volume data in step 4 of the present invention includes:

[0022] Step 4-1: Establish a three-dimensional Cartesian coordinate system with the center or a corner of the target imaging area as the origin, and the directions of the X, Y, and Z axes are the same as those of the positioning system described in step 2;

[0023] Step 4-2, continuously allocating voxels within the target imaging area in a three-dimensional Cartesian coordinate system;

[0024] Step 4-3, interpolating the pixel values ​​of several rows close to the lateral focal depth of the probe in each two-dimensional ultrasound image in the two-dimensional ultrasound image sequence into voxels in the three-dimensional reconstruction coordinate system by a weighted average method;

[0025] Step 4-4, perform hole filling to fill the vacancies of some voxels after interpolation.

[0026] The physical size of the voxel described in step 4-2 of the present invention is selected based on the accuracy and speed of three-dimensional reconstruction; the physical size of the voxel is intuitively the resolution of the three-dimensional reconstruction. Assuming that the physical size of the voxel is 1mm, the minimum resolution of the three-dimensional reconstruction is 1mm. Obviously, the smaller the physical size of the voxel, the clearer the reconstructed image, but relatively, the longer the reconstruction time is. The lower limit of the physical size of the voxel is sub-pixel (i.e., 1 / 2 of the physical size of the pixel), and in principle there is no upper limit for N, but if N is too large, the reconstructed three-dimensional image will be too blurry and meaningless. The physical size of the voxel is selected based on the user's requirements for reconstruction accuracy and speed, and the minimum is not less than 1 / 2 of the physical size corresponding to the pixel of the two-dimensional ultrasound image.

[0027] The several rows described in step 4-3 of the present invention, that is, the physical depth range taken in step 2 is converted into the pixel range of the image coordinate system, which is set to n pixels. Then, with the row in the image corresponding to the lateral focus depth as the center, n / 2 rows are taken upward and downward, for a total of n+1 rows.

[0028] The calibration plate in step 2-1 of the present invention is composed of a plurality of circular spots with known radius and arrangement in the middle and four ArUco calibration blocks at the four corners.

[0029] Beneficial effects:

[0030] The present invention proposes an ultrasonic imaging method for improving lateral and transverse resolution. The method uses a common single-row one-dimensional ultrasonic probe, fully utilizes the sound field distribution properties of the ultrasonic sensor, and combines it with three-dimensional ultrasonic imaging technology. By allowing the lateral focal depth of the sensor to cover the imaging target area layer by layer, the lateral resolution is significantly improved without increasing the cost and system complexity. At the same time, by scanning from two orthogonal directions, the axial direction with high resolution is used to compensate for the lateral direction with low resolution, thereby improving the lateral resolution, thereby changing the shortcomings of low lateral and transverse resolution of common ultrasonic imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more apparent.

[0032] Figure 1 It is a schematic flow diagram of the present invention.

[0033] Figure 2 It is a schematic diagram of the direction of the ultrasonic sensor.

[0034] Figure 3 It is a schematic diagram of ultrasound resolution in various directions.

[0035] Figure 4 It is the sound field distribution diagram of the ultrasonic sensor in the lateral direction.

[0036] Figure 5 is a schematic diagram of rotating the probe laterally.

[0037] Figure 6 This is a schematic diagram of the calibration plate style. DETAILED DESCRIPTION

[0038] The present invention provides an ultrasonic imaging method for improving transverse and lateral resolution, which is used to improve the transverse and lateral resolution of common one-dimensional probe ultrasonic imaging. The present invention makes full use of the acoustic field distribution properties of the ultrasonic sensor and effectively improves the transverse and lateral resolution without increasing the cost and system complexity. Figure 2 As shown in the figure, it is a schematic diagram of the definition of each direction of the ultrasonic sensor. The transverse direction refers to the direction in which the sensor array is arranged, the axial direction refers to the direction perpendicular to the sensor plane, and the direction perpendicular to both the transverse and axial directions is called the lateral direction. Figure 3 The figure shows the ultrasonic resolution in each direction. For the axial and lateral directions, the resolution is the minimum distance between two objects that can be distinguished in that direction, while the lateral resolution is the minimum distance between the object and the central axis of the sensor that can distinguish an object that is not in the scanning plane. The present invention discloses an ultrasonic imaging method that improves the lateral and transverse resolutions. Figure 1FIG. 5 is a flow chart of the method of the present invention, which comprises the following steps:

[0039] Step 1: Use a single-row one-dimensional array ultrasonic sensor probe, such as a one-dimensional linear, curved, or phased array sensor probe, control the sensor array to transmit and receive ultrasonic echo signals through an ultrasonic collector, set the transverse focus depth to be the same as the lateral focus depth of the probe, and reconstruct a two-dimensional ultrasonic grayscale image to complete a single scan of the target imaging area;

[0040] Step 2: By hand-held or mechanically moving the probe, the scanning process of step 1 is continuously repeated, and the spatial position and angle information corresponding to each two-dimensional ultrasound image is synchronously acquired until a certain depth range of the target imaging area is overscanned near the lateral focus depth of the ultrasound probe (assuming the lateral focus depth is N mm, the range is [Nx, N+x], where x is usually a value in the millimeter range (1-10 mm)). Figure 4 The figure shows the lateral sound field distribution of the ultrasonic sensor. The depth of the sound field focus is the lateral focus depth of the sensor. For the same single-row one-dimensional ultrasonic sensor, the lateral focus depth is a certain value. The lateral resolution capability gradually decreases from the lateral focus depth to the deep and shallow sides. The depth range of the target area can take different values, such as 20mm. The smaller the range, the stronger the resolution capability, but more scans are required.

[0041] Step 3: Move the probe so that the lateral focus depth of the ultrasound probe is within another depth range of the target imaging area, and repeat step 2 to complete scanning of another depth range of the target imaging area near the lateral focus depth;

[0042] Step 4, repeating step 3 until all depth ranges of the target imaging area are overscanned near the lateral focus depth of the ultrasound probe;

[0043] Step 5: Rotate the probe or imaging target about 90 degrees laterally, as shown in Figure 5 The figure shows a schematic diagram of keeping the imaging target stationary and rotating the probe 90 degrees laterally, so that the axis of the current probe in the imaging target coordinate system is parallel to the transverse direction of the original probe, and repeating the layer-by-layer scanning process of steps 1 to 4;

[0044] Step 6: Based on the two-dimensional ultrasound image sequence and its corresponding spatial position and angle information, a suitable origin is selected, such as the center of the reconstruction area, and a suitable resolution is selected, i.e., the physical size corresponding to the voxel, such as 0.5 mm × 0.5 mm × 0.5 mm for each voxel. A three-dimensional reconstruction coordinate system is established, and the pixel values ​​of several rows in each two-dimensional ultrasound image close to the lateral focal depth of the probe, i.e., the physical depth range obtained in step 2 is converted into a pixel range of the image coordinate system. Assuming that there are n pixels, the pixel values ​​of the row in the image corresponding to the lateral focal depth are taken as the center, and n / 2 rows are taken upward and downward. The pixel values ​​of these n+1 rows are interpolated to the voxels in the three-dimensional reconstruction coordinate system by weighted averaging. Post-processing is then performed, such as hole filling, i.e., a spherical kernel with a certain radius, such as 3 pixels, is selected, and the vacant voxels in the reconstructed volume data are traversed. The voxel is used as the center of the spherical kernel, and the vacant voxel is assigned a value using the average value of the voxel values ​​of the non-vacant voxels in the spherical kernel or the distance weighted value, thereby completing the reconstruction of the three-dimensional volume data.

[0045] Step 7: extracting cross sections from the 3D ultrasound volume data.

[0046] Preferably, the probe is moved in step 2 by hand, and the method for synchronously acquiring the spatial position and angle information of each two-dimensional ultrasound image includes:

[0047] (1) Fix a calibration plate on the ultrasonic probe. The calibration plate consists of a glass substrate and a metal surface, such as Figure 6 The figure shows the surface style of the calibration plate. The fixing method is to design and 3D print a fixture according to the shape of the probe.

[0048] (2) Obtaining the relative spatial position and angle information between the calibration plate and the two-dimensional ultrasound image through ultrasound probe calibration;

[0049] (3) Fix a camera in space and use image detection algorithm to identify the spatial position and angle information of the calibration plate in the camera coordinate system;

[0050] (4) Combining the above two sets of spatial position and angle information, the two-dimensional ultrasound image is transformed into the camera coordinate system.

[0051] The present invention provides a method and concept for improving transverse and lateral resolution in ultrasonic imaging. Numerous methods and approaches exist for implementing this technical solution. The foregoing merely represents a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Any components not specified in this embodiment may be implemented using existing technologies.

Claims

1. A method for ultrasonic imaging with improved transverse and lateral resolution, characterized in that: The following steps are involved: Step 1: Use an ultrasonic sensor to perform a single scan of the target imaging area; control the ultrasonic sensor probe to transmit and collect ultrasonic signals through the ultrasonic collector, and set the transverse focusing depth of the ultrasonic sensor probe to be the same as the lateral focusing depth of the ultrasonic sensor probe, reconstruct a two-dimensional ultrasonic image, and complete a single scan of the target imaging area; Step 2: Scan layer by layer and move the ultrasonic sensor probe layer by layer, continuously repeating the single scanning process described in Step 1, and using the positioning system to synchronously obtain the spatial position and angle information of each two-dimensional ultrasonic image until all depth ranges of the target imaging area are overscanned near the lateral focus depth of the ultrasonic sensor probe, thereby obtaining a two-dimensional ultrasonic image sequence and its corresponding spatial position and angle information; Step 3: Rotate the probe or imaging target 90 degrees laterally and repeat steps 1 to 2. Step 4: Reconstructing the three-dimensional volume data by combining the two-dimensional ultrasound image sequence obtained in step 2 and its corresponding spatial position and angle information; Step 5: extract cross sections from the three-dimensional volume data to obtain ultrasound imaging results with high transverse resolution and high lateral resolution.

2. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 1, characterized in that: The method of moving the ultrasonic sensor probe in step 2 is free-hand holding.

3. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 2, characterized in that: The method for synchronously acquiring the spatial position and angle information of each two-dimensional ultrasound image in step 2 includes: Step 2-1, fix a calibration plate on the ultrasonic sensor probe; Step 2-2, by calibrating the ultrasonic sensor probe, obtaining the relative spatial position and angle information of the calibration plate and the two-dimensional ultrasonic image; Step 2-3: Fix a camera in space and use image detection algorithm to identify the spatial position and angle information of the calibration plate in the camera coordinate system; Step 2-4, combining the above two sets of spatial position and angle information, transforming the two-dimensional ultrasound image into the camera coordinate system.

4. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 3, characterized in that: The ultrasonic sensor described in step 1 is a single-row one-dimensional array ultrasonic sensor.

5. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 4, characterized in that: The image detection algorithm described in steps 2-3 identifies the spatial position and angle information of the calibration plate in the camera coordinate system, including: Step 2-3-1: Use the ArUco calibration block detection algorithm to obtain the image coordinates of the four ArUco calibration blocks, and use the rectangle formed by them to crop the original image A taken by the camera to obtain the image B containing the calibration plate; Step 2-3-2, use the contour detection algorithm to extract the contours of all circular spots in image B and calculate their respective center coordinates; Step 2-3-3, based on the center coordinates of all circular spot outlines in step 2-3-2 and the positional relationship between the center coordinates of each circular spot known on the calibration plate, solve the PnP equation to obtain the rotation matrix and translation vector of the calibration plate relative to the camera.

6. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 5, characterized in that: The method for reconstructing the three-dimensional volume data in step 4 includes: Step 4-1: Establish a three-dimensional Cartesian coordinate system with the center or a corner of the target imaging area as the origin, and the directions of the X, Y, and Z axes are the same as those of the positioning system described in step 2; Step 4-2, continuously allocating voxels within the target imaging area in a three-dimensional Cartesian coordinate system; Step 4-3, interpolating the pixel values ​​of several rows close to the lateral focal depth of the probe in each two-dimensional ultrasound image in the two-dimensional ultrasound image sequence into voxels in the three-dimensional reconstruction coordinate system by a weighted average method; Step 4-4, perform hole filling to fill the vacancies of some voxels after interpolation.

7. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 6, characterized in that: The physical size of the voxel in step 4-2 is selected based on the accuracy and speed of three-dimensional reconstruction.

8. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 7, characterized in that: The number of rows described in step 4-3, that is, the physical depth range obtained in step 2 is converted into the pixel range of the image coordinate system, which is set to n pixels. Then, with the row in the image corresponding to the lateral focus depth as the center, n / 2 rows are taken upward and downward, for a total of n+1 rows.

9. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 8, characterized in that: The calibration plate described in step 2-1 consists of several circular spots with known radius and arrangement in the middle and four ArUco calibration blocks at the four corners.

10. The ultrasonic imaging method for improving transverse and lateral resolution according to claim 9, characterized in that: In step 2-3-3, the PnP equation is solved using the Levenberg-Marquardt iteration method.

Images