Clock synchronization method for ultrasonic image and tracking signal acquisition

Abstract

The invention discloses a clock synchronization method for ultrasonic image and tracking signal acquisition. The method comprises the following steps: constructing an ultrasonic image and tracking signal synchronous acquisition system; synchronous acquisition and storage of ultrasonic images and spatial pose data are carried out; extracting motion features of the ultrasonic image; extracting tracking pose motion features; and delay matching calculation is carried out. According to the invention, the sub-millisecond-level clock synchronization precision is realized, and the consistency and reliability of the ultrasonic image and the spatial pose data are obviously enhanced. Through an accurate time alignment method, a solid foundation is provided for three-dimensional reconstruction, and the high accuracy of data is ensured. The application and development of the three-dimensional ultrasonic technology in the medical field are further promoted. According to the invention, the cost and technical threshold of the system are reduced, so that synchronous acquisition of high-precision ultrasonic images and spatial pose data becomes more popular and easy to realize.

Description

Technical Field

[0001] The invention belongs to the technical field of ultrasonic imaging and data acquisition, and in particular relates to a clock synchronization method for ultrasonic imaging and tracking signal acquisition. Background Art

[0002] Ultrasound imaging is safe, portable, low-cost and highly real-time, and is widely used in medical fields such as imaging diagnosis, biometry and interventional therapy. Traditional two-dimensional ultrasound scanning requires the operating physician to have rich experience and expertise to construct a three-dimensional anatomical structure in his mind, which places high demands on the physician's skills. In contrast, three-dimensional ultrasound imaging provides more detailed spatial information. By collecting corresponding ultrasound images and spatial postures for accurate three-dimensional reconstruction, it not only helps with the precise measurement and repeated biopsy of lesions, but also shows important value in areas such as image fusion and surgical navigation.

[0003] However, in the actual ultrasound acquisition process, the ultrasound device and the posture tracking device usually use different clock systems, which may lead to significant time deviation between the two. This deviation is caused by different sampling frequencies and phase differences, which will cause time alignment errors between the ultrasound image and the posture data, thereby affecting the accuracy of ultrasound 3D reconstruction. Clock synchronization is to obtain the values ​​of multiple physical properties of the system at the same time to ensure accurate time alignment between the ultrasound device and the posture tracking device.

[0004] In order to solve the above problems, existing technical means mostly use timestamp-based methods, that is, considering the inherent delay between the two systems to calculate the optimal frame offset, so as to minimize the synchronization error and achieve the most accurate clock synchronization. Although such methods can achieve clock synchronization to a certain extent, their synchronization accuracy is limited by the sampling frequency, and there is still room for improvement in efficiency.

[0005] Therefore, the present invention proposes an improved clock synchronization method for ultrasound image and tracking signal acquisition. The method aims to improve the data quality and application performance of three-dimensional ultrasound reconstruction, ensure adaptability in different application scenarios, and strive to achieve higher synchronization accuracy and efficiency without increasing the complexity of existing hardware and software. This method will help overcome the limitations of the existing technology and promote the wider application and development of three-dimensional ultrasound reconstruction technology in the medical field. Summary of the invention

[0006] The present invention is proposed to solve the problems existing in the prior art, and its purpose is to provide a clock synchronization method for collecting ultrasonic images and tracking signals.

[0007] The technical solution of the present invention is: a clock synchronization method for ultrasonic image and tracking signal acquisition, comprising the following steps:

[0008] A. Construct a synchronous acquisition system for ultrasonic images and tracking signals;

[0009] B. Perform synchronous acquisition and storage of ultrasonic images and spatial pose data;

[0010] C. Extract the motion features of ultrasonic images;

[0011] D. Extract the motion features of the tracking pose;

[0012] E. Perform delay matching calculation.

[0013] Furthermore, in step A, to construct a synchronous acquisition system for ultrasonic images and tracking signals, the specific process is as follows:

[0014] First, during the acquisition process, make the ultrasonic probe perform periodic and stable motion in the vertical direction in the water tank to ensure the stability and repeatability of the ultrasonic probe motion, so as to obtain high-quality and continuous ultrasonic image sequences;

[0015] Then, fixedly connect a reflective marker on the ultrasonic probe that can be recognized by the binocular optical tracking device, and ensure that the reflective marker is within the recognition range of the binocular optical tracking device;

[0016] Finally, establish communication connections between the ultrasonic probe and the binocular optical tracking device and the image processing workstation respectively to achieve real-time data transmission.

[0017] Furthermore, the synchronous acquisition system for ultrasonic images and tracking signals in step A includes an ultrasonic imaging device, an ultrasonic probe, a binocular pose tracking device, a tracking coordinate frame, a UR3 robotic arm, an acquisition water tank, and an image processing workstation.

[0018] Furthermore, in step B, to perform synchronous acquisition and storage of ultrasonic images and spatial pose data, the specific process is as follows:

[0019] First, the main process is responsible for initializing the acquisition program and sequentially establishing data transmission pipelines from the binocular optical tracking device and the ultrasonic imaging device to the image processing workstation;

[0020] Then, the image processing workstation triggers a synchronous acquisition command through a queue signal;

[0021] Next, two sub-processes run simultaneously, independently obtaining data from different data sources to ensure the real-time and accuracy of the data, and performing classified storage.

[0022] Furthermore, during the data acquisition process, the start and stop commands for controlling the acquisition are uniformly triggered by the main process to ensure the synchronization of the entire acquisition process and the consistency of the acquired data.

[0023] Further, in step C, the motion features of the ultrasonic image are extracted, and the specific process is as follows:

[0024] c1. Read the ultrasonic video file and decode the video frame by frame into an image sequence;

[0025] c2. Convert it into a grayscale image;

[0026] c3. Extract the edge information in the image based on the Canny operator;

[0027] c4. Apply the Hough transform algorithm to detect the line features;

[0028] c5. Screen, sort and remove artifacts;

[0029] c6. Obtain the minimum image feature coordinate value and acquire the corresponding timestamp;

[0030] c7. Obtain the set of motion coordinates that can characterize the change of line features in the video over time.

[0031] Further, in step D, the motion features of the tracking pose are extracted, and the specific process is as follows:

[0032] Establish a world coordinate system, a tracking coordinate system, and the coordinate system of the first frame of the image;

[0033] d1. Read and parse the pose file. After filtering and string splitting, obtain the pose matrices collected in the world coordinate system at all times;

[0034] d2. Obtain the spatial pose transformation matrix from the image coordinate system to the tracking coordinate system through spatial calibration;

[0035] d3. Calculate the pose of the image in the world coordinate system at any moment;

[0036] d4. Calculate the relative pose of the image with respect to the image at the moment of collecting the first frame at any moment;

[0037] d5. Convert all the obtained relative pose matrices into a six-degree-of-freedom representation form;

[0038] d6. Calculate the maximum motion distance of the translation and judge the main motion direction;

[0039] d7. Obtain the set of motion coordinates of this motion component changing over time.

[0040] Further, in step E, the delay matching calculation is performed, which is to plot the motion coordinates of the image features and the spatial motion coordinates of the tracked ultrasonic probe in the same coordinate system, and calculate the optimal delay based on the least squares method.

[0041] Further, the specific process of performing the delay matching calculation in step E is as follows:

[0042] e1. Normalize the motion coordinates of the image features and the spatial motion coordinates of the tracked probe;

[0043] e2. Perform cubic spline interpolation resampling on the spatial motion coordinates of the tracked probe to create a one-dimensional interpolation function object between time and motion coordinates;

[0044] e3. Generate new interpolated time points with a delay;

[0045] e4. Calculate the pose feature coordinates at the new interpolation points to generate new interpolation points corresponding one-to-one to the number of images;

[0046] e5. Calculate the distance error between non-empty corresponding points based on the least squares method;

[0047] e6. Repeat the error calculation at intervals of the unit time within a certain range of the delay duration, and calculate the delay duration corresponding to the minimum error value;

[0048] e7. Compensate the calculated delay time before the sub-process of synchronous acquisition, so that the pose acquisition has an appropriate delay relative to the image acquisition, ensuring that the acquired ultrasonic images match the real spatial pose at that moment;

[0049] e8. Perform interpolation resampling on all probe motion components to obtain six-degree-of-freedom parameters corresponding one-to-one to the ultrasonic images;

[0050] e9. Re-convert the resampled pose data into a pose transformation matrix to accurately restore the spatial position and orientation of the ultrasonic probe at each time point.

[0051] The beneficial effects of the present invention are as follows:

[0052] The present invention achieves a clock synchronization accuracy of sub-millisecond level, significantly enhancing the consistency and reliability of ultrasonic images and spatial pose data. Through a precise time alignment method, the present invention provides a solid foundation for three-dimensional reconstruction, ensuring a high degree of accuracy of the data.

[0053] The process design of the present invention is simple, the operation is convenient, the calculation process is efficient, and it is applicable to the application scenarios of continuous ultrasonic image acquisition and three-dimensional reconstruction. Without increasing the complexity of the existing hardware, the present invention realizes a high degree of synchronization between ultrasonic images and spatial pose data through software algorithm optimization.

[0054] The present invention further promotes the application and development of three-dimensional ultrasonic technology in the medical field. It reduces the cost and technical threshold of the system, making the synchronous acquisition of high-precision ultrasonic images and spatial pose data more popular and easier to achieve. Description of the Drawings

[0055] Figure 1It is a schematic diagram of the clock synchronization acquisition system in the present invention;

[0056] Figure 2 It is a schematic diagram of the synchronization control acquisition process in the present invention;

[0057] Figure 3 It is a schematic diagram of the image feature extraction process in the present invention;

[0058] Figure 4 It is a schematic diagram of the calculation of the relative spatial pose in the present invention. Specific Embodiments

[0059] Hereinafter, the present invention will be described in detail with reference to the accompanying drawings and embodiments:

[0060] As Figures 1 to 4 shown, a clock synchronization method for ultrasonic image and tracking signal acquisition includes the following steps:

[0061] A. Construct a synchronous acquisition system for ultrasonic images and tracking signals;

[0062] B. Perform synchronous acquisition and storage of ultrasonic images and spatial pose data;

[0063] C. Extract the motion features of the ultrasonic image;

[0064] D. Extract the motion features of the tracking pose;

[0065] E. Perform delay matching calculation.

[0066] Step A constructs a synchronous acquisition system for ultrasonic images and tracking signals, and the specific process is as follows:

[0067] First, during the acquisition process, the ultrasonic probe is made to perform periodic and stable motion in the vertical direction in the water tank to ensure the stability and repeatability of the motion of the ultrasonic probe, so as to obtain a high-quality and continuous ultrasonic image sequence;

[0068] Then, a reflective marker that can be recognized by the binocular optical tracking device is fixedly connected to the ultrasonic probe, and it is tested that the reflective marker is within the recognition range of the binocular optical tracking device;

[0069] Finally, the ultrasonic probe and the binocular optical tracking device are respectively connected to the image processing workstation to establish a communication connection to realize real-time data transmission.

[0070] The synchronous acquisition system for ultrasonic images and tracking signals in step A includes an ultrasonic image device, an ultrasonic probe, a binocular pose tracking device, a tracking coordinate frame, a UR3 robotic arm, an acquisition water tank, and an image processing workstation.

[0071] Step B performs synchronous acquisition and storage of ultrasonic images and spatial pose data, and the specific process is as follows:

[0072] First, the main process is responsible for initializing the acquisition program and sequentially establishing data transmission pipelines from the binocular optical tracking device and the ultrasonic imaging device to the image processing workstation;

[0073] Then, the image processing workstation triggers a synchronous acquisition command through a queue signal;

[0074] Subsequently, the two subprocesses run simultaneously, independently obtaining data from different data sources to ensure the real-time and accuracy of the data, and performing classified storage.

[0075] During the data acquisition process, the commands for controlling the start and stop of the acquisition are uniformly triggered by the main process to ensure the synchronization of the entire acquisition process and the consistency of the acquired data.

[0076] Step C extracts the motion features of the ultrasonic image, and the specific process is as follows:

[0077] c1. Read the ultrasonic video file and decode the video frame by frame into an image sequence;

[0078] c2. Convert it into a grayscale image;

[0079] c3. Extract the edge information in the image based on the Canny operator;

[0080] c4. Apply the Hough transform algorithm to detect the line features;

[0081] c5. Screen, sort, and remove artifacts;

[0082] c6. Obtain the minimum image feature coordinate value and acquire the corresponding timestamp;

[0083] c7. Obtain a set of motion coordinates that can characterize the change of the line features in the video over time.

[0084] Step D extracts the motion features of the tracking pose, and the specific process is as follows:

[0085] Establish a world coordinate system, a tracking coordinate system, and the coordinate system of the first frame of the image;

[0086] d1. Read and parse the pose file, and after filtering and string splitting, obtain the pose matrices collected in the world coordinate system at all times;

[0087] d2. Obtain the spatial pose transformation matrix from the image coordinate system to the tracking coordinate system through spatial calibration;

[0088] d3. Calculate the pose of the image in the world coordinate system at any moment;

[0089] d4. Calculate the relative pose of the image with respect to the image at the moment of collecting the first frame at any moment;

[0090] d5. Convert all the obtained relative pose matrices into a six - degree - of - freedom representation form;

[0091] d6. Calculate the maximum movement distance of translation and determine the main movement direction;

[0092] d7. Obtain the set of motion coordinates of this motion component changing with time.

[0093] In step E for delay matching calculation, the motion coordinates of the image features and the spatial motion coordinates of the tracked ultrasound probe are plotted in the same coordinate system, and the optimal delay is calculated based on the least - squares method.

[0094] The specific process of performing delay matching calculation in step E is as follows:

[0095] e1. Normalize the motion coordinates of the image features and the spatial motion coordinates of the tracked probe;

[0096] e2. Perform cubic spline interpolation resampling on the spatial motion coordinates of the tracked probe to create a one - dimensional interpolation function object between time and motion coordinates;

[0097] e3. Generate new interpolated time points after delay;

[0098] e4. Calculate the pose feature coordinates at the new interpolated points to generate new interpolated points corresponding one - to - one with the number of images;

[0099] e5. Calculate the distance error between non - empty corresponding points based on the least - squares method;

[0100] e6. Repeat the error calculation at intervals of a unit time within a certain range of the delay duration, and calculate the delay duration corresponding to the minimum error;

[0101] e7. Compensate the calculated delay time before the subprocess of synchronous acquisition, so that the pose acquisition has an appropriate delay relative to the image acquisition, ensuring that the acquired ultrasound image matches the real - space pose at this moment;

[0102] e8. Perform interpolation resampling on all probe motion components to obtain six - degree - of - freedom parameters corresponding one - to - one with the ultrasound images;

[0103] e9. Re - convert the resampled pose data into a pose transformation matrix to accurately restore the spatial position and orientation of the ultrasound probe at each time point.

[0104] Embodiment 1

[0105] A clock synchronization method for ultrasonic imaging and tracking signal acquisition, comprising the following steps:

[0106] A. Construct a synchronous acquisition system for ultrasonic imaging and tracking signals;

[0107] B. Synchronously collect and store ultrasonic images and spatial pose data;

[0108] C. Extract the motion features of ultrasonic images;

[0109] D. Extract the motion features of the tracking pose;

[0110] E. Perform delay matching calculations.

[0111] Specifically, step A constructs a synchronous acquisition system for ultrasonic images and tracking signals, as follows:

[0112] First, as shown in the appendix Figure 1 The acquisition system includes an ultrasonic imaging device, an ultrasonic probe, a binocular optical pose tracking device, a tracking coordinate frame, a UR3 robotic arm, an acquisition water tank, an image processing workstation, etc.

[0113] Then, during the data acquisition process, the UR3 robotic arm is used to clamp the ultrasonic probe, so that the ultrasonic probe performs periodic and stable motion in the water tank in the vertical direction.

[0114] In the above way, the stability and repeatability of the ultrasonic probe motion are ensured, so as to obtain high-quality and continuous ultrasonic image sequences.

[0115] At the same time, a reflective marker that can be recognized by the binocular optical tracking device is fixedly connected to the ultrasonic probe, and it is ensured that the reflective marker is within the recognition range of the binocular optical tracking device.

[0116] Finally, the ultrasonic imaging device transmits the image data to the image processing workstation through a video capture card. The binocular optical tracking device establishes a communication connection with the image processing workstation through the IP address and port number to realize the real-time transmission of spatial pose data.

[0117] The organic combination of the above devices constitutes a complete synchronous acquisition system, providing a basis for realizing the synchronous acquisition of ultrasonic images and spatial poses.

[0118] The data acquisition method in step B is based on a multi-process framework to realize the synchronous acquisition and storage of ultrasonic images and spatial pose data. The specific process is as follows:

[0119] As shown in the appendix Figure 2 The main process is responsible for initializing the acquisition program and sequentially establishing data transmission pipelines from the binocular optical tracking device and the ultrasonic imaging device to the image processing workstation. After the acquisition system is initialized, the image processing workstation triggers a synchronous acquisition command through a queue signal.

[0120] ① Sub-process 1: Continuously monitor the communication signal. Once a collection command is received, read the ultrasonic image data obtained through the video capture card, crop it to the imaging area (442 * 432 pixels), and save the video file at a sampling frame rate of 30 fps.

[0121] ② Sub-process 2: Continuously monitor the communication signal. Once a collection command is received, obtain the spatial pose data of the coordinate frame collected by the binocular optical tracking system and write it to the pose file in real time.

[0122] The two sub-processes run simultaneously, independently obtaining data from different data sources to ensure the real-time and accuracy of the data and classify and store it. During the data collection process, the start and stop commands for controlling the collection are uniformly triggered by the main process to ensure the synchronization of the entire collection process and the consistency of the collected data.

[0123] Step C extracts the motion features of the ultrasonic image, specifically as follows:

[0124] c1. Read the ultrasonic video file and decode each frame of the video into an image sequence. For each frame image in the video file, extract the position of the feature points.

[0125] c2. Convert the RGB image to a grayscale image to simplify the subsequent processing steps.

[0126] c3. Based on the Canny operator, extract the edge information in the image, set the threshold range to 100 - 200 pixel values, and obtain a binary image containing feature information.

[0127] c4. Apply the Hough transform algorithm to detect line features. Since the resolution of the ultrasonic image is low, the line features may be slightly deformed in the image. Set the distance resolution in the Hough space to 2 pixels and the angle resolution to 1 degree. To retain the main feature lines, set the minimum number of intersection points to 100, and output a list L of the detected line segment information. Each line segment consists of the coordinates of its starting point and ending point.

[0128] c5. During the ultrasonic imaging process, the rubber material at the front end of the transducer may introduce artifacts in the image, which will interfere with the subsequent feature extraction. For all line segments with a vertical coordinate less than 10 in the coordinate values, they should be regarded as artifacts caused by the rubber material and removed from the line segment list.

[0129] c6. Sort and filter the detected lines to obtain the motion features of the line closest to the probe. Set the upper left corner of the image as the coordinate origin, sort the y-axis coordinate points of all line segments, and the minimum value of the obtained vertical coordinates is the feature coordinate y of the i-th image. i 。

[0130] c7. Record the feature coordinate y extracted from all image sequences. i, and obtain the corresponding timestamp t i By performing the above processing and screening on the straight lines in each frame of the image, a set of motion coordinates Y that can characterize the change of the straight line features in the video over time is finally obtained img .

[0131] Step D extracts the tracking pose motion features, specifically as follows:

[0132] Parse the collected pose tracking data to extract the actual motion changes of the ultrasonic probe. As shown in the schematic diagram of the spatial relative pose calculation Figure 4 , where O w -X w Y w Z w is the world coordinate system, O d -X d Y d Z d is the tracking coordinate system, O I -X I Y I Z I is the coordinate system of the first frame of the image

[0133] d1. Read and parse the pose file. After filtering and string splitting, obtain the pose matrices collected in the world coordinate system at all times

[0134] d2. Obtain the spatial pose transformation matrix from the image coordinate system to the tracking coordinate system through the N-line calibration method

[0135] d3. Calculate the pose of the image in the world coordinate system at the initial moment and any subsequent moment j

[0136]

[0137] In the formula, represents the pose of the image in the world coordinate system at the initial j0 moment, represents the pose of the image in the world coordinate system at the j moment

[0138] d4. Calculate the relative pose of the image at any moment with respect to the image collected at the first frame j0 moment. Then, the pose of the j moment with respect to the j0 moment image in the O I coordinate system can be expressed as

[0139]

[0140] d5. Convert all the obtained relative pose matrices into the representation form of six degrees of freedom θ = {t x , t y , tz , r x , r y , r z}, including three translational components t x , t y , t z and three rotational components r x , r y , r z . The specific calculation is as follows:

[0141] Relative pose matrix can be expressed as

[0142]

[0143] It is known that the probe only performs translational motion and does not rotate under the control of the robotic arm. Therefore, only the translational pose change needs to be calculated, and the translational components of the probe can be calculated as

[0144]

[0145] d6. Calculate the maximum movement distance of each translational component, determine the direction of the maximum movement distance as the main movement direction of the ultrasonic probe, and record the set of movement coordinates of this movement component changing with time. The specific calculation is as follows:

[0146] Y pos = getvec(N) (5)

[0147] In the formula, where n represents the collected pose data of the length, and getvec(·) represents obtaining the column of data with the largest eigenvalue variance in the column vector.

[0148] Step E performs delay matching calculation, as follows:

[0149] Plot the image feature movement coordinates Y img and the tracked probe spatial movement coordinates Y pos in the same coordinate system, and calculate the optimal delay based on the least squares method. The specific process is as follows:

[0150] e1. To ensure consistency in scale, normalize the two sets of data Y img and Y pos to obtain the normalized data values and The formula for calculation is as follows:

[0151]

[0152] where min and max represent taking the minimum and maximum values in the array respectively.

[0153] e2. Since and the sampling frequencies of the two groups of data may be inconsistent, resample using cubic spline interpolation. Let the initial value of the delay be β, and create a one-dimensional interpolation function object f within the time range from 0 to (t + β), i.e.:

[0154] Y = f(t) (7)

[0155] where f(·) represents the fitting function between time t and .

[0156] e3. Generate new interpolated time points after the delay

[0157] t k = t min + k·Δx (k = 0, 1,..., n - 1) (8)

[0158] where Δx = (t max - t min ) / (m - 1), m is the number of corresponding Y img , and tmin represents the initial time.

[0159] e4. Calculate the pose feature coordinates at the new interpolation points, and generate new interpolation points corresponding one-to-one to the number of

[0160]

[0161] Specifically, due to the delay between t k and the image timestamp t i , eliminate the i values that exceed the time range of t .

[0162] e5. Calculate the distance error between non-empty corresponding points with a delay of β based on the least squares method:

[0163]

[0164] where l is the number of valid corresponding point pairs, Y i pos represents the normalized coordinate of the y-axis of the probe spatial movement of the i-th corresponding point, and Y i img is the normalized coordinate of the y-axis of the image feature movement of the i-th corresponding point.

[0165] ​e6. Repeat steps (e2 - e5) every 1 ms within the range of -500 ms to +500 ms for the delay duration β, and record the delay duration and the corresponding error. Then, the delay duration corresponding to the minimum error is the desired optimal clock synchronization delay β min .

[0166] e6. To ensure perfect synchronization between the ultrasound image and its corresponding spatial pose data, the calculated delay time β min is compensated before the subprocess of synchronous acquisition.

[0167] Specifically, a precision timer is added before the main program sends instructions to two subroutines respectively, which is used to adjust the trigger timing of pose acquisition to introduce an appropriate delay relative to image acquisition. This measure ensures that each acquired ultrasound image matches the real spatial pose at that moment.

[0168] e7. To obtain the six - degree - of - freedom spatial relative pose corresponding to each frame of the synchronized ultrasound image, for the 6 calculated probe motion components θ i ={t x ,t y ,t z ,r x ,r y ,r z} i Steps e2 - e4 are executed for all of them, and at this time the delay value is β = β min , obtaining 6 sets of motion component sequences consistent with the number of ultrasound video frames. They are combined according to their physical meanings into a new θ i ′={t x ′,t y ′,t z ′,r x ′,r y ′,r z ′} i , and they are in one - to - one correspondence with the ultrasound images.

[0169] e8. By re - converting the resampled pose data into a homogeneous transformation matrix, the spatial position and orientation of the ultrasound probe at each time point can be accurately restored. This precise time - alignment method not only enhances the consistency and reliability of the data but also provides a solid foundation for 3D reconstruction.

[0170] In summary, the clock synchronization method and acquisition system for ultrasonic imaging and tracking signal acquisition of the present invention are convenient to operate, simple to calculate, and the clock synchronization accuracy can reach sub-milliseconds, which is suitable for the application scenario of collecting continuous ultrasonic images for three-dimensional reconstruction. Through the fine compensation of the delay time and the interpolation resampling of the pose data, the present invention realizes the high synchronization of the ultrasonic image and the spatial pose data without increasing the complexity of the existing hardware, thereby ensuring that the acquired data can accurately reflect the actual physical state and further promoting the application and development of three-dimensional ultrasound technology in the medical field.

[0171] The above description of the present invention is merely illustrative and not restrictive. Therefore, the implementation mode of the present invention is not limited to the above specific implementation mode. If those of ordinary skill in the art are inspired by it and make other changes or modifications without departing from the spirit and scope of the present invention as protected by the claims, they all belong to the protection scope of the present invention.

Claims

1. A clock synchronization method for ultrasonic image and tracking signal acquisition, characterized in that: The following steps are involved: A. Construct a synchronous acquisition system for ultrasound images and tracking signals; B. Synchronous acquisition and storage of ultrasound images and spatial posture data; C. Extract motion features of ultrasound images; D. Extract tracking posture motion features; E. Perform delay matching calculations.

2. A clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 1, characterized in that: Step A: Constructing a synchronous acquisition system for ultrasonic images and tracking signals. The specific process is as follows: First, during the acquisition process, the ultrasound probe is made to perform periodic and stable motion in the vertical direction in the water tank to ensure the stability and repeatability of the ultrasound probe motion, thereby obtaining high-quality, continuous ultrasound image sequences; Then, a reflective mark that can be recognized by a binocular optical tracking device is fixedly connected to the ultrasonic probe, and the reflective mark is tested to be within the recognition range of the binocular optical tracking device; Finally, the ultrasound probe and binocular optical tracking device are respectively connected to the image processing workstation to achieve real-time data transmission.

3. The clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 1, characterized in that: The ultrasonic image and tracking signal synchronous acquisition system in step A includes ultrasonic imaging equipment, ultrasonic probe, binocular degree posture tracking equipment, tracking coordinate frame, UR3 robotic arm, acquisition tank, and image processing workstation.

4. The clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 1, characterized in that: Step B performs synchronous acquisition and storage of ultrasound images and spatial posture data. The specific process is as follows: First, the main process is responsible for initializing the acquisition program and establishing the data transmission pipeline from the binocular optical tracking device and the ultrasonic imaging device to the image processing workstation in turn; Then, the image processing workstation triggers the synchronous acquisition command through the queue signal; Afterwards, the two sub-processes run simultaneously, each independently obtaining data from different data sources to ensure the real-time and accuracy of the data, and store them in a classified manner.

5. A clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 4, characterized in that: During the data collection process, the start and stop commands of the control collection are uniformly triggered by the main process to ensure the synchronization of the entire collection process and the consistency of the collected data.

6. The clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 1, characterized in that: Step C extracts the motion features of the ultrasound image. The specific process is as follows: c1. Read the ultrasound video file and decode the video frame by frame into an image sequence; c2. Convert to grayscale image; c3. Extract edge information in the image based on the Canny operator; c4. Apply Hough transform algorithm to detect straight line features; c5. Screening, sorting and removing artifacts; c6. Get the minimum image feature coordinate value and the corresponding timestamp; c7. Obtain a set of motion coordinates that can characterize the temporal changes of straight line features in the video.

7. The clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 1, characterized in that: Step D extracts the tracking posture motion features. The specific process is as follows: Establish the world coordinate system, tracking coordinate system, and coordinate system of the first frame image; d1. Read and parse the pose file, and obtain the pose matrix collected in the world coordinate system at all times through filtering and string segmentation; d2. Obtain the spatial pose transformation matrix from the image coordinate system to the tracking coordinate system through spatial calibration; d3. Calculate the position and posture of the image in the world coordinate system at any moment; d4. Calculate the relative position of the image at any moment relative to the image at the time of collecting the first frame; d5. Convert all obtained relative pose matrices into a six-degree-of-freedom representation; d6. Calculate the maximum movement distance of translation and determine the main movement direction; d7. Obtain a set of motion coordinates of the motion component that changes with time.

8. The clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 1, characterized in that: Step E performs delay matching calculation, which is to plot the image feature motion coordinates and the tracked ultrasound probe spatial motion coordinates in the same coordinate system, and calculate the optimal delay based on the least squares method.

9. The clock synchronization method for ultrasonic image and tracking signal acquisition according to claim 1, characterized in that: Step E performs delay matching calculation, and the specific process is as follows: e1. Normalize the image feature motion coordinates and the tracked probe space motion coordinates; e2. Perform cubic spline interpolation resampling on the tracked probe space motion coordinates, and create a one-dimensional interpolation function object between time and motion coordinates; e3. Generate a new interpolation time point after the delay; e4. Calculate the pose feature coordinates under the new interpolation point and generate new interpolation points corresponding to the number of images; e5. Calculate the distance error between non-empty corresponding points based on the least squares method; e6. Repeat the calculation of the error at each unit time interval within a certain range of the delay time, and calculate the delay time corresponding to the minimum error value; e7. Compensate the calculated delay time to the subprocess of synchronous acquisition, so that the posture acquisition has an appropriate delay relative to the image acquisition, ensuring that the acquired ultrasound image matches the real spatial posture at that moment; e8. Interpolation resampling is performed on all probe motion components to obtain six-degree-of-freedom parameters corresponding to the ultrasound image; e9. Convert the resampled pose data back into a pose transformation matrix to accurately restore the spatial position and pose of the ultrasound probe at each time point.

Images