CN100450445C - Real-time, freedom-arm, three-D ultrasonic imaging system and method therewith - Google Patents

Abstract

The invention relates to the technical field of ultrasonic imaging, in particular to a real-time free-arm three-dimensional ultrasonic imaging system and a method thereof. The system includes: a PC; a B-ultrasound instrument; a video acquisition card; a positioning device; the PC is respectively connected to the B-ultrasonic instrument and the positioning device. The method comprises: (1) the video acquisition card continuously collects the analog video signal output by the B-ultrasound instrument; (2) the PC continuously reads the 6-degree-of-freedom parameter between the position transmitter and the position receiver from the positioning device; 3) The PC matches the corresponding orientation information with the newly acquired 2D B-ultrasound image, and then performs 3D reconstruction based on the 2D image; (4) After the PC performs 3D reconstruction of each 2D image, calculate the whole volume The reconstruction rate and the reconstruction rate increment, and then dynamically update the reconstruction rate, and drive the real-time volume rendering of the scanned reconstructed body according to the reconstruction rate increment. The invention is used for real-time and interactive three-dimensional ultrasonic imaging of the diagnosis part in clinical diagnosis.

Description

A real-time free-arm three-dimensional ultrasound imaging system and method thereof

technical field

The invention relates to the technical field of ultrasonic imaging, in particular to a real-time free-arm three-dimensional ultrasonic imaging system and a method thereof.

Background technique

The method of obtaining three-dimensional ultrasound images is called three-dimensional ultrasound imaging method, which is another leap in the history of medical diagnostic technology since two-dimensional B-ultrasound imaging. The three-dimensional imaging methods currently used in the three-dimensional ultrasound imaging system include direct three-dimensional ultrasound imaging method (one-time imaging using a three-dimensional volume array probe) and reconstruction three-dimensional ultrasound imaging method. Since the hardware of the direct 3D ultrasound imaging system is relatively expensive and is not suitable for 3D ultrasound imaging of large areas, most of the 3D ultrasound imaging systems currently used at home and abroad still use reconstruction 3D ultrasound imaging methods, specifically including 3D ultrasound imaging of manipulators. Three-dimensional ultrasonic imaging, hand-held mechanical scanning and free-arm three-dimensional ultrasonic imaging with a positioning device. Among them, the free-arm three-dimensional ultrasonic imaging method only needs to fix the receiver of a positioning device on the handle end of the traditional two-dimensional B-ultrasound probe to perform three-dimensional scanning imaging of the target body, which can upgrade any conventional two-dimensional ultrasonic instrument into a three-dimensional ultrasound device. Because of its convenience and flexibility, it has become a very active part in the field of 3D ultrasound imaging research and clinical application.

The research on free-arm three-dimensional ultrasound imaging at home and abroad has a history of more than ten years and has achieved certain results. In clinical diagnosis, clinicians need to obtain as detailed information as possible of the diagnosis site in order to obtain more accurate diagnosis results. Therefore, clinicians hope that while scanning, they can know which parts of the diagnostic part have been scanned and which parts need to be scanned, so that they can interactively scan based on this information to obtain a relatively complete three-dimensional image of the diagnostic part. However, the existing free-arm 3D ultrasound imaging system separates scanning, 3D reconstruction and visualization, or reconstructs and displays after collecting all the scanning data, or collects and reconstructs while scanning, and then performs reconstruction after scanning is completed. Display, which cannot provide the necessary interactive information, clinicians can only see the results of scanning imaging after the reconstruction is completed and displayed, but cannot grasp the information of scanning imaging in real time during the scanning process, so that the actual diagnosis cannot be made The need for real-time interactive scanning to obtain a complete three-dimensional image.

Contents of the invention

The real-time free-arm three-dimensional ultrasonic imaging system of the present invention takes a traditional B-ultrasound instrument, a video acquisition card and a positioning device as an interface, takes a real-time free-arm three-dimensional ultrasonic imaging method as the core, and reconstructs the obtained two-dimensional B-ultrasound image in real time It generates 3D ultrasound images and drives real-time volume rendering of scanned reconstructed volumes based on reconstruction rate increments.

According to the present invention, the system is composed of two parts, the first part is a real-time free-arm three-dimensional ultrasonic imaging hardware system; the second part is a real-time free-arm three-dimensional ultrasonic imaging software system based on the first part.

In the present invention, the system continuously collects two-dimensional B-ultrasound images and reads the 6-degree-of-freedom orientation information between the transmitter and the receiver of the positioning device, and matches the corresponding orientation information for the newly collected B-ultrasound images in real time. This 2D image is then subjected to fast real-time 3D reconstruction. After 3D reconstruction of each 2D image, the system will calculate the reconstruction rate and the reconstruction rate increment of the whole volume, then dynamically update and display the reconstruction rate of the whole volume, and drive the real-time volume rendering of the scanned reconstructed volume according to the reconstruction rate increment. Because clinicians can not only qualitatively see the visualization results of scan reconstruction while scanning the diagnostic part, but also quantitatively see the reconstruction rate of the whole body, so that they can scan and reconstruct the diagnostic part according to the needs of actual diagnosis. complete 3D image. And because the system adopts a data-driven three-dimensional reconstruction method, it can combine the characteristics of respiration to overcome the influence of respiration on the scan reconstruction and improve the imaging quality. The invention is used for real-time and interactive three-dimensional ultrasonic imaging of the diagnosis part in clinical diagnosis.

Description of drawings

Fig. 1 is a diagram of the real-time free-arm three-dimensional ultrasonic imaging system of the present invention.

Fig. 2 is a flowchart of the real-time free-arm three-dimensional ultrasonic imaging method of the present invention.

Fig. 3 is a frame diagram of the 3D reconstruction of the real-time free-arm 3D ultrasound imaging system.

Fig. 4 is a mathematical model diagram for coordinate transformation of two-dimensional image pixels adopted in the present invention.

Fig. 5 is a B-ultrasound image matching diagram using a circular queue buffer structure and three processing threads in the present invention.

Fig. 6 is a grid diagram of the reconstructed volume space by the system before scanning and reconstruction in the present invention.

Fig. 7 is a dynamic process diagram of a real-time reconstruction experiment using the present invention.

Detailed ways

According to the present invention, the system structure of the real-time free-arm three-dimensional ultrasonic imaging hardware system is shown in FIG. 1 . The hardware system includes the following parts: The real-time free-arm three-dimensional ultrasonic imaging system takes the real-time free-arm three-dimensional ultrasonic imaging software system as the core, and uses the traditional B-ultrasound instrument, video acquisition card and positioning device as the interface to realize the diagnosis of the diagnosis site in clinical diagnosis. Perform real-time interactive scan imaging.

The real-time free-arm three-dimensional ultrasonic imaging system of the present invention of Fig. 1 comprises:

A PC, the PC has a slot for installing a video capture card and can support the work of the video capture card. Its communication interface adopts USB interface for connecting with the USB interface of the positioning device, or adopts a serial interface for connecting with the serial interface of the positioning device, and can read the 6-freedom between the position transmitter and the position receiver from the positioning device. degree parameter.

A B-ultrasound instrument, which can continuously collect two-dimensional B-ultrasound images of diagnostic parts, and can output the collected two-dimensional B-ultrasound images in the form of analog video signals; The ranging function of two-dimensional B-ultrasound images.

A video capture card, the video capture card is in the PC; it is used to collect analog video signal output, and convert the collected analog video signal into a two-dimensional digital image that can be processed by the PC. The analog video output port of the B-ultrasound instrument is connected, which can continuously collect the analog video signal output by the B-ultrasound instrument, and can send the converted two-dimensional digital image into the memory of the PC.

A positioning device, which has at least a pair of transmitters and receivers, the receiver of the positioning device is fixed on the handle end of the probe of the B-ultrasound instrument, and the transmitter is fixed at a certain position from the diagnosis site. The positioning device can Connected to the USB port or serial interface of the PC through the connecting line, the 6-degree-of-freedom parameters between the transmitter and the receiver can be continuously measured, and sent to the output buffer for the PC to read. The PC is respectively connected to the B-ultrasonic instrument and the positioning device.

The PC has a slot for installing a video capture card and can support the work of the video capture card, and its communication interface adopts a USB interface for connecting with the USB interface of the positioning device, or adopts a serial interface for connecting with the serial interface of the positioning device Connected, and can read the 6-DOF parameters between the position transmitter and the position receiver from the positioning device.

The B-ultrasonic instrument can continuously collect two-dimensional B-ultrasonic images, and can output the collected two-dimensional B-ultrasonic images in the form of analog video signals; Function.

The video acquisition card can be connected with the analog video output port of the B-ultrasonic instrument through the connecting wire, can continuously collect the analog video signal output by the B-ultrasonic instrument, and can send the converted two-dimensional digital image into the memory of the PC .

The receiver of the positioning device is fixed on the holding end of the probe of the B-ultrasound instrument, and the transmitter is fixed at a certain position away from the diagnosis site; the positioning device can be connected to the USB port or the serial interface of the PC through a connecting line , can continuously measure the 6-DOF parameters between the transmitter and the receiver, and send them to the output buffer for PC to read.

Fig. 2 is a flowchart of the real-time free-arm three-dimensional ultrasonic imaging method of the present invention. While scanning, on the one hand, the video acquisition card continuously collects the analog video signal output by the B-ultrasound instrument, and sends the converted two-dimensional digital image into the memory of the computer; on the other hand, the computer continuously reads the locator's Azimuth information between transmitter and receiver, both of which are carried out in parallel. The computer matches the corresponding azimuth information to the newly acquired two-dimensional B-ultrasound image in real time, and performs real-time three-dimensional reconstruction on the two-dimensional image. After each 3D reconstruction of a 2D image is completed, the volume reconstruction rate and the reconstruction rate increment are recalculated, and then the reconstruction rate of the displayed volume is updated, and the re-volume rendering of the scanned and reconstructed volume is driven according to the reconstruction rate increment.

According to the present invention, the real-time free-arm three-dimensional ultrasonic imaging software system is implemented in a PC, and the method flow is shown in FIG. 2 . The free-arm three-dimensional ultrasonic imaging method of the software system includes the following steps:

(1) The video acquisition card continuously collects the analog video signal output by the B-ultrasound instrument, and sends the converted two-dimensional digital image into the memory of the PC;

(2) The PC continuously reads the 6-DOF parameters between the position transmitter and the position receiver from the positioning device;

(3) The PC matches the corresponding orientation information for the newly acquired two-dimensional B-ultrasound image, and then performs three-dimensional reconstruction on the two-dimensional image;

(4) After the PC performs 3D reconstruction of each 2D image, calculate the reconstruction rate and the reconstruction rate increment of the whole body, then dynamically update and display the reconstruction rate of the whole body, and drive the scanning of the reconstructed body according to the reconstruction rate increment Real-time volume rendering.

The described method also includes the following steps:

The three-dimensional reconstruction is performed only when the newly acquired two-dimensional B-ultrasound image is successfully matched with the orientation information, otherwise the two-dimensional B-ultrasound image is discarded.

Also includes the following steps:

The scan can be paused at any time during the scan and rebuild process, the pause will not damage the scanned and reconstructed data, and the scan and rebuild can be continued afterwards.

Using the interpolation method of the front and back azimuth data closest to the acquisition time of the two-dimensional B-ultrasound image to match the upper azimuth information for the latest acquired two-dimensional B-ultrasound image.

Using the data-driven three-dimensional reconstruction method based on the two-dimensional B-ultrasound image, the three-dimensional reconstruction of the two-dimensional image is possible only when the two-dimensional B-ultrasound image is collected.

The clinician selects the diagnostic part less affected by breathing for continuous scanning and imaging to achieve better imaging quality.

According to the present invention, the three-dimensional reconstruction process can generally be divided into two stages: coordinate transformation of two-dimensional image pixels and volume sampling in the reconstructed volume space.

The framework of 3D reconstruction is shown in Fig. 3. Before scanning and reconstruction, the system pre-selects the size of voxels according to needs or according to certain criteria, and grids the reconstructed volume space into a regular voxel field, in which each voxel is a cuboid block with a certain size. During the scanning and reconstruction process, the system obtains the two-dimensional image and the orientation information related to the two-dimensional image, and uses a series of coordinate transformations to transform each pixel of the two-dimensional image into the reconstruction space coordinate system. In the regular voxel field, volume sampling is performed on each pixel in the reconstructed volume space coordinate system, that is, the pixel values of these pixels are assigned to the appropriate voxel in the voxel field according to certain criteria, thus completing the two-dimensional image 3D reconstruction of . As the scanning continues, the system continuously performs three-dimensional reconstruction of two-dimensional images, and finally obtains a regular volume data set, that is, reconstructed volume data.

The mathematical model for coordinate transformation of two-dimensional image pixels adopted in the present invention is shown in FIG. 4 . There are 4 coordinate systems in the figure. P is a coordinate system established on the B-ultrasound scanning plane, and its X-axis and Y-axis are both in the scanning plane, respectively along the horizontal and vertical directions of the two-dimensional scanning image. Because the plane of each two-dimensional B-ultrasound image is consistent with the scanning plane, the Z coordinate value of any pixel in the P coordinate system is zero. R is the coordinate system established on the receiver, and T is the coordinate system established on the transmitter. These two coordinate systems respectively adopt the receiver and the transmitter of the positioning device of the 6-DOF locator. The default coordinate system. C is a coordinate system established on the reconstructed volume space. Generally, the coordinate origin of C is selected at a vertex of the reconstructed volume space, and its three axes are respectively on the three edges passing through this vertex. The coordinate transformation process of the two-dimensional B-ultrasound image pixels from the P coordinate system to the C coordinate system can be referred to the movement path of the white line in Fig. 4 . Its coordinate transformation relationship can be expressed in the form of multiplication of three coordinate transformation matrices: C _X = ^C T _T · ^T T _R · ^R T _P · ^P X, where

^P X=[s _x μs _y v 01] ^T is the homogeneous form of the coordinate values of the two-dimensional image pixels in the P coordinate system. S _x and S _y are the horizontal and vertical intervals of two-dimensional image pixels respectively, which can be obtained according to the ranging function of the B-ultrasound instrument, and v and μ are the row and column values of the two-dimensional array used to store the two-dimensional image respectively Index value, ^R T _P is the coordinate transformation matrix from P coordinate system to R coordinate system, ^T T _R is the coordinate transformation matrix from R coordinate system to T coordinate system, ^C T _T is the coordinate transformation matrix from T coordinate system to C coordinate system , C _X is the homogeneous form of the coordinate value of the transformed pixel in the C coordinate system. ^R T _P , ^T T _R and ^C T _T can be written in a unified matrix form:

$\mathrm{<span\; class="notranslate">\; T</span>}_{\mathrm{<span\; class="notranslate">\; I</span>}}^{\mathrm{<span\; class="notranslate">\; J</span>}}<span\; class="notranslate">\; =</span>\left[\begin{array}{cccc}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}& \mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}& \mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}& \mathrm{<span\; class="notranslate">\; x</span>}\\ \mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}& \mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}& \mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}& \mathrm{<span\; class="notranslate">\; the\; y</span>}\\ <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}& \mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}& \mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&gamma;</span>}& \mathrm{<span\; class="notranslate">\; z</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right]<span\; class="notranslate">\; ,</span>$

Indicates the coordinate transformation matrix from the I coordinate system to the J coordinate system. x, y, z, α, β, γ are 6 degrees of freedom parameters between the two coordinate systems.

From the formula C _X = ^C T _T · ^T T _R · ^R T _P · ^P X, it can be seen that to transform the pixels of a two-dimensional image from the P coordinate system to the C coordinate system, it is necessary to know a series of transformation matrices related to the two-dimensional image. ^R T _P can be obtained through spatial calibration, and ^C T _T is a matrix parameter that is interactively selected according to actual scanning needs, both of which can be obtained in advance. Therefore, it is only necessary to obtain _the ^TTR related to the two-dimensional image to transform the pixels of the two-dimensional image into the C coordinate system. This requires ^TTR to be collected at the same time _as the two-dimensional image is collected. But in fact, this simultaneity cannot be achieved, because the computer cannot collect these two data at the same time.

Fig. 5 is a B-ultrasound image matching diagram using a circular queue buffer structure and three processing threads in the present invention. The location acquisition thread continuously stores the orientation data read from the locator and its time stamp into the circular queue buffer. The matching thread searches the circular queue buffer for the orientation data matching the two-dimensional image in real time according to the time stamp of the latest collected two-dimensional image.

The present invention adopts a circular buffer structure and three processing threads to match the corresponding orientation information on the newly collected two-dimensional B-ultrasound images, and the matching process is shown in FIG. 5 . The length of the circular buffer is N, which is used to circularly store the latest read orientation data and its time stamp. The method of circular storage is as follows:

(1) Open up a data cache with a length of N, and set an index to the cache, denoted as n,

And 0<n<N-1.

(2) Before each additional data is stored, set n=n+1, if n=N, set n=0, and

The data is put into the data cache at the location indexed by n.

The use of circular cache can not only store unknown amount of data in a limited memory space, but also avoid the extra overhead caused by memory transfer, which obviously improves the efficiency in such a system that requires real-time.

The three processing threads of the present invention are respectively a position acquisition thread, a two-dimensional image acquisition thread and a matching thread, and the specific implementation methods are as follows:

(1) The position acquisition thread continuously reads the orientation data between the R coordinate system and the T coordinate system from the locator, and at the same time marks the acquisition time for it, which is recorded as Tp. And store the 6-degree-of-freedom parameters and their time stamps in the circular buffer;

(2) The two-dimensional image acquisition thread continuously acquires the two-dimensional B-ultrasound image, and stores it in the memory of the PC, and marks the acquisition time for it at the same time, which is recorded as Ti;

(3) The matching thread searches in the circular buffer according to the time stamp of the latest collected two-dimensional image in real time. If Tp1<Ti<Tp2, the orientation data corresponding to Tp1 and Tp2 are taken out, and the orientation data matching the two-dimensional image is obtained by linear interpolation. If Ti is greater than the maximum Tp in the circular buffer, immediately add an orientation data read-in, use the orientation data corresponding to the maximum Tp in the circular buffer and the additional read-in orientation data, and linearly interpolate the orientation that matches the two-dimensional image data. If Ti is smaller than the minimum Tp in the circular buffer, the two-dimensional image is discarded and not reconstructed. Since the acquisition time of image data is longer than that of azimuth data, the third case generally does not occur.

According to the present invention, before scanning and reconstruction, the system has gridded the reconstructed volume space into a regular voxel field, each voxel is a cuboid block, and all voxels are equal in size, the regular voxel Fields are represented internally by the computer as three-dimensional arrays. And set the initial value of all voxels to 0. The regular voxel field is shown in Fig. 6, and the regular voxel field is expressed in the form of a three-dimensional array in the PC. From the matching thread, the system obtains the newly acquired two-dimensional B-ultrasound image and the 6-degree-of-freedom parameters between the R and T coordinate systems matched with it. After the 6-DOF parameters are transformed into _{the matrix TTR} ^, the two-dimensional image pixels can be transformed into the C coordinate system by using the formula C _X = ^C T _T · ^T T _R · ^R T _P · ^P X . Volume sampling is performed on each pixel transformed into the C coordinate system, and the reconstruction result of the pixel in the regular voxel field is obtained. In the present invention, the nearest neighbor sampling method is adopted to perform volume sampling on the pixel points. The method is as follows:

(1) Let the position of the pixel p in the C coordinate system be P, and the pixel value be _Vp ;

(2) Set the space area occupied by the voxel v under the C coordinate system as R, and the voxel value is V _v ;

(3) If the pixel p _i is within the range _{of R j ,} and the voxel value of voxel v _j is 0,

and has not been assigned a value, then make V _vj =V _pi .

Fig. 7 is a dynamic process diagram of a real-time reconstruction experiment using the present invention. The object in the picture is the fetal model used in the experiment. During the whole scanning process, the system acquires 2D image data and its orientation information in real time, performs 3D reconstruction calculation in real time, and drives the real-time volume rendering of the scanned reconstructed volume according to the reconstruction rate increment. The figure shows the volume rendering results when 30%, 60%, 85% and 92% of the whole volume scan reconstruction are completed, respectively.

According to the present invention, after performing three-dimensional reconstruction on each two-dimensional image, calculate the reconstruction rate and the reconstruction rate increment of the whole body, then dynamically update and display the reconstruction rate of the whole body, and drive the scanning of the reconstructed body according to the reconstruction rate increment The dynamic visualization process of real-time volume rendering and real-time reconstruction is shown in Figure 7. The specific implementation includes the following steps:

(1) Let the total number of voxels in the regular voxel field be V _total ;

(2) Assuming that in the regular voxel field, the number of currently filled voxels (assigned during the 3D reconstruction process) is V _current , and the initial V _current =0;

(3) Assume that the number of filled voxels in the regular voxel field is V _origin when the reconstructed body is displayed in the last 3D volume, and the initial V _origin =V _current =0;

(4) Set the reconstruction rate increment threshold of the real-time volume rendering of the driving scanning reconstruction volume as R (such as 5%);

(5) During the three-dimensional reconstruction of each two-dimensional image, if a new voxel is filled, then V _current = V _current +1;

(6) After performing three-dimensional reconstruction on each two-dimensional image, calculate the reconstruction rate R _total of the whole body =100*V _current /V _total ;

(7) After 3D reconstruction of each 2D image, calculate the reconstruction rate increment

R _add =100*(V _current -V _origin )/V _total ;

(8) If R _add >R, drive the real-time volume rendering of the scanned reconstructed volume, and set V _origin =V _current .

According to the present invention, clinicians are required to perform real-time and interactive scanning and imaging of diagnostic parts that are less affected by breathing, so as to achieve better imaging quality.

In the present invention, since the data-driven three-dimensional reconstruction method based on two-dimensional images is adopted, the scanning can be suspended at any time during the scanning and reconstruction process, and the pause will not damage the scanned and reconstructed data, and the scanning can be continued afterwards reconstruction. According to the advantages of the three-dimensional reconstruction method, combined with the characteristics of respiration, the influence of respiration on scan reconstruction can be overcome, and the imaging quality can be improved. A method for overcoming the influence of respiration on scan reconstruction includes the following steps:

Specific steps are as follows:

(1) Instruct the patient to take a deep breath and hold his breath;

(2) Start to scan and image the diagnostic site, and stop after a few seconds;

(3) Instruct the patient to take a breath, then take another deep breath and hold it;

(4) Start to scan and image the diagnostic site again, and stop after a few seconds;

(5) The cycle is repeated until the scanning and imaging of the diagnosis site is completed.

Claims (10)

1. A real-time free-arm three-dimensional ultrasonic imaging system, characterized in that it comprises: a PC; A B-ultrasound machine, used to generate two-dimensional B-ultrasound images of the diagnosis site; A video acquisition card, used to collect the analog video signal output of the B-ultrasound instrument, and convert the collected analog video signal into a two-dimensional digital image that can be processed by the PC, and the video acquisition card is in the PC; A positioning device, the device has at least a pair of transmitter and receiver, used to measure the parameters of 6 degrees of freedom between the transmitter and receiver; The PC is connected to the positioning device, and the 6-degree-of-freedom parameters between the transmitter and the receiver are read from the positioning device; the PC is connected to the B-ultrasound instrument, and the two-dimensional B-ultrasound image is collected from the B-ultrasound instrument; The PC matches the corresponding 6-degree-of-freedom parameters to each two-dimensional B-ultrasound image, reconstructs it into a three-dimensional ultrasound image in real time, and displays the reconstruction rate of the scanned and reconstructed volume and the image of the volume rendering in real time. 2. system according to claim 1, it is characterized in that: described PC has the slot that video capture card is installed and can support the work of video capture card, and the communication interface of PC adopts USB interface to be used for and the USB of positioning device The interface is connected, or the serial interface is used to connect with the serial interface of the positioning device, and the 6-degree-of-freedom parameters between the transmitter and the receiver can be read from the positioning device. 3. system according to claim 1, is characterized in that: described B supersonic instrument can gather two-dimensional B supersonic image continuously, and can the two-dimensional B supersonic image that gathers be output with the form of analog video signal; The B-ultrasound instrument has a distance-measuring function for two-dimensional B-ultrasound images. 4. system according to claim 1, is characterized in that: described video acquisition card can link to each other with the analog video output port of B-ultrasonic instrument by connecting wire, can gather the analog video signal of B-ultrasonic instrument output continuously, and The converted two-dimensional digital image can be sent to the memory of the PC. 5. system according to claim 1, is characterized in that: the receiver of described locating device is fixed on the holding end of B-ultrasonic instrument probe, and transmitter is fixed on the place of certain orientation away from diagnosis site; Said locating device It can be connected to the USB port or serial interface of the PC through the connection line, and can continuously measure the 6-degree-of-freedom parameters between the transmitter and the receiver, and send them to the output buffer for the PC to read. 6. A real-time free-arm three-dimensional ultrasonic imaging method, the method comprising the steps of: (1) The video acquisition card continuously collects the analog video signal output by the B-ultrasound instrument, and sends the converted two-dimensional digital image into the memory of the PC; (2) The PC continuously reads the 6-DOF parameters between the position transmitter and the position receiver from the positioning device; (3) The PC matches the corresponding orientation information for the newly acquired two-dimensional B-ultrasound image, and then performs three-dimensional reconstruction on the two-dimensional image; (4) After the PC performs 3D reconstruction of each 2D image, calculate the reconstruction rate and the reconstruction rate increment of the whole body, then dynamically update and display the reconstruction rate of the whole body, and drive the scanning of the reconstructed body according to the reconstruction rate increment Real-time volume rendering. 7. The method of claim 6, further comprising the steps of: The three-dimensional reconstruction is performed only when the newly acquired two-dimensional B-ultrasound image is successfully matched with the orientation information, otherwise the two-dimensional B-ultrasound image is discarded. 8. The method of claim 6, further comprising the steps of: The scan can be paused at any time during the scan and rebuild process, the pause will not damage the scanned and reconstructed data, and the scan and rebuild can be continued afterwards. 9. The method according to claim 6, characterized in that: with the two-dimensional B-ultrasonic image acquisition time closest before and after the method of two orientation data interpolation for the latest two-dimensional B-ultrasonic image acquired to match the upper orientation information. 10. The method according to claim 6, characterized in that: a data-driven three-dimensional reconstruction method based on two-dimensional B-ultrasound images is adopted, and the three-dimensional reconstruction of the two-dimensional images can only be performed if the two-dimensional B-ultrasound images are collected.

Images