CN110251231A - The method and device that ultrasonic three-dimensional is rebuild - Google Patents
The method and device that ultrasonic three-dimensional is rebuild Download PDFInfo
- Publication number
- CN110251231A CN110251231A CN201910509854.5A CN201910509854A CN110251231A CN 110251231 A CN110251231 A CN 110251231A CN 201910509854 A CN201910509854 A CN 201910509854A CN 110251231 A CN110251231 A CN 110251231A
- Authority
- CN
- China
- Prior art keywords
- dimensional
- data
- pixel
- ultrasonic image
- bezier
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
- A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
- A61B34/20—Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
- A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
- A61B2034/101—Computer-aided simulation of surgical operations
- A61B2034/105—Modelling of the patient, e.g. for ligaments or bones
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
- A61B34/20—Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
- A61B2034/2046—Tracking techniques
- A61B2034/2063—Acoustic tracking systems, e.g. using ultrasound
Abstract
The embodiment of the present invention provides a kind of method and device that ultrasonic three-dimensional is rebuild, the described method includes: the information based on two-dimensional ultrasonic image sequence, determine that the dimensions of reconstructed volumetric data, the dimensions include: the physical separation between the coordinate origin of volume data, the dimension size of volume data and voxel；Each pixel in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence is traversed, each pixel in the two-dimensional ultrasound plane is mapped in three-dimensional data to row interpolation processing of going forward side by side using Bezier interpolation algorithm；Empty completion is carried out to the three-dimensional data using kernel regression algorithm.The method and device that ultrasonic three-dimensional provided in an embodiment of the present invention is rebuild, Bezier interpolation algorithm is combined with kernel regression algorithm, can effectively improve the reconstruction quality of three-D ultrasonic.
Description
Technical field
The present invention relates to the Technology of surgery navigation fields based on ultrasonic guidance, more particularly, to a kind of ultrasonic three-dimensional weight
The method and device built.
Background technique
With the rapid development of minimally invasive surgery and accurate medicine, the surgical navigation systems based on image are increasingly becoming outer
Section's operation mainstream, mainly in the preoperative medical image data of patient (CT, MRI, DSA and PET multi-modal data) and art
The specific location of organ-tissue is connected by computer, and is combined with associated surgical instrument, can accurately show organ
The three-dimensional space position of tissue and spatial relation with adjacent vital tissue remain in the case where various complicated
It is accurately positioned, guidance operation safety carries out, and saves the time, and largely improve the self-confidence of patient, can
The generation for avoiding the severe complications such as the damage of important feature improves the success rate of operation.
Ultrasound image relies on its real-time, hurtless measure, without advantages such as ionising radiations, becomes clinical the most commonly used imaging at present
One of equipment, ultrasonic guidance Interventional procedure Clinics are also because its important clinical use value obtains extensive research.It passes
The two-dimensional ultrasonic imaging of system compared to CT scan (CT) and nuclear magnetic resonance (MR), have in real time, facilitate operation,
The advantages such as radiationless and cheap.However two-dimensional ultrasound is a tangent plane picture for showing tissue, effect is inadequate
Intuitively, operator needs to have enough understanding to the anatomical knowledge of human body, is envisioned as where the slice by its experience
Tissue three-dimensional structural information has very strong subjectivity.
Third party's positioning device is mounted on ultrasonic probe by ultrasonic hand-held method for reconstructing, the hand-holdable ultrasound of clinician
Probe observes interested human body, obtains the ultrasound image of any angle and orientation, which meets clinic
The operating habit of doctor, and it is very small to be mounted on locator on ultrasonic probe, is suitble to clinical operation environment, thus this method obtains
Extensive concern and application.Handheld three-dimensional ultrasound method for reconstructing is sparse, interpolation algorithm due to the ultrasound image of acquisition
Quality will directly affect rebuild effect superiority and inferiority.
Summary of the invention
The undesirable problem of effect is rebuild in order to improve when existing handheld three-dimensional ultrasound reconstructing system carries out three-dimensional reconstruction,
The embodiment of the present invention provides a kind of method and device that ultrasonic three-dimensional is rebuild.
In a first aspect, the embodiment of the present invention provides a kind of method that ultrasonic three-dimensional is rebuild, comprising:
Based on the information of two-dimensional ultrasonic image sequence, determine that the dimensions of reconstructed volumetric data, the dimensions include:
Physical separation between the coordinate origin of volume data, the dimension size of volume data and voxel；
Each pixel in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence is traversed, shellfish plug is utilized
Each pixel in the two-dimensional ultrasound plane is mapped in three-dimensional data row interpolation processing of going forward side by side by your interpolation algorithm；
Empty completion is carried out to the three-dimensional data using kernel regression algorithm.
Second aspect, the embodiment of the present invention provide a kind of device that ultrasonic three-dimensional is rebuild, comprising:
Volume data specification determining module determines the ruler of reconstructed volumetric data for the information based on two-dimensional ultrasonic image sequence
Very little specification, the dimensions include: the physical separation between the coordinate origin of volume data, the dimension size of volume data and voxel；
Voxel maps module, it is each in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence for traversing
Each pixel in the two-dimensional ultrasound plane is mapped to three-dimensional data using Bezier interpolation algorithm by a pixel
In go forward side by side row interpolation processing；
Empty completion module, for carrying out empty completion to the three-dimensional data using kernel regression algorithm.
The third aspect, the embodiment of the present invention provides a kind of electronic equipment, including memory, processor and is stored in memory
Computer program that is upper and can running on a processor, is realized when the processor executes described program as first aspect provides
Ultrasonic three-dimensional rebuild method the step of.
Fourth aspect, the embodiment of the present invention provide a kind of non-transient computer readable storage medium, are stored thereon with calculating
Machine program realizes the step for the method that the ultrasonic three-dimensional as provided by first aspect is rebuild when the computer program is executed by processor
Suddenly.
The method and device that ultrasonic three-dimensional provided in an embodiment of the present invention is rebuild calculates Bezier interpolation algorithm and kernel regression
Method combines, and can effectively improve the reconstruction quality of three-D ultrasonic.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the flow diagram for the method that ultrasonic three-dimensional provided in an embodiment of the present invention is rebuild；
Fig. 2 is the structural schematic diagram of ultrasonic hand-held three-dimensional reconstruction system；
Fig. 3 is the schematic illustration of determining reconstructed volumetric data dimensions provided in an embodiment of the present invention；
Fig. 4 is in two-dimensional ultrasound plane corresponding to the traversal two-dimensional ultrasonic image sequence provided in an embodiment of the present invention
Each pixel, each pixel in the two-dimensional ultrasound plane is mapped to three-dimensional using Bezier interpolation algorithm
The flow diagram of the step of row interpolation of going forward side by side in volume data processing；
Fig. 5 is the structural schematic diagram for the device that ultrasonic three-dimensional provided in an embodiment of the present invention is rebuild；
Fig. 6 is the entity structure schematic diagram of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
As shown in Figure 1, the flow diagram for the method rebuild for ultrasonic three-dimensional provided in an embodiment of the present invention, comprising:
Step 100, the information based on two-dimensional ultrasonic image sequence determine the dimensions of reconstructed volumetric data, the size
Specification includes: the physical separation between the coordinate origin of volume data, the dimension size of volume data and voxel；
Specifically, the method that ultrasonic three-dimensional provided by the embodiment of the present invention is rebuild is suitable for ultrasonic hand-held three-dimensional reconstruction
System, as shown in Fig. 2, being the structural schematic diagram of ultrasonic hand-held three-dimensional reconstruction system.By third party's magnetic orientation device (such as Fig. 2
In magnetic orientation tracking equipment) be mounted on ultrasonic probe, the hand-holdable ultrasonic probe of clinician observes interested human body portion
Position, to obtain the two-dimensional ultrasonic image in any angle and orientation, then work station carries out three-dimensional super according to two-dimensional ultrasonic image
It is low voice speaking to build.
The first step of three dimentional reconstruction is to determine the dimensions of reconstructed volumetric data according to two-dimensional ultrasonic image information,
Dimensions specifically includes the physical separation between the coordinate origin of volume data, dimension size and voxel.
Volume data is made of multiple voxels, and voxel is the abbreviation of volume element, and voxel is the minimum unit in segmentation
Wherein, the information based on two-dimensional ultrasonic image sequence, the step of determining the dimensions of reconstructed volumetric data, specifically
Are as follows:
The sector boundary coordinate that every two-dimensional ultrasound is sliced is mapped in real space, by searching for x-axis, y-axis and z-axis
The maximin in three directions determines minimum bounding box；
Using the smallest point of the minimum bounding box as the origin of volume data；
The dimension size of volume data, and the physics being arranged between voxel are obtained by calculating the x-axis, y-axis and z-axis of volume data
Interval.
It is understood that the present invention rapidly and easily determines the dimensions of volume data using bounding box technology.
One bounding box is only by its smallest point (Xmin,Ymin,Zmin) and maximum point (Xmax,Ymax,Zmax) definition.By by every two dimension
The sector boundary coordinate of ultrasonic wafer slicing is mapped in real space, and the maximin for finding three directions of x-axis, y-axis and z-axis comes
Determine minimum bounding box.Once establishing the bounding box of volume, so that it may determine the intrinsic parameter of volume.The origin of volume data is
The smallest point of bounding box.As shown in figure 3, the principle for determining reconstructed volumetric data dimensions provided in an embodiment of the present invention is illustrated
Figure.X-axis, y-axis, the z-axis of volume data are calculated by the vector of point 0 to point 3, point 0 to point 1, point 0 to point 4 respectively.Voxel interval
It oneself can be defined according to demand by operator, voxel interval has decided on whether that finer image can be obtained.
Each pixel in two-dimensional ultrasound plane corresponding to step 101, the traversal two-dimensional ultrasonic image sequence,
Each pixel in the two-dimensional ultrasound plane is mapped in three-dimensional data and is carried out using Bezier interpolation algorithm
Interpolation processing；
Specifically, the second step of three dimentional reconstruction is redistributed to the pixel on two-dimensional surface, that is, traverses two
Each pixel in plane of ultrasound, and the transformation matrix of the location information according to them are tieed up, pixel-map to said three-dimensional body
In data.If more than one pixel drops into the same voxel simultaneously, need according to certain rule, it is suitable to choose
Value, such as average value, maximum value, at first value of (last) arrival etc..
The embodiment of the present invention specifically uses Bezier interpolation algorithm by each pixel in the two-dimensional ultrasound plane
Row interpolation processing of going forward side by side is mapped in three-dimensional data.
As shown in figure 4, each in two-dimensional ultrasound plane corresponding to the traversal two-dimensional ultrasonic image sequence
Each pixel in the two-dimensional ultrasound plane is mapped in three-dimensional data by pixel using Bezier interpolation algorithm
Row interpolation of going forward side by side processing the step of include:
Step 400 makees any one pixel in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence
For current pixel point, the pixel for generating Bezier is controlled using a control window comprising N frame two-dimensional ultrasonic image
Point number；
For example, the embodiment of the present invention may include the control window of four frame two-dimensional ultrasonic images using one to control and generate
The pixel number namely N=4 of Bezier.Assuming that P1~P4Respectively indicate 4 frame two-dimensional ultrasonic images in control window
On pixel, and their corresponding positions in respective image are identical.
Step 401, calculate current pixel point location information transformation matrix, using the transformation matrix by the control
The current pixel point on N frame two-dimensional ultrasonic image in window is transformed into the volume elements of three-dimensional grid, obtains N number of tissue points；
By P1~P4It is transformed into the volume elements of three-dimensional grid, they are V in the corresponding tissue points of three-dimensional space1~V4。
Wherein, the step of calculating the transformation matrix of the location information of current pixel point, specifically:
According to the setting of handheld three-dimensional ultrasonic scanning coordinate system, the position of current pixel point is calculated using following formula
The transformation matrix of information:
In formula,It indicates from world coordinate system W to the transformation of reconstructed volumetric data coordinate system C,It indicates to fill from magnetic orientation
The transformation of coordinate system M to world coordinate system W is set,It indicates from magnetic on ultrasound image plane of scanning motion coordinate system US to ultrasonic probe
The transformation matrix of positioning device coordinate system M, XiIndicate the two-dimensional coordinate of ith pixel point in two dimensional image, YiIt indicates X-Y scheme
Ith pixel point is mapped to the three-dimensional coordinate in three-dimensional space as in.
Wherein, transformation matrixIt is unknown, but can be obtained by spatial calibration method, the i.e. scaling method of ultrasonic probe
, and immobilize once after calibration.Transformation matrixIt can be directly obtained in real time from tracking system, this is one
The matrix of real-time dynamic change.Transformation matrixEffect be to be aligned coordinate system W and coordinate system C, after facilitating reconstruction
The operation and display of volume data.For the unified operation operation convenient for matrix, the embodiment of the present invention uniformly uses 4 × 4 homogeneous change
Change matrix.
Step 402, according to the coordinate of N number of tissue points, construct a location-based Bezier；
It is defined according to Bezier, by the coordinate of 4 tissue points, a location-based Bezier can be constructed
(Bezier) curve, expression formula are as follows:
Step 403, according to the gray value of N number of tissue points, construct a Bezier based on gray value；
Wherein, N is the natural number greater than 1.
Specifically, by the gray value of 4 tissue points, the Bezier based on voxel gray values can be constructed, is had:
V (t)=V1(1-t)3+3V2t(1-t)2+3V3t2(1-t)+V4t3 (3)
Wherein, variable t is equal to interpolation voxel V (t) and first control point V1Offset in the z-axis direction to whole
Origin of curve V1With terminal V4Between offset in the z-axis direction be normalized resulting as a result, having
It similarly, can be in the hope of the value of the variable t in formula (2).
Step 102 carries out empty completion to the three-dimensional data using kernel regression algorithm.
Specifically, since the two-dimensional ultrasonic image sequence of scanning is discontinuous, after interpolation three-dimensional data
Can still there be the region that do not fill up, needs further empty completion and then obtain complete volume data.In the embodiment of the present invention
Using kernel regression algorithm, to be repaired to empty portions.
Described the step of empty completion is carried out to the three-dimensional data using kernel regression algorithm, specifically:
Search needs the volume data repaired, and determines the tissue points of vacancy, finds the voxel neighborhood of a point of the vacancy,
In, neighborhood is the square centered on the tissue points of the vacancy；
The tissue points being not zero are found in the neighborhood, calculate the tissue points that are not zero to the centre of neighbourhood x, y,
The offset in the direction z carries out empty completion using kernel regression algorithm based on the offset.
Volume data after interpolation can indicate are as follows:
Yi=r (Xi)+εi, i=1,2 ..., P (5)
Wherein, r () is regression equation, Xi=(Xi1,Xi2,Xi3)TIt is the three-dimensional coordinate vector of three-dimensional voxel, εiIt is independent
With the mean value noise of distribution.When voxel X is very close to sample point XiWhen, regression equation can be subjected to N rank Taylor expansion:
Following optimization problem is solved using least square method:
Wherein, K () is kernel function, and for punishing the farther away voxel of distance center point, smoothing parameter h (bandwidth) control is punished
The intensity penalized.Particularly, function K is the symmetric function for reaching maximum value at 0, meets following formula:
Wherein, c is a constant value.The selection of the particular form of function K be usually Gaussian form, exponential form or other
Form, these forms all meet above-mentioned constraint condition.Since the selection of core influences very little to the precision of estimation, the present invention
Embodiment chooses the higher Gaussian kernel of computation complexity.
Kernel function K is the function of 3 variables now, provides nonsingular 3 × 3 bandwidth matrices H of positive definite, is defined as follows:
Under normal conditions, each covariant has identical mean value and variance, then uses kernel
Wherein, K is arbitrarily one-dimensional core, therefore only one single bandwidth parameter h.
As target value X=(X1,X2,X3)TWhen, local quadratic sum is as follows:
Wherein,Estimated value isIt is to make to put down
The square the smallest β of weighted sum=(β0,…,β3)TValue.The value to be estimated are as follows:
Wherein,
WxIt is a diagonal matrix, its (i, i) element is Wi(X).So that
Wherein, e1It is a column vector, other than first element is 1, other elements are all 0,It is finally to rebuild
The value of volume data.
The method that ultrasonic three-dimensional provided in an embodiment of the present invention is rebuild, Bezier interpolation algorithm and kernel regression algorithm are mutually tied
It closes, can effectively improve the reconstruction quality of three-D ultrasonic.
Content based on the above embodiment, it is described to traverse two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence
On each pixel, each pixel in the two-dimensional ultrasound plane is mapped to three using Bezier interpolation algorithm
It ties up in volume data, and the step of carrying out empty completion to the three-dimensional data using kernel regression algorithm executes in GPU.
Specifically, it is contemplated that calculation amount is huge during three-dimensional reconstruction, is rebuild using CPU and expends overlong time, and GPU has
There is powerful parallel floating point computing capability, the embodiment of the present invention carries out operation using GPU during three-dimensional reconstruction, can be very big
Shorten the time that ultrasonic three-dimensional is rebuild in ground.
There are a large amount of duplicate calculating processes during the three-dimensional reconstruction of ultrasound, uses the mode efficiency of CPU serial process
It is very low, it is difficult to meet clinical demand.In view of the characteristic that ultrasonic three-dimensional is rebuild uses GPU in the part of interpolation and empty completion
It carries out operation, based on entire work flow GPU, the performance of optimized reconstruction is carried out supplemented by CPU, while guaranteeing precision, shortens and rebuild
Time.
During Bezier interpolation, it is assumed that the size of ultrasound image is M × N, then circulates in every time and create M on GPU
× N number of thread includes four pixels in per thread, it is identical to respectively represent 4 continuous ultrasound images in same control window
Pixel at position.Pixel information include pixel value size and corresponding image coordinate system to world coordinate system rotation
Torque battle array: the former is the voxel value in order to calculate three-dimensional data, and the latter is then the coordinate of thread where calculating the pixel, into
And calculate the location information that pixel is mapped in three-dimensional world coordinate system.
During empty completion, it is assumed that after reconstruction in the effective coverage of volume data, there are K unassigned voxels, then
K thread is created on GPU, includes voxel value interior near 1 unassignable voxel and this voxel neighborhood in per thread
It is not 0 L voxel.The parallel calculating vacancy voxel value of per thread completes the three-dimensional reconstruction of ultrasound finally by empty completion
Process.
The method that ultrasonic three-dimensional provided in an embodiment of the present invention is rebuild carries out carrying out voxel maps and cavity using GPU
Completion operation highly shortened the time of ultrasonic three-dimensional reconstruction, can preferably meet clinical demand.
As shown in figure 5, the structural schematic diagram for the device rebuild for ultrasonic three-dimensional provided in an embodiment of the present invention, comprising: body
Data requirement determining module 501, voxel maps module 502 and empty completion module 503, wherein
Volume data specification determining module 501, for the information based on two-dimensional ultrasonic image sequence, determines reconstructed volumetric data
Dimensions, the dimensions include: between the physics between the coordinate origin of volume data, the dimension size of volume data and voxel
Every；
Voxel maps module 502, for traversing in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence
Each pixel in the two-dimensional ultrasound plane is mapped to said three-dimensional body using Bezier interpolation algorithm by each pixel
In data；
Empty completion module 503, for carrying out empty completion to the three-dimensional data using kernel regression algorithm.
The device that ultrasonic three-dimensional is rebuild is used to execute the method that the ultrasonic three-dimensional in above method embodiment is rebuild.Therefore,
The description and definition in method that the ultrasonic three-dimensional provided by foregoing embodiments is rebuild, can be used in the embodiment of the present invention
The understanding for the device that ultrasonic three-dimensional is rebuild, details are not described herein.
Wherein, the voxel maps module 502 is specifically used for:
Using any one pixel in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence as current picture
Vegetarian refreshments controls the pixel number for generating Bezier using a control window comprising N frame two-dimensional ultrasonic image；
The transformation matrix for calculating the location information of current pixel point, will be in the control window using the transformation matrix
Current pixel point on N frame two-dimensional ultrasonic image is transformed into the volume elements of three-dimensional grid, obtains N number of tissue points；
According to the coordinate of N number of tissue points, a location-based Bezier is constructed；
According to the gray value of N number of tissue points, a Bezier based on gray value is constructed；
Wherein, N is the natural number greater than 1.
The device that ultrasonic three-dimensional provided in an embodiment of the present invention is rebuild, Bezier interpolation algorithm and kernel regression algorithm are mutually tied
It closes, can effectively improve the reconstruction quality of three-D ultrasonic.
Fig. 6 is the entity structure schematic diagram of electronic equipment provided in an embodiment of the present invention, as shown in fig. 6, the electronic equipment
It may include: processor (processor) 610,620, memory communication interface (Communications Interface)
(memory) 630 and communication bus 640, wherein processor 610, communication interface 620, memory 630 pass through communication bus 640
Complete mutual communication.Processor 610 can call the meter that is stored on memory 630 and can run on processor 610
Calculation machine program, the method to execute the reconstruction of ultrasonic three-dimensional provided by above-mentioned each method embodiment, for example, super based on two dimension
The information of acoustic image sequence, determines the dimensions of reconstructed volumetric data, the dimensions include: volume data coordinate origin,
Physical separation between the dimension size and voxel of volume data；It is flat to traverse two-dimensional ultrasound corresponding to the two-dimensional ultrasonic image sequence
Each pixel in the two-dimensional ultrasound plane is mapped to by each pixel on face using Bezier interpolation algorithm
Row interpolation of going forward side by side in three-dimensional data processing；Empty completion is carried out to the three-dimensional data using kernel regression algorithm.
In addition, the logical order in above-mentioned memory 630 can be realized by way of SFU software functional unit and conduct
Independent product when selling or using, can store in a computer readable storage medium.Based on this understanding, originally
The technical solution of the inventive embodiments substantially part of the part that contributes to existing technology or the technical solution in other words
It can be expressed in the form of software products, which is stored in a storage medium, including some instructions
With so that computer equipment (can be personal computer, server or the network equipment an etc.) execution present invention is each
The all or part of the steps of embodiment the method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory
(ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk
Etc. the various media that can store program code.
The embodiment of the present invention also provides a kind of non-transient computer readable storage medium, is stored thereon with computer program,
The computer program realizes the method that the ultrasonic three-dimensional that above-mentioned each method embodiment provides is rebuild when being executed by processor, such as wraps
Include: the information based on two-dimensional ultrasonic image sequence determines that the dimensions of reconstructed volumetric data, the dimensions include: body number
According to coordinate origin, the physical separation between the dimension size of volume data and voxel；It is right to traverse the two-dimensional ultrasonic image sequence institute
Each pixel in two-dimensional ultrasound plane answered, will be each in the two-dimensional ultrasound plane using Bezier interpolation algorithm
A pixel is mapped in three-dimensional data row interpolation processing of going forward side by side；The three-dimensional data is carried out using kernel regression algorithm empty
Hole completion.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case where, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (10)
1. a kind of method that ultrasonic three-dimensional is rebuild characterized by comprising
Based on the information of two-dimensional ultrasonic image sequence, determine that the dimensions of reconstructed volumetric data, the dimensions include: body number
According to coordinate origin, the physical separation between the dimension size of volume data and voxel；
Each pixel in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence is traversed, is inserted using Bezier
Each pixel in the two-dimensional ultrasound plane is mapped in three-dimensional data row interpolation processing of going forward side by side by value-based algorithm；
Empty completion is carried out to the three-dimensional data using kernel regression algorithm.
2. the method according to claim 1, wherein the information based on two-dimensional ultrasonic image sequence, determines
The step of dimensions of reconstructed volumetric data, specifically:
The sector boundary coordinate that every two-dimensional ultrasound is sliced is mapped in real space, search x-axis, y-axis and z-axis three is passed through
The maximin in direction determines minimum bounding box；
Using the smallest point of the minimum bounding box as the origin of volume data；
The dimension size of volume data is obtained by calculating the x-axis, y-axis and z-axis of volume data, and between the physics being arranged between voxel
Every.
3. the method according to claim 1, wherein corresponding to the traversal two-dimensional ultrasonic image sequence
Each pixel in two-dimensional ultrasound plane, using Bezier interpolation algorithm by each picture in the two-dimensional ultrasound plane
Vegetarian refreshments is mapped in three-dimensional data the step of going forward side by side row interpolation processing, specifically:
Using any one pixel in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence as current pixel point,
The pixel number for generating Bezier is controlled using a control window comprising N frame two-dimensional ultrasonic image；
Calculate current pixel point location information transformation matrix, using the transformation matrix by it is described control window in N frame
Current pixel point on two-dimensional ultrasonic image is transformed into the volume elements of three-dimensional grid, obtains N number of tissue points；
According to the coordinate of N number of tissue points, a location-based Bezier is constructed；
According to the gray value of N number of tissue points, a Bezier based on gray value is constructed；
Wherein, N is the natural number greater than 1.
4. according to the method described in claim 3, it is characterized in that, the transformation square of the location information for calculating current pixel point
The step of battle array, specifically:
According to the setting of handheld three-dimensional ultrasonic scanning coordinate system, the location information of current pixel point is calculated using following formula
Transformation matrix:
In formula,It indicates from world coordinate system W to the transformation of reconstructed volumetric data coordinate system C,It indicates to sit from magnetic orientation device
The transformation of mark system M to world coordinate system W,It indicates from ultrasound image plane of scanning motion coordinate system US to magnetic orientation device coordinate system
The transformation matrix of M；XiIndicate the two-dimensional coordinate of ith pixel point in two dimensional image, YiIt indicates ith pixel in two dimensional image
The three-dimensional coordinate that point is mapped in three-dimensional space.
5. the method according to claim 1, wherein it is described using kernel regression algorithm to the three-dimensional data into
The step of row cavity completion, specifically:
Search needs the volume data repaired, and determines the tissue points of vacancy, finds the voxel neighborhood of a point of the vacancy；
The tissue points being not zero are found in the neighborhood, calculate the tissue points that are not zero to the x, y, z side of the centre of neighbourhood
To offset, based on it is described offset using kernel regression algorithm carry out cavity completion.
6. the method according to claim 1, wherein corresponding to the traversal two-dimensional ultrasonic image sequence
Each pixel in two-dimensional ultrasound plane, using Bezier interpolation algorithm by each picture in the two-dimensional ultrasound plane
Vegetarian refreshments is mapped in three-dimensional data, and the step of carrying out empty completion to the three-dimensional data using kernel regression algorithm exists
It is executed in GPU.
7. the device that a kind of ultrasonic three-dimensional is rebuild characterized by comprising
Volume data specification determining module determines the size rule of reconstructed volumetric data for the information based on two-dimensional ultrasonic image sequence
Lattice, the dimensions include: the physical separation between the coordinate origin of volume data, the dimension size of volume data and voxel；
Voxel maps module, for traversing each picture in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence
Each pixel in the two-dimensional ultrasound plane is mapped in three-dimensional data simultaneously by vegetarian refreshments using Bezier interpolation algorithm
Carry out interpolation processing；
Empty completion module, for carrying out empty completion to the three-dimensional data using kernel regression algorithm.
8. device according to claim 7, which is characterized in that the voxel maps module is specifically used for:
Using any one pixel in two-dimensional ultrasound plane corresponding to the two-dimensional ultrasonic image sequence as current pixel point,
The pixel number for generating Bezier is controlled using a control window comprising N frame two-dimensional ultrasonic image；
Calculate current pixel point location information transformation matrix, using the transformation matrix by it is described control window in N frame
Current pixel point on two-dimensional ultrasonic image is transformed into the volume elements of three-dimensional grid, obtains N number of tissue points；
According to the coordinate of N number of tissue points, a location-based Bezier is constructed；
According to the gray value of N number of tissue points, a Bezier based on gray value is constructed；
Wherein, N is the natural number greater than 1.
9. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes the ultrasound three as described in any one of claim 1 to 6 when executing described program
The step of tieing up the method rebuild.
10. a kind of non-transient computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer
The step of method that the ultrasonic three-dimensional as described in any one of claim 1 to 6 is rebuild is realized when program is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910509854.5A CN110251231A (en) | 2019-06-13 | 2019-06-13 | The method and device that ultrasonic three-dimensional is rebuild |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910509854.5A CN110251231A (en) | 2019-06-13 | 2019-06-13 | The method and device that ultrasonic three-dimensional is rebuild |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110251231A true CN110251231A (en) | 2019-09-20 |
Family
ID=67917983
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910509854.5A Pending CN110251231A (en) | 2019-06-13 | 2019-06-13 | The method and device that ultrasonic three-dimensional is rebuild |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110251231A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110779989A (en) * | 2019-11-04 | 2020-02-11 | 北京理工大学 | Material monitoring system and method based on ultrasonic three-dimensional reconstruction |
| CN111950639A (en) * | 2020-08-14 | 2020-11-17 | 四川维思模医疗科技有限公司 | Imaging method for synchronously displaying ultrasonic and tomography anatomical images in real time |
| CN112022345A (en) * | 2020-09-11 | 2020-12-04 | 上海交通大学医学院附属第九人民医院 | Three-dimensional visual preoperative planning method, system and terminal for abdominal wall defect reconstruction |
| CN112102315A (en) * | 2020-11-02 | 2020-12-18 | 腾讯科技（深圳）有限公司 | Medical image processing method, medical image processing device, computer equipment and storage medium |
| CN112345646A (en) * | 2020-10-30 | 2021-02-09 | 安徽理工大学 | Method for reconstructing microdefect in slender rod |
| CN114663410A (en) * | 2022-03-31 | 2022-06-24 | 清华大学 | Heart three-dimensional model generation method, device, equipment and storage medium |
| CN117422813A (en) * | 2023-11-08 | 2024-01-19 | 上海卓昕医疗科技有限公司 | Medical image three-dimensional reconstruction method, device, equipment and medium based on pixels |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103761767A (en) * | 2014-01-02 | 2014-04-30 | 华南理工大学 | Quick three-dimensional ultrasound image reconstruction method based on sparse data |
-
2019
- 2019-06-13 CN CN201910509854.5A patent/CN110251231A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103761767A (en) * | 2014-01-02 | 2014-04-30 | 华南理工大学 | Quick three-dimensional ultrasound image reconstruction method based on sparse data |
Non-Patent Citations (2)
| Title |
|---|
| 范阳阳: "基于光学定位的自由式三维超声成像系统图像重建的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 陈镇平: "基于GPU的实时自由臂三维超声成像", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110779989A (en) * | 2019-11-04 | 2020-02-11 | 北京理工大学 | Material monitoring system and method based on ultrasonic three-dimensional reconstruction |
| CN111950639A (en) * | 2020-08-14 | 2020-11-17 | 四川维思模医疗科技有限公司 | Imaging method for synchronously displaying ultrasonic and tomography anatomical images in real time |
| CN111950639B (en) * | 2020-08-14 | 2024-03-19 | 四川维思模医疗科技有限公司 | Imaging method for synchronously displaying ultrasonic and tomographic anatomical images in real time |
| CN112022345A (en) * | 2020-09-11 | 2020-12-04 | 上海交通大学医学院附属第九人民医院 | Three-dimensional visual preoperative planning method, system and terminal for abdominal wall defect reconstruction |
| CN112022345B (en) * | 2020-09-11 | 2021-09-21 | 上海交通大学医学院附属第九人民医院 | Three-dimensional visual preoperative planning method, system and terminal for abdominal wall defect reconstruction |
| CN112345646A (en) * | 2020-10-30 | 2021-02-09 | 安徽理工大学 | Method for reconstructing microdefect in slender rod |
| CN112345646B (en) * | 2020-10-30 | 2022-03-08 | 安徽理工大学 | Method for reconstructing microdefect in slender rod |
| CN112102315A (en) * | 2020-11-02 | 2020-12-18 | 腾讯科技（深圳）有限公司 | Medical image processing method, medical image processing device, computer equipment and storage medium |
| CN112102315B (en) * | 2020-11-02 | 2021-02-19 | 腾讯科技（深圳）有限公司 | Medical image processing method, medical image processing device, computer equipment and storage medium |
| CN114663410A (en) * | 2022-03-31 | 2022-06-24 | 清华大学 | Heart three-dimensional model generation method, device, equipment and storage medium |
| CN117422813A (en) * | 2023-11-08 | 2024-01-19 | 上海卓昕医疗科技有限公司 | Medical image three-dimensional reconstruction method, device, equipment and medium based on pixels |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110251231A (en) | The method and device that ultrasonic three-dimensional is rebuild | |
| Kainz et al. | Fast volume reconstruction from motion corrupted stacks of 2D slices | |
| US11783451B2 (en) | Systems and methods for reducing colored noise in medical images using deep neural network | |
| JP4512586B2 (en) | Volume measurement in 3D datasets | |
| US9025858B2 (en) | Method and apparatus for automatically generating optimal 2-dimensional medical image from 3-dimensional medical image | |
| US6545678B1 (en) | Methods, systems, and computer program products for generating tissue surfaces from volumetric data thereof using boundary traces | |
| JP4152648B2 (en) | Method for segmenting a 3D image contained in an object | |
| US10380735B2 (en) | Image data segmentation | |
| US20110262015A1 (en) | Image processing apparatus, image processing method, and storage medium | |
| CN109598697A (en) | The determination of two-dimentional mammary gland radiography data group | |
| JP5194138B2 (en) | Image diagnosis support apparatus, operation method thereof, and image diagnosis support program | |
| KR20220032067A (en) | Data processing methods, devices, devices and storage media | |
| Furst et al. | Marching cores: A method for extracting cores from 3D medical images | |
| CN107516314A (en) | The super voxel dividing method of medical image and device | |
| KR20130010732A (en) | Method and apparatus for generating 3d volume panorama based on a plurality of 3d volume images | |
| KR20130016942A (en) | Method and apparatus for generating 3d volume panorama | |
| CN111445479B (en) | Method and device for segmenting region of interest in medical image | |
| Perez-Gonzalez et al. | Deep learning spatial compounding from multiple fetal head ultrasound acquisitions | |
| CA2873918C (en) | Method and system for the three-dimensional reconstruction of structures | |
| CN109255806B (en) | Two-dimensional and three-dimensional image registration method, system, storage medium and registration device for medical images | |
| Jin et al. | A new approach of arc skeletonization for tree-like objects using minimum cost path | |
| CN108064148A (en) | The embryo transfer guided for image in vitro fertilization | |
| CN114299096A (en) | Outline delineation method, device, equipment and storage medium | |
| US20230343026A1 (en) | Method and device for three-dimensional reconstruction of brain structure, and terminal equipment | |
| Gillam et al. | List-mode image reconstruction for positron emission tomography using tetrahedral voxels |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190920 |
|
| RJ01 | Rejection of invention patent application after publication |