Summary of the invention
The technical problem to be solved by the present invention is to determine brain image coronal scan direction, preferably to meet brain
The needs of image scanning.
To solve the above-mentioned problems, according to an aspect of the invention, there is provided a kind of determining brain image Coronal side
To method, comprising the following steps:
According to brain image reconstructed center sagittal view;
Brain stem area is partitioned into the figure of the central sagittal face;
Brain stem back edge direction is determined in the brain stem area;
The direction for taking brain stem back edge direction or being parallel to brain stem back edge direction is the Coronal side
To.
In one embodiment, described to be partitioned into brain stem area in the figure of central sagittal face and include:
Seed point is determined in corresponding brain stem area in the figure of the central sagittal face;
Brain stem area growth is carried out by basic point of the seed point, obtains the brain stem area.
In one embodiment, the determining seed point includes:
Index point is detected in the figure of the central sagittal face;The index point is located at corresponding in the figure of the central sagittal face
In brain stem area or near brain stem area;
The index point is deviated, the position of the seed point is obtained;
Obtain the gray value of the seed point.
In one embodiment, the index point is the tie point of midbrain and pons.
In one embodiment, the detection index point includes: to determine the index point by region template matching.
In one embodiment, the offset index point includes: that the index point is deviated N toward brain stem internal direction
A location of pixels, the value range of the N are 1~12.
In one embodiment, the gray value for obtaining the seed point includes:
K candidate point is chosen in the brain stem area;The candidate point in brain stem internal direction with the index point
It is distributed from the near to the distant;
The gray value of each candidate point is obtained, and is ranked up by gray value size increasing or decreasing;
If K be odd number, choose gray value be ordered as (K+1)/2 candidate point gray value as the seed point
Gray value;
If K is even number, selection gray value is ordered as the gray value of the candidate point of K/2 or (K/2)+1 as the seed
The gray value of point.
In one embodiment, brain stem area growth includes: that will meet the picture of growth conditions around the seed point
Vegetarian refreshments is merged into the same area;The growth conditions are as follows: the difference of the gray value of the pixel and the gray value of the seed point
In first threshold range.
In one embodiment, the first threshold value range is 70 ~ 85.
In one embodiment, it is described be partitioned into brain stem area in the figure of central sagittal face before, further includes: to described
Central sagittal face carries out edge enhancing.
In one embodiment, the edge enhancing is carried out using edge detection operator.
In one embodiment, the edge detection operator include: Robert operator, Sobel operator, Prewitt operator,
Kirsch operator, Laplacian operator, Gaussian-Laplacian operator and/or Canny operator.
In one embodiment, described to determine that brain stem back edge direction includes: in brain stem area
Pixel data in the brain stem area interception brain stem upper half;
Pivot analysis is carried out, brain stem back edge direction is obtained.
In one embodiment, the pixel data of the brain stem upper half are as follows: the m row since the brain stem area top
In n pixel data of every a line on rear side of human body;The m is the positive integer between 20 ~ 30, and the n is between 4 ~ 8
Positive integer.
According to another aspect of the present invention, a kind of device in determining brain image Coronal direction is additionally provided, comprising:
Reconstruction unit, according to brain image reconstructed center sagittal view;
Cutting unit is partitioned into brain stem area in the figure of the central sagittal face;
It determines direction unit, brain stem back edge direction is determined in the brain stem area;
Direction unit is established, the direction for taking brain stem back edge direction or being parallel to brain stem back edge direction is
The Coronal direction.
In one embodiment, the cutting unit includes:
Seed point determination unit, in the figure of the central sagittal face in corresponding brain stem area determine seed point position and
Gray value;
Region growing unit carries out brain stem area growth by basic point of the seed point, obtains the brain stem area.
In one embodiment, the seed point determination unit includes:
Index point detection unit detects index point in the figure of the central sagittal face;The index point is located at the center
In sagittal view in corresponding brain stem area or near brain stem area;
Seed point location obtaining unit deviates the index point, obtains the position of the seed point;
Seed point gray value obtaining unit, obtains the gray value of the seed point.
In one embodiment, described device further include:
Edge enhancement unit before being partitioned into brain stem area in the figure of central sagittal face, carries out the central sagittal face
Edge enhancing.
In one embodiment, the determining direction unit includes:
Interception unit, the pixel data in the brain stem area interception brain stem upper half;
Pivot analysis unit carries out pivot analysis to the pixel data of the brain stem upper half, obtains the brain stem back
Edge direction.
Compared with prior art, technical solution of the present invention has the advantage that
1, the present invention first determines brain stem direction, then determines coronal scan direction based on brain stem direction, it is ensured that this is coronal
Position direction is parallel with brain stem back edge direction, can preferably meet the needs of brain imaging scanning.
2, in optinal plan, the characteristics of gray value is closer in the picture using same tissue, in brain imaging
Seed point is determined in brain stem area in figure, to around seed point with pixel of the gray scale difference value of seed point in threshold range into
Row region growing, the brain stem area being partitioned into are more objective and accurate.
3, in optinal plan, the right along pixel data of brain stem top half is intercepted, pivot analysis is carried out, determines brain stem back
Edge direction eliminates influence of the brain stem lower curved portion to Coronal direction is determined, more objective and accurate compared with the prior art.
4, by first finding the more apparent index point of visual signature in brain stem area, index point is past in optinal plan
Brain stem internal direction adjustment, it is ensured that seed point is located in brain stem area, and is determined by the gray scale I d median of multiple pixels
The gray value of seed point, obtained seed point are more accurate and reasonable.
5, in optinal plan, before determining seed point, edge increase first is carried out to center sagittal view, makes brain stem area
Gray value differences with tissue around have cut off brain stem area and the possible connection of tissue around, have prevented life away from becoming apparent
It is long to overflow, reduce the error of region growing, identified brain stem back edge direction is more accurate.
Specific embodiment
In the following description, numerous specific details are set forth in order to facilitate a full understanding of the present invention.But the present invention can be with
Much it is different from other way described herein to implement, those skilled in the art can be without prejudice to intension of the present invention the case where
Under do similar popularization, therefore the present invention is not limited to the specific embodiments disclosed below.
Secondly, the present invention is described in detail using schematic diagram, when describing the embodiments of the present invention, for purposes of illustration only, institute
Stating schematic diagram is example, should not limit the scope of protection of the invention herein.
In order to solve the technical problem in background technique, the present invention provides a kind of determining brain image Coronal directions
Method.Fig. 5 is the flow diagram for determining the first embodiment of method in brain image Coronal direction.As shown in figure 5, this reality
Apply example the following steps are included:
Step S11 is executed, according to brain image reconstructed center sagittal view.It will be understood by those skilled in the art that existing
A variety of prior arts can realize the reconstruction of central sagittal face figure, such as: gradient declines optimization, center line fitting process etc..It is above-mentioned
Method all can be applied to the present invention, and the present invention is not especially limited this.
Step S12 is executed, is partitioned into brain stem area in the figure of central sagittal face.It specifically, can be by being sweared at the center
Seed point is determined in the figure of shape face in corresponding brain stem area.Brain stem area growth is carried out by basic point of the seed point, obtains institute
State brain stem area.
The seed point can observe by the naked eye determination, or be determined by the detection of Region Matching technology.The present invention is to true
The mode for determining seed point is not especially limited, it is only necessary to ensure that seed point is located at brain stem area inside.
The brain stem area growth includes: that the pixel for meeting growth conditions around the seed point is merged into same area
Domain.This is because each neighbor pixel often has and is closer in a target area (such as: brain stem area) in imaging of medical
Like attribute, such as: intensity, gray level, texture color etc..Accordingly, it is determined that a seed point in target area, region
Growth can gradually be expanded toward surrounding field from seed point according to like attribute, will there is the adjacent picture of like attribute with seed point
Element is merged into seed point region, to obtain a more complete target area (being partitioned into brain stem area).Specifically, originally
The like attribute that embodiment selects is gray value.
Determine whether to determine by the threshold value of setting area growth conditions for the adjacent pixel with like attribute,
That is: when the difference of certain attribute value of adjacent pixel attribute value corresponding with seed point is less than or equal to the threshold value, it is believed that this is adjacent
Pixel and seed point have like attribute, combinable in region growing to arrive same target area.Specifically, region growing condition
Threshold value can be arranged according to different application.The applicant passes through many experiments, obtains the preference threshold of brain stem area growth conditions
Value is 70 ~ 85, it may be assumed that will be merged into where seed point with adjacent pixel of the difference of seed point gray value less than or equal to 70 ~ 85
Region obtains the brain stem area.It is overflowed it is preferred that threshold value 70 ~ 85 not only can avoid growing caused by due to threshold value setting is excessive, but also
Can avoid because threshold value be arranged it is too small due to grow formation brain stem area it is too small, influence the accurate of subsequent determining brain stem back edge direction
Degree.
It will be understood by those skilled in the art that other methods can also be used to be partitioned into brain stem area in the figure of central sagittal face
Domain, such as: connection domain method, template matching method, watershed algorithm etc., the present invention is not especially limited this.
Step S13 is executed, brain stem back edge direction is determined in brain stem area.It specifically, can be by the brain stem area
The pixel data of the domain interception brain stem upper half;Pivot analysis is carried out to the pixel data, obtains brain stem back edge direction.
Since brain stem area and brain stem material object morphologically have a similitude, the right of the upper half along relatively flat,
Lower half is bending part, therefore the present embodiment only intercepts the pixel data progress pivot analysis of the brain stem upper half, determines brain
On the one hand dry back edge direction can avoid interference of the lower half bending part to brain stem back edge direction is determined, on the other hand may be used
Unnecessary data calculation amount is reduced, arithmetic speed is accelerated.Specifically, it counts and shows through many experiments result, from brain stem area
Top starts 20 ~ 30 row pixel of interception, can cover the upper half comprehensively substantially, then intercept in these row pixels every a line near
4 ~ 8 pixels on rear side of person of modern times's body, can cover the region close to the right edge of brain stem of the upper half comprehensively substantially.
By pivot analysis, dimension can be reduced under the premise of not sacrificing accuracy, can effectively be found out in data
Most important element and structure remove noise and redundancy, and by original complex data dimensionality reduction, announcement is hidden in complex data behind
Simple structure.In the present embodiment, pivot analysis is carried out based on the pixel data to the brain stem area upper half, it may be determined that brain stem
Back edge direction.
Step S14 is executed, the direction for taking brain stem back edge direction or being parallel to brain stem back edge direction is Coronal side
To.Specifically, can be based on brain stem back edge direction, with brain stem back edge direction itself or be parallel to brain stem back edge direction and
The multiple parallel directions not waited with brain stem back edge direction distance are Coronal direction.
It will be understood by those skilled in the art that the technical program can determine brain stem back using a variety of data analysing methods
Edge direction, such as: pivot analysis, frontier tracing method or Wavelet Transform etc..Pca method, but this are used in the present embodiment
Invention is not especially limited this.
The present embodiment, which passes through, determines brain stem back edge direction, so that it is determined that brain coronal scan direction, it is ensured that the hat
Shape position direction is parallel with brain stem back edge direction, can preferably meet the needs of brain imaging scanning, preferably observation brain stem shape
Shape improves the correctness and timeliness of brain lesions clinical diagnosis.
Fig. 6 is the flow diagram of the second embodiment of the method in determination brain image Coronal direction of the invention.With
Unlike previous embodiment, the step of determining seed point by index point is had been shown in particular in the present embodiment, and swears to center
Shape face carries out the step of edge increase.As shown in fig. 6, the present embodiment the following steps are included:
Step S21 is executed, according to brain image reconstructed center sagittal view.Concrete mode is detailed in previous embodiment
Illustrate, details are not described herein again.
Step S22 is executed, edge enhancing is carried out to center sagittal plane.Although as previously mentioned, each pixel of same organization internal
Point usually has the like attribute being closer to, but due to organizing the complicated connection between tissue to close when actual scanning imaging
System, its hetero-organization being connected with the tissue also may have like attribute with the tissue in the pixel of junction.These belong to
Its hetero-organization but with the tissue have like attribute pixel, will cause during subsequent region growings growth overflow,
It may cause finally determining brain stem back edge direction and error occur.Specifically, edge detection operator pair is used in the present embodiment
Central sagittal face figure carries out edge enhancing, so that image local intensity is changed most significant operative gap and becomes apparent, cuts off
Brain stem area and tissue around presumable connection.
It will be understood by those skilled in the art that there are many edge detection operators can be used for edge enhancing operation, and such as: base
In the Robert operator of first differential, Sobel operator, Prewitt operator, Kirsch operator etc.;Based on second-order differential
Laplacian operator, Gaussian-Laplacian operator, Canny operator etc..The combination of any of the above-described kind of operator or operator is all
It can be applied to the present invention.
Preferably, the Laplacian operator using the Sobel operator based on first differential, or based on second-order differential.Cause
It is several operators based on first differential all with directional (i.e. horizontal edge, vertical edge or oblique 45 degree of edges etc.), application
People has found after many tests: the case where growth is overflowed occurs mainly in the marginal position of vertical direction, therefore preferably using detection
The Sobel operator of vertical edge.In addition, Laplacian operator can also be used for the edge detection of vertical direction, through examining
Laplacian operator and Sobel operator are in effect without too big difference.And other Second Order Differential Operators are compared with Laplacian
For operator, there are direction is single or the secondary adaptability for the reasons such as calculation amount is larger.
Step S23 is executed, in the figure of central sagittal face, detects index point;The index point is located in the figure of central sagittal face
In corresponding brain stem area or near brain stem area.
Different from a upper embodiment, the present embodiment does not directly determine seed point, but first determines a mark in brain stem area
Point determines seed point according to index point.Since index point often has visual signature outstanding, it is thus determined that index point is compared with kind
Son point is more convenient and accurate.Simultaneously as being necessary to ensure that seed point is located in brain stem area, so the selection of index point also needs
Consider its positional relationship with brain stem area.Specifically, index point can be located in brain stem area, or examine for the precision of images
Consider, is suitably relaxed near brain stem area.It is described nearby to fall in brain stem so that the index point to be deviated to the position after several pixels
It is limited in region.
In the present embodiment, selecting midbrain-pons point of contact (Midbrain-Pons Junction, MPJ) is index point,
The point is located at the top of brain stem left margin recess, and usually naked eyes are i.e. distinguishable, can also be determined by modes such as region template matchings
Midbrain-pons point of contact (MPJ), the present invention are not especially limited the mode of detection index point.In other embodiments,
Other index points may be selected, such as: midbrain top edge, pons-medulla oblongata tie point etc..
Step S24 is executed, drift marker point obtains the position of seed point.By in index point-in this present embodiment
Brain-pons point of contact (MPJ) is located on the boundary of brain stem and surrounding tissue, to ensure that seed point is located in brain stem area, will mark
Will point deviates several location of pixels toward brain stem internal direction, obtains the position of seed point.In view of index point and brain stem area
Positional relationship and image resolution ratio, the location of pixels of offset is advisable with 1~12.
Step S25 is executed, the gray value of seed point is obtained.The gray value of brain stem fringe region respective pixel point in order to prevent
It is too small or excessive, with the comparison standard of influence area growth conditions, the gray average of multiple pixels in brain stem area may be selected,
As the gray value of seed point, to reduce error.Specifically, several candidate points can be chosen in brain stem area.These are candidate
Point is distributed in brain stem internal direction from the near to the distant since the index point.The gray value of each candidate point is obtained, and presses gray scale
Value size increasing or decreasing is ranked up.Take the gray value of the pixel of median as the gray value of seed point.That is: K is chosen
A candidate point, if K be odd number, choose gray value be ordered as (K+1)/2 candidate point gray value as the seed point
Gray value;If K is even number, selection gray value is ordered as the gray value of the candidate point of K/2 or (K/2)+1 as the seed
The gray value of point.
In the present embodiment, 5 candidate seed points are selected in the internal direction of brain stem area.This 5 candidate seed points with
The distance of the MPJ point is respectively 1,3,5,7,9 pixel.Obtain the gray value of 5 candidate seed points and sequence.Selection comes
Gray value of the gray value of 3rd candidate seed point as seed point (Seed).
Step S26 is executed, brain stem area growth is carried out by basic point of seed point, obtains brain stem area.It is raw in the present embodiment
Elongate member threshold value is 75.Concrete mode is described in detail in previous embodiment, and details are not described herein again.
Step S27 is executed, the pixel data of the brain stem upper half is intercepted in brain stem area.In the present embodiment, brain stem is intercepted
Region 6 pixels of every a line on rear side of human body since 25 rows top, for pivot analysis.
Step S28 is executed, pivot analysis is carried out, obtains brain stem back edge direction.Concrete mode is in previous embodiment
It is described in detail, details are not described herein again.
Step S29 is executed, the direction for taking brain stem back edge direction or being parallel to brain stem back edge direction is Coronal side
To.Concrete mode is described in detail in previous embodiment, and details are not described herein again.
In the present embodiment, before determining seed point, first to center sagittal view carry out edge increase, make brain stem area with
The gray value differences of tissue around have cut off brain stem area and the possible connection of tissue around, have prevented growth away from becoming apparent
It overflows, reduces the error of region growing, identified brain stem back edge direction is more accurate.On the other hand, by first finding
The more apparent index point MPJ of visual signature in brain stem area adjusts index point toward brain stem internal direction, it is ensured that seed point
In in brain stem area, and determine by the gray scale I d median of multiple pixels the gray value of seed point, obtained seed point
It is more accurate and reasonable.
Second embodiment is described further with reference to the accompanying drawing.
Fig. 7 ~ Figure 12 is the result schematic diagram in second embodiment of the invention each stage.
Shown in Fig. 7 is the central sagittal face figure obtained after brain image is rebuild in the present embodiment.The central sagittal
Vertical irregular portion is divided into brain stem area 1 in the figure of face.On the top of brain stem left margin recess, the position in scissors mouth shape is
The index point MPJ(midbrain-pons point of contact of the present embodiment, Midbrain-Pons Junction).Pass through area in the present embodiment
Domain template matching determines the index point.
Then, edge enhancing is carried out to central sagittal face figure using Sobel operator.Shown in Fig. 8 is edge enhancing
Central sagittal face figure afterwards.As shown in figure 8, the edge that Fig. 8 deutocerebral region is respectively organized is divided relatively sharp compared to Fig. 7.
Shown in Fig. 9 is candidate seed point and finally determining seed point.As shown in figure 9, the present embodiment selection is in brain stem
On the x of internal direction, y-axis distance marker point MPJ be respectively 1,3,5,7,9 pixel 5 points as candidate seed point, obtain
The gray value and sequence, selection gray value of 5 candidate seed points are ordered as the candidate seed point of the 3rd (i.e. I d median) as true
Fixed seed point seed, it may be assumed that with the index point MPJ toward candidate kind of 5 pixels of brain stem internal blas in x, y-axis
Sub- point.The position of the seed point seed is the position of the candidate seed point, and the gray value of the seed point seed is the candidate
The gray value of seed point.
Then, using seed point seed as basic point, brain stem area growth is carried out, obtains brain stem area.In the present embodiment, it will give birth to
First threshold in elongate member is set as 75, it may be assumed that will be less than or equal to 75 adjacent picture with the difference of seed point seed gray value
Vegetarian refreshments is merged together, and ultimately forms brain stem area.Shown in Fig. 10 is the brain stem area obtained after region growing.
Then, in the brain stem area being partitioned into, the pixel data of the brain stem upper half is intercepted.In the present embodiment, it intercepts
6 pixels of every a line on rear side of human body since 25 rows the top of brain stem area, for pivot analysis.Figure 11 is shown
The brain stem upper half pixel data that intercepts out.
Then, pivot analysis is carried out to 25 × 6 pixel datas of interception, obtains brain stem back edge direction.Such as figure
Rectilinear direction shown in 12 is the brain stem back edge direction determined after pivot analysis.
Finally, determine brain Coronal direction be brain stem back edge direction or with the brain stem back edge side
To parallel direction.
It should be noted that the picture that the present embodiment is based on 256 × 256 is not departing from this hair for other sizes picture
In bright spirit and scope, when can make a little modification and perfect, therefore protection scope of the present invention is when with claims institute
Subject to defining.
It should be noted that by the description of the above specific embodiment, those skilled in the art can be clearly
Solving some or all of the application can realize by software and in conjunction with required general hardware platform.Based on such reason
Solution, substantially the part that contributes to existing technology can body in the form of software products in other words for the technical solution of the application
Reveal and, which may include machine readable Jie of one or more for being stored thereon with machine-executable instruction
Matter, these instructions can make when being executed by one or more machines such as computer, computer network or other electronic equipments
The one or more machine embodiment according to the present invention is obtained to execute operation.Machine readable media may include, but be not limited to, soft
Disk, CD, CD-ROM(compact-disc-read-only memory), magneto-optic disk, ROM(read-only memory), RAM(random access memory),
EPROM(Erasable Programmable Read Only Memory EPROM), EEPROM(electrically erasable programmable read-only memory), magnetic or optical card, sudden strain of a muscle
Deposit or suitable for store machine-executable instruction other kinds of medium/machine readable media.
The application can be used in numerous general or special purpose computing system environments or configuration.Such as: personal computer, service
Device computer, handheld device or portable device, laptop device, multicomputer system, microprocessor-based system, top set
Box, programmable consumer-elcetronics devices, network PC, minicomputer, mainframe computer, including any of the above system or equipment
Distributed computing environment etc..
The application can describe in the general context of computer-executable instructions executed by a computer, such as program
Module.Generally, program module includes routines performing specific tasks or implementing specific abstract data types, programs, objects, group
Part, data structure etc..The application can also be practiced in a distributed computing environment, in these distributed computing environments, by
Task is executed by the connected remote processing devices of communication network.In a distributed computing environment, program module can be with
In the local and remote computer storage media including storage equipment.
The present invention also provides a kind of devices in determining brain image Coronal direction.Figure 13 is to determine that brain image is coronal
The structural schematic diagram of the first embodiment of the device in position direction.As shown in figure 13, the present embodiment includes:
Reconstruction unit U11 rebuilds central sagittal face figure for the brain Scout scan image according to acquisition;
The cutting unit U12 being connected with the reconstruction unit U11, for being partitioned into brain stem in the figure of the central sagittal face
Region;
The determination direction unit U13 being connected with the cutting unit U12, for after the brain stem area always determines brain stem
Edge direction.
Direction unit U14 is established, for taking brain stem back edge direction or being parallel to brain stem back edge direction
Direction be the Coronal direction.
Figure 14 is the structural schematic diagram of the second embodiment of the device in determination brain image Coronal direction of the invention.Such as
Shown in Figure 14, the present embodiment includes:
Reconstruction unit U11 rebuilds central sagittal face figure for the brain Scout scan image according to acquisition;
The edge enhancement unit U11 ' being connected with the reconstruction unit U11, for being partitioned into brain in the figure of central sagittal face
Before dry region, edge enhancing is carried out to the central sagittal face.
The cutting unit U12 includes: the seed point determination unit U121 being connected with the edge enhancement unit U11 ', is used
In the position and the gray value that determine seed point in corresponding brain stem area in the figure of the central sagittal face.And with the seed
Point determination unit U121 connected region growing unit U122, for carrying out brain stem area growth by basic point of the seed point,
Obtain the brain stem area.
The seed point determination unit U121 may further comprise: the index point being connected with the edge enhancement unit U11 '
Detection unit (not shown), for detecting index point in the figure of the central sagittal face;The index point is located at center arrow
In the figure of shape face in corresponding brain stem area or near brain stem area.And the seed point being connected with the index point detection unit
Obtaining unit (not shown) is set, for deviating the index point, obtains the position of the seed point.And with the seed point
The connected seed point gray value obtaining unit (not shown) of obtaining unit is set, for obtaining the gray value of the seed point.
The determining direction unit U13 includes: the interception unit U131 being connected with the region growing unit U122, is used for
Pixel data in the brain stem area interception brain stem upper half.And the pivot analysis list being connected with the interception unit U131
First U132 carries out pivot analysis for the pixel data to the brain stem upper half, obtains brain stem back edge direction.
Direction unit U14 is established with what the pivot analysis unit U132 was connected, for the brain stem area binary map
The pixel data of the middle upper half carries out pivot analysis, obtains brain stem back edge direction.
It should be noted that it will be understood by those skilled in the art that above-mentioned members can be such as: programmable array
Logic (Programmable Array Logic, PAL), Universal Array Logic (Generic Array Logic, GAL), scene
Programmable gate array (Field-Programmable Gate Array, FPGA), Complex Programmable Logic Devices (Complex
Programmable Logic Device, CPLD) etc. one of programmable logic device or a variety of, but the present invention is to this
It is not specifically limited.
The present invention first determines brain stem direction, then determines coronal scan direction based on brain stem direction, it is ensured that the Coronal
Direction is parallel with brain stem back edge direction, can preferably meet the needs of brain imaging scanning.
The characteristics of present invention also uses same tissue gray value is closer in the picture, in brain imaging figure
Brain stem area in determine seed point, carry out area with pixel of the gray scale difference value of seed point in threshold range to around seed point
Domain growth, the brain stem area being partitioned into are more objective and accurate.
The present invention also passes through the right along pixel data of interception brain stem top half, carries out pivot analysis, determines brain stem back
Edge direction eliminates influence of the brain stem lower curved portion to Coronal direction is determined, more objective and accurate compared with the prior art.
The present invention, which also passes through, first finds the more apparent index point of visual signature in brain stem area, by index point in brain stem
Portion direction adjustment, it is ensured that seed point is located in brain stem area, and determines seed point by the gray scale I d median of multiple pixels
Gray value, obtained seed point is more accurate and reasonable.
The present invention first carries out edge increase to center sagittal view before determining seed point, makes brain stem area near
The gray value differences of tissue have cut off brain stem area and the possible connection of tissue around away from becoming apparent, and prevent growth and overflow,
Reduce the error of region growing, identified brain stem back edge direction is more accurate.
Although the invention has been described by way of example and in terms of the preferred embodiments, but it is not for limiting the present invention, any this field
Technical staff without departing from the spirit and scope of the present invention, may be by the methods and technical content of the disclosure above to this hair
Bright technical solution makes possible variation and modification, therefore, anything that does not depart from the technical scheme of the invention, and according to the present invention
Technical spirit any simple modifications, equivalents, and modifications to the above embodiments, belong to technical solution of the present invention
Protection scope.