CN106023144B - Divide the method for femur in fault image - Google Patents
Divide the method for femur in fault image Download PDFInfo
- Publication number
- CN106023144B CN106023144B CN201610297197.9A CN201610297197A CN106023144B CN 106023144 B CN106023144 B CN 106023144B CN 201610297197 A CN201610297197 A CN 201610297197A CN 106023144 B CN106023144 B CN 106023144B
- Authority
- CN
- China
- Prior art keywords
- region
- femoral head
- femur
- joint space
- concentric circles
- 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.)
- Expired - Fee Related
Links
- 210000000689 upper leg Anatomy 0.000 title claims abstract description 53
- 238000000034 method Methods 0.000 title claims abstract description 43
- 210000000988 bone and bone Anatomy 0.000 claims abstract description 57
- 238000003325 tomography Methods 0.000 claims abstract description 49
- 210000004394 hip joint Anatomy 0.000 claims abstract description 8
- 230000011218 segmentation Effects 0.000 claims description 13
- 210000002391 femur head Anatomy 0.000 claims description 12
- 210000001981 hip bone Anatomy 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 6
- 239000012141 concentrate Substances 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims 1
- 206010003246 arthritis Diseases 0.000 abstract description 3
- 238000003745 diagnosis Methods 0.000 abstract description 3
- 238000011540 hip replacement Methods 0.000 abstract description 3
- 210000005067 joint tissue Anatomy 0.000 abstract description 3
- 238000004088 simulation Methods 0.000 abstract description 3
- 238000001356 surgical procedure Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 6
- 230000011731 head segmentation Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 210000004872 soft tissue Anatomy 0.000 description 3
- 238000000638 solvent extraction Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 210000001624 hip Anatomy 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 210000001519 tissue Anatomy 0.000 description 2
- 210000000845 cartilage Anatomy 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007850 degeneration Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 210000000527 greater trochanter Anatomy 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 210000000528 lesser trochanter Anatomy 0.000 description 1
- 210000003041 ligament Anatomy 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 239000004570 mortar (masonry) Substances 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The present invention provides a kind of in fault image divides the method for femur, comprising: obtain include hip joint tomography picture；Area of bone tissue is extracted from the tomography picture；Mark the joint space region between femoral head and acetabular bone；Calculate the reconciliation field isopleth in the joint space region；Femoral head is partitioned into according to reconciliation field isopleth.By the way that joint space region is marked, obtain rough joint space regional scope, reconciliation field isopleth is calculated to joint space region again, obtain exact joint space zone boundary, the boundary of femoral head is thus also obtained, in turn femoral head efficiently, accurately divide, and high degree of automation, be had great significance for subsequent applications such as auxiliary diagnosis, the hip replacement surgery simulations of joint tissue wear assessment, coxitis.
Description
Technical field
The present invention relates to field of image processing more particularly to a kind of methods for dividing femur in fault image.
Background technique
Hip joint is one of important joint of human body, it mainly by a spherical femoral head, the acetabular bone of nest shape and
Ligament, cartilage between the two is formed；And femur is in addition to the femoral head said including front, further includes connecting with femoral head
Large and small rotor, its hetero-organization such as neck of femur.On the other hand, computerized tomography image technology, that is, the CT technology being commonly called as, it can
Show the sightless organizational information of inside of human body naked eyes through skin histology, therefore as a kind of computer aided technique, CT skill
Art is widely used in clinical treatment；Such as auxiliary diagnosis, the hip replacement hand of joint tissue wear assessment, coxitis
Art simulation etc..
Tomography picture can be understood as scanning a series of equidistant tissues cross caused by specific organization using CT technology
Section picture, region intensity value with higher corresponding to middle-high density tissue are on the contrary then intensity value is smaller.Hip is closed
For tomography picture at section, since the intensity value at bone tissue is higher, the intensity value at soft tissue is lower, therefore, if by force
The contrast of angle value is good, then can be easily disengaged femur with hipbone using the methods of general Threshold segmentation and come.So
And in practice, due to degeneration, sick and wounded, acquisition equipment precision of skeleton etc., the contrast of above-mentioned intensity value is logical
Perfect condition is often not achieved.In other words, the joint space region between the femoral head and acetabular bone that tomography picture is shown, usually
There are interference, and in some instances, which can even disappear, so that conventional method can not accurately obtain the side of femoral head
Boundary's information.In order to obtain the exact boundry of femoral head, it is usually necessary to use the modes of manual interaction successively to tomography for existing method
Partitioning boundary in picture is edited.This mode will expend a large amount of human cost, and have one for the experience of operator
Fixed requirement.
Therefore, it how to be accurately and efficiently partitioned into femoral head from computerized tomography image it is one and urgently to be solved asks
Topic.
Summary of the invention
A kind of divide femur based on reconciling field in fault image the technical problems to be solved by the present invention are: providing
Method can accurately and efficiently be partitioned into femoral head.
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention are as follows:
A method of dividing femur in fault image, comprising:
Obtain the tomography picture comprising hip joint；
Area of bone tissue is extracted from the tomography picture；
Mark the joint space region between femoral head and acetabular bone；
Calculate the reconciliation field isopleth in the joint space region；
Femoral head is partitioned into according to reconciliation field isopleth.
The beneficial effects of the present invention are: femoral head, acetabular region are obtained by extracting area of bone tissue, and to femoral head
Joint space region between acetabular bone is marked, and obtains rough joint space regional scope, further according between label joint
Tomography picture behind gap region calculates the reconciliation field isopleth in the joint space region, obtains exact femoral head boundary, in turn
Femoral head efficiently, accurately divide, it is inaccurate to efficiently solve the unobvious caused femoral head segmentation in joint space region
True problem.
Detailed description of the invention
Fig. 1 is the flow chart of the method for dividing femur in fault image of the embodiment of the present invention；
Fig. 2 is the flow chart of the method for dividing femur in fault image of the embodiment of the present invention one；
Fig. 3 is that the method for dividing femur in fault image of the embodiment of the present invention one extracts area of bone tissue schematic diagram；
Fig. 4 is that the method for dividing femur in fault image of the embodiment of the present invention one marks the signal in joint space region
Figure；
Fig. 5 is the schematic diagram one of the method modifying gradient value for dividing femur in fault image of the embodiment of the present invention one；
Fig. 6 is the schematic diagram two of the method modifying gradient value for dividing femur in fault image of the embodiment of the present invention one；
Fig. 7 is the schematic diagram of the method reconciliation field isopleth for dividing femur in fault image of the embodiment of the present invention one；
Fig. 8 is the schematic diagram on the method femoral head boundary for dividing femur in fault image of the embodiment of the present invention one；
Fig. 9 is the schematic diagram of the method femur segmentation result for dividing femur in fault image of the embodiment of the present invention one.
Label declaration:
1, region one；2, region two；3, region three；4, region four；5, region five；6, region six；7, region seven；8, region
Eight；9, region nine；10, region ten；11, region 11.
Specific embodiment
To explain the technical content, the achieved purpose and the effect of the present invention in detail, below in conjunction with embodiment and cooperate attached
Figure is explained.
The most critical design of the present invention is: the joint space region between femoral head and acetabular bone is marked, and right
Joint space region after label carries out reconciliation field equivalence line computation, goes out femur according to field isopleth is reconciled as boundary segmentation
Head.
Explanation of technical terms of the present invention:
Fig. 1 is please referred to, the present invention provides a kind of method for dividing femur in fault image, comprising:
S1, the tomography picture comprising hip joint is obtained；
S2, area of bone tissue is extracted from the tomography picture；
Joint space region between S3, label femoral head and acetabular bone；
S4, the reconciliation field isopleth for calculating the joint space region；
S5, femoral head is partitioned into according to reconciliation field isopleth.
As can be seen from the above description, the beneficial effects of the present invention are: by the way that joint space region is marked, obtain big
The joint space regional scope of cause obtains exact joint space zone boundary in all tomography pictures by matching algorithm, then
Reconciliation field computation is carried out to joint space region, obtains its isopleth, as the boundary of femoral head, and then femoral head is carried out high
Effect, accurately segmentation efficiently solve the problems, such as the unobvious caused femoral head segmentation inaccuracy in joint space region, reduce
User's interaction times improve the intelligent level of dividing method.
Further, after described " being partitioned into femoral head according to reconciliation field isopleth " further include:
S6, the region in addition to femoral head is partitioned into femur using region growing methods；
The femoral head and the region in addition to femoral head that S7, combination and segmentation go out obtain femur.
As can be seen from the above description, after the completion of femoral head segmentation, then it is partitioned into the region other than the femoral head, front is divided
The region merging technique other than femoral head and above-mentioned femoral head out to get arrive femur.
Further, described " extracting area of bone tissue from the tomography picture " specifically:
The largest connected region that will there are intensity values in the tomography picture using the thresholding method based on histogram
It extracts, obtains binary map of the area of bone tissue being connected using femur with hipbone as prospect, other regions for background.
As can be seen from the above description, the joint space area since femur and hipbone belong to bone tissue, between femoral head and hipbone
There are soft tissues in domain, and the intensity value difference of bone tissue and soft tissue in tomography picture is larger, therefore to realize femoral head
Segmentation, the identification in joint space region and using being the key that one of to solve the problems, such as；Obtain the bone being connected with femur with hipbone
Tissue regions are prospect, the binary map that other regions are background, that is, complete the extraction of area of bone tissue.
Further, the tomography picture is two or more, and every tension fault picture all has two-value corresponding thereto
Figure, it is described " the joint space region between label femoral head and acetabular bone " specifically:
Two initial concentric circles are drawn, wherein inside and outside circle is located at the boundary of femoral head and its surrounding acetabular bone；
It since layer where initial concentric circles, is successively identified, obtains all existing simultaneously the disconnected of femoral head and acetabular bone
Layer picture, and mark corresponding joint space region.
Further, described " binary map to be identified, all tomographies for existing simultaneously femoral head and acetabular bone are obtained
Picture, and mark corresponding joint space region " specifically:
To current layer binary map M, first using initial inside and outside circle as constraint, by the side for changing its central point and radius
Formula obtains multiple groups concentric circles collection, then according to statistics receptance functionAnd it is greedy
Algorithm from the concentric circles concentrate to obtain with most matched two concentric circles in the boundary of femoral head and acetabular bone, as current layer joint
The final label of gap area；Wherein, σ is parameter preset, FΦAnd TΦBefore respectively representing in the binary map within the scope of the Φ of region
The number of scene element and total pixel.
As can be seen from the above description, since the boundary of femoral head and acetabular region is generally circular in tomography picture, with two
Value figure and user mark initial concentric circles be input constraint condition, using based on statistics receptance function greedy algorithm carry out by
Layer identification, can obtain tomograph on piece joint space zone marker.
Further, described " the reconciliation field isopleth for calculating the joint space region " specifically:
It obtains the tomograph piece and is based on the modified gradient data of directional information；
According to formulaReconciliation field value corresponding to obtained each pixel of tomography pictureWherein,L be according between each pixel syntople and the obtained m × m rank of the modifying gradient data
Laplacian Matrix, m are the pixel number of above-mentioned joint space marked region；
Wherein, n is obligatory point number, and if only if u (1≤u≤n) a obligatory point in joint space area pixel serial number v (1≤v
≤ m) when cu,v=1, it is 0 in the case of remaining；B when u (1≤u≤n) a obligatory point is inner circular corresponding pixel pointsu
=1, it is otherwise 0；γ is predetermined coefficient；
According to reconciliation field valueObtain reconciliation field isopleth.
As can be seen from the above description, being first depending on pass to obtain the reconciliation field that can accurately reflect femoral head boundary information
The center position information for saving gap area, is modified the gradient intensity data of the tomography picture, excludes or reduce stock
The interference of inside bone cancellous bone and adjacent acetabular bone boundary to reconciliation field data；Since femur head contour is endless in tomography picture
It is complete rounded, and the boundary information is frequently subjected to its interior cancellous bone and adjacent acetabular region and influences, by joint space region
Syntople description and the modifying gradient data between each pixel, obtain the reconciliation field for accurately reflecting femoral head boundary information
And its isopleth.
Further, " femoral head is partitioned into according to reconciliation field isopleth " specifically:
The reconciliation field isopleth of default number of branches is successively extracted as boundary candidate line set；
Similarity in conjunction with the modifying gradient data and between adjacent layer femoral head boundary line is as standard, to time
Every reconciliation field isopleth in selected works conjunction scores；
Boundary line of the highest reconciliation field isopleth of scoring as femoral head is chosen, and is carried out from the center of circle of the concentric circles
Region increases, and obtains the femur head region that segmentation obtains.
As can be seen from the above description, by choosing boundary of the optimal reconciliation field isopleth as femoral head, so that femoral head
Boundary it is the most accurate, so as to accurately, effectively be partitioned into femur head region.
Further, described " using the region in region growing methods segmentation femur in addition to femoral head " specifically:
Femur head region is removed from the area of bone tissue extracted；
Use region growing methods from the area of bone tissue for eliminating femur head region divide femur in except femoral head with
Outer region.
As can be seen from the above description, removal femur head region after, in femur the boundary of other regions and acetabular bone with regard to apparent,
So as to the region being easily partitioned into femur in addition to femoral head.
Further, pass through the femoral head being partitioned into described in boolean sum operation merging and the region in addition to femur.
Referring to figure 2. to Fig. 9, the embodiment of the present invention one are as follows:
A method of dividing femur in fault image, comprising:
S1, the tomography picture that multiple include hip joint is obtained；The hip joint includes femoral head and acetabular bone, and femoral head with
The connection such as neck of femur, greater trochanter, lesser trochanter in femur；Therefore, the tomography picture comprising hip joint include at least femoral head with
And the region in femur in addition to femoral head；
Specifically, being loaded into faultage image data from the formatted files such as DICOM, and is visualized and put down by computer graphics
Platform (such as OpenGL) carries out visualization processing, obtains the tomography picture of multilayer, and show in visual interactive interface；
S2, area of bone tissue is extracted from the tomography picture；
Specifically, the most Dalian that will there are intensity values in above-mentioned tomography picture using the thresholding method based on histogram
Logical extracted region comes out, obtain it is corresponding using femur and hipbone linking area as prospect, other regions be background binary map
M, and colour-coded prospect, transparency process background are carried out to binary map M, then the binary map is superimposed on original tomography picture
On shown, obtain connected region shown in region 1 in such as Fig. 3；
Joint space region between S3, label femoral head and acetabular bone；
Specifically, substantially being marked by hand by two by user first in the tomography picture that said extracted goes out area of bone tissue
The joint space region of correlativity between femoral head and acetabular bone is represented determined by a initial concentric circles, be located at two circles it
Between region be joint space region, such as annular region shown in region 22 in Fig. 3；Then according to statistics receptance functionAssess a series of candidate circles C (x, r) determined by initial concentric circles with
The matching degree of femoral head or acetabular bone borderline region in tomography picture M obtains and femoral head or hip in current layer tomography picture
The most matched one group of concentric circles in mortar boundary, as shown in region 33 in Fig. 4；Again using this group of concentric circles as next layer of tomography picture
Initial concentric circles, successively automatically identify in all tomography pictures for existing simultaneously femoral head and acetabular bone and the tomography picture
Joint space region, the processing of such loop iteration is until respective function value corresponding to the candidate circles of all matching femorals head
Both less than 0 or all matching acetabular bones candidate circles corresponding to until respective function value is both greater than 0；Wherein, σ=3, FΦWith
TΦRespectively represent the number of foreground pixel and total pixel within the scope of the Φ of region in M；R (M, x, r) is bigger, and C (x, r) and femur are in front
The matching degree on boundary is better；R (M, x, r) is smaller, and C (x, r) and the matching degree on acetabular bone boundary are better；
The match cognization process in above-mentioned joint space region using with constraint conditions greedy algorithm, i.e., with it is given just
Beginning concentric circles is constraint, is preferentially found and femoral head boundary in a series of candidate circles set determined by initial concentric circles
Optimum Matching circle C (c, r), find again with C (c, r) concyclic heart and with the optimal circle C (c, R) of acetabular bone Boundary Match；In addition, should
Initial the specified of concentric circles follows deduction induction, i.e., it is participated in specifying initial concentric circles at the beginning by user, in addition to this, when
The optimal concentric circles that front layer tomography picture is recognized is by the initial value as its adjacent layer；The candidate circles set is with initial
On the basis of the inside and outside circle of concentric circles, obtained by way of changing its central point and radius；
S4, the reconciliation field isopleth for calculating the joint space region；
Specifically, preparing for the calculating of subsequent reconciliation field and its scoring of isopleth, the gradient of tomography picture is first obtained
Data；In the gradient map of tomography picture, since surface cortex bone and interior cancellous bone are shown as by force respectively in tomography picture
The higher region of angle value and the lower region of intensity value are directed toward in femoral head so that the borderline region of femoral head nearby exists simultaneously
The larger ladder in portion's (such as contour line shown in region 44 in Fig. 5) and external (such as contour line shown in region 55 in Fig. 5)
Spend information；In addition, acetabular bone boundary close to femur is same (such as to belong to the profile of acetabular bone shown in region 66 in Fig. 5
Line).In order to retain the larger gradient information outside femoral head boundary direction, and other useless larger gradient informations are excluded to it
Interference (such as in reserved graph 5 contour line shown in region 55 and remove contour line shown in region 44 and region 6 6), adopt
Gradient data is modified with based on locality information；Specifically, using the acquired concentric circles center of circle relative to
The orientation of current pixel and the information of current pixel gradient direction are modified, and revised gradient value is as shown in Figure 6；
Then reconciliation field value is carried out to calculate.It solves image reconciliation field and mathematically solves Poisson's equation(Δ represents
Discrete Laplace operator)；The Poisson's equation can be solved using the least square method with edge-restraint condition, used here as generation
Pixel corresponding to the inside and outside circle of table joint space zone boundary is as obligatory point, and original equation is converted at this time
Form；Wherein,For reconcile field value,Wherein L is Laplacian Matrix, if two circles it
Between region in have m pixel, then L is m × m rank matrix, and the i-th row jth column element (1≤i, j≤m)Wherein, E indicates that there are syntoples between pixel；ω is the power of Laplace operator
Important influence is distributed with for reconciliation field in value；Here it usesThe larger area of gradient is distributed in capture
The partitioning boundary in domain；Wherein, γijThe above-mentioned big Mr. Yu of modifying gradient value for two-valued variable, at pixel i or j
Value takes a biggish numerical value, such as 1000 when threshold value, is otherwise 1；piIndicate the pixel value at pixel i；lenijIndicate picture
Length between vegetarian refreshments i, j；C and b' is the matrix and vector for representing constraint condition in reconciliation field respectively, if total n constraint
Point, then C and b' can be expressed as follows:
C when u (1≤u≤n) a obligatory point is in joint space area pixel serial number v (1≤v≤m)u,v=
1, it is 0 in the case of remaining；B when u (1≤u≤n) a obligatory point is inner circular corresponding pixel pointsu=1, otherwise for
0；γ=1000；Non trivial solution is found out using the kit with Large Scale Sparse coefficient matrix linear equation A Φ=b is solved, i.e.,
For reconciliation field value corresponding to each pixel；Reconciliation field isopleth can be obtained according to each reconciliation field value；As shown in Figure 7；
S5, femoral head is partitioned into according to reconciliation field isopleth；
Specifically, firstly, carrying out equivalent line extraction processing to reconciliation field；The lines as shown in region 77 in Fig. 7 are to adjust
With the isopleth of field, they are by the Candidate Set as final partitioning boundary；Then, in conjunction with modified gradient data and adjacent layer
Shape similarity these two types index between femoral head boundary line can obtain an optimal closed contour as femoral head
In boundary line, such as Fig. 8 shown in 88 curve of region；Finally, carrying out region growth from the identified concentric circles center of circle, can obtain
To the segmentation result of femur head region, as shown in region 99 in Fig. 8.
S6, the region in addition to femoral head is partitioned into femur using region growing methods；
The femur head region obtained based on reconciliation field is removed from the above-mentioned tomography picture M for obtaining best isopleth, i.e., will
Femoral head corresponding region is changed into background by prospect in this layer of two-value picture M, can be by remaining femoral component and hipbone part point
It is segmented into M mutual disconnected foreground area, therefore the segmentation result that region growth method obtains remaining femoral region can be used, such as
In Fig. 8 shown in region 10；
The femoral head and the above-mentioned region in addition to femoral head that S7, combination and segmentation go out obtain femur；
Specifically, the femur head region obtained based on reconciliation field and remaining femoral region are carried out boolean sum operation, obtain
Final femur segmentation result, as shown in the region Fig. 9 11.
In conclusion the method provided by the invention for dividing femur in fault image, by femoral head and acetabular bone it
Between joint space region be marked and identify that and the reconciliation field value by calculating the joint space region obtains accurate
Femoral head boundary, so as to accurately and efficiently obtain the femoral region in computerized tomography image as a result, and the degree of automation
Height has emphatically the subsequent applications such as auxiliary diagnosis, the hip replacement surgery simulation of joint tissue wear assessment, coxitis
The meaning wanted.
The above description is only an embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalents made by bright specification and accompanying drawing content are applied directly or indirectly in relevant technical field, similarly include
In scope of patent protection of the invention.
Claims (5)
1. a kind of method for dividing femur in fault image characterized by comprising
Obtain the tomography picture comprising hip joint；
Area of bone tissue is extracted from the tomography picture；
Mark the joint space region between femoral head and acetabular bone；
Calculate the reconciliation field isopleth in the joint space region；
Femoral head is partitioned into according to reconciliation field isopleth；
Described " extracting area of bone tissue from the tomography picture " specifically:
There is the largest connected extracted region of intensity values to go out the tomography picture using the thresholding method based on histogram
Come, obtains the binary map using femur and hipbone linking area as prospect, other regions for background；
The tomography picture is two or more, and every tension fault picture all has binary map corresponding thereto, " the label stock
Joint space region between bone and acetabular bone " specifically:
Two initial concentric circles are drawn, wherein inside and outside circle is located at the boundary of femoral head and its surrounding acetabular bone；
Since layer where initial concentric circles, successively the binary map is identified, obtain it is all exist simultaneously femoral head and
The tomography picture of acetabular bone, and mark corresponding joint space region；
It is described " binary map to be identified, all tomography pictures for existing simultaneously femoral head and acetabular bone, and label pair are obtained
The joint space region answered " specifically:
To current layer binary map M using initial inside and outside circle as constraint, change the initially central point of inside and outside circle and radius
Obtain multiple groups concentric circles collection；Then according to statistics receptance functionAnd greedy calculation
Method from the concentric circles concentrate to obtain with most matched two concentric circles in the boundary of femoral head and acetabular bone, as between current layer joint
The final label in gap region；Wherein, σ is parameter preset, FΦAnd TΦRespectively represent in the binary map prospect within the scope of the Φ of region
The number of pixel and total pixel, x are the center of circle of concentric circles, and r is the radius of concentric circles, and Ring (x, r- σ, r) represents concentric circles C
Annular region between (x, r- σ) and C (x, r), Ring (x, r, r+ σ) represent the ring between concentric circles C (x, r) and C (x, r+ σ)
Shape region；
" the reconciliation field isopleth for calculating the joint space region " specifically:
It obtains the tomograph piece and is based on the modified gradient data of directional information；
According to formulaReconciliation field value corresponding to obtained each pixel of tomography pictureWherein,L be according between each pixel syntople and the obtained m × m rank of the modifying gradient data
Laplacian Matrix, m are the pixel number of the joint space marked region；
Wherein, n is obligatory point number, c when u-th of obligatory point is in joint space area pixel serial number vu,v=1, remaining
In the case of be 0；B when u-th of obligatory point is inner circular corresponding pixel pointsu=1, it is otherwise 0；γ is predetermined coefficient,
Wherein, 1≤u≤n, 1≤v≤m；
According to reconciliation field valueObtain reconciliation field isopleth.
2. the method according to claim 1 for dividing femur in fault image, which is characterized in that described " according to described in
The field isopleth that reconciles is partitioned into femoral head " after further include:
The region in addition to femoral head is partitioned into femur using region growing methods；
The femoral head and the region in addition to femoral head that combination and segmentation goes out obtain femur.
3. the method according to claim 1 for dividing femur in fault image, which is characterized in that " according to the reconciliation
Field isopleth is partitioned into femoral head " specifically:
The reconciliation field isopleth of default number of branches is successively extracted as boundary candidate line set；
It is equivalent to every reconciliation field in conjunction with the similarity between the modified gradient data and adjacent layer femoral head boundary line
Line scores；
Boundary line of the highest boundary candidate as femoral head of scoring is chosen, and carries out region from the center of circle of the concentric circles
Increase, obtains the femur head region that segmentation obtains.
4. the method according to claim 2 for dividing femur in fault image, which is characterized in that described " to use region
Growing method divides the region in femur in addition to femoral head " specifically:
Femur head region is removed from the area of bone tissue extracted；
Using region growing methods from the area of bone tissue for eliminating femur head region divide femur in addition to femoral head
Region.
5. the method according to claim 2 for dividing femur in fault image, which is characterized in that operated by boolean sum
The femoral head being partitioned into described in merging and the region in addition to femoral head.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610297197.9A CN106023144B (en) | 2016-05-06 | 2016-05-06 | Divide the method for femur in fault image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610297197.9A CN106023144B (en) | 2016-05-06 | 2016-05-06 | Divide the method for femur in fault image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106023144A CN106023144A (en) | 2016-10-12 |
| CN106023144B true CN106023144B (en) | 2019-05-31 |
Family
ID=57081348
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610297197.9A Expired - Fee Related CN106023144B (en) | 2016-05-06 | 2016-05-06 | Divide the method for femur in fault image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106023144B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108648193B (en) * | 2018-06-06 | 2023-10-31 | 南方医科大学 | Biological tissue image identification method and system and computer storage medium thereof |
| CN109377478B (en) * | 2018-09-26 | 2021-09-14 | 宁波工程学院 | Automatic grading method for osteoarthritis |
| CN110210481A (en) * | 2019-06-06 | 2019-09-06 | 福建师范大学 | A kind of soleplate automatic separation method based on spill gap region recognition |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254317A (en) * | 2011-03-25 | 2011-11-23 | 苏州迪凯尔医疗科技有限公司 | Method for automatically extracting dental arch curved surface in dental implantation navigation |
| CN102968783A (en) * | 2012-10-15 | 2013-03-13 | 深圳市旭东数字医学影像技术有限公司 | Method and system for automatically segmenting bones from abdomen image data |
| CN103854288A (en) * | 2014-03-11 | 2014-06-11 | 深圳市旭东数字医学影像技术有限公司 | Cruciate ligament segmentation method and system |
| CN103854287A (en) * | 2014-03-11 | 2014-06-11 | 深圳市旭东数字医学影像技术有限公司 | Meniscus segmentation method and device based on magnetic resonance image |
| CN103871057A (en) * | 2014-03-11 | 2014-06-18 | 深圳市旭东数字医学影像技术有限公司 | Magnetic resonance image-based bone segmentation method and system thereof |
| CN104091365A (en) * | 2014-07-12 | 2014-10-08 | 大连理工大学 | Acetabulum tissue model reconstruction method for serialization hip joint CT image |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2002348241A1 (en) * | 2001-11-24 | 2003-06-10 | Image Analysis, Inc. | Automatic detection and quantification of coronary and aortic calcium |
| US9613426B2 (en) * | 2014-10-21 | 2017-04-04 | Toshiba Medical Systems Corporation | Segmentation and matching of structures in medical imaging data |
-
2016
- 2016-05-06 CN CN201610297197.9A patent/CN106023144B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254317A (en) * | 2011-03-25 | 2011-11-23 | 苏州迪凯尔医疗科技有限公司 | Method for automatically extracting dental arch curved surface in dental implantation navigation |
| CN102968783A (en) * | 2012-10-15 | 2013-03-13 | 深圳市旭东数字医学影像技术有限公司 | Method and system for automatically segmenting bones from abdomen image data |
| CN103854288A (en) * | 2014-03-11 | 2014-06-11 | 深圳市旭东数字医学影像技术有限公司 | Cruciate ligament segmentation method and system |
| CN103854287A (en) * | 2014-03-11 | 2014-06-11 | 深圳市旭东数字医学影像技术有限公司 | Meniscus segmentation method and device based on magnetic resonance image |
| CN103871057A (en) * | 2014-03-11 | 2014-06-18 | 深圳市旭东数字医学影像技术有限公司 | Magnetic resonance image-based bone segmentation method and system thereof |
| CN104091365A (en) * | 2014-07-12 | 2014-10-08 | 大连理工大学 | Acetabulum tissue model reconstruction method for serialization hip joint CT image |
Non-Patent Citations (3)
| Title |
|---|
| Visualization in biomedical computer;R.A.Robb;《Parallel Computing》;19991231;第2067-2110页 |
| 基于变分等值线方法的图像分割技术;蔡力等;《计算数学》;20060228;第28卷(第1期);第43-52页 |
| 面向支撑喉镜手术的虚拟仿真系统研究;许天春;《中国优秀硕士学位论文全文数据库 信息科技辑》;20070515(第5期);第27-33页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106023144A (en) | 2016-10-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cebral et al. | From medical images to anatomically accurate finite element grids | |
| CN106485704B (en) | Method for extracting center line of blood vessel | |
| CN110738681A (en) | automatic pedicle screw operation path planning method based on deep learning network | |
| US11132801B2 (en) | Segmentation of three-dimensional images containing anatomic structures | |
| US10258412B2 (en) | Method and node for manufacturing a surgical kit for cartilage repair | |
| US11626212B2 (en) | Systems and methods for automated segmentation of patient specific anatomies for pathology specific measurements | |
| CN107895364B (en) | A kind of three-dimensional reconstruction system for the preoperative planning of virtual operation | |
| CN106023144B (en) | Divide the method for femur in fault image | |
| CN106127753B (en) | CT images body surface handmarking's extraction method in a kind of surgical operation | |
| CN109801268B (en) | CT radiography image renal artery segmentation method based on three-dimensional convolution neural network | |
| CN106846346A (en) | Sequence C T image pelvis profile rapid extracting methods based on key frame marker | |
| Shiffman et al. | Medical image segmentation using analysis of isolable-contour maps | |
| Basavaprasad et al. | A comparative study on classification of image segmentation methods with a focus on graph based techniques | |
| KR101223681B1 (en) | Automatic Segmentation device and method of Cartilage in Magnetic Resonance Image | |
| Egger et al. | Nugget-cut: a segmentation scheme for spherically-and elliptically-shaped 3D objects | |
| Lee et al. | Segmentation of anterior cruciate ligament in knee MR images using graph cuts with patient-specific shape constraints and label refinement | |
| Zou et al. | Semi-automatic segmentation of femur based on harmonic barrier | |
| CN111080556A (en) | Method, system, equipment and medium for strengthening trachea wall of CT image | |
| Tessamma et al. | A semi-automatic method for segmentation and 3D modeling of glioma tumors from brain MRI | |
| Umadevi et al. | Enhanced Segmentation Method for bone structure and diaphysis extraction from x-ray images | |
| Pardo et al. | A snake for model-based segmentation of biomedical images | |
| Chen et al. | Multimodality‐based knee joint modelling method with bone and cartilage structures for total knee arthroplasty | |
| CN105469432A (en) | Improved-tree-shaped-model-based automatic medical image segmentation method | |
| Song et al. | Computer-aided modeling and morphological analysis of hip joint | |
| Oliveira et al. | Automatic Couinaud Liver and Veins Segmentation from CT Images. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190531 |