CN107730505A - Image processing apparatus and image processing method - Google Patents
Image processing apparatus and image processing method Download PDFInfo
- Publication number
- CN107730505A CN107730505A CN201710337843.4A CN201710337843A CN107730505A CN 107730505 A CN107730505 A CN 107730505A CN 201710337843 A CN201710337843 A CN 201710337843A CN 107730505 A CN107730505 A CN 107730505A
- Authority
- CN
- China
- Prior art keywords
- section
- mentioned
- image processing
- processing apparatus
- curve
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/174—Segmentation; Edge detection involving the use of two or more images
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20101—Interactive definition of point of interest, landmark or seed
-
- 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
- G06T2207/30061—Lung
Abstract
Embodiment is related to image processing apparatus and image processing method, there is provided a kind of more to carry out accurately and advantageously the image processing apparatus and image processing method of region segmentation, the image processing apparatus of embodiment possesses selector and extension.Selector selected from 3 d image data as defined in section.Extension forms curve corresponding with the curve being depicted on above-mentioned defined section on away from the section that above-mentioned defined section is prescribed limit, so as to form curved surface on above-mentioned 3 d image data.
Description
The reference of related application
The application enjoys the Chinese application numbers 201610662482.6 filed an application for 12nd of August in 2016 and 2016
The interests of the Japanese patent application numbering 2016-239517 to file an application December 9 priority, quoting in this application should
The full content of Japanese patent application.
Technical field
Embodiment is related to a kind of image processing apparatus and image processing method.
Background technology
In recent years, along with the development of image recognition technology, the medical imaging for having internal organs to shooting carries out region segmentation
Technology attracts attention.Such as the planning of positioning and the operation of leaf volume-diminished of lobe of the lung domain decomposition technique for substantive lesion etc.
It is all extremely important.
So-called lobe of the lung domain decomposition technique is in CT (Computerized Tomography：Computer X-ray tomography is made
Shadow art) shooting of the medical imaging harvester such as device medical imaging in by detecting interlobar fissure by lung region segmentation into 5 phases
Answer the technology of the lobe of the lung.
Interlobar fissure is low resolution and the clear structure of obscure boundary on CT images.The factors such as peripheral vessels, picture noise
So that the detection of interlobar fissure becomes more difficult.
Conventionally, as carrying out region segmentation automatically hardly results in expected result, therefore, in general,
Need to first pass through physician in view experience and the position of interlobar fissure is positioned on image.
For example, a kind of combination two dimension segmentation and three-dimensional segmentation are disclosed in patent document 1 (US2014/0298270A1)
Domain decomposition technique, wherein, a point is selected in the initial slice (section) of lung by user, so as to two dimension segmentation module base
Calculate the interlobar fissure being used as by the curve of the point on two dimension slicing automatically in selected point, and then, user ask into
During row three-dimensional segmentation, segmentation result of the three-dimensional segmentation module based on the initial slice carries out three-dimensional simulation, obtains 3-D view time
Choosing., can be by multiple 3-D view candidates for more so generating come corrected Calculation result in patent document 1.
In addition, a kind of region segmentation result amendment dress is disclosed in patent document 2 (Japanese Unexamined Patent Publication 2012-45256)
Put, wherein, the segmentation interface of zone boundary is automatically generated as on 3-D view, and user can be accepted in 3-D view
To the manual modification at segmentation interface in data.
In patent document 1, no matter two dimension segmentation or three-dimensional segmentation, be all based on what user inputted in single section
Anchor point, calculated according to pixel, so as to form segmentation curve.But the precision of this dividing method is not high, also without
Method distinguishes lobe of the lung classification.Also, if the interlobar fissure of photography target show it is incomplete, it is likely that desired result can not be obtained.
In addition, the automatic division method in patent document 2 equally exists larger error, it is necessary to which user enters to generation result
Row is corrected and calculating is repeated, so as to need more man-machine interactions.Especially for the infull person of interlobar fissure, statistical representation
There are about existing in the image of 80% patient interlobar fissure partly or entirely disappear and on image sightless situation, so as to automatic
It is difficult to find the segmentation interface between region during progress region segmentation.Therefore, this be segmented in automatically can not under existing technology
Obtain desired result.
The content of the invention
The problem to be solved in the present invention be to provide it is a kind of can be at image that is more accurate and advantageously carrying out region segmentation
Manage device and image processing method.
The image processing apparatus of embodiment possesses selector and extension.Selector selects rule from 3 d image data
Fixed section.Extension forms and is depicted in above-mentioned defined section on away from the section that above-mentioned defined section is prescribed limit
On curve corresponding to curve, so as to forming curved surface on above-mentioned 3 d image data.
Effect：
According to the image processing apparatus and image processing method of embodiment, can be more accurate and region be advantageously carried out
Segmentation.
Brief description of the drawings
Fig. 1 is the structured flowchart for representing the image processing apparatus that first embodiment is related to.
Fig. 2 is the flow chart for representing the region segmentation processing that first embodiment is related to.
Fig. 3 is the exemplary plot for representing the display interface in the region segmentation processing that first embodiment is related to.
Fig. 4 is the structured flowchart for representing the image processing apparatus that second embodiment is related to.
Fig. 5 is the flow chart for representing the region segmentation processing that second embodiment is related to.
Fig. 6 is the schematic diagram of lung mechanics.
Fig. 7 is the structured flowchart for representing the image processing apparatus that the 3rd embodiment is related to.
Fig. 8 is the flow chart for representing the region segmentation processing that the 3rd embodiment is related to.
Fig. 9 is the explanation figure for representing region segmentation processing.
Embodiment
The mode of the application is a kind of image processing apparatus, including：Selector, rule are selected from 3 d image data
Fixed section；And extension, formed and be depicted on away from the section that defined section is prescribed limit on defined section
Curve corresponding to curve, so as to forming curved surface on 3 d image data.
Another mode of the application is a kind of image processing method, including：From 3 d image data as defined in selection
The step of section；And by being formed and being depicted on defined section on away from the section that defined section is prescribed limit
Curve corresponding to curve, so as on 3 d image data formed curved surface the step of.
According to presently filed embodiment, using the teaching of the invention it is possible to provide it is a kind of can image that is more accurate and advantageously carrying out region segmentation
Processing unit and image processing method, user can more easily be entered edlin with less operation, only pass through the volume of two dimension
The result of intuitively 3D display can just be obtained by collecting.Moreover, in the image processing apparatus of embodiment, can carry out anti-in real time
Present and further improve the precision of region segmentation.
Embodiment is related to a kind of image processing apparatus handled image, the image processing apparatus can by with
Independent computer of the image gathering devices such as X-ray apparatus connection etc. has CPU (central process unit：Center
Processor) equipment perform the software of each function with image processing apparatus and realize, can also be used as and be able to carry out figure
Realized as the circuit of each function of processing unit in a manner of hardware.Also, the image processing apparatus of present embodiment
The above can be pre-installed in as the part in the medical imaging harvester such as CT devices or MR imaging apparatus
In medical imaging harvester.
Preferred embodiment in the embodiment being related to hereinafter, with reference to the accompanying drawings of the application.In each embodiment
In, so that photography target includes lung, carries out the processing of lobe of the lung segmentation (extraction interlobar fissure) for lung images as an example, according to each
Embodiment, carry out the explanation of region segmentation processing.
In addition, in different embodiments, identical reference is used for identical part, and suitably omit and repeat
Explanation.
(first embodiment)
Fig. 1 is the structured flowchart for representing the image processing apparatus that first embodiment is related to.As shown in figure 1, image procossing
Device 100 includes selector 10, receiving unit 20 and extension 30.
Selector 10 selects crucial section (defined section) from 3 d image data.Specifically, selector 10 from
The image collecting devices such as CT devices in lung's 3 d image data for being shot and collected as the chest of photography target,
The section of selection one or more is as crucial section, so that the display device such as monitor connected in image processing apparatus 100
On using selected crucial section as crucial two dimensional image of cutting into slices be subject to two-dimentional display.In addition, it is following, selected with selector 10
Multiple sections illustrate as the situation of key section as an example.Selector 10 can be that by function above
Circuit or software module.
Specifically, selector 10 can carry out crucial section using the anatomic information or geometry information of subject
Selection.Exemplified by the 4 crucial sections of left lung regional choice, preferably left lung 3-D view sagittal plane
Select following four sections being parallel to each other to be cut into slices as crucial in (sagittal plane) successively, i.e., select successively from inside to outside
The arch of aorta is selected just to disappear and the ratio of two dimensional image area that the also existing section of heart, lung areas account for section is maximum cuts
The section and the section in the centre position in remaining section that piece, heart are just wholly absent are cut into slices as crucial.In other words, select
Select portion 10 and not including actively in the section close with the section comprising the arch of aorta is selected from the section parallel to sagittal plane
Arcus haemalis but section, the area in lung region comprising heart reach maximum section, the section adjacent with the section comprising heart
In the section not comprising heart and the section not comprising heart but comprising lung in positioned at middle section as crucial
Section.
Can also be that selector 10 selects four along sagittal axis in the 3-D view of lung, with the plan range of equalization
Section cut into slices as crucial.That is, the selection of selector 10 is parallel with sagittal plane and there are 4 keys at certain interval to cut
Piece.This system of selection is effectively in the case where not knowing lobe of the lung classification.
In addition it is also possible to it is, according to the average segmentation result for counting obtained lung in advance and corresponding crucial section position
Put, produce generally segmentation template and stored, this, which is generally split, has preselected crucial section in template, so as to selector
The 3 d image data of the lung when selecting crucial section for some 3 d image data, is mapped directly to the lung by 10
Usual segmentation template, to select crucial section corresponding with the crucial section in the usual segmentation template.
Receiving unit 20 accepts the operation for describing curve in crucial section.For example, receiving unit 20 is used to accept successively selecting
Select on the two dimensional image of the crucial section selected by portion 10 to for distinguishing retouching for the curve of different zones (segmentation curve) progress
Paint.In the case where splitting to the lobe of the lung, the segmentation curve represents the interlobar fissure by lung segmentation into multiple lobes of the lung.Receiving unit
20 can be that by the circuit or software module of function above.
Specifically, doctor can show on the display apparatus, the X-Y scheme of crucial section selected by selector 10
As upper, according to the performance of two dimensional image and experience, describe the segmentation curve for representing interlobar fissure on 2d, so as to receiving unit
20 accept the segmentation curve of the description.
Doctor can describe the segmentation curve in each crucial section successively on the two dimensional image of multiple crucial sections, from
And receiving unit 20 accepts respective segmentation curve corresponding with each key section successively.
Extension 30 by away from crucial section for prescribed limit section (it is following, be also recited as prescribed limit section,
Or contiguous slices) on form curve corresponding with the curve being depicted in crucial section, the thus shape on 3 d image data
Into curved surface.That is, extension 30 by the curve extension in crucial cut into slices by, to the section of prescribed limit, being consequently formed curved surface.Example
Such as, the segmentation curve that extension 30 is accepted based on receiving unit 20, the segmentation curve extension in crucial section is cut to key
It is bent so as to form adjacent segmentation corresponding with segmentation curve on contiguous slices on contiguous slices in the adjacent prescribed limit of piece
Line, using key section and contiguous slices, the part three formed respectively including the segmentation interface as three-dimensional interlobar fissure
Image is tieed up, so as to which the segmentation curve described in crucial section and the adjacent segmentation expanded on contiguous slices adjacent thereto are bent
Line spatially forms three-dimensional surface, forms the segmentation interface for splitting the part lobe of the lung.Extension 30 can be that by the above
The circuit or software module of function.
In the example of the sagittal plane section selected by selector 10, extension 30 cuts doctor (user) in two-dimentional key
The curve extension that two dimension is described on piece performs 2D segmentations on each contiguous slices, recycles each section on contiguous slices
Form three-dimensional segmentation.
The scope (prescribed limit) that the extension of portion 30 is expanded as contiguous slices can be crucial cuts into slices axially back and forth
Untill predetermined distance, or the midpoint extended also between two contiguous sliceses.
Extension 30 can use the mapping method between existing various different sections.Can also be with the following method.
That is, multiple key points are extracted on the segmentation curve that extension 30 is depicted in crucial section, such as extraction starting point,
Terminal and between maximum two points of Curvature varying as key point, and then, positioning and crucial section on contiguous slices
On the corresponding adjacent key point of key point., can be using the pixel of existing different layers on the confirmation of adjacent key point
Between various mapping methods obtain.In addition, it is following, the key point in crucial section or contiguous slices is simply recited as
Point.
After adjacent key point is determined, by each adjacent key point on same contiguous slices according to institute in key section
Corresponding key point is linked in sequence, and the curve being formed by connecting is as between leaf corresponding with the segmentation curve in crucial section
Split.
As described above, extension 30 selects the point of specified quantity from the curve described in crucial section, from adjacent
Point corresponding with the point selected is calculated in section, and the point calculated is linked, so as to the forming curves on contiguous slices.As
Connected mode, both can be straight line connection or to carry out curve connection along certain path planning between 2 points.For example,
If the quantity of the key point of selection is enough, can be connected and whole with straight line between points according to a certain direction
Body forms a curve.In other words, the straight line of extension 30 links two points of the adjoining in adjacent key point, so as to
Forming curves on contiguous slices.Or water channel principium is calculated according to gradation of image between two neighbor points, along being counted
The water channel principium calculated is attached.In other words, extension 30 links phase with the water channel principium related to pixel value
Two points of the adjoining in adjacent key point, so as to the forming curves on contiguous slices.Here, water channel principium refers to
Such as in multiple paths for linking of two points will abut against, based on the pixel in path and between the pixel that adjoins each other
The difference of pixel value between the difference of value or each pixel in path and adjacent key point or the length in path etc. form
Originally minimum path is reached.
In addition, linking point on contiguous slices with straight line in the case of forming curves etc., contiguous slices is being formed at
On curve in include so-called broken line.That is, the curve being formed on contiguous slices is not limited to its entirety by without angle
But the situation of the line composition continuously bent or its overall or part are made up of straight line or broken line.Equally, retouch
The curve being painted in crucial section can also be that its overall or part is made up of straight line or broken line.In addition, it is same, formed
Curved surface on 3 d image data can also be that its overall or part is made up of one or more panel data.
In addition, on extension prescribed limit (spreading range) selected or, by this key section and lung
The distance between edge or the position of next crucial section are used as largest extension distance, in the range of the ultimate range, by
Gradually to external expansion.Also, extend outwardly to every time on a contiguous slices, just calculate this key section on key point with it is adjacent
The pixel value difference between adjacent key point in section, it is more than or equal to the situation of threshold value set in advance in the pixel value difference
Under, stop extension, and enter the processing of next crucial section.The threshold value set in advance can according to required precision and
The data of key section arbitrarily determine.
Or largest extension distance can not also be set, and the threshold value of pixel value difference is only set, so, work as extension
Segmentation curve extension on contiguous slices to next crucial section, i.e., next crucial section are turned into the last crucial phase cut into slices by 30
In the case of neighbour's section, it can also cancel and the curve of the next crucial section is described, or automatic modification is cut into slices to the key
Selection.
Both can be with identical to the spreading range of different crucial section settings, can also be different.
Extension 30 is by being forming curves on the contiguous slices of prescribed limit away from crucial section, so as in 3-D view
Curved surface is formed in data.For example, the curve carried out by extension 30 is formed, can be based on each crucial section generation with it is each
Partial 3-D image corresponding to key section, and the partial 3-D image includes the segmentation curve and adjacent segmentation by describing
The three-dimensional segmentation interface that curve is formed.So as to can show that the part after segmentation on user oriented monitor or display
3-D view is for reference.Also, after some 3-D views are generated, by showing these partial 3-D figures simultaneously
Picture, it can be seen that whole lung mechanics, 3-D view with interlobar fissure.
In the first embodiment, the selector that selector 10 corresponds in claims, receiving unit 20 correspond to power
Receiving unit in sharp claim, the extension that extension 30 corresponds in claims.Below in conjunction with accompanying drawing 2 and Fig. 3 explanations
The flow for the region segmentation processing that image processing apparatus 100 is carried out.
Fig. 2 is the flow chart for representing the region segmentation processing that first embodiment is related to.As shown in Fig. 2 proceeding by
During segmentation, first, selector 10 automatically selects multiple sections as crucial section (step in the 3 d image data of lung
S201)。
As one, the situation that dividing processing is carried out to the 3-D view to left lung shown in Fig. 3 illustrates.Fig. 3 is
Represent the exemplary plot of the display interface in region segmentation processing.As shown in figure 3, selector 10 is selected on 3 d image data
The crucial section of sagittal plane be shown in block 301 in a manner of two dimensional image.
Doctor can be depicted on the two-dimensional image as the white line M shown on the crucial section in Fig. 3
Split curve, so as to which receiving unit 20 accepts the description (step S202) to the segmentation curve in key section.
Then, into step S203, extension 30 is by before and after the segmentation curve extension in key section to key section
Contiguous slices on, so as to form the part three being directed under the visual angle of coronal-plane as shown in the frame 304 of Fig. 3 right side bottom
Tie up image (step S204).The partial 3-D image is shown in a manner of three-dimensional in block 304.Also, no matter having an X-rayed
Position with segmentation interface K can be confirmed in the case of non-perspective.
In the example in figure 3, except frame 301 and the display for showing the crucial curve cut into slices and accept user's description
Outside the frame 304 of partial 3-D image, the edit instruction frame 305 also for input instruction, the partial 3-D figure based on generation
As and simulate the two-dimentional display box 302,303 of interface image of the interlobar fissure on horizontal plane and coronal-plane respectively.Thus, image
Processing unit 100 can provide the user the spreading result of interlobar fissure in real time.Also, pass through the extension that each key is cut into slices
As a result combination is shown in a display box, and image processing apparatus 100 can provide more directly perceived and accurate lung segmentation knot
Fruit.
Certainly, the display interface of display device is not limited in Fig. 3 layout, and the position of each frame can change, also,
Two-dimensional display image frame 302,303 can also be omitted.
In present embodiment, selector 10 selects multiple sections as crucial section, so as to which extension 30 can distinguish pin
To each crucial section generating portion 3-D view so that the position of interlobar fissure is more accurate, will not be due to a certain bar curve
Mistake is described and influenceed overall.Also, user can carry out the description of whole curve on the image of two dimension, on two dimensional image
Describe more directly perceived and facility, the precision of accepted description curve can be made further to improve.
Particularly, photography target be interlobar fissure partial disappearance it is incomplete between split in the case of, doctor can be according to warp
Test higher in definition and curve description is more intuitively carried out on two dimension slicing, even if so as to interlobar fissure disappearance or non-norm
Paste, also can substantially determine the position of interlobar fissure, and then it is contemplated that multiple sections on axial direction on diverse location
Split curve, further improve the description precision of interlobar fissure.
(variation)
In the first embodiment, selector 10 selects be parallel to each other as sagittal plane 4 in 3-D view to cut
Piece is cut into slices as crucial, and still, the quantity of key section is not limited to four, can arbitrarily set.For example, selector 10 selects
Following multiple sections include not wrapping in the section close with the section comprising the arch of aorta as crucial section, the plurality of section
Section, the area in lung region containing the arch of aorta but comprising heart reach maximum section and with the section comprising heart
The section not comprising heart in adjacent section.In addition, for example, the selection of selector 10 approaches with the section comprising the arch of aorta
Section in do not include section, the Yi Jiyu that the arch of aorta but section, the area in lung region comprising heart reach maximum
The section not comprising heart in the adjacent section of section comprising heart is cut into slices as crucial, and then also selection does not include heart
And it is used as crucial cut into slices positioned at middle section in the section including lung.In addition, selected crucial section is not limited to swear
The face in other directions such as shape face or coronal-plane (coronal plane).
In addition, the situation for the multiple crucial sections for selecting to be parallel to each other although the description of selector 10, but selected pass
Key section is also not necessarily limited to be parallel to each other, and there may also be certain angle, in this case, extension 30 between neighboring slice
Both parallel as described above can be extended so that exist between some 3-D views it is overlapping, can also be in different height
Different extended boundaries or threshold value are set on degree, to carry out the generation of partial 3-D image.
In addition, in first embodiment, selector 10 selects multiple sections as crucial section, so that extension 30 is successively
The plurality of crucial section is extended.It however, it can be, selector 10 only selects a section to expand as crucial section
Exhibition portion 30 forms curved surface based on the curve being depicted in a crucial section on 3 d image data.That is, extension 30 also may be used
To be only by the situation of a crucial section extension.For example, only doctor is closed without integrally carrying out region segmentation to lung
A part of region of the lung of the heart carry out region segmentation just it is enough in the case of etc., can by only by a crucial section expansion
Reduce the operation that doctor is carried out.Or or, selector 10 first select that a crucial section march line drawing paints by
Reason, after the segmentation curve described in the crucial section selected to this of extension 30 is extended, selector 10 further according to
Describe for the first time and the result of extension, the further crucial section of selection second, by that analogy repeatedly aforesaid operations, until completing
Untill the extension of whole lung, in such manner, it is possible to avoid the disabled problem of key section of selection.So as to save system resource.
(second embodiment)
Second embodiment is that second implements on the basis of first embodiment with the difference of first embodiment
In mode, image processing apparatus 200 also has identification part 40.Below mainly for second embodiment and first embodiment
Difference illustrates, and suitably the repetitive description thereof will be omitted.
Fig. 4 is the structured flowchart for representing the image processing apparatus that second embodiment is related to.As shown in figure 4, image procossing
Device 200 includes selector 10, receiving unit 20 and extension 30 and identification part 40.Wherein, selector 10, receiving unit 20,
The effect of extension 30 is identical with first embodiment, so as to omit detailed description.
The species of curve of the identification part 40 to describing in crucial section is identified.That is, identification part 40 can identify by
The classification for the segmentation curve that reason portion 20 is accepted.Identification part 40 can be that by the circuit or software mould of function above
Block.
Illustrated by taking the anatomical structure of lung as an example.Fig. 6 is the schematic diagram of lung mechanics.Lung includes left lung and the right side
Lung, wherein, left lung includes upper lobe and lower lobe, and right lung includes upper lobe, middle lobe and lower lobe.In general, lung
Interlobar fissure include three classifications, specifically, including upper leaf and the left fissura obliqua pulmonis gap A of inferior lobe will be divided in left lung, and will
Right lung is divided into leaf, middle period, the right lung oblique segmentation gap C of inferior lobe and level and splits B.
For a user, it is comparatively laborious for each curve editing curve category, in the present embodiment, identifies
The automatic identification segmentation curve of portion 40 belongs to the classification of which kind of above-mentioned interlobar fissure, and the curve is belonged to corresponding interlobar fissure
Classification and prompted to user.
Identification part 40 can identify pulmo according to the curve described in the overall position of lung images, due to left lung
Classification including upper lobe and lower lobe, therefore interlobar fissure only has left fissura obliqua pulmonis gap A, and this is a kind of, therefore will be identified as left lung
The category division of curve of cut-off rule be left fissura obliqua pulmonis gap.
Because right lung includes upper lobe, middle lobe and lower lobe, thus the classification of interlobar fissure have horizontal fissure of right lung gap B and
Right lung oblique segmentation gap C both, therefore, identified in identification part 40 be right lung interlobar fissure in the case of, based on anatomical structure,
It further to identify is horizontal fissure of right lung gap B and right lung oblique segmentation gap to utilize the position distance of described segmentation curve and tracheae tree
C which kind of.
In addition it is also possible to be, identification part 40 is entered using the different characteristics between horizontal fissure of right lung gap and right lung oblique segmentation gap
Row automatic identification, for example, distinguishing horizontal fissure of right lung gap or right lung oblique segmentation gap according to the angle of inclination in crack.
In this second embodiment, the selector that selector 10 corresponds in claims, receiving unit 20 correspond to power
Receiving unit in sharp claim, the extension that extension 30 corresponds in claims, identification part 40 corresponds to claim
Identification part in book.The flow that the region segmentation carried out below in conjunction with the explanation image processing apparatus 200 of accompanying drawing 5 is handled.
Fig. 5 is the flow chart for representing the region segmentation processing that second embodiment is related to.As shown in figure 5, proceeding by
During segmentation, first, selector 10 automatically selects multiple sections as crucial section (step in the 3 d image data of lung
S501)。
The segmentation curve that different zones are split is described on the two-dimensional image that the key is cut into slices by doctor
(shown herein as the interlobar fissure of the lobe of the lung), so as to which receiving unit 20 accepts the description (step to the segmentation curve in key section
S502)。
Then, into step S503, the classification for the segmentation curve that identification part 40 is accepted to receiving unit 20 is identified simultaneously
The result of identification is corresponding with segmentation curve foundation.
Then, into step S504, extension 30 is by before and after the segmentation curve extension in key section to key section
Contiguous slices on, so as to form part 3-D view (step S505).By between the partial 3-D image and the leaf identified
The classification split is prompted to user together.
According to present embodiment, can also obtain and first embodiment identical technique effect.
Further, it is possible to the classification of automatic identification segmentation curve, so as to which doctor need not be manually entered classification and spend the time, energy
It is enough more efficiently to carry out region segmentation processing.
(the 3rd embodiment)
3rd embodiment is that the 3rd implements on the basis of first embodiment with the difference of first embodiment
In mode, image processing apparatus 300 also has display part 50 and combination section 60.In addition, in the receiving unit 20 of the 3rd embodiment
Processing a part of difference.Illustrated below mainly for the difference of the 3rd embodiment and first embodiment, and
It is appropriate that the repetitive description thereof will be omitted.
Fig. 7 is the structured flowchart for representing the image processing apparatus that the 3rd embodiment is related to.As shown in fig. 7, image procossing
Device 300 includes selector 10, receiving unit 20, extension 30, display part 50 and combination section 60.
Selector 10 collects from image collecting devices such as CT devices to being shot as the chest of photography target
In lung's 3 d image data, multiple sections are selected as crucial section, so as to the monitoring connected in image processing apparatus 300
It is subject to two-dimentional display using selected crucial section as key section two dimensional image in the display devices such as device.Selector 10 can be with
It is that by the circuit or software module of function above.
In addition, receiving unit 20 be used to accept successively on the two dimensional image of the crucial section selected by selector 10 to for
Distinguish the description that the segmentation curve of different zones is carried out.In the case where splitting to the lobe of the lung, the segmentation curve is represented lung
Part is cut into the interlobar fissure of multiple lobes of the lung.Receiving unit 20 can be that by the circuit or software module of function above.
The segmentation curve that extension 30 is accepted based on receiving unit 20, by the segmentation curve extension in crucial section to pass
It is corresponding adjacent point with segmentation curve so as to be formed on contiguous slices on the contiguous slices that key is cut into slices in adjacent prescribed limit
Curve is cut, using key section and contiguous slices, the portion formed respectively including the segmentation interface as three-dimensional interlobar fissure
Divide 3-D view.That is, extension 30 according to each crucial section by forming curved surface, so as to according to each crucial section generation shape
Into the partial 3-D view data for having curved surface, the partial 3-D view data is being cut into slices with key in 3 d image data entirety
And part corresponding to contiguous slices.Moreover, extension 30 is formed according to each crucial section is based on partial 3-D view data
Partial 3-D image.Extension 30 can be that by the circuit or software module of function above.
Display part 50 is shown to the 3-D view based on the 3 d image data formed with curved surface.For example, whenever certain
After the description of individual crucial section is completed, display part 50 makes all partial 3-D image real-time displays generated.That is, Mei Dang
When description has curve in key section, extension 30 is being forming curves on the contiguous slices of prescribed limit away from crucial section, from
And curved surface is formed on 3 d image data, and by display part 50 to formed with the range of curved surface based on 3 d image data
3-D view shown.Display part 50 can be that by the circuit or software module of function above.
View data is mutually combined by combination section 60, specifically, for combining some generated by extension 30
3-D view, so as to form complete lung mechanics.Particularly, narrower in set spreading range and formation partial 3-D
Between image and it is discontinuous in the case of, i.e., the part when combination section 60 is combined as lung's entirety 3-D view
In the case of space being present between partial 3-D image, combination section 60 fills above-mentioned space by interpolation, complete so as to be formed
Lung's 3-D view.That is, the one side of combination section 60 carries out interpolation to the space between partial 3-D view data, while will be according to
The partial 3-D view data that each key is cut into slices and generated is combined.Combination section 60 can be that by function above
Circuit or software module.
For example, the set half for axially expanding the distance between the surface that distance is shorter than between crucial section, so as to neighbouring
The contiguous slices that two crucial sections extend respectively can not link up, and the partial 3-D image formed has space.This
In the case of, combination section 60 to the space between two partial 3-D images by carrying out interpolation, to connect different partial 3-Ds
Image, so as to form lung's 3-D view including complete segmentation interface (interlobar fissure) to be prompted to user.The mode of interpolation
Existing various interpolating methods can be used.The extended method that can also be illustrated by first embodiment, from partial 3-D
The edge section of image starts further to be extended to space to plug the gap.
Receiving unit 20 accepts the amendment for forming the curved surface on 3 d image data.That is, receiving unit 20 can be accepted
The amendment that user is carried out on the 3-D view including segmentation interface to segmentation interface.Receiving unit 20 can either be accepted to aobvious
Show the amendment that the segmentation interface on the partial 3-D image of the real-time display of portion 50 is carried out, can also accept and combination section 60 is generated
Complete three dimensional image in segmentation interface carry out amendment.
The partial 3-D image of real-time display is abandoned indicating in addition, receiving unit 20 can also accept user, so as to by
Reason portion 20 accepts the description to splitting curve again.Receiving unit 20 can be that by the circuit or software mould of function above
Block.
Receiving unit 20 can accept it is above-mentioned abandon instruction when or, selector 10 is previously stored with a variety of keys
Section selection mode, when the 3 d image data for same photography target regenerates segmentation curve, switch different passes
Key section selection mode.Specifically, selector 10 stores multiple positions of crucial section based on past selection etc. first
Selection mode.Next, the defined selection mode in multiple selection modes of the selector 10 based on storage, selects multiple keys
Section.Then, in 10 forming curves again of selector, based on different from defined selection mode in multiple selection modes
Selection mode, select multiple crucial sections.
In the third embodiment, the selector that selector 10 corresponds in claims, receiving unit 20 correspond to power
Receiving unit in sharp claim, the extension that extension 30 corresponds in claims, identification part 40 corresponds to claim
Identification part in book, the display part that display part 50 corresponds in claims, combination section 60 correspond in claims
Combination section.Below in conjunction with Fig. 8 and Fig. 9, illustrate the flow that the region segmentation that image processing apparatus 300 is carried out is handled.
Fig. 8 is the flow chart for representing the region segmentation processing that the 3rd embodiment is related to.As shown in figure 8, proceeding by
During segmentation, first, selector 10 automatically selects multiple sections as crucial section (step in the 3 d image data of lung
S801)。
As one, the situation that dividing processing is carried out to the 3-D view to left lung shown in Fig. 9 illustrates.Such as Fig. 9
Shown, selector 10 selects four sagittal plane being parallel to each other sections to be cut as key successively in the 3 d image data of left lung
Piece 1, crucial section 2, crucial section 3, crucial section 4.Wherein, the three-dimensional under the visual angle for sagittal plane is shown on the left of Fig. 9
Image, selector 10 select four sagittal plane being parallel to each other sections as crucial section in the 3-D view.In Fig. 9
Centre shows each crucial section, also, show in the lower section of each crucial section key cut into slices in 3-D view and
The relative distance between position and crucial section in the vertical axial direction of sagittal plane.
Show the two-dimensional image of crucial section 1 on the display apparatus first, make user on the two-dimensional image
Depict and split curve M1 as the black line shown on the crucial section 1 in Fig. 9, so as to which receiving unit 20 is accepted to the key
The description (step S802) of segmentation curve M1 in section 1.
Then, into step S803, extension 30 by the key cut into slices on segmentation curve M1 expand to crucial section before
It is corresponding with key section 1 under the visual angle for coronal-plane as shown in Fig. 9 top so as to be formed on contiguous slices afterwards
Partial 3-D image.In the partial 3-D image, as the boundary of different zones, with different gray scales or color displays not same district
Domain, so as to as shown in figure 9, also can clearly distinguish the position at segmentation interface in 3-D view.
Display part 50 makes the partial 3-D image real-time display, and doctor can be by the part three that shows on the display apparatus
The segmentation result for the partial 3-D image that dimension image confirming is generated.
Then, display part 50 shows 2 (step S804) of next crucial section, includes in the extension section of crucial section 1
In the case of next crucial section 2, the segmentation expanded from crucial section 1 can be also shown in shown crucial section 2
Curve, doctor can judge whether the segmentation curve that expands and whether receive the pass according to the display of display part 50
Spreading result (step S805) in key section 2.
Crucial section 2 it is not related in the extension of crucial section 1, i.e., it is bent not form extension in crucial section 2
In the case of line, or in the case that receiving unit 20 accepts the instruction for not receiving the spreading result (step S805 is "No"),
Return to step S802, receiving unit 20 accept the description to the segmentation curve M2 of key section 2.
Receiving unit 20 accept receive the instruction of spreading result in the case of (step S805 is "Yes"), proceed to step
S806, display device show next crucial section 3, so as to processing procedure for crucial section 3 return to step S802, accept pair
The segmentation curve M3 of key section 3 description.Formed and key 3 corresponding partial 3-D images of section.
When being shown to key section 2 or crucial section 3, display part 50 can also make and crucial section 2 or crucial
Partial 3-D image real-time display corresponding to section 3, and the corresponding partial 3-D image of each crucial section with showing before
Show simultaneously, so as to form the partial 3-D image as shown in Fig. 9 tops.
In the example shown in Fig. 9, it is assumed that for key section 2 and crucial section 3, receiving unit 20, which is all accepted, not to be received
Instruction (the step S805 of spreading result：It is no), so as to be painted to key 2,3 march line drawings of section, also, the expansion of crucial section 3
With the opening up range set larger lung edge so as to extend to including crucial section 4, it is only aobvious in the example shown in Fig. 9
Show the extension segmentation curve of crucial section 4, the instruction for whether receiving extension segmentation curve directly accepted by receiving unit 20, with
In the case that family receives extension segmentation curve, no longer key 4 march line drawings of section are painted.
By handling successively each crucial section as described above, far right images institute above such as Fig. 9 of display part 50
Show the form after the merging of real-time display various pieces 3-D view.
And then combination section 60 carries out the combination adjustment of three partial 3-D images, by carrying out interpolation to gap, or it is right
Overlapping part is overlapped processing, and the segmentation interface formed is finely adjusted, puts down the border between partial 3-D image
Along connection (step S807).Now, user can also be modified by receiving unit 20 to the segmentation interface after connection.So as to
3-D view to after the segmentation as shown in Fig. 9 rightmost sides.
In addition, in flow example more than, after the segmentation curve in each crucial section is extended respectively
Be combined adjustment, but naturally it is also possible to often generate partial 3-D image be therewith previous existence into 3-D view carry out
Combination adjustment, so as to which user can be intuitive to see the segmentation backward three-dimensional viewing that each stage combination finishes.
According to present embodiment, can also obtain and first embodiment identical technique effect.
In addition, according to present embodiment, can be in real time to partial 3-D figure by combining display part 50 and receiving unit 20
As being modified, so, can show the 3-D view in a manner of three-dimensional is confirmed doctor on the display apparatus,
When doctor does not receive the spreading result at the segmentation interface on the partial 3-D image, corresponding crucial section can be carried out again
Describe, so as to which the segmentation curve that extension 30 is again based on newly describing is extended.
In addition, according to present embodiment, especially for the lung images that interlobar fissure is not complete, doctor can be made multiple two
Described and judged in dimension section, interface is split according to combined result real time modifying, even if so as to which part interlobar fissure is invisible,
Also relatively accurate segmentation result can be obtained.
In addition, according to present embodiment, it can judge that being unsatisfied with some key in user cuts to each crucial section
During the analog result of piece, the two dimension description of interlobar fissure is re-started, with completing to reform again after the segmentation on whole 3-D view
Prior art compare, calculating cost can be saved in the present embodiment, being capable of more efficient Ground Split difference lobe of the lung area
Domain.
(other variations)
In the embodiment that the application is related to, the tissue segmentation of photography target can be provided a kind of more clear and intuitively
Instruct.User can complete editing with less action, even for the lung that interlobar fissure is not complete, can also improve
The precision of segmentation.But the embodiment that the application is related to is not limited to each embodiment of mistake described above, can also carry out
Various deformation.
For example, in embodiment of above, region segmentation processing is carried out using display part 50 and combination section 60, but
Any part therein can be omitted.Only apply a certain item function therein.In addition, for example, receiving unit 20 can also be not bound by
Amendment for forming the curved surface on 3 d image data.
In addition, between in the above-described embodiment, including lung with photography target, the different lobes of the lung being carried out for lung images
Segmentation (extraction interlobar fissure) processing exemplified by be illustrated, here, the 3 d image data of embodiment can include lung
Whole, the part of lung can also be included.For example, 3 d image data can also only include right lung.In addition, but the application relate to
And embodiment can also be used to segmentation and include other organs of multiple subregions.Such as when suitable for the segmentation of liver, energy
The enough anatomy according to liver and geometric information are split.That is, 3 d image data can also include the complete of liver
Portion or a part.
And then the embodiment that the application is related to also can apply to extract the region carried out during whole organ in the picture
In dividing processing.Such as it also can be applied to extract the situation of whole lung or whole heart respectively on the CT images of chest.This
When the segmentation curve depicted can be a closed contour that surround organ.
The image processing apparatus that embodiment is related to can also be as the work(illustrated by can realizing in each embodiment
The circuit of energy is arranged in medical equipment, can also be stored in disk (floppy disk as the program that can perform computer
(floppy, log in trade mark), hard disk etc.), CD (CD-ROM, DVD etc.), photomagneto disk (MO), the storage such as semiconductor memory be situated between
Matter and issue.
Moreover, OS (the operation systems that the instruction based on the program that computer is installed on from storage medium operates on computers
System), database management language, the MW (middleware) etc. of network software etc. can also be performed for realizing each of above-mentioned embodiment
A part for processing.
Several embodiments of the invention is explained above, but these embodiments are to propose as an example, not
It is intended to limit invention scope.These new embodiments can be implemented in a manner of other are various, can not depart from hair
Various omission, substitution, and alteration are carried out in the range of bright purport.These embodiments or its deformation are included in invention model
Enclose or purport in, and be also contained in described in right invention and its equalization scope in.
Claims (16)
1. a kind of image processing apparatus, wherein, including：
Selector, the section as defined in selection from 3 d image data；And
Extension, formed and be depicted on away from the section that above-mentioned defined section is prescribed limit on above-mentioned defined section
Curve corresponding to curve, so as to form curved surface on above-mentioned 3 d image data.
2. image processing apparatus according to claim 1, wherein,
The point of specified quantity is selected on the curve that above-mentioned extension is described on above-mentioned defined section, in above-mentioned prescribed limit
Point corresponding with the point selected is calculated on section, and the point calculated is linked, so as to the shape on the section of above-mentioned prescribed limit
Into curve.
3. image processing apparatus according to claim 1, wherein,
It is also equipped with accepting the receiving unit of the amendment for above-mentioned curved surface.
4. image processing apparatus according to claim 1, wherein,
It is also equipped with the identification part that the species of the curve to being depicted on above-mentioned defined section is identified.
5. image processing apparatus according to claim 1, wherein,
Display part is also equipped with, the display part is carried out to the 3-D view based on the above-mentioned 3 d image data formed with above-mentioned curved surface
It has been shown that,
The above-mentioned multiple above-mentioned defined sections of selector selection,
Whenever when description has curve on above-mentioned defined section, above-mentioned display part is to formed with upper in the range of above-mentioned curved surface
3-D view is stated to be shown.
6. image processing apparatus according to claim 1, wherein,
Multiple above-mentioned defined sections that above-mentioned selector selection is parallel to each other.
7. image processing apparatus according to claim 1, wherein,
Above-mentioned selector selection is parallel with sagittal plane and has multiple above-mentioned defined sections at certain interval.
8. image processing apparatus according to claim 1, wherein,
Above-mentioned 3 d image data includes all or part of lung.
9. image processing apparatus according to claim 8, wherein,
Above-mentioned selector selects from the section parallel to sagittal plane in above-mentioned 3 d image data：With including the arch of aorta
Section that not including in the close section of section, the arch of aorta and section, the area in lung region comprising heart reached maximum,
And the section not comprising heart in the section abutted with the section comprising heart, as above-mentioned defined section.
10. image processing apparatus according to claim 9, wherein,
Above-mentioned selector also selects the section positioned at centre in the section not comprising heart and comprising lung as above-mentioned defined
Section.
11. image processing apparatus according to claim 1, wherein,
Above-mentioned 3 d image data includes all or part of liver.
12. image processing apparatus according to claim 1, wherein,
The combination section for being mutually combined view data is also equipped with,
The above-mentioned multiple above-mentioned defined sections of selector selection,
Above-mentioned extension by according to it is each it is above-mentioned as defined in section form above-mentioned curved surface, so as to according to it is each it is above-mentioned as defined in cut open
Look unfamiliar and be shaped as the partial 3-D view data of above-mentioned curved surface, the partial 3-D view data is above-mentioned 3 d image data
Part corresponding with the section of above-mentioned defined section and above-mentioned prescribed limit in entirety,
Combinations thereof portion to the space between above-mentioned partial 3-D view data while carry out interpolation, while will pass through above-mentioned extension
Multiple above-mentioned partial 3-D view data combinations of portion's generation.
13. image processing apparatus according to claim 1, wherein,
Above-mentioned selector is based on defined selection mode and selects multiple above-mentioned defined sections, when forming above-mentioned curved surface again,
Based on the selection mode selection multiple above-mentioned defined sections different from above-mentioned defined selection mode.
14. image processing apparatus according to claim 2, wherein,
Above-mentioned extension straight line links two points of the adjoining in the above-mentioned point calculated, so as in above-mentioned prescribed limit
Forming curves on section.
15. image processing apparatus according to claim 2, wherein,
The above-mentioned extension water channel principium relevant with pixel value links two of the adjoining in the above-mentioned point calculated
Point, so as to the forming curves on the section of above-mentioned prescribed limit.
16. a kind of image processing method, wherein, comprise the following steps：
The section as defined in selection from 3 d image data；And
By forming and being depicted in the song on above-mentioned defined section on away from the section that above-mentioned defined section is prescribed limit
Curve corresponding to line, so as to form curved surface on above-mentioned 3 d image data.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2016106624826 | 2016-08-12 | ||
| CN201610662482 | 2016-08-12 | ||
| JP2016239517A JP6900180B2 (en) | 2016-08-12 | 2016-12-09 | Image processing device and image processing method |
| JP2016-239517 | 2016-12-09 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107730505A true CN107730505A (en) | 2018-02-23 |
Family
ID=61194638
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710337843.4A Pending CN107730505A (en) | 2016-08-12 | 2017-05-15 | Image processing apparatus and image processing method |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP6900180B2 (en) |
| CN (1) | CN107730505A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109598734A (en) * | 2018-12-29 | 2019-04-09 | 上海联影智能医疗科技有限公司 | The method and system of heart and lobe of the lung segmentation |
| CN109658425A (en) * | 2018-12-12 | 2019-04-19 | 上海联影医疗科技有限公司 | A kind of lobe of the lung dividing method, device, computer equipment and storage medium |
| CN114332087A (en) * | 2022-03-15 | 2022-04-12 | 杭州电子科技大学 | Three-dimensional cortical surface segmentation method and system for OCTA image |
| US11436720B2 (en) | 2018-12-28 | 2022-09-06 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for generating image metric |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101326547A (en) * | 2005-10-12 | 2008-12-17 | 卡尔斯特里姆保健公司 | Method and system for processing an image of body tissues |
| CN102446349A (en) * | 2010-09-29 | 2012-05-09 | 奥林巴斯株式会社 | Image processing apparatus and image processing method |
| CN103841894A (en) * | 2011-04-18 | 2014-06-04 | 开创治疗股份有限公司 | Image segmentation of organs and anatomical structures |
| CN103927732A (en) * | 2013-01-11 | 2014-07-16 | 上海联影医疗科技有限公司 | Method for detecting chest wall lines |
| CN103959331A (en) * | 2011-10-11 | 2014-07-30 | 皇家飞利浦有限公司 | A workflow for ambiguity guided interactive segmentation of lung lobes |
| CN103971372A (en) * | 2014-05-21 | 2014-08-06 | 南华大学 | Fast artificial bone forming method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0728976A (en) * | 1993-07-07 | 1995-01-31 | Toshiba Corp | Picture display device |
| EP2244633A4 (en) * | 2008-02-14 | 2013-10-23 | Penn State Res Found | Medical image reporting system and method |
| JP5194138B2 (en) * | 2011-03-02 | 2013-05-08 | 富士フイルム株式会社 | Image diagnosis support apparatus, operation method thereof, and image diagnosis support program |
| US20130211231A1 (en) * | 2012-02-14 | 2013-08-15 | Manivannan Sundarapandian | Method and system for visualization of treatment volumes |
| US10991102B2 (en) * | 2016-08-12 | 2021-04-27 | Canon Medical Systems Corporation | Image processing apparatus and image processing method |
-
2016
- 2016-12-09 JP JP2016239517A patent/JP6900180B2/en active Active
-
2017
- 2017-05-15 CN CN201710337843.4A patent/CN107730505A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101326547A (en) * | 2005-10-12 | 2008-12-17 | 卡尔斯特里姆保健公司 | Method and system for processing an image of body tissues |
| CN102446349A (en) * | 2010-09-29 | 2012-05-09 | 奥林巴斯株式会社 | Image processing apparatus and image processing method |
| CN103841894A (en) * | 2011-04-18 | 2014-06-04 | 开创治疗股份有限公司 | Image segmentation of organs and anatomical structures |
| CN103959331A (en) * | 2011-10-11 | 2014-07-30 | 皇家飞利浦有限公司 | A workflow for ambiguity guided interactive segmentation of lung lobes |
| CN103927732A (en) * | 2013-01-11 | 2014-07-16 | 上海联影医疗科技有限公司 | Method for detecting chest wall lines |
| CN103971372A (en) * | 2014-05-21 | 2014-08-06 | 南华大学 | Fast artificial bone forming method |
Non-Patent Citations (1)
| Title |
|---|
| 曾峦等: "《侦查图像获取与融合技术》", 31 May 2015, 国防工业出版社 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109658425A (en) * | 2018-12-12 | 2019-04-19 | 上海联影医疗科技有限公司 | A kind of lobe of the lung dividing method, device, computer equipment and storage medium |
| CN109658425B (en) * | 2018-12-12 | 2021-12-28 | 上海联影医疗科技股份有限公司 | Lung lobe segmentation method and device, computer equipment and storage medium |
| US11436720B2 (en) | 2018-12-28 | 2022-09-06 | Shanghai United Imaging Intelligence Co., Ltd. | Systems and methods for generating image metric |
| CN109598734A (en) * | 2018-12-29 | 2019-04-09 | 上海联影智能医疗科技有限公司 | The method and system of heart and lobe of the lung segmentation |
| CN114332087A (en) * | 2022-03-15 | 2022-04-12 | 杭州电子科技大学 | Three-dimensional cortical surface segmentation method and system for OCTA image |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2018023709A (en) | 2018-02-15 |
| JP6900180B2 (en) | 2021-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6129310B2 (en) | Image processing apparatus and image processing method | |
| CN106663309B (en) | Method and storage medium for user-guided bone segmentation in medical imaging | |
| US10991102B2 (en) | Image processing apparatus and image processing method | |
| JP5814853B2 (en) | Stereo model data generation apparatus and method, and program | |
| US8532359B2 (en) | Biodata model preparation method and apparatus, data structure of biodata model and data storage device of biodata model, and load dispersion method and apparatus of 3D data model | |
| CN104573309B (en) | Device and method for computer-aided diagnosis | |
| JP2020512129A (en) | System and method for synthesizing low-dimensional image data from high-dimensional image data using object grid augmentation | |
| CN107730505A (en) | Image processing apparatus and image processing method | |
| US20070116334A1 (en) | Method and apparatus for three-dimensional interactive tools for semi-automatic segmentation and editing of image objects | |
| JP5399225B2 (en) | Image processing apparatus and method, and program | |
| US9129391B2 (en) | Semi-automated preoperative resection planning | |
| US9547906B2 (en) | System and method for data driven editing of rib unfolding | |
| CN103679810A (en) | Method for three-dimensional reconstruction of liver computed tomography (CT) image | |
| CN111667491B (en) | Breast tumor block diagram generation method with boundary marking information based on depth countermeasure network | |
| JP5946029B2 (en) | Tooth graph cut-based interactive segmentation method in 3D CT volumetric data | |
| JP2015513945A6 (en) | Tooth graph cut-based interactive segmentation method in 3D CT volumetric data | |
| JP5105997B2 (en) | Medical image processing apparatus, medical image diagnostic apparatus, and program | |
| Nowinski et al. | A 3D model of human cerebrovasculature derived from 3T magnetic resonance angiography | |
| CN107106110A (en) | Image processing apparatus and image processing method | |
| CN107077718A (en) | Reformatted when considering the anatomical structure of object to be checked | |
| EP3933848A1 (en) | Vrds 4d medical image processing method and product | |
| CN106920282B (en) | Editing method and system of blood vessel digital model | |
| Choi et al. | Automatic initialization active contour model for the segmentation of the chest wall on chest CT | |
| JP6061991B2 (en) | Information processing apparatus, information processing method, information processing system, program, and storage medium | |
| Gillmann et al. | Visual exploration in surgery monitoring for coronary vessels |
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 |