CN106952273A - The dividing method and device of pancreas in medical image - Google Patents
The dividing method and device of pancreas in medical image Download PDFInfo
- Publication number
- CN106952273A CN106952273A CN201710138713.8A CN201710138713A CN106952273A CN 106952273 A CN106952273 A CN 106952273A CN 201710138713 A CN201710138713 A CN 201710138713A CN 106952273 A CN106952273 A CN 106952273A
- Authority
- CN
- China
- Prior art keywords
- pancreas
- medical image
- belly
- segmentation result
- area
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- 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
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses the dividing method and device of pancreas in a kind of medical image, it the described method comprises the following steps:The belly medical image of those who are investigated is obtained, and the belly medical image includes pancreas；Determine the area-of-interest for including pancreas tail end in belly medical image；Pancreas in area-of-interest is split on cross section, initial segmentation result is obtained；Since pancreas tail end, the part pancreas in initial segmentation result is split in sagittal plane, the start reference layer for the segmentation of the pancreas in belly medical image is obtained；From start reference layer respectively to pancreas head end and pancreas tail end, the pancreas in belly medical image is split in sagittal plane, the pancreas segmentation result in belly medical image is obtained.Methods and apparatus of the present invention can realize that pancreas is full-automatic, precisely split.
Description
【Technical field】
The present invention relates to field of medical images, more particularly, to a kind of dividing method of pancreas in medical image.
【Background technology】
Computer aided radio therapy is one of three big therapies of current main flow, one of its successful treatment cancer substantially before
It is to confirm knub position and size to carry, while protecting perilesional critical organ.Therefore, accurate high efficiency extraction critical organ profile
It is the committed step of adaptive radiation therapy and surgical navigational etc. with Gross tumor volume (GTV, gross tumor volume),
With very important Research Significance.Pancreas and liver, stomach, small intestine, spleen etc. it is in close relations, once one of swell
Knurl is easy to occur tumor invasion etc., moreover, inherently a kind of grade malignancy of cancer of pancreas is very high, diagnosis and treatment are all highly difficult
Malignant tumor of digestive tract.Therefore, for the patient that lesion occurs for belly, the accurate high efficiency extraction of pancreas is significant.Pancreas
The diagnosis rate of gland cancer early stage is not high, and operative mortality rate is higher, and cure rate is very low.Because the span of pancreas is larger, and with surrounding
Connected close, the picture contrast bottom around organ of histoorgan, different patient's pancreas sizes, differences in shape all very greatly, essence
Accurate splitting will be especially difficult.
【The content of the invention】
The technical problems to be solved by the invention are to provide the dividing method and device of pancreas in a kind of medical image, and obtain
The segmentation result taken has the higher degree of accuracy.
The present invention is for the technical scheme that uses of solution above-mentioned technical problem：The segmentation side of pancreas in a kind of medical image
Method, it comprises the following steps:The belly medical image of those who are investigated is obtained, and the belly medical image includes pancreas；It is determined that
The area-of-interest for including pancreas tail end in belly medical image；Pancreas in area-of-interest is divided on cross section
Cut, obtain initial segmentation result；Since pancreas tail end, the part pancreas in initial segmentation result is divided in sagittal plane
Cut, obtain the start reference layer for the segmentation of the pancreas in belly medical image；From start reference layer respectively to pancreas head end
With pancreas tail end, the pancreas in belly medical image is split in sagittal plane, the pancreas in belly medical image is obtained
Segmentation result.
Optionally, the area-of-interest only includes part pancreas, and the pancreas tail end is according in belly medical image
Spleen positioning or manually locate.
Optionally, the part pancreas in area-of-interest is split on cross section, obtains initial segmentation knot
Really, including：Pancreas in area-of-interest is successively split on cross section, several cross-sectional images are obtained；From cross-sectional view
Connected domain to be selected is determined as in；Pancreas reference layer is determined in connected domain to be selected.
Optionally, it is described to determine connected domain to be selected from cross-sectional image, including：To the pancreas in area-of-interest
Each cross-sectional image enters row threshold division, obtains the connected domain in the range of given threshold；Or/and remove close to cross section
The connected domain of image top edge, lower edge or right hand edge；Or/and remove the connected domain that length-width ratio in cross-sectional image is less than 1.
Optionally, the determination pancreas reference layer in connected domain to be selected, including：Retain bag in connected domain to be selected
The connected domain on the border containing Far Left, selects cross section most long connected domain from the connected domain of reservation.
Optionally, it is described since pancreas tail end, the part pancreas in initial segmentation result is divided in sagittal plane
Cut, obtain the start reference layer for the segmentation of the pancreas in belly medical image, including：To the part in initial segmentation result
Pancreas is successively split in sagittal plane, obtains segmentation result again；Retain in segmentation result again has friendship with initial segmentation result
The connected domain of collection；The maximum connected domain of area is chosen from the connected domain for having common factor；To the boundary point of the maximum connected domain of area
Number is counted, and is chosen border and is counted most connected domains as start reference layer.
Optionally, it is described from start reference layer respectively to pancreas head end and pancreas tail end, to the pancreas in belly medical image
Gland is split in sagittal plane, including：Start reference layer is expanded to the coarse segmentation result after first size as current layer；System
The histogram distribution of coarse segmentation result corresponding grey scale figure is counted, adding and subtracting twice of variance using the average of statistics is used as current layer pancreas
Threshold range, the mark being not belonging in the threshold range is；The coarse segmentation result of current layer is expanded into the second size
Afterwards as the area-of-interest of current layer pancreas；Centered on the center of gravity of coarse segmentation result, pancreas boundary point is traveled through with ray, is penetrated
The place that line runs into dark areas is pancreas border, and the part without dark areas is left a blank, to part adjacent two border difference of leaving a blank,
It is exactly the current layer pancreas segmentation result to obtain closed curve.
Optionally, also including being rotated to the belly medical image, and the belly medical image is rotated
Angle be with horizontal line be in 15 ° -45 °, the medical image be 3D CT images or 3D MR images.
The present invention is for the technical scheme that uses of solution above-mentioned technical problem：The segmentation of pancreas in a kind of medical image
Device, described device includes processor, and it includes:Image acquisition unit, the belly medical image for obtaining those who are investigated, and
The belly medical image includes pancreas；Area-of-interest determining unit, for determining to include pancreas in belly medical image
The area-of-interest of tail end；First cutting unit, for splitting to the pancreas in area-of-interest on cross section, is obtained
Initial segmentation result；Second cutting unit, for since pancreas tail end, to the part pancreas in initial segmentation result in sagittal
Split on face, obtain the start reference layer for the segmentation of the pancreas in belly medical image；From start reference layer respectively
To pancreas head end and pancreas tail end, the pancreas in belly medical image is split in sagittal plane, belly medical science figure is obtained
Pancreas segmentation result as in.
Optionally, the belly medical image is 3D CT images or 3D MR images, can be set by computer tomography
Standby or MR imaging apparatus performs scanning to the belly for being scanned object and obtained.
Present invention contrast prior art has following beneficial effect：The present invention takes full advantage of arrow of the pancreas in medical image
Shape face is continuous and adjacent layer changes slow knowledge, has used in medical image pancreas in cross section and the difference of sagittal plane
Feature, is used respectively, and finally both direction information is carried out comprehensively, to make use of the dark space around pancreas on medical image
Domain, operative constraint pancreas border can be achieved that pancreas is full-automatic, precisely segmentation.
【Brief description of the drawings】
Technical scheme in order to illustrate more clearly the embodiments of the present invention, below will be attached to what is used needed for embodiment
Figure is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for this area
For those of ordinary skill, without having to pay creative labor, it can also be obtained according to these accompanying drawings other attached
Figure.
Fig. 1 is the structural representation of the computer equipment of some embodiments of the invention；
Fig. 2 is the processor structure schematic diagram of the embodiment of the present invention；
Fig. 3 for the embodiment of the present invention a kind of medical image in pancreas segmentation method flow chart；
Fig. 4 is that belly medical image rotates forward and backward schematic diagram；
Fig. 5 is segmentation result of the pancreas on cross section in area-of-interest；
Fig. 6 is segmentation result of the pancreas in sagittal plane in area-of-interest；
Fig. 7 for the embodiment of the present invention medical image in pancreas segmentation result.
【Embodiment】
In order to be better understood from technical scheme, the embodiment of the present invention is retouched in detail below in conjunction with the accompanying drawings
State.
It will be appreciated that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art obtained under the premise of creative work is not made it is all its
Its embodiment, belongs to the scope of protection of the invention.
The term used in embodiments of the present invention is the purpose only merely for description specific embodiment, and is not intended to be limiting
The present invention." one kind ", " described " and "the" of singulative used in the embodiment of the present invention and appended claims
It is also intended to including most forms, unless context clearly shows that other implications.
It should be appreciated that term "and/or" used herein is only a kind of incidence relation for describing affiliated partner, represent
There may be three kinds of relations, for example, A and/or B, can be represented：Individualism A, while there is A and B, individualism B these three
Situation.In addition, character "/" herein, it is a kind of relation of "or" to typically represent forward-backward correlation object.
Flow chart is used herein to be used for illustrating the operation performed by system according to an embodiment of the invention.It should manage
Solution, before or operation below not necessarily accurately carry out in sequence.On the contrary, can handle each according to inverted order or simultaneously
Plant step.It is also possible to which other operations are added to during these, or a certain step or number step behaviour are removed from these processes
Make.
Fig. 1 is the structural representation of the computer equipment of some embodiments of the invention.Computer 100 can be used to realize
Implement the ad hoc approach and device disclosed in some embodiments of the invention.Specific device in the present embodiment utilizes functional block diagram exhibition
A hardware platform for including display module is shown.In certain embodiments, computer 100 can be by its hardware device, soft
Part program, firmware and combinations thereof realize the specific implementation of some embodiments of the invention.In certain embodiments, calculate
Machine 100 can be the computer of a general purpose, or a computer for having a specific purpose.
As shown in figure 1, computer 100 can include internal communication bus 101, processor (processor) 102 is read-only
Memory (ROM) 103, random access memory (RAM) 104, COM1 105, input output assembly 106, hard disk 107, with
And user interface 108.Internal communication bus 101 can realize the data communication of the inter-module of computer 100.Processor 102 can be with
Calculated, judged and sent prompting.In certain embodiments, processor 102 can be made up of one or more subelements, and
It is respectively completed corresponding processing function.COM1 105 can realize computer 100 and miscellaneous part (not shown) example
Such as：Enter row data communication between external equipment, image capture device, database, external storage and image processing workstations etc..
In certain embodiments, computer 100 can send and receive information and data by COM1 105 from network.Input/defeated
The input/output data stream gone out between the support computer 100 of component 106 and miscellaneous part.User interface 108 can realize calculating
Interaction and information between machine 100 and user are exchanged.Computer 100 can also include various forms of program storage units and
Data storage element, such as hard disk 107, read-only storage (ROM) 103, random access memory (RAM) 104 can store meter
The various data files that the processing of calculation machine and/or communication are used, and the possible programmed instruction performed by processor 102.
As shown in figure 1, processor 102 includes image data acquisition unit 1021, area-of-interest determining unit 1022, the
One cutting unit 1023 and the second cutting unit 1024.Image data acquisition unit 1021 is adopted with image processing workstations or image
Collect equipment connection, the belly medical image for obtaining those who are investigated, and the belly medical image includes pancreas.Region of interest
Domain determining unit 1022 is connected with image data acquisition unit 1021, for automatically determining to include pancreas in belly medical image
The area-of-interest of gland tail end.First cutting unit (cross section cutting unit) 1023 connects with area-of-interest determining unit 1022
Connect, for splitting to the part pancreas in area-of-interest on cross section, obtain initial segmentation result.Second segmentation is single
First (sagittal plane cutting unit) 1024 is connected with the first cutting unit 1023, for existing to the part pancreas in initial segmentation result
Split in sagittal plane, obtain the start reference layer for the segmentation of the pancreas in belly medical image；Then from starting ginseng
Layer is examined respectively to pancreas head end and pancreas tail end, carries out (fine) segmentation in sagittal plane to the pancreas in belly medical image,
Obtain the pancreas segmentation result in belly medical image.
Alternatively, processor 102 can be specialized application integrated circuit (Application Specific
Integrated Circuit, ASIC), dedicated instruction processor (Application Specific Instruction Set
Processor, ASIP), concurrent physical processor (Physics Processing Unit, PPU), digital signal processor
(Digital Processing Processor, DSP), field programmable gate array (Field-Programmable
Gate Array, FPGA), one kind or several in PLD (Programmable Logic Device, PLD) etc.
The combination planted, for performing the segmentation to pancreas in medical image.
Fig. 3 be some embodiments of the invention medical image in pancreas dividing method flow chart.Referring to Fig. 3, perform
Step S21, obtains the belly medical image of those who are investigated, and the belly medical image includes pancreas；Step S22 is performed, really
Determine the area-of-interest for including pancreas tail end in belly medical image；Step 23 is performed, to the part pancreas in area-of-interest
Gland is split on cross section, obtains initial segmentation result；Step 24 is performed, since pancreas tail end, to initial segmentation knot
Part pancreas in fruit carries out (thick) segmentation in sagittal plane, obtains the starting of the segmentation for the pancreas in belly medical image
Reference layer；Step 25 is performed, from start reference layer respectively to pancreas head end and pancreas tail end, to the pancreas in belly medical image
(fine) segmentation is carried out in sagittal plane, the pancreas segmentation result in belly medical image is obtained.
The belly medical image is 3D CT images or 3D MR images, can by computer tomography equipment (CT) or
The image capture devices such as person's MR imaging apparatus (MR) perform scanning to the belly for being scanned object and obtained, can also be from
The view data obtained with image capture device is stored in data storage cell.The belly medical image includes pancreas, spleen
Dirty, liver and kidney and other organs information.
For the ease of the segmentation to image, belly medical image can be pre-processed, be had based on pancreas unique special
Property, it is elongated, extend since spleen to liver direction, and extension trend somewhat compares inclined；And utilize pancreas
This characteristic, the torsion of certain angle is done to belly medical image, makes pancreas in image by horizontal pendulum position or substantially water
Yaw position.Optionally, the angle rotated to the belly medical image is in 15 ° -45 ° with horizontal line.The present invention is implemented
The anglec of rotation used is in 30 ° with horizontal line in example.By taking a patient data as an example, Fig. 4 is that the rotation of belly medical image is forward and backward
Schematic diagram.
The area-of-interest only includes part pancreas (since the part pancreas section pancreas tail end), and the cauda pancreatis
Hold the spleen positioning in belly medical image or manually locate.For example, can be done based on spleen information one it is less
There is one section of pancreas in frame constituency (bounding box), the frame favored area, and include pancreas tail end.Input module can also be passed through
(such as mouse) directly determines above-mentioned frame favored area.
The step S23 is split on cross section to the part pancreas in area-of-interest, obtains initial segmentation knot
Really, including following sub-step：
Step S231, successively splits to the part pancreas in area-of-interest on cross section, obtains several cross-sectional views
Picture；
Step S232, determines connected domain to be selected from cross-sectional image；
Step S233, determines pancreas reference layer in connected domain to be selected.
In the step S232, threshold value is carried out to each cross-sectional image of the part pancreas in area-of-interest first
Segmentation, obtains the connected domain in the range of given threshold.Need first to determine the tonal range where pancreas, CT images in the present invention
Middle pancreas parameter (CT values) range of definition is 0-80HU (Hounsfield Unit).The mark that gray scale is not belonging to the scope is
Region, Threshold segmentation result removes liver,spleen,kidney part.It (is, for example, less than 150mm that next, which removes area especially small,2), or remove
On image, the connected domain of lower edge or right hand edge；Again connected domain length-width ratio be less than 1 and connected domain circularity it is small
Also remove in 0.6 very rough.
Pancreas middle part connects together with stomach sometimes, and usual pancreas afterbody farther out, is split with other tissue distances
Accurately, therefore, pancreas width is determined from pancreas afterbody.If pancreas width occurs more than 30mm, ginseng is used as using most long a line
Examine, retain the row each 15mm up and down scope, the wide position can be modified.
In the step S233, because pancreas area-of-interest is blocked from right to left, left boundary part is necessarily wrapped
Containing pancreas, so, retain the connected domain for including Far Left border in segmentation result, cross section is selected from the connected domain of reservation most
Long connected domain.Segmentation result of the pancreas on cross section such as Fig. 5 in area-of-interest.
The step S24 is divided the part pancreas in initial segmentation result since pancreas tail end in sagittal plane
Cut, obtain for the segmentation of the pancreas in belly medical image start reference layer；Including following sub-step：
S241, is successively split to the part pancreas in initial segmentation result in sagittal plane, obtains segmentation result again；
S242, retains the connected domain for having common factor in segmentation result again with initial segmentation result；
S243, chooses the maximum connected domain of area from the connected domain for having common factor；
S244, the border points to the maximum connected domain of area are counted, and choose the connected domains work for counting most in border
For start reference layer.
In step S241, according to the continuity features of pancreas in sagittal plane, one layer is first looked in sagittal plane as reference layer,
Then successively split.1/3rd of current connected domain lateral length is taken to do new pancreas region of interest ROI from pancreas tail end,
Pancreas segmentation is done in the ROI and determines reference layer.In sagittal plane to pancreas Threshold segmentation scope it is 10- in the embodiment of the present invention
80HU.Since pancreas tail end, Threshold segmentation is done in sagittal plane, the mark for being not belonging to the threshold range is, and does
Morphologic closed operation, kernel function is the disk that yardstick is 3；Obtain the segmentation result again in each (individual) corresponding sagittal plane.Point
Segmentation method is similar in cross section segmentation to foregoing pancreas, will not be described here.
In step S242, for the segmentation result again in each corresponding sagittal plane, including one or several corresponding (the
Two) connected domain, retains one or several connected domains having common factor in segmentation result with initial segmentation result again.
In step S243, it is performed according to step S242 in the several connected domains obtained afterwards, selects the correspondence sagittal plane
On the maximum connected domain of area again in segmentation result.
In step S244, after terminating in new pancreas region of interest ROI in sagittal plane to all layers of pancreas segmentation,
Mask borders are taken to count most one layers as start reference layer.
Step S25, including (essence) segmentation first is carried out to start reference layer, after to from start reference layer respectively to pancreas head end
With pancreas tail end, (essence) segmentation is carried out in sagittal plane to the pancreas in belly medical image.
Further, step S25 carries out (essence) segmentation to start reference layer, comprises the following steps：
Step S251, first size is expanded (such as on the basis of the border of start reference layer, upward by start reference layer
Extend 2mm) after as current layer coarse segmentation result；
Count the histogram distribution of coarse segmentation result corresponding grey scale figure, using the average of statistics add and subtract twice of variance as
The threshold range of current layer (start reference layer) pancreas, the mark being not belonging in the threshold range is；
The coarse segmentation result of current layer is expanded into the region of interest after the second size (such as 5mm) as current layer pancreas
Domain；
Centered on the center of gravity of coarse segmentation result, with 360 ° of traversal pancreas boundary points of ray, ray runs into the ground of dark areas
Side is pancreas border, and the part without dark areas leaves a blank, and to part of leaving a blank with adjacent two interpolating on sides, obtaining closed curve is exactly
The current layer pancreas segmentation result.Segmentation result of the pancreas on cross section such as Fig. 6 in new area-of-interest.
Further, step S25 is split to the pancreas in addition to start reference layer, comprises the following steps：
Successively split to pancreas two from start reference layer, when dark areas or reference layer area are less than 15mm2, or currently
Layer pancreas area is less than 0.3 times of pancreas area before ten layers, then direction segmentation is just terminated.If both direction all splits completion simultaneously
And the pancreas total length being partitioned into is less than 120mm, then finds out a ginseng from current segmentation result again according to step S24 methods
Layer is examined, then second of pancreas fine segmentation is done according to step S25.The segmentation result of whole pancreas is finally given, such as Fig. 7 institutes
Show.
The present invention can be automatically positioned out cauda pancreatis using metastable position relationship between pancreas tail end and spleen
End.Pancreas be lying in posterior wall of abdomen this anatomical structure of silkworm shape body of gland so that pancreas in the sagittal plane of CT images adjacent layer it
Between be continuous, and because CT images generally resolution ratio in sagittal plane and coronal-plane is higher, adjacent interlayer difference
Very little, can carry out integrated application to both direction information.The present invention just precisely efficiently realizes pancreas using these features
Full-automatic dividing.
It should be noted that through the above description of the embodiments, those skilled in the art can be understood that
Part or all of to the application can be realized by software and the required general hardware platform of combination.Understood based on such,
The part that the technical scheme of the application substantially contributes to prior art in other words can be embodied in the form of software product
Out, the computer software product may include the one or more machine readable medias for being stored thereon with machine-executable instruction,
These instructions may be such that this when being performed by one or more machines such as computer, computer network or other electronic equipments
One or more machine embodiments in accordance with the present invention perform operation.Machine readable media may include, but be not limited to, floppy disk,
CD, CD-ROM (compact-disc-read-only storage), magneto-optic disk, ROM (read-only storage), RAM (random access memory),
EPROM (Erasable Programmable Read Only Memory EPROM), EEPROM (Electrically Erasable Read Only Memory), magnetic or optical card, sudden strain of a muscle
Deposit or other kinds of medium/machine readable media suitable for storage machine-executable instruction.
The application can be used in numerous general or special purpose computing system environments or configuration.For example：Personal computer, service
Device computer, handheld device or portable set, laptop device, multicomputer system, the system based on microprocessor, top set
Box, programmable consumer-elcetronics devices, network PC, minicom, mainframe computer including any of the above system or equipment
DCE etc..
The application can be described in the general context of computer executable instructions, such as program
Module.Usually, program module includes execution particular task or the routine for realizing particular abstract data type, program, object, group
Part, data structure etc..The application can also be put into practice in a distributed computing environment, in these DCEs, by
Remote processing devices connected by communication network perform task.In a distributed computing environment, program module can be with
Positioned at including in the local and remote computer-readable storage medium including storage device.
The main advantage of the present invention is to be automatically positioned out pancreas tail end using spleen information, while using between adjacent layer
Continuity and pancreas around low gray level areas, realize full-automatic dividing successively.Pancreas in the medical image of the present invention
The segmentation of gland is full automatic.The present invention takes full advantage of that pancreas is continuous in the sagittal plane of CT images and adjacent layer change is slow
Slow knowledge.The present invention has used pancreas in CT images in cross section and the different characteristic of sagittal plane, to be used respectively, finally
Both direction information is carried out comprehensive.Patent of the present invention takes full advantage of the dark areas around pancreas, operative constraint on CT images
Pancreas border.
Although the present invention is disclosed as above with preferred embodiment, so it is not limited to the present invention, any this area skill
Art personnel, without departing from the spirit and scope of the present invention, when a little modification can be made and perfect, therefore the protection model of the present invention
Enclose when by being defined that claims are defined.
Claims (10)
1. the dividing method of pancreas in a kind of medical image, it is characterised in that comprise the following steps:
The belly medical image of those who are investigated is obtained, and the belly medical image includes pancreas；
Determine the area-of-interest for including pancreas tail end in belly medical image；
Pancreas in area-of-interest is split on cross section, initial segmentation result is obtained；
Since pancreas tail end, the part pancreas in initial segmentation result is split in sagittal plane, obtained for belly
The start reference layer of the segmentation of pancreas in medical image；
From start reference layer respectively to pancreas head end and pancreas tail end, the pancreas in belly medical image is carried out in sagittal plane
Segmentation, obtains the pancreas segmentation result in belly medical image.
2. the dividing method of pancreas in medical image according to claim 1, it is characterised in that the area-of-interest is only
Comprising part pancreas, and spleen positioning of the pancreas tail end in belly medical image or manually locate.
3. the dividing method of pancreas in medical image according to claim 1, it is characterised in that described to area-of-interest
In part pancreas split on cross section, obtain initial segmentation result, including：
Pancreas in area-of-interest is successively split on cross section, several cross-sectional images are obtained；
Connected domain to be selected is determined from cross-sectional image；
Pancreas reference layer is determined in connected domain to be selected.
4. the dividing method of pancreas in medical image according to claim 3, it is characterised in that described from cross-sectional image
In determine connected domain to be selected, including：Row threshold division is entered to each cross-sectional image of the pancreas in area-of-interest, obtained
Take the connected domain in the range of given threshold；Or/and remove the connection of close cross-sectional image top edge, lower edge or right hand edge
Domain；Or/and remove the connected domain that length-width ratio in cross-sectional image is less than 1.
5. the dividing method of pancreas in medical image according to claim 3, it is characterised in that described to treat selection connection
Pancreas reference layer is determined in domain, including：Retain the connected domain for including Far Left border in connected domain to be selected, from the company of reservation
Cross section most long connected domain is selected in logical domain.
6. the dividing method of pancreas in medical image according to claim 1, it is characterised in that described to be opened from pancreas tail end
Begin, the part pancreas in initial segmentation result is split in sagittal plane, obtained for the pancreas in belly medical image
Segmentation start reference layer, including：
Part pancreas in initial segmentation result is successively split in sagittal plane, segmentation result again is obtained；
Retain the connected domain for having common factor in segmentation result again with initial segmentation result；
The maximum connected domain of area is chosen from the connected domain for having common factor；
Border points to the maximum connected domain of area are counted, and are chosen the most connected domains of border points and are used as start reference
Layer.
7. the dividing method of pancreas in medical image according to claim 1, it is characterised in that described from start reference layer
Respectively to pancreas head end and pancreas tail end, the pancreas in belly medical image is split in sagittal plane, including：
Start reference layer is expanded to the coarse segmentation result after first size as current layer；
The histogram distribution of coarse segmentation result corresponding grey scale figure is counted, twice of variance is added and subtracted as current using the average of statistics
The threshold range of layer pancreas, the mark being not belonging in the threshold range is；
The coarse segmentation result of current layer is expanded into the area-of-interest after the second size as current layer pancreas；
Centered on the center of gravity of coarse segmentation result, pancreas boundary point is traveled through with ray, the place that ray runs into dark areas is pancreas
Border, the part without dark areas is left a blank, and to part adjacent two border difference of leaving a blank, it is exactly the current layer to obtain closed curve
Pancreas segmentation result.
8. the dividing method of pancreas in medical image according to claim 1, it is characterised in that also including to the belly
Medical image is rotated, and the angle rotated to the belly medical image is in 15 ° -45 °, the doctor with horizontal line
It is 3D CT images or 3D MR images to learn image.
9. a kind of device of the segmentation of pancreas in medical image, described device includes processor, it is characterised in that the processor
Including:
Image acquisition unit, the belly medical image for obtaining those who are investigated, and the belly medical image include pancreas；
Area-of-interest determining unit, for determining the area-of-interest for including pancreas tail end in belly medical image；
First cutting unit, for splitting to the pancreas in area-of-interest on cross section, obtains initial segmentation result；
Second cutting unit, for since pancreas tail end, being carried out to the part pancreas in initial segmentation result in sagittal plane
Segmentation, obtains the start reference layer for the segmentation of the pancreas in belly medical image；From start reference layer respectively to caput pancreatis
End and pancreas tail end, split to the pancreas in belly medical image in sagittal plane, obtain the pancreas in belly medical image
Gland segmentation result.
10. the segmenting device of pancreas in medical image according to claim 9, it is characterised in that the belly medical science figure
, can be by computer tomography equipment or MR imaging apparatus to scanned pair as being 3D CT images or 3D MR images
The belly of elephant performs scanning and obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710138713.8A CN106952273B (en) | 2017-03-09 | 2017-03-09 | The dividing method and device of pancreas in medical image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710138713.8A CN106952273B (en) | 2017-03-09 | 2017-03-09 | The dividing method and device of pancreas in medical image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106952273A true CN106952273A (en) | 2017-07-14 |
| CN106952273B CN106952273B (en) | 2019-11-19 |
Family
ID=59466897
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710138713.8A Active CN106952273B (en) | 2017-03-09 | 2017-03-09 | The dividing method and device of pancreas in medical image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106952273B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830852A (en) * | 2018-07-13 | 2018-11-16 | 上海深博医疗器械有限公司 | Three-D ultrasonic tumour auxiliary measurement system and method |
| CN110717915A (en) * | 2019-09-25 | 2020-01-21 | 武汉联影智融医疗科技有限公司 | Segmentation method, segmentation device, computer equipment and storage medium |
| CN111275720A (en) * | 2020-01-20 | 2020-06-12 | 浙江大学 | Full end-to-end small organ image identification method based on deep learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070116345A1 (en) * | 2005-11-23 | 2007-05-24 | Peterson Samuel W | Automatic aortic detection and segmentation in three-dimensional image data |
| CN102697526A (en) * | 2012-06-15 | 2012-10-03 | 华东医院 | Ultrasonic scanning tomography device for volumes of superficial tissues and organs |
| CN103932694A (en) * | 2014-05-07 | 2014-07-23 | 霍云龙 | Method and device for accurately diagnosing FFR |
| CN105979865A (en) * | 2014-02-27 | 2016-09-28 | 金伯利-克拉克环球有限公司 | Methods for assessing health conditions using single coil magnetic induction tomography imaging |
| CN106097374A (en) * | 2016-06-24 | 2016-11-09 | 西安电子科技大学 | 3D MRI pancreas dividing method based on sparse low-rank Yu Atlas collection |
-
2017
- 2017-03-09 CN CN201710138713.8A patent/CN106952273B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070116345A1 (en) * | 2005-11-23 | 2007-05-24 | Peterson Samuel W | Automatic aortic detection and segmentation in three-dimensional image data |
| CN102697526A (en) * | 2012-06-15 | 2012-10-03 | 华东医院 | Ultrasonic scanning tomography device for volumes of superficial tissues and organs |
| CN105979865A (en) * | 2014-02-27 | 2016-09-28 | 金伯利-克拉克环球有限公司 | Methods for assessing health conditions using single coil magnetic induction tomography imaging |
| CN103932694A (en) * | 2014-05-07 | 2014-07-23 | 霍云龙 | Method and device for accurately diagnosing FFR |
| CN106097374A (en) * | 2016-06-24 | 2016-11-09 | 西安电子科技大学 | 3D MRI pancreas dividing method based on sparse low-rank Yu Atlas collection |
Non-Patent Citations (2)
| Title |
|---|
| 张欢 等: "3D增强MRA在恶性胰腺肿瘤术前评估中的应用", 《中国医学计算机成像杂志》 * |
| 杜磊: "核磁共振图像中的 3D 胰腺分割", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830852A (en) * | 2018-07-13 | 2018-11-16 | 上海深博医疗器械有限公司 | Three-D ultrasonic tumour auxiliary measurement system and method |
| CN110717915A (en) * | 2019-09-25 | 2020-01-21 | 武汉联影智融医疗科技有限公司 | Segmentation method, segmentation device, computer equipment and storage medium |
| CN110717915B (en) * | 2019-09-25 | 2022-05-27 | 武汉联影智融医疗科技有限公司 | Segmentation method, segmentation device, computer equipment and storage medium |
| CN111275720A (en) * | 2020-01-20 | 2020-06-12 | 浙江大学 | Full end-to-end small organ image identification method based on deep learning |
| CN111275720B (en) * | 2020-01-20 | 2022-05-17 | 浙江大学 | Full end-to-end small organ image identification method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106952273B (en) | 2019-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109636811B (en) | Integration method and device for lung lobe segment segmentation of CT (computed tomography) image | |
| CN105976367B (en) | Image partition method, pulmonary nodule detection method and its computer-aided detection system | |
| CN110111313A (en) | Medical image detection method and relevant device based on deep learning | |
| US8600125B2 (en) | System and method for computer aided polyp detection | |
| CN110648337A (en) | Hip joint segmentation method, hip joint segmentation device, electronic apparatus, and storage medium | |
| Peng et al. | 3D liver segmentation using multiple region appearances and graph cuts | |
| WO2020028352A1 (en) | Methods and systems for segmenting organs in images using a cnn-based correction network | |
| Xing et al. | Lesion segmentation in ultrasound using semi-pixel-wise cycle generative adversarial nets | |
| CN107203741A (en) | Vessel extraction method, device and its system | |
| CN107545584A (en) | The method, apparatus and its system of area-of-interest are positioned in medical image | |
| US20120148123A1 (en) | Segmenting an organ in a medical digital image | |
| CN110706241B (en) | Three-dimensional focus region extraction method and device | |
| Tong et al. | Improved U-net MALF model for lesion segmentation in breast ultrasound images | |
| CN106952273B (en) | The dividing method and device of pancreas in medical image | |
| WO2022213654A1 (en) | Ultrasonic image segmentation method and apparatus, terminal device, and storage medium | |
| CN103562961A (en) | Apparatus and method for generating an attenuation correction map | |
| CN106909947A (en) | CT image metal artifacts removing method and elimination system based on Mean Shift algorithms | |
| CN108053401A (en) | A kind of B ultrasound image processing method and device | |
| CN116797612B (en) | Ultrasonic image segmentation method and device based on weak supervision depth activity contour model | |
| CN113450359A (en) | Medical image segmentation, display, model training methods, systems, devices, and media | |
| CN113139948A (en) | Organ contour line quality evaluation method, device and system | |
| Wang et al. | Accurate tumor segmentation via octave convolution neural network | |
| Kiraly et al. | 3D human airway segmentation for virtual bronchoscopy | |
| CN116416221A (en) | Ultrasonic image analysis method | |
| WO2023092124A1 (en) | Machine-learning based segmentation of biological objects in medical images |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 201807 Shanghai city Jiading District Industrial Zone Jiading Road No. 2258 Patentee after: Shanghai Lianying Medical Technology Co., Ltd Address before: 201807 Shanghai city Jiading District Industrial Zone Jiading Road No. 2258 Patentee before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |