CN106683090B - The localization method and its system of rib cage in medical image - Google Patents
The localization method and its system of rib cage in medical image Download PDFInfo
- Publication number
- CN106683090B CN106683090B CN201611268139.XA CN201611268139A CN106683090B CN 106683090 B CN106683090 B CN 106683090B CN 201611268139 A CN201611268139 A CN 201611268139A CN 106683090 B CN106683090 B CN 106683090B
- Authority
- CN
- China
- Prior art keywords
- image
- rib cage
- connected domain
- value
- obtains
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
-
- 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
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
The present invention discloses the localization method and its system of rib cage in a kind of medical image, including：If obtaining the medical image for including dried layer two dimensional image；Skeletal tissue is extracted according to medical image is stated, obtains the first image；Described first image is projected along first direction, obtains projected image；Vertebral region is determined according to the projected image, obtains mask images；In described first image, the second image is obtained based on the mask images；Connected domain analysis is carried out to second image, extracts rib cage positioning result.Localization method disclosed by the invention and its system, are efficiently modified existing rib cage localization method, and influence of the removal rib cage to vessel extraction improves the accuracy of medical consultations.
Description
【Technical field】
The present invention relates to image processing fields, localization method more particularly to rib cage in a kind of medical image and its are
System.
【Background technology】
Angiography includes CT angiographies (CTA, CT angiography), magnetic resonance angiography
(Magnetic Resonance Angiography, MRA) and the non-contrast imagings of MR etc..Angiography is currently to examine
One of the important method of disconnected vascular diseases, is clinically mainly used for a variety of vascular diseases of diagnosis and treatment, as aneurysm, hemadostewnosis,
Angiosteosis etc..
Since the CT values of its hetero-organization such as CTA images medium vessels and bone are Chong Die, the form of the skeletal tissues such as rib cage in addition
It is also tubular structure to learn, close with blood vessel, current main blood vessel segmentation method such as region growth, dynamic outline etc. be easy to by
The tissues such as bone are mistakened as into vessel extraction, cause over-segmentation, or minute blood vessel etc. is removed as non-vascular tissue, cause
Less divided influences subsequent medical consultations work.
Therefore, it is necessary to existing rib cage localization method be improved, to improve the accuracy of subsequent applications.
【Invention content】
A kind of vessel extraction method of technical problems to be solved by the invention proposition and its system, can be efficiently modified existing
Rib cage localization method, influence of the removal rib cage to vessel extraction.
In order to solve the above technical problems, the present invention provides a kind of localization method of rib cage in medical image, including：
If obtaining the medical image for including dried layer two dimensional image；
Skeletal tissue is extracted according to medical image is stated, obtains the first image；
Described first image is projected along first direction, obtains projected image；
Vertebral region is determined according to the projected image, obtains mask images；
In described first image, the second image is obtained based on the mask images；
Connected domain analysis is carried out to second image, extracts rib cage positioning result.
Further, according to medical image extraction skeletal tissue is stated, the first image is obtained, including：
Enhanced based on first threshold and Laplce and handle the medical image, obtains described first image；Or
Skeletal tissue is extracted in the medical image based on second threshold, obtains described first image.
Further, the first threshold value range is that image value is more than 120, and the second threshold value range is figure
Picture value is more than 600.
Further, described that vertebral region is determined according to the projected image, mask images are obtained, including：
Obtain several connected domains in the mask images；
The vertebra point for the connected domain for taking image value mean value big is seed point；
It is grown with adaptive threshold based on the seed point, determines vertebral region, obtain the mask images.
Further, the vertebra point, be the central point of the big connected domain of described image value mean value, barycenter, center of gravity or
Image value maximum point.
Further, described in described first image, the second image is obtained based on the mask images, including：
According to the vertebral region in the mask images, successively removes in described first image and correspond to vertebral region, obtain
Second image.
Further, described that connected domain analysis is carried out to second image, rib cage positioning result is extracted, including：
Obtain several connected domains in second image；
Confirm that connected domain meets following one or more conditions, that is, determine that the connected domain is rib cage, obtains rib cage positioning knot
Fruit：
As maximum width is more than threshold value a in the connected domain；Or
As the Ratio of long radius to short radius of the connected domain is more than threshold value b；Or
Picture such as the connected domain in spatial position apart from background area less than threshold value c pixels number and the connected domain
The ratio between vegetarian refreshments sum is more than threshold value d；Or
The ratio between the area of the connected domain and circumscribed rectangular area are less than threshold value e；
Otherwise the connected domain is removed.
Further, the value range of the threshold value a be more than 20, the value range of the threshold value b be more than 1, it is described
It is less than 50mm that the value range of threshold value c, which is less than value range, and the value range of the threshold value d is the decimal more than 0.5, institute
The value range for stating threshold value e is the decimal less than 0.7.
In order to solve the above technical problems, the present invention also provides a kind of positioning systems of rib cage to include：
First extraction unit, for according to medical image extraction skeletal tissue is stated, obtaining the first image；
Projecting cell obtains projected image for projecting described first image along first direction；
Mask unit obtains mask images for determining vertebral region according to the projected image；
Second extraction unit, in described first image, the second image to be obtained based on the mask images；
Positioning unit extracts rib cage positioning result for carrying out connected domain analysis to second image.
Further, the mask unit, including seed point determination subelement；And growth subelement, for being based on institute
It states seed point to grow with adaptive threshold, determines vertebral region.
Compared with the prior art, the advantages of the present invention are as follows：
The present invention is by being based on the methods of threshold method, gradient method, and/or the enhancing of Laplce edge to obtain including theme bone
First image in bone region improves subsequent processing speed to reduce the treating capacity and radiography of follow-up rib cage positioning；By being positioned to
Except vertebral region, the second image of intercostal space approximate location is obtained；Second image is connected according to costal stigmata
Logical domain analysis, extracts rib cage positioning result.The localization method of rib cage provided in this embodiment being capable of quick, accurate positionin rib cage area
Domain, strong robustness are suitable for all kinds of medical images, and influence of the removal rib cage to vessel extraction improves the accuracy of medical consultations.
【Description of the drawings】
Fig. 1 is the structural schematic diagram of rib cage positioning system in one embodiment of the invention；
Fig. 2 is the schematic diagram of data processing system in one embodiment of the invention；
Fig. 3 is the schematic diagram of rib cage locating module in one embodiment of the invention；
Fig. 4 is rib cage localization method flow chart in one embodiment of the invention；
Fig. 5 a~5d are the result schematic diagram of each step in rib cage localization method of the present invention.
【Specific implementation mode】
Many details are elaborated in the following description in order to fully understand the present invention.But the present invention can be with
Much implement different from other manner described here, those skilled in the art can be without prejudice to intension of the present invention the case where
Under do similar popularization, therefore the present invention is not limited to the specific embodiments disclosed below.
Secondly, the present invention is described in detail using schematic diagram, when describing the embodiments of the present invention, for purposes of illustration only, institute
It is embodiment to state schematic diagram, should not limit the scope of protection of the invention herein.Make above-mentioned purpose, the feature and excellent of the present invention
Point can become apparent it is understandable, with reference to the accompanying drawings and examples to the present invention specific implementation mode be described in detail.
As shown in Fig. 1 according to the schematic diagram of some embodiments of the present application rib cage positioning system；Rib cage is fixed
Position system 100 may include an imaging device 110, a controller 120, a data processing system 130, and an input/
140, memories 150 of output equipment and a network 160.
Different component/parts in imaging system 100 can be in communication with each other.For example, data processing system 130 can be with net
Network 160 is connected with each other or communication, or directly with imaging system 100 or part of it (for example, imaging device 110, controller
120 etc.) interconnection or communication or the combination of two ways.Imaging device 110, data processing system 130, memory 150,
Data communication between the miscellaneous equipment that controller 120 and imaging system 100 may include can pass through data line, network
160 equal or aforesaid way combinations are realized.
Imaging device 110 can be scanned target object, and obtain associated data (for example, scan data
Deng).Imaging device 110 can be an equipment or an equipment group.In some embodiments, imaging device 110 can be one
A medical information collecting device, for example, a Positron emission computed tomography (Positron emission
Tomography (PET)) equipment, CT scan (Computed Tomography (CT)) equipment, one
Magnetic resonance imaging (Magnetic Resonance Imaging (MRI)) equipment etc..The equipment can be single use, and also may be used
To be used in combination.Imaging device 110 can be a PET-CT equipment, a PET-MRI equipment or a SPECT-MRI equipment
Deng.The scan data can be the one of CT scan data, MRI scan data, ultrasound scan data, X-ray scan data etc.
Kind or a variety of combinations.
Controller 120 can be used for controlling imaging device 110, input-output apparatus 140 and/or data processing system
130.In some embodiments, controller 120 may be controlled to as in equipment 110 X-ray generation unit and/or X-ray visit
Survey unit.Controller 120 can connect from control imaging device 110, input-output apparatus 140 and/or data processing system 130
It collects mail and ceases, or information is sent out to above-mentioned system/device.In some embodiments, controller 120 can be received from imaging device
110 picture signal or with the relevant data of target object.Controller 120 can send image letter to data processing system 130
Number or with the relevant data of target object.Controller 120 can receive the processed data from data processing system 130
Or reconstruction image.Controller 120 can send processed data or reconstruction image to input-output apparatus 140.At some
In embodiment, controller 120 may include computer, program, algorithm, software, storage device, interface etc..
Data processing system 130 can handle from imaging device 110, controller 120, network 160 and/or input/it is defeated
Go out the information of equipment 140.In some embodiments, data processing system 130 can be based on described information and generate one or more
CT images.Data processing system 130 can send image to input-output apparatus 140.Data processing system 130 can be completed
With the relevant each generic operation of data processing, for example, data prediction, data conversion processing, data cleansing processing, at data fitting
Reason, one or more combination such as data weighting processing.Data processing system 130 can be based on algorithms of different program and realize data
Processing, for example, Fourier transform principle, filtered back projection algorithm, iterative approximation, histogram expansion calculate, image data letter
Count the one or more of optimization, level set function calculating etc..In some embodiments, data processing system 130 can have been handled
Close the data of blood-vessel image.For example, data processing system 130 can from identified in image the bone of thorax abdomen, abdominal vascular, under
Limb blood vessel, vessel centerline or the blood vessel at other positions.In some embodiments, data processing system 130 can be based on a variety of
Algorithm routine or method carry out blood vessel or bone positioning to image, for example, positioning rib cage, ilium, rumpbone, shin bone or common iliac artery
Deng.In some embodiments, data processing system 130 can store data to memory 150 or read number from memory 150
According to.These data can itself be generated directly or indirectly from data, the data processing system 130 in imaging device 110
Ephemeral data or non-provisional data or the data etc. that image procossing is carried out for auxiliary data processing system 130.
Input-output apparatus 140 can receive, send or show information.In some embodiments, input-output apparatus
140 may include in keyboard, touch control device, mouse, audio input device, image input device and remote control equipment etc.
It is one or more.Input-output apparatus 140 can send information to terminal (for example, display screen, projecting apparatus, printer, depositing
Store up one or more combinations such as equipment, computing device) carry out corresponding operating.In some embodiments, input-output apparatus
140 may include graphic user interface, and the information or processing result image of displaying imaging process stage are (for example, image histogram
Figure, bone mask, blood vessel mask, one or more combinations through the transformed blood vessel associated picture of image etc.).The figure
User interface can give user's input parameter and prompt, can also allow user participate in data handling procedure (for example, start or
Stop processing procedure, selection or modification operational parameter, selection or modification algorithm, modification program, logs off, system maintenance, is
System upgrading or system update etc.).
Network 160 can be the combination of single network or multiple and different networks.For example, network 160 may be a local
Net (local area network (LAN)), wide area network (wide area network (WAN)), public switch telephone network
Any group of (public switched telephone network (PSTN)), virtual network (VN) or above-mentioned network
It closes.Network 160 can also include multiple network access points, and can use landline network infrastructures, radio network frame and have
Line/wireless network hybrid structure.Cable network may include utilizing one or more lines such as metallic cable, compound cable, optical cable
The mode of cable combination.Wireless network may include bluetooth (Bluetooth), wireless network (Wi-Fi), purple honeybee (ZigBee), near field
It is a kind of or more to communicate (Near Field Communication (NFC)), cellular network (including GSM, CDMA, 3G, 4G etc.) etc.
The combination of kind network mode.Network 160 can be adapted in range described herein, but be not limited to the description.
Memory 150 can be used for storing data.These data can be the data that imaging system 100 is generated or obtained,
Such as scan data, the one or more components of imaging system 100 when running the data that generate, user set by input/output
The data etc. that the data of standby 140 input, user are obtained by network 160 from other data source (not shown in figure 1)s.Memory
150 can be the combination of an equipment/component or several equipment/components with store function.In some embodiments, it stores
Device 150 may include one or more independent equipment with data storage function, such as a computer or a service
Device etc..Memory 150 may include local storage or remote memory (for example, cloud storage of the framework in network 160
Deng).In some embodiments, memory 150 may include the component with data storage function in an autonomous device, such as
One disk or a disk array etc..Memory 150 may include that any equipment in imaging system 100 (is set for example, being imaged
Standby 110, data processing system 130, input-output apparatus 140 etc.) in the component with store function.
In some embodiments, network 160 can be used between each equipment/component in imaging system 100 transmitting
Information.In some embodiments, network 160 can be used for imaging system 100 and obtain information from a remote server.Network
160 can be the combination of an independent network or multiple and different networks.For example, network 160 may include LAN
(local area network (LAN)), wide area network (wide area network (WAN)), public switch telephone network
(public switched telephone network (PSTN)), virtual network (Virtual Network (VN)) etc. or
The combination of above-mentioned network.Network 160 may include multiple network access points.Network 160 can use landline network infrastructures, wireless
Network architecture and wire/radio network hybrid structure.Cable network may include utilizing metallic cable, compound cable, optical cable
Etc. the mode of one or more Wire/cable contact members.The transmission mode of wireless network may include bluetooth (Bluetooth), wireless network
(Wi-Fi), purple honeybee (ZigBee), near-field communication (Near Field Communication (NFC)), cellular network (including
GSM, CDMA, 3G, 4G etc.) etc..
According to some embodiments of the present application, Fig. 2 show a schematic diagram of data processing system 130.Data processing
System 130 may include one or more data acquisition modules 210, one or more memory modules 220, one or more displays
Module 230 and one or more rib cage locating modules 240.(and/or indirect) connection directly with one another between shown module.
Data acquisition module 210 can obtain data.The data of the acquisition can come from imaging device 110 and/or control
Device 120 processed.In some embodiments, data can be obtained by network 160 from external data source.The data can be three-dimensional
Image data and/or two-dimensional image data.The data can be the data of the privileged site set by detection demand,
For example, the panoramic scanning of target object, chest, lung, lower limb, bronchus, bone, blood vessel, Substance P of target object etc.
It is one or more.In some embodiments, the data can be angiographic data.In some embodiments, data acquisition
Module 210 can obtain blood-vessel image initial data, the parameter etc. of blood-vessel image treated data or blood-vessel image processing.
Memory module 220 can store data or information.The data or information of the storage can come from imaging device
110 and/or controller 120 and/or data processing system 130 in other module/units (data acquisition module 210, displays
Module 230, rib cage locating module 240 or other correlation module (not shown)).The data or information of storage can be various shapes
Formula, for example, the combination that the relevant information of numerical value, signal, image, target object, order, algorithm, program etc. are one or more.
In some embodiments, the data of storage can be blood-vessel image, blood-vessel image parameter, blood-vessel image treated data or blood
Program that pipe image procossing is applied and/or algorithm etc..
Display module 230 can be with display data.The data information of the display can come from data acquisition module 210, deposit
Store up module 220 and/or rib cage locating module 240 etc..The data of the display can be sent to input-output apparatus 140.
In some embodiments, the image data that rib cage locating module 240 obtains can be sent to terminal and shown by display module 230.
In some embodiments, display module 230, which can directly display, comes automatic network 160 or the related data in memory module 220
Information.The data of the display can be various forms, including the relevant information of numerical value, signal, image, target object, order,
One or more combination such as algorithm, program.In some embodiments, display module 230 can be shown including bone information
Image (for example, histogram, bone gray level image, to a mask image, bone coarse segmentation image, bone subdivision cut image etc.).
Rib cage locating module 240 can carry out the processing of related data, and build image, including base according to corresponding data
Vascular tissue is extracted in data.The data can come from data acquisition module 210, memory module 220 and/or other be not shown
Module；It can also be the external data resource obtained by network 160.The respective image of the structure can be sent to display
Module 230 etc..The data of the processing of rib cage locating module 240 can be the data of the special part about target object, for example, greatly
One or more combination such as brain, heart, blood vessel, lung, liver,spleen,kidney, bone.
It should be noted that the above description for data processing system 130 only for convenience of description can not be this Shen
It please be limited within the scope of illustrated embodiment.It is appreciated that for those skilled in the art, in the original for understanding the system
After reason, arbitrary combination may be carried out to modules, or constitute subsystem and other without departing substantially from this principle
Module connects, to implementing the various modifications and variations of the above method and systematic difference field in form and details.
Shown in Fig. 3 is a schematic diagram according to the rib cage locating module 240 of some embodiments of the present application.The rib
Bone locating module 240 includes that first extraction unit 310, a projecting cell 320, a mask unit 330, second carries
Take unit 340, a positioning unit 350.
First extraction unit 310, for according to medical image extraction skeletal tissue is stated, obtaining the first image；It is described
Medical image can derive from memory module 220, data acquisition module 210.
The projecting cell 320 obtains projected image for projecting described first image along first direction.
The mask unit 330, including seed point determination subelement 331；And growth subelement 332, for being based on institute
It states seed point to grow with adaptive threshold, determines vertebral region.
The mask unit 330 obtains mask images for determining vertebral region according to the projected image.
Second extraction unit 340, in described first image, the second figure to be obtained based on the mask images
Picture.
The positioning unit 350 extracts rib cage positioning result for carrying out connected domain analysis to second image
It should be noted that the above description for rib cage locating module 240 and its each unit, only for convenience of description,
The application can not be limited within the scope of illustrated embodiment.It is appreciated that for those skilled in the art,
After the principle for solving the system, changing in arbitrary combination, separation or details can be carried out to each unit of rib cage locating module 240
Become.Rib cage locating module 240 can remove one or more units.
In order to solve the above technical problems, providing a kind of positioning of rib cage in medical image according to some embodiments of the present application
A kind of exemplary process diagram of method.The method can be executed by data processing system 130.The output image of the generation can
To be shown in a display device (for example, input-output apparatus 140, controller 120 etc.), or it is stored in storage device (example
Such as, memory 150 etc.) in.
Since the CT values of its hetero-organization such as medical image medium vessels and bone are Chong Die, the skeletal tissues such as rib cage and vascular morphology
It is close, cause the difficulty of vessel extraction.In order to solve the above technical problems, the present embodiment provides rib cages in a kind of medical image to determine
Position method, the method is as shown in figure 4, include：
Step S410 is executed, if obtaining the medical image for including dried layer two dimensional image, the medical image includes the doctor
It includes but not limited to obtain three-dimensional or two dimensional image by the imaging system scanning collections of all kinds of mode to learn image medicine, and described three
Can also be image archiving and communication system by such as storing if tieing up the volume data that the two dimensional image that image includes dried layer is constituted
Storage system transmission obtains inside or outside (Picture Archiving and Communication Systems, PACS) etc.
.The mode includes but not limited to magnetic resonance imaging (MRI), magnetic resonance angiography (MRA), computed tomography
(CT), CT angiographic images (CTA, CT angiography), positron emission computerized tomography (Positron Emission
Tomography, PET) etc. one or more combinations, can also be longitudinal relaxation (T1), transverse relaxation (T2) and quickly overturning
Restore not homotactic magnetic resonance image, the present invention such as sequence (STIR) to be not specifically limited this, preferred ridge in the present embodiment
Column magnetic resonance image；The volume data is usually to be obtained in scan plan to obtain the coarse scanning that accurate scanning parameter carries out
The sequence image that ten is several layers of or if the dried layer two dimensional image such as tens layers is constituted.Fig. 5 a are to be obtained by Angiography
Abdominal CT A images, wherein including rib cage, vertebra and abdomen Major Vessels.
Step S420 is executed, skeletal tissue is extracted according to the medical image, obtains the first image；Since rib cage is blood vessel
Common noise in extraction process, it is therefore desirable to improve the accuracy of positioning rib cage.This step, by being extracted in medical image
Major skeletal region, to be conducive to, subsequent step is quick, accurately extracts blood vessel.In this step, threshold method, gradient can be based on
The processing of the methods of method, and/or the enhancing of Laplce edge extracts skeletal tissue according to the medical image.
For example, can be constructed according to skeletal tissue's boring, and internal blood vessel is homogeneous structure, is handled by gradient method
The medical image.The gradient method can handle the medical image in conjunction with drawing pula color to enhance.Aforementioned various methods can be with
One or more combinations obtain the first image as shown in Figure 5 b.
Illustratively, it is based on first threshold and handles the medical image acquisition coarse segmentation image, then pass through Laplce
Edge enhances the coarse segmentation image, obtains the Laplacian values of each point in the coarse segmentation image, and it is big to choose Laplacian values
In 120 region, described first image is obtained；The first threshold value range is that image value is more than 120, the Laplce
Enhancing can value obtains the Laplacian values of the point after 6 field differences around each point in the medical image by calculating.
Illustratively, skeletal tissue is extracted in the medical image based on second threshold, obtains described first image, institute
It is that image value is more than 600 to state second threshold value range.
In the present embodiment, the first threshold, second threshold and Laplce's threshold value can pass through data processing system 130
Or input-output apparatus 140 is default, and described image value can be the gray value of image, for CT CTA images, the figure
Picture value can refer to CT values.
It should be noted that since bone is hollow tissue, it, can be to described first image for the ease of subsequent processing
In several connected domains hollow area carry out tamping operations.
Step S430 is executed, described first image is projected along first direction, obtains projected image.The first party
To can drill Z-direction, the Z axis corresponding direction can be the direction from human foot to head.The projection pattern
Can be volume intensity projection technology, maximum intensity projection's technology, minimum intensity projection technology, average Intensity Projection technology etc.,
Or the arbitrary combination of above-mentioned technology.The projected image can be by projecting the cross section obtained to described first image along Z axis
Image.
Step S440 is executed, vertebral region is determined according to the projected image, obtains mask images.Due to vertebra directly with
Rib cage is connected, and is spatially that 33 pieces or so vertebras are formed by connecting by interverbebral disc and joint, and vertebra is in medical image
Image value is higher, in the projected image obtained by step S40 processing, due in Z-direction multiple canalis spinalis it is cumulative,
It should be in brightest area on cross section.Therefore, it can be determined and be waited according to the image mean value of each connected domain in the projected image
The vertebral region of choosing determines that vertebra point is seed point according to the vertebral region of the candidate.Based on the seed point with adaptive
Threshold value is grown, and is determined vertebral region, is obtained the mask images.
It is exemplary, obtain several connected domains in the mask images；The vertebra point for the connected domain for taking image value mean value big
For seed point.The vertebra point can be central point, barycenter, center of gravity or the image value of the big connected domain of described image value mean value
Maximum point.Adaptive threshold is carried out with the seed point.Region growing is carried out for initial threshold, such as the growth result obtained
Width is close with vertebra normal size or default vertebral dimensions, such as general vertebra in cross section is directly 6cm or so, then
Otherwise the growth result obtained with present threshold value increases threshold value continued growth for the vertebral region, until the grown junction obtained
The width of fruit is close with vertebra normal size or default vertebral dimensions.
Step S450 is executed, in described first image, the second image is obtained based on the mask images.Illustratively,
It can position in described first image according to the vertebral region in the mask images and correspond to vertebral region in every layer of two dimensional image
Connected domain sets the region to background area, realizes the purpose for removing the connected domain.It is removal vertebral region as shown in Figure 5 c
The second image afterwards.
Step S460 is executed, second image carries out connected domain analysis, extracts rib cage positioning result.Due to the spy of rib cage
Sign is proximate to human skin tissue, is in prolate shape in space, can carry out connected domain and analysis accordingly.The connected domain analysis can
With based on the average area and adjacent bone degree of closeness, connected domain letter on threshold range, xy axis range, yz axis range, x/y plane
Cease (such as connected domain points, cross-sectional area, cross-sectional area ratio, cross section Ratio of long radius to short radius, middle layer connected domain number), number
It is closed according to extractable (for example, growth etc.) or not extractable information, distance field information, with background degree of closeness, institutional framework position
The combination of one or more information such as system.X-axis corresponding direction can be from the direction to left side on the right side of human body, y-axis corresponding direction
Can be direction from human body shirtfront to back and z-axis corresponding direction can be direction from human foot to head.It is described
Xy axis range, yz axis range, x/y plane are cross section, coronal-plane and sagittal plane respectively.
Illustratively, several connected domains in second image are obtained；According to one or more combinations of the following conditions
Connected domain analysis is carried out to second image, obtains the positioning result of rib cage, as fig 5d.
As maximum width is more than threshold value a in the connected domain；The maximum width can go in the connected domain distance most
The wire length of two remote points.Since rib cage is in space prolate shape, the connected domain can be screened by maximum width
Whether it is the rib cage.The value range of the threshold value a is more than 20 pixel distances, and 30 are chosen in the present embodiment.
As the Ratio of long radius to short radius of the connected domain is more than threshold value b；Since rib cage is in prolate shape, major diameter value is more than section diameter value, can
Whether to be the rib cage according to connected domain described in rib cage this Feature Selection.The value range of the threshold value b is more than 1, originally
2 are chosen in embodiment.
Picture such as the connected domain in spatial position apart from background area less than threshold value c pixels number and the connected domain
The ratio between vegetarian refreshments sum is more than threshold value d；Due to rib cage on spatial position close to tissue region, can Feature Selection institute accordingly
State whether connected domain is the rib cage.The value range of the threshold value c is to be chosen for 40mm in the present embodiment more than 30；It is described
The value range of threshold value d is to be chosen for 0.5 in the present embodiment more than 0.4.
The area of the connected domain is less than threshold value e with circumscribed area ratio of holding；Since rib cage is in prolate shape, rib cage place
The area of connected domain levels off to its circumscribed area ratio, otherwise removes the connected domain.It can connected domain described in Feature Selection accordingly
Whether it is the rib cage.The value range of the threshold value e is the decimal less than 0.7, and 0.6 is chosen in the present embodiment.
In some embodiments, for rapidly connected domain analysis, adjacent n-layer X-Y scheme in described first image can be chosen
Connected domain as in makees above-mentioned connected domain analysis.
It should be noted that the condition of the connected domain analysis can be set by data processing system 130 or input/output
Standby 140 is default, and described image value can be the gray value of image, and for CT CTA images, described image value can refer to CT
Value.
It should be noted that the above description for rib cage positioning flow, can not be limited in the application and lift implementation
Within the scope of example.It is appreciated that for those skilled in the art, it, can be to the stream after the principle for understanding the flow
One or more operations of journey carry out the change in details.For example, adjusting operation order, mixing operation, partition operation, removing one
A or multiple operations, the one or more operations of increase etc..These change all without departing from scope of the claims.
The localization method of rib cage in the present embodiment traditional Chinese medicine figure, by being based on threshold method, gradient method, and/or Laplce
The methods of edge enhancing obtains the first image for including theme bony areas, to reduce the treating capacity of follow-up rib cage positioning and make
Shadow improves subsequent processing speed；Vertebral region is removed by positioning, obtains the second image of intercostal space approximate location；According to
Costal stigmata carries out connected domain analysis for second image, extracts rib cage positioning result.Rib cage provided in this embodiment
Localization method can quickly, be accurately positioned rib region, strong robustness, be suitable for all kinds of medical images, removal rib cage to blood vessel
The influence of extraction improves the accuracy of medical consultations.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
With within principle, any modification, equivalent substitution, improvement and etc. done should all be included in the protection scope of the present invention god.
Claims (10)
1. the localization method of rib cage in a kind of medical image, which is characterized in that including：
If obtaining the medical image for including dried layer two dimensional image；
Skeletal tissue is extracted according to the medical image, obtains the first image；
Described first image is projected along first direction, obtains projected image；
Candidate vertebral region is determined according to the image mean value of each connected domain in the projected image, according to the vertebra of the candidate
Bone region determines that vertebra point is seed point；It is grown with adaptive threshold based on the seed point, determines vertebral region, obtain mask
Image；
In described first image, the second image is obtained based on the mask images；
Connected domain analysis is carried out to second image, extracts rib cage positioning result.
2. the localization method of rib cage according to claim 1, which is characterized in that extract bone group according to the medical image
It knits, obtains the first image, including：
Enhanced based on first threshold and Laplce and handle the medical image, obtains described first image；Or
Skeletal tissue is extracted in the medical image based on second threshold, obtains described first image.
3. the localization method of rib cage according to claim 2, which is characterized in that the value range of the first threshold is figure
Picture value is more than 120, and the value range of the second threshold is that image value is more than 600.
4. the localization method of rib cage according to claim 1, which is characterized in that the acquisition mask images, including：
Obtain several connected domains in the mask images；
The vertebra point for the connected domain for taking image value mean value big is seed point；
It is grown with adaptive threshold based on the seed point, determines vertebral region, obtain the mask images.
5. the localization method of rib cage according to claim 4, which is characterized in that the vertebra point is described image value mean value
Central point, barycenter, center of gravity or the image value maximum point of big connected domain.
6. the localization method of rib cage according to claim 1, which is characterized in that it is described in described first image, it is based on
The mask images obtain the second image, including：
According to the vertebral region in the mask images, corresponding vertebral region in described first image is successively removed, obtains institute
State the second image.
7. the localization method of rib cage according to claim 1, which is characterized in that described to be connected to second image
Rib cage positioning result is extracted in domain analysis, including：
Obtain several connected domains in second image；
Confirm that connected domain meets following one or more conditions, that is, determines that the connected domain is rib cage, obtain rib cage positioning result：
As maximum width is more than threshold value a in the connected domain；Or
As the Ratio of long radius to short radius of the connected domain is more than threshold value b；Or
The pixel of pixel number and the connected domain such as the connected domain in spatial position apart from background area less than threshold value c
The ratio between point sum is more than threshold value d；Or
The ratio between the area of the connected domain and circumscribed rectangular area are less than threshold value e；
Otherwise the connected domain is removed.
8. the localization method of rib cage in the medical image described in claim 7, which is characterized in that the value range of the threshold value a
To be more than 20, the value range of the threshold value b be more than 1, the value range of the threshold value c be less than 50mm, the threshold value d's
Value range is the decimal more than 0.5, and the value range of the threshold value e is the decimal less than 0.7.
9. a kind of positioning system of rib cage, which is characterized in that including：
First extraction unit obtains the first image for extracting skeletal tissue according to medical image；
Projecting cell obtains projected image for projecting described first image along first direction；
Mask unit obtains mask images for determining vertebral region according to the projected image；
Second extraction unit determines candidate vertebral region with according to the image mean value of each connected domain in the projected image,
Determine that vertebra point is seed point according to the vertebral region of the candidate；It is grown, is determined with adaptive threshold based on the seed point
Vertebral region obtains the mask images；In described first image, the second image is obtained based on the mask images；
Positioning unit extracts rib cage positioning result for carrying out connected domain analysis to second image.
10. the positioning system of rib cage according to claim 9, which is characterized in that
The mask unit, including seed point determination subelement；And growth subelement, for being based on the seed point with adaptive
It answers threshold value to grow, determines vertebral region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611268139.XA CN106683090B (en) | 2016-12-31 | 2016-12-31 | The localization method and its system of rib cage in medical image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611268139.XA CN106683090B (en) | 2016-12-31 | 2016-12-31 | The localization method and its system of rib cage in medical image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106683090A CN106683090A (en) | 2017-05-17 |
| CN106683090B true CN106683090B (en) | 2018-10-09 |
Family
ID=58848903
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611268139.XA Active CN106683090B (en) | 2016-12-31 | 2016-12-31 | The localization method and its system of rib cage in medical image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106683090B (en) |
Families Citing this family (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107977971A (en) * | 2017-11-09 | 2018-05-01 | 哈尔滨理工大学 | The method of vertebra positioning based on convolutional neural networks |
| CN108389210A (en) * | 2018-02-28 | 2018-08-10 | 深圳天琴医疗科技有限公司 | A kind of medical image cutting method and device |
| CN110555850B (en) * | 2018-06-04 | 2023-11-03 | 青岛海信医疗设备股份有限公司 | Method, device, electronic equipment and storage medium for identifying rib area in image |
| CN110619621B (en) * | 2018-06-04 | 2023-10-27 | 青岛海信医疗设备股份有限公司 | Method, device, electronic equipment and storage medium for identifying rib area in image |
| CN110555860B (en) * | 2018-06-04 | 2023-11-14 | 青岛海信医疗设备股份有限公司 | Method for labeling rib areas in medical image, electronic equipment and storage medium |
| CN108924432B (en) * | 2018-06-12 | 2021-01-05 | 东软医疗系统股份有限公司 | Method, device and equipment for correcting signal deviation |
| CN110599494B (en) * | 2018-06-12 | 2021-11-02 | 中国科学院深圳先进技术研究院 | Rib image reconstruction system and method, terminal and readable storage medium |
| CN109124662B (en) * | 2018-07-13 | 2021-12-03 | 上海皓桦科技股份有限公司 | Rib center line detection device and method |
| CN109740602B (en) * | 2019-01-10 | 2023-06-30 | 上海联影医疗科技股份有限公司 | Pulmonary artery stage blood vessel extraction method and system |
| CN111968069B (en) * | 2019-05-20 | 2023-06-27 | 杭州依图医疗技术有限公司 | Image processing method, processing device and computer readable storage medium |
| CN112017148B (en) * | 2019-05-31 | 2024-03-22 | 杭州三坛医疗科技有限公司 | Method and device for extracting single-segment skeleton contour |
| CN110544245B (en) * | 2019-08-30 | 2020-11-27 | 北京推想科技有限公司 | Image processing method, image processing device, computer-readable storage medium and electronic equipment |
| CN110599505A (en) * | 2019-09-17 | 2019-12-20 | 上海微创医疗器械（集团）有限公司 | Organ image segmentation method and device, electronic equipment and storage medium |
| CN110807770A (en) * | 2019-10-30 | 2020-02-18 | 杭州依图医疗技术有限公司 | Medical image processing, recognizing and displaying method and storage medium |
| CN112890736B (en) * | 2019-12-03 | 2023-06-09 | 精微视达医疗科技（武汉）有限公司 | Method and device for obtaining field mask of endoscopic imaging system |
| CN112349391A (en) * | 2020-11-10 | 2021-02-09 | 山东大学齐鲁医院(青岛) | Optimized rib automatic labeling method |
| CN112489005B (en) * | 2020-11-26 | 2021-11-09 | 推想医疗科技股份有限公司 | Bone segmentation method and device, and fracture detection method and device |
| CN113139954B (en) * | 2021-05-11 | 2023-06-20 | 上海杏脉信息科技有限公司 | Medical image processing device and method |
| CN113506241A (en) * | 2021-05-25 | 2021-10-15 | 首都医科大学附属北京友谊医院 | Method for processing images of the ossicular chain and related product |
| CN113344920B (en) * | 2021-08-04 | 2022-03-22 | 浙江太美医疗科技股份有限公司 | Method and device for correcting medical image positioning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2557525A1 (en) * | 2011-08-08 | 2013-02-13 | Instytut Biologii Doswiadczalnej IM.M. Nenckiego Pan | A method and a system for processing an image comprising dendritic spines |
| CN105678746B (en) * | 2015-12-30 | 2018-04-03 | 上海联影医疗科技有限公司 | The localization method and device of liver scope in a kind of medical image |
| CN106127819B (en) * | 2016-06-30 | 2019-10-08 | 上海联影医疗科技有限公司 | The method and device thereof of vessel centerline are extracted in medical image |
-
2016
- 2016-12-31 CN CN201611268139.XA patent/CN106683090B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106683090A (en) | 2017-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106683090B (en) | The localization method and its system of rib cage in medical image | |
| CN106682636B (en) | Blood vessel extraction method and system | |
| CN106485704B (en) | Method for extracting center line of blood vessel | |
| CN106600609B (en) | Spine segmentation method and system in medical image | |
| WO2018023917A1 (en) | Method and system for extracting lower limb blood vessel | |
| CN107203741B (en) | Blood vessel extraction method, device and system | |
| CN112508965B (en) | Automatic outline sketching system for normal organs in medical image | |
| US7148688B2 (en) | Magnetic resonance imaging apparatus and method of controlling magnetic resonance imaging apparatus | |
| CN111161241B (en) | Liver image identification method, electronic equipment and storage medium | |
| CN107545584A (en) | The method, apparatus and its system of area-of-interest are positioned in medical image | |
| JP2011000439A (en) | Apparatus and method for isolating region in image | |
| CN106960439B (en) | A kind of vertebrae identification device and method | |
| US20110052028A1 (en) | Method and system of liver segmentation | |
| CN112862833A (en) | Blood vessel segmentation method, electronic device and storage medium | |
| JP5388638B2 (en) | Image processing apparatus and image processing program | |
| Küstner et al. | Semantic organ segmentation in 3D whole-body MR images | |
| CN106709925A (en) | Method and device for locating vertebral block in medical image | |
| Selle et al. | Mathematical methods in medical imaging: analysis of vascular structures for liver surgery planning | |
| CN116664635B (en) | Method, computing device and medium for constructing multi-dimensional dynamic model of target object | |
| CN111627023B (en) | Method and device for generating coronary artery projection image and computer readable medium | |
| EP3951433A1 (en) | Line structure extraction device, method and program, and learned model | |
| Sharma et al. | Importance of deep learning models to perform segmentation on medical imaging modalities | |
| Larralde et al. | Evaluation of a 3D segmentation software for the coronary characterization in multi-slice computed tomography | |
| CN113538419B (en) | Image processing method and system | |
| Chan et al. | SAM3D: Zero-Shot Semi-Automatic Segmentation in 3D Medical Images with the Segment Anything Model |
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 No. 2258 Chengbei Road, Jiading Industrial Zone, Jiading District, Shanghai. Patentee after: Shanghai Lianying Medical Technology Co., Ltd Address before: 201807 No. 2258 Chengbei Road, Jiading Industrial Zone, Jiading District, Shanghai. Patentee before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |