CN109118501A - Image processing method and system - Google Patents

Image processing method and system Download PDF

Info

Publication number
CN109118501A
CN109118501A CN201810878707.0A CN201810878707A CN109118501A CN 109118501 A CN109118501 A CN 109118501A CN 201810878707 A CN201810878707 A CN 201810878707A CN 109118501 A CN109118501 A CN 109118501A
Authority
CN
China
Prior art keywords
image
contour line
dimensional
correcting
image data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201810878707.0A
Other languages
Chinese (zh)
Inventor
叶震
汤雪华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Electric Group Corp
Original Assignee
Shanghai Electric Group Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Electric Group Corp filed Critical Shanghai Electric Group Corp
Priority to CN201810878707.0A priority Critical patent/CN109118501A/en
Publication of CN109118501A publication Critical patent/CN109118501A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/12Edge-based segmentation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00Three dimensional [3D] modelling, e.g. data description of 3D objects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/181Segmentation; Edge detection involving edge growing; involving edge linking
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/50Depth or shape recovery
    • G06T7/55Depth or shape recovery from multiple images
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00ICT specially adapted for the handling or processing of medical images
    • G16H30/20ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00ICT specially adapted for the handling or processing of medical images
    • G16H30/40ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10072Tomographic images
    • G06T2207/10081Computed x-ray tomography [CT]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30004Biomedical image processing
    • G06T2207/30008Bone

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Computer Graphics (AREA)
  • Geometry (AREA)
  • Software Systems (AREA)
  • Image Analysis (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a kind of image processing method and system, described image processing method includes: to obtain the first image data of the first image comprising process object;Processing is split to the first image according to the first image data, obtains the second image data of the process object in the first image;Obtain the second image of the process object corresponding with second image data;Automatically determine the image that correcting process is needed in second image.In the present invention, after being split processing and three-dimensional reconstruction by the image to process object, the two dimensional image for needing to be modified in image after judging automatically out dividing processing, and the image for needing correcting process filtered out is modified, the workload for efficiently reducing user, improves work efficiency;Meanwhile obtaining the 3-D image of process object again in conjunction with correction result, improve the precision and accuracy rate of processing result image.

Description

Image processing method and system
Technical field
The present invention relates to product image processing technology field, in particular to a kind of image processing method and system.
Background technique
With the continuous development of the technologies such as image processing techniques, visual computer technology, machine learning, calculates auxiliary and examine Treatment brings great convenience for doctor, as biggish mention has been obtained in terms of medical image segmentation and three-dimensional reconstruction effect It rises.But while there is also some shortcomings.On the one hand, there is certain limitation since computer automatically processes algorithm itself Property, it is not able to satisfy the demand of all possibilities, meanwhile, according to machine learning techniques, algorithm itself has self-learning function, but The result of study depends on the data for training, so that computer is automatically processed, there are certain limitations;On the other hand, For the data of same process object, such as CT data (electronic computer break tomographic data), there may be not by different doctors Same opinion.Therefore, when carrying out medical diagnosis using area of computer aided, after completing medical image segmentation and three-dimensional reconstruction, doctor It is raw still to carry out confirmation and modified demand.
But since the medical image quantity after rebuilding is more, for hip joint and knee joint reconstructing three-dimensional model CT data instance, sequence may up to 400 images, every image is successively checked and is corrected, is brought to doctor huge Workload also reduces working efficiency.
Summary of the invention
The technical problem to be solved by the present invention is to after medical image segmentation in the prior art and three-dimensional reconstruction, there are medicine figures Picture quantity is more, and every image is successively checked and corrected, brings huge workload to doctor, reduces working efficiency etc. and lacks It falls into, and it is an object of the present invention to provide a kind of image processing method and system.
The present invention is to solve above-mentioned technical problem by following technical proposals:
The present invention provides a kind of image processing method, and described image processing method includes:
S1, the first image data for obtaining the first image comprising process object;
S2, processing is split to the first image according to the first image data, obtained in the first image The process object the second image data;
S3, the second image for obtaining the process object corresponding with second image data;
S4, the image that correcting process is needed in second image is automatically determined.
Preferably, second image data includes the first two-dimensional image data, second image includes the first two dimension Image;And/or
Second image data includes the first 3 d image data, and second image includes the first 3-D image;
Wherein, first two dimensional image includes at least one in cross sectional image, coronal image and sagittal view picture Kind.
Preferably, working as, second image includes first two dimensional image and first two dimensional image includes the cross When cross-sectional image, step S4 is specifically included:
According to the cross sectional image, the first contour line of process object described in the cross sectional image is obtained automatically, And obtain the second contour line of process object described in the first image;
The first contour line be aligned comparing with second contour line, judge the first contour line with it is described Whether the second contour line is overlapped, if not, it is determined that second image is the image for needing correcting process.
Preferably, step S4 is specifically included when second image includes first two dimensional image:
Automatically the probability that first two dimensional image needs correcting process is calculated;
Determine that the probability is greater than image of first two dimensional image for setting probability threshold value to need correcting process.
Preferably, step S4 is specifically included when second image includes first 3-D image:
Obtain the abnormal point in first 3-D image;
Obtain first two dimensional image corresponding with abnormal point described in first 3-D image;
Determine that first two dimensional image is the image for needing correcting process.
Preferably, after step S4 further include:
S5, to needing the image of correcting process to be modified processing in second image, after obtaining correcting process the Two two dimensional images;
S6, according to after correcting process the corresponding image data of second two dimensional image and second image in be not required to The first image data for wanting the image of correcting process obtain the second 3-D image of the process object.
Preferably, step S5 includes:
Second image is compared with corresponding the first image, obtains and needs to correct in second image Correcting region;
The contour line in the correcting region is deleted, and repaints the corresponding contour line of the correcting region, until institute The contour line stated in correcting region is overlapped with contour line corresponding in the first image, obtains obtaining second after correcting process Two dimensional image;Or,
The profile curvature of a curve in the correcting region is adjusted, until the contour line and described first in the correcting region Corresponding contour line is overlapped in image, obtains obtaining the second two dimensional image after correcting process.
Preferably, the process object includes bone.
The present invention also provides a kind of image processing system, described image processing system includes the first image data acquisition mould Block, the second image data acquisition module, image collection module and automatically determine module;
The first image data acquisition module is used to obtain the first image data of the first image comprising process object;
The second image data acquisition module is for dividing the first image according to the first image data Processing is cut, the second image data of the process object in the first image is obtained;
Described image obtains the second figure that module is used to obtain the process object corresponding with second image data Picture;
The module that automatically determines is for automatically determining the image for needing correcting process in second image.
Preferably, second image data includes the first two-dimensional image data, second image includes the first two dimension Image;And/or
Second image data includes the first 3 d image data, and second image includes the first 3-D image;
Wherein, first two dimensional image includes at least one in cross sectional image, coronal image and sagittal view picture Kind.
Preferably, the module that automatically determines includes contour line acquiring unit, judging unit and the first determination unit;
When second image includes first two dimensional image and first two dimensional image includes the cross-sectional view When picture, the contour line acquiring unit is used to be obtained automatically described in the cross sectional image according to the cross sectional image The first contour line of object is managed, and obtains the second contour line of process object described in the first image;
The judging unit with second contour line for be aligned comparing the first contour line, described in judgement Whether first contour line is overlapped with second contour line, if it is not, then calling the determination unit;
First determination unit is for determining that second image is the image for needing correcting process.
Preferably, the module that automatically determines includes probability calculation unit and the second determination unit;
When second image includes first two dimensional image, the probability calculation unit is for described in calculating automatically First two dimensional image needs the probability of correcting process;
Second determination unit be used to determine the probability be greater than set first two dimensional image of probability threshold value as Need the image of correcting process.
Preferably, the module that automatically determines includes abnormal point acquiring unit, image acquisition unit and third determination unit;
When second image includes first 3-D image, the abnormal point acquiring unit is for obtaining described the Abnormal point in one 3-D image;
Described image acquiring unit is for obtaining corresponding with abnormal point described in first 3-D image described first Two dimensional image;
The third determination unit is for determining that first two dimensional image is the image for needing correcting process.
Preferably, described image processing system includes correction module and acquiring three-dimensional images module;
The correction module is used for needing the image of correcting process to be modified processing in second image, and acquisition is repaired Just treated the second two dimensional image;
The acquiring three-dimensional images module is used for according to the corresponding picture number of second two dimensional image after correcting process According to the first image data with second image for not needing correcting process in image, the of the process object is obtained Two 3-D images.
Preferably, the correction module includes correcting region acquiring unit and amending unit;
The correcting region acquiring unit is obtained for comparing second image with corresponding the first image It takes in second image and needs modified correcting region;
The amending unit is used to delete the contour line in the correcting region, and it is corresponding to repaint the correcting region Contour line, until the correcting region in contour line be overlapped with contour line corresponding in the first image, obtained The second two dimensional image after correcting process;Or,
The amending unit is used to adjust the profile curvature of a curve in the correcting region, until in the correcting region Contour line is overlapped with contour line corresponding in the first image, obtains obtaining the second two dimensional image after correcting process.
Preferably, the process object includes bone.
The positive effect of the present invention is that:
In the present invention, after being split processing and three-dimensional reconstruction by the image to process object, judges automatically out and divide It needs to be modified the two dimensional image of processing in treated image, and the image for needing correcting process filtered out is repaired Just, the workload for efficiently reducing user, improves work efficiency;Meanwhile obtaining process object again in conjunction with correction result 3-D image improves the precision and accuracy rate of processing result image.
Detailed description of the invention
Fig. 1 is the flow chart of the image processing method of the embodiment of the present invention 1.
Fig. 2 is the image segmentation result schematic diagram of the image processing method of the embodiment of the present invention 1.
Fig. 3 is the flow chart of the image processing method of the embodiment of the present invention 2.
Fig. 4 is the flow chart of the image processing method of the embodiment of the present invention 3.
Fig. 5 is the flow chart of the image processing method of the embodiment of the present invention 4.
Fig. 6 is the flow chart of the image processing method of the embodiment of the present invention 5.
Fig. 7 is the flow chart of the image processing method of the embodiment of the present invention 6.
Fig. 8 is the module diagram of the image processing system of the embodiment of the present invention 7.
Fig. 9 is the module diagram of the image processing system of the embodiment of the present invention 8.
Figure 10 is the module diagram of the image processing system of the embodiment of the present invention 9.
Figure 11 is the module diagram of the image processing system of the embodiment of the present invention 10.
Figure 12 is the module diagram of the image processing system of the embodiment of the present invention 11.
Figure 13 is the module diagram of the image processing system of the embodiment of the present invention 12.
Specific embodiment
The present invention is further illustrated below by the mode of embodiment, but does not therefore limit the present invention to the reality It applies among a range.
Embodiment 1
As shown in Figure 1, the image processing method of the present embodiment includes:
S101, the first image data for obtaining the first image comprising process object;
Wherein, process object includes bone, but is not limited to bone, can also include other kinds of process object.
First image is the CT scan image of skeletal sites, is indicated in the form of a sequence, picture format DICOM (Digital Imaging and Communications in Medicine, digital imaging and communications in medicine).DICOM is doctor The international standard (ISO 12052) for learning image and relevant information defines the data exchange that can be used for for being able to satisfy clinical needs Medical Image Format analyzes the numerical value for obtaining characterization tissue according to dicom standard.
Specifically, CT scan is using accurate X-ray beam, gamma-rays, ultrasonic wave etc., together with the detector high with sensitivity Make profile scanning one by one around a certain position of human body.One sequence refers to multiple comprising all information of disease sites The set of tomoscan image.Further, a sequence refers to all CT scan picture numbers comprising target body tissue site According to one of DICOM file set.
S102, processing is split to the first image according to the first image data, obtains the process object in the first image The second image data;
Specifically, FCN (Full Convolutional Network, full convolutional Neural are used according to the first image data Network) algorithm carries out coarse segmentation processing to the first image, Level Set is then used according to coarse segmentation treated image data Method (Level Set Method) carries out precision processing processing, obtains the second image data of the process object in the first image.
S103, the second image for obtaining process object corresponding with the second image data;
Wherein, by set display parameters, the process object region in the second image that dividing processing is obtained with A kind of color is shown that the region except process object in the first image is shown with another color (such as Fig. 2 institute Show, the region a indicates bony areas, and the region b indicates other regions in CT scan image other than bony areas), it is partitioned into Deal with objects corresponding image.
Wherein, the second image data includes the first two-dimensional image data, and the second image includes the first two dimensional image;And/or
Second image data includes the first 3 d image data, and the second image includes the first 3-D image;
Wherein, at using Marching Cubes algorithm (a kind of iso-surface patch algorithm) to the first two-dimensional image data Reason obtains the first 3 d image data;The first 3-D image of process object is obtained further according to the first 3 d image data.
Specifically, include: using the specific steps that Marching Cubes algorithm finally obtains the first 3-D image
1) the first two-dimensional image data is read in into three-dimensional array;
2) volume elements is extracted from three-dimensional array, becomes current voxel, while obtaining all information of the volume elements, such as 8 The value on a vertex, coordinate etc.;
3) value on 8 vertex of current voxel is compared with the value of given contour surface, obtains the state of the volume elements;
4) according to the state index of current voxel, the volume elements seamed edge intersected with contour surface is found out, and using linear interpolation Method calculates the position coordinates of each intersection point;
5) calculate the plane by each seamed edge intersection point in the volume elements, using the plane normal orientation, each seamed edge intersection point as Vertex, the first 3 d image data;
6) the first 3-D image dealt with objects according to the first 3 d image data Isosurface construction.
In addition, the first two dimensional image includes at least one of cross sectional image, coronal image and sagittal view picture.? The contour line for sketching the contours of segmentation result edge in cross sectional image with lines, checks convenient for user and confirms segmentation result.
General user's multi-angle for convenience checks process object, shows the cross of process object simultaneously in software interactive interface Cross-sectional image, coronal image, sagittal view picture and the first 3-D image four open image, each image position root when display It is replaced according to actual demand.
S104, the image that correcting process is needed in the second image is automatically determined.
In the present embodiment, processing is split by the image to process object, the image after automatically determining dividing processing The middle image for needing to be modified, efficiently reduces the workload of user, improves work efficiency.
Embodiment 2
As shown in figure 3, the image processing method of the present embodiment is the further improvement to embodiment 1, specifically:
When the second image includes the first two dimensional image and the first two dimensional image includes cross sectional image, step S104 is specific Include:
S1041, according to cross sectional image, it is automatic to obtain the first contour line dealt with objects in cross sectional image, and obtain The second contour line dealt with objects in first image;
S1042, first contour line be aligned comparing with the second contour line, judge first contour line and the second contour line Whether it is completely coincident, if not, it is determined that the second image is the image for needing correcting process.
In the present embodiment, processing is split by the image to process object, the cross-sectional view after obtaining dividing processing As in the first contour line that deals with objects and with the second contour line for being dealt with objects in the first image, judge first contour line and the Whether two contour lines are completely coincident to determine whether the second image is the image for needing correcting process, to efficiently reduce user Workload, improve work efficiency.
Embodiment 3
As shown in figure 4, the image processing method of the present embodiment is the further improvement to embodiment 1, specifically:
When the second image includes the first two dimensional image, step S104 is specifically included:
S1043, the automatic probability for calculating the first two dimensional image and needing correcting process;
Specifically, the method for probability is calculated are as follows: calculate two figures adjacent with the sequence number of current first two dimensional image The differential index (di) of the segmentation result of picture, the probability for needing correcting process of the size of the differential index (di) and current first two dimensional image It is positively correlated.
S1044, determine that probability is greater than image of the first two dimensional image for setting probability threshold value to need correcting process.
In the present embodiment, processing is split by the image to process object, every second after obtaining dividing processing Image needs the probability of correcting process, picks out probability and is greater than the second image of setting probability threshold value as needing correcting process Image improves work efficiency to efficiently reduce the workload of user.
Embodiment 4
As shown in figure 5, the image processing method of the present embodiment is the further improvement to embodiment 1, specifically:
When the second image includes the first 3-D image, step S104 is specifically included:
Abnormal point in S1045, the first 3-D image of acquisition;
Wherein, abnormal point, which refers to, obviously is not belonging to deal with objects the corresponding point in region of itself.
S1046, the first two dimensional image corresponding with abnormal point in the first 3-D image is obtained;
Specifically, it according to the mapping relations of the coordinate system of displaing coordinate system and threedimensional model where process object, will show Show that the target abnormal point of interactive process selection in coordinate system is converted to the space coordinate point on threedimensional model coordinate system;
According to the mapping relations between threedimensional model coordinate system and two-dimensional image sequence, by the space on threedimensional model coordinate system Coordinate points are scaled the point in the second image of two-dimensional image sequence, and obtain the sequence number of corresponding second image.In addition, Since the sequence number of two-dimensional image sequence is discrete integer value, the sequence number to convert may be not present in really Two-dimensional image sequence in, at this point, two two dimensional images nearest from the point when rebuilding threedimensional model are as needing Corrections Division The image of reason.
S1047, determine that the first two dimensional image is the image for needing correcting process.
In the present embodiment, processing is split by the image to process object, obtains and deals with objects corresponding three-dimensional figure Abnormal point as in obtains the first two dimensional image corresponding with abnormal point in the first 3-D image, determines that the first two dimensional image is The image of correcting process is needed, to efficiently reduce the workload of user, is improved work efficiency.
Embodiment 5
As shown in fig. 6, the image processing method of the present embodiment is the further improvement to embodiment 1, specifically:
After step S104 further include:
S105, the image that correcting process is needed in the second image is modified to processing, second after obtaining correcting process Two dimensional image;
S106, according to after correcting process the corresponding image data of the second two dimensional image and the second image in do not need to correct First image data of the image of processing obtains the second 3-D image of process object.
Finally, display is to process object (such as bone) corresponding two dimensional image segmentation result (cross sectional image, coronal-plane figure Picture and sagittal view picture) and 3-D image, until user thinks that all confirmation and amendment work are completed.
In the present embodiment, processing is split by the image to process object, the image after automatically determining dividing processing The middle image for needing to be modified, and the image for needing correcting process filtered out is modified, to efficiently reduce use The workload at family, improves work efficiency;Meanwhile obtaining the 3-D image of process object again in conjunction with correction result, it improves The precision and accuracy rate of processing result image.
Embodiment 6
As shown in fig. 7, the image processing method of the present embodiment is the further improvement to embodiment 5, specifically:
Step S105 includes:
S1051, the second image is compared with corresponding first image, obtains in the second image and needs modified amendment Region;
Specifically, left mouse button dragging is pinned, is unclamped after reaching target position, it is left with left mouse button depressed position, mouse Key released position is in cornerwise rectangular area (i.e. correcting region), and the contour line for being included is the profile for choosing part.
Left mouse button dragging is pinned, makes to drag track formation closed curve, the contour line that encapsulation curvilinear inner is included is i.e. For the profile for choosing part;If not formed encapsulation curve when dragging, that is, presses the coordinate of left mouse button and unclamp left mouse button Position is inconsistent, continuous with straight line, is closed curve.
Contour line in S1052, deletion correcting region, and the corresponding contour line of correcting region is repainted, until amendment Contour line in region is completely coincident with corresponding contour line in the first image, obtains obtaining the second X-Y scheme after correcting process Picture;
Specifically, the mode of the corresponding contour line of correcting region is repainted are as follows: pin left mouse button dragging, mouse is mobile Track is new contour line, unclamps left mouse button after the completion.
Or, adjustment correcting region in profile curvature of a curve, until correcting region in contour line with it is right in the first image The contour line answered is completely coincident, and obtains obtaining the second two dimensional image after correcting process.
It chooses and needs to need modified correcting region in the second image of correcting process, left mouse button is moved to the amendment area Position point, presses left mouse button and does not unclamp among profile in domain, mobile mouse position, release when meeting the requirements.It moves When dynamic, new curve is to choose profile two o'clock as curve starting point and terminal, while mouse point is the secondary of third point Spline curve;
Desired curvature of curve is inputted after choosing, computer is automatically generated to choose profile two o'clock as curve starting point and end Point, while using input value as the curve of curvature, choose partial contour line to delete as new contour line, and by original.
In the present embodiment, processing is split by the image to process object, the image after automatically determining dividing processing The middle image for needing to be modified, and the image for needing correcting process filtered out is modified, to efficiently reduce use The workload at family, improves work efficiency;Meanwhile obtaining the 3-D image of process object again in conjunction with correction result, it improves The precision and accuracy rate of processing result image.
Embodiment 7
As shown in figure 8, the image processing system of the present embodiment includes the first image data acquisition module 1, the second picture number According to acquisition module 2, image collection module 3 and automatically determine module 4.
First image data acquisition module 1 is used to obtain the first image data of the first image comprising process object;
Wherein, process object includes bone, but is not limited to bone, can also include other kinds of process object.
First image is the CT scan image of skeletal sites, is indicated in the form of a sequence, picture format DICOM (Digital Imaging and Communications in Medicine, digital imaging and communications in medicine).DICOM is doctor The international standard (ISO 12052) for learning image and relevant information defines the data exchange that can be used for for being able to satisfy clinical needs Medical Image Format analyzes the numerical value for obtaining characterization tissue according to dicom standard.
Specifically, CT scan is using accurate X-ray beam, gamma-rays, ultrasonic wave etc., together with the detector high with sensitivity Make profile scanning one by one around a certain position of human body.One sequence refers to multiple comprising all information of disease sites The set of tomoscan image.Further, a sequence refers to all CT scan picture numbers comprising target body tissue site According to one of DICOM file set.
Second image data acquisition module 2 is used to be split processing to the first image according to the first image data, obtains Second image data of the process object in the first image;
Specifically, FCN (Full Convolutional Network, full convolutional Neural are used according to the first image data Network) algorithm carries out coarse segmentation processing to the first image, Level Set is then used according to coarse segmentation treated image data Method (Level Set Method) carries out precision processing processing, obtains the second image data of the process object in the first image.
Image collection module 3 is used to obtain the second image of process object corresponding with the second image data;
Wherein, by set display parameters, the process object region in the second image that dividing processing is obtained with A kind of color is shown that the region except process object in the first image is shown with another color (such as Fig. 2 institute Show, the region a indicates bony areas, and the region b indicates other regions in CT scan image other than bony areas), it is partitioned into Deal with objects corresponding image.Wherein, the second image data includes the first two-dimensional image data, and the second image includes the first two dimension Image;And/or
Second image data includes the first 3 d image data, and the second image includes the first 3-D image;
Wherein, at using Marching Cubes algorithm (a kind of iso-surface patch algorithm) to the first two-dimensional image data Reason obtains the first 3 d image data;The first 3-D image of process object is obtained further according to the first 3 d image data.
Specifically, include: using the specific steps that Marching Cubes algorithm finally obtains the first 3-D image
1) the first two-dimensional image data is read in into three-dimensional array;
2) volume elements is extracted from three-dimensional array, becomes current voxel, while obtaining all information of the volume elements, such as 8 The value on a vertex, coordinate etc.;
3) value on 8 vertex of current voxel is compared with the value of given contour surface, obtains the state of the volume elements;
4) according to the state index of current voxel, the volume elements seamed edge intersected with contour surface is found out, and using linear interpolation Method calculates the position coordinates of each intersection point;
5) calculate the plane by each seamed edge intersection point in the volume elements, using the plane normal orientation, each seamed edge intersection point as Vertex, the first 3 d image data;
6) the first 3-D image dealt with objects according to the first 3 d image data Isosurface construction.
In addition, the first two dimensional image includes at least one of cross sectional image, coronal image and sagittal view picture.? The contour line for sketching the contours of segmentation result edge in cross sectional image with lines, checks convenient for user and confirms segmentation result.
General user's multi-angle for convenience checks process object, shows the cross of process object simultaneously in software interactive interface Cross-sectional image, coronal image, sagittal view picture and the first 3-D image four open image, each image position root when display It is replaced according to actual demand.Module 4 is automatically determined for automatically determining the image for needing correcting process in the second image.
In the present embodiment, processing is split by the image to process object, the image after automatically determining dividing processing The middle image for needing to be modified, efficiently reduces the workload of user, improves work efficiency.
Embodiment 8
As shown in figure 9, the image processing system of the present embodiment is the further improvement to embodiment 7, specifically:
Automatically determining module 4 includes contour line acquiring unit 41, judging unit 42 and the first determination unit 43.
When the second image includes the first two dimensional image and the first two dimensional image includes cross sectional image, contour line obtains single Member 41 obtains the first figure for obtaining the first contour line dealt with objects in cross sectional image automatically according to cross sectional image The second contour line dealt with objects as in;
Judging unit 42 judges first contour line and for be aligned comparing first contour line with the second contour line Whether two contour lines are completely coincident, if it is not, then calling determination unit;
First determination unit 43 is for determining that the second image is the image for needing correcting process.
In the present embodiment, processing is split by the image to process object, the cross-sectional view after obtaining dividing processing As in the first contour line that deals with objects and with the second contour line for being dealt with objects in the first image, judge first contour line and the Whether two contour lines are completely coincident to determine whether the second image is the image for needing correcting process, to efficiently reduce user Workload, improve work efficiency.
Embodiment 9
As shown in Figure 10, the image processing system of the present embodiment is the further improvement to embodiment 7, specifically:
Automatically determining module 4 includes probability calculation unit 44 and the second determination unit 45.
When the second image includes the first two dimensional image, probability calculation unit 44 is needed for calculating the first two dimensional image automatically Want the probability of correcting process;
Specifically, the difference of the segmentation result of two image adjacent with the sequence number of current first two dimensional image is calculated Index, the size of the differential index (di) and the probability for needing correcting process of current first two dimensional image are positively correlated.Second determines Unit 45 is used to determine that probability to be greater than image of the first two dimensional image for setting probability threshold value to need correcting process.
In the present embodiment, processing is split by the image to process object, every second after obtaining dividing processing Image needs the probability of correcting process, picks out probability and is greater than the second image of setting probability threshold value as needing correcting process Image improves work efficiency to efficiently reduce the workload of user.
Embodiment 10
As shown in figure 11, the image processing system of the present embodiment is the further improvement to embodiment 7, specifically:
Automatically determining module 4 includes abnormal point acquiring unit 46, image acquisition unit 47 and third determination unit 48.
When the second image includes the first 3-D image, abnormal point acquiring unit 46 is for obtaining in the first 3-D image Abnormal point;
Wherein, abnormal point, which refers to, obviously is not belonging to deal with objects the corresponding point in region of itself.Image acquisition unit 47 is used In acquisition the first two dimensional image corresponding with abnormal point in the first 3-D image;
Specifically, it according to the mapping relations of the coordinate system of displaing coordinate system and threedimensional model where process object, will show Show that the target abnormal point of interactive process selection in coordinate system is converted to the space coordinate point on threedimensional model coordinate system;
According to the mapping relations between threedimensional model coordinate system and two-dimensional image sequence, by the space on threedimensional model coordinate system Coordinate points are scaled the point in the second image of two-dimensional image sequence, and obtain the sequence number of corresponding second image.In addition, Since the sequence number of two-dimensional image sequence is discrete integer value, the sequence number to convert may be not present in really Two-dimensional image sequence in, at this point, two two dimensional images nearest from the point when rebuilding threedimensional model are as needing Corrections Division The image of reason.
Third determination unit 48 is for determining that the first two dimensional image is the image for needing correcting process.
In the present embodiment, processing is split by the image to process object, obtains and deals with objects corresponding three-dimensional figure Abnormal point as in obtains the first two dimensional image corresponding with abnormal point in the first 3-D image, determines that the first two dimensional image is The image of correcting process is needed, to efficiently reduce the workload of user, is improved work efficiency.
Embodiment 11
As shown in figure 12, the image processing system of the present embodiment is the further improvement to embodiment 7, specifically:
Image processing system includes correction module 5 and acquiring three-dimensional images module 6.
Correction module 5 is used to for the image for needing correcting process in the second image to be modified processing, obtains correcting process The second two dimensional image afterwards;
Acquiring three-dimensional images module 6 is used for according to the corresponding image data of the second two dimensional image and second after correcting process The first image data that the image of correcting process is not needed in image obtains the second 3-D image of process object.
Finally, display is to process object (such as bone) corresponding two dimensional image segmentation result (cross sectional image, coronal-plane figure Picture and sagittal view picture) and 3-D image, until user thinks that all confirmation and amendment work are completed.
In the present embodiment, processing is split by the image to process object, the image after automatically determining dividing processing The middle image for needing to be modified, and the image for needing correcting process filtered out is modified, to efficiently reduce use The workload at family, improves work efficiency;Meanwhile obtaining the 3-D image of process object again in conjunction with correction result, it improves The precision and accuracy rate of processing result image.
Embodiment 12
As shown in figure 13, the image processing system of the present embodiment is the further improvement to embodiment 11, specifically:
Correction module 5 includes correcting region acquiring unit 51 and amending unit 52.
Correcting region acquiring unit 51 obtains the second image for comparing the second image with corresponding first image It is middle to need modified correcting region;
Specifically, left mouse button dragging is pinned, is unclamped after reaching target position, it is left with left mouse button depressed position, mouse Key released position is in cornerwise rectangular area (as correcting region), and the contour line for being included is the wheel for choosing part It is wide.
Left mouse button dragging is pinned, makes to drag track formation closed curve, the contour line that encapsulation curvilinear inner is included is i.e. For the profile for choosing part;If not formed encapsulation curve when dragging, that is, presses the coordinate of left mouse button and unclamp left mouse button Position is inconsistent, continuous with straight line, is closed curve.
Amending unit 52 is used to delete the contour line in correcting region, and repaints the corresponding contour line of correcting region, Until the contour line in correcting region is completely coincident with corresponding contour line in the first image, the after obtaining acquisition correcting process Two two dimensional images;
Specifically, the mode of the corresponding contour line of correcting region is repainted are as follows: pin left mouse button dragging, mouse is mobile Track is new contour line, unclamps left mouse button after the completion.
Or, amending unit 52 is used to adjust the profile curvature of a curve in correcting region, until the contour line in correcting region It is completely coincident with contour line corresponding in the first image, obtains obtaining the second two dimensional image after correcting process.
It chooses and needs to need modified correcting region in the second image of correcting process, left mouse button is moved to the amendment area Position point, presses left mouse button and does not unclamp among profile in domain, mobile mouse position, release when meeting the requirements.It moves When dynamic, new curve is to choose profile two o'clock as curve starting point and terminal, while mouse point is the secondary of third point Spline curve;
Desired curvature of curve is inputted after choosing, computer is automatically generated to choose profile two o'clock as curve starting point and end Point, while using input value as the curve of curvature, choose partial contour line to delete as new contour line, and by original.
In the present embodiment, processing is split by the image to process object, the image after automatically determining dividing processing The middle image for needing to be modified, and the image for needing correcting process filtered out is modified, to efficiently reduce use The workload at family, improves work efficiency;Meanwhile obtaining the 3-D image of process object again in conjunction with correction result, it improves The precision and accuracy rate of processing result image.
Although specific embodiments of the present invention have been described above, it will be appreciated by those of skill in the art that these It is merely illustrative of, protection scope of the present invention is defined by the appended claims.Those skilled in the art is not carrying on the back Under the premise of from the principle and substance of the present invention, various changes or modifications can be made to these embodiments, but these are changed Protection scope of the present invention is each fallen with modification.

Claims (16)

1. a kind of image processing method, which is characterized in that described image processing method includes:
S1, the first image data for obtaining the first image comprising process object;
S2, processing is split to the first image according to the first image data, obtains the institute in the first image State the second image data of process object;
S3, the second image for obtaining the process object corresponding with second image data;
S4, the image that correcting process is needed in second image is automatically determined.
2. image processing method as described in claim 1, which is characterized in that second image data includes the first X-Y scheme As data, second image includes the first two dimensional image;And/or
Second image data includes the first 3 d image data, and second image includes the first 3-D image;
Wherein, first two dimensional image includes at least one of cross sectional image, coronal image and sagittal view picture.
3. image processing method as claimed in claim 2, which is characterized in that when second image includes first two dimension Image and when first two dimensional image includes the cross sectional image, step S4 is specifically included:
According to the cross sectional image, the first contour line of process object described in the cross sectional image is obtained automatically, and is obtained Take the second contour line of process object described in the first image;
The first contour line be aligned comparing with second contour line, judges the first contour line and described second Whether contour line is overlapped, if not, it is determined that second image is the image for needing correcting process.
4. image processing method as claimed in claim 2, which is characterized in that when second image includes first two dimension When image, step S4 is specifically included:
Automatically the probability that first two dimensional image needs correcting process is calculated;
Determine that the probability is greater than image of first two dimensional image for setting probability threshold value to need correcting process.
5. image processing method as claimed in claim 2, which is characterized in that when second image includes described first three-dimensional When image, step S4 is specifically included:
Obtain the abnormal point in first 3-D image;
Obtain first two dimensional image corresponding with abnormal point described in first 3-D image;
Determine that first two dimensional image is the image for needing correcting process.
6. image processing method as described in claim 1, which is characterized in that after step S4 further include:
S5, to needing the image of correcting process to be modified processing in second image, the two or two after obtaining correcting process Tie up image;
S6, according to after correcting process the corresponding image data of second two dimensional image and second image in do not need to repair The first image data of the image just handled obtain the second 3-D image of the process object.
7. image processing method as claimed in claim 6, which is characterized in that step S5 includes:
Second image is compared with corresponding the first image, obtains in second image and needs modified repair Positive region;
The contour line in the correcting region is deleted, and repaints the corresponding contour line of the correcting region, until described repair Contour line in positive region is overlapped with contour line corresponding in the first image, obtains obtaining the second two dimension after correcting process Image;Or,
The profile curvature of a curve in the correcting region is adjusted, until contour line and the first image in the correcting region In corresponding contour line be overlapped, obtain obtaining the second two dimensional image after correcting process.
8. the image processing method as described in any one of claims 1 to 7, which is characterized in that the process object includes bone Bone.
9. a kind of image processing system, which is characterized in that described image processing system includes the first image data acquisition module, the Two image data acquisition modules, image collection module and automatically determine module;
The first image data acquisition module is used to obtain the first image data of the first image comprising process object;
The second image data acquisition module is used to be split place to the first image according to the first image data Reason obtains the second image data of the process object in the first image;
Described image obtains the second image that module is used to obtain the process object corresponding with second image data;
The module that automatically determines is for automatically determining the image for needing correcting process in second image.
10. image processing system as claimed in claim 9, which is characterized in that second image data includes the first two dimension Image data, second image include the first two dimensional image;And/or
Second image data includes the first 3 d image data, and second image includes the first 3-D image;
Wherein, first two dimensional image includes at least one of cross sectional image, coronal image and sagittal view picture.
11. image processing system as claimed in claim 10, which is characterized in that the module that automatically determines includes that contour line obtains Take unit, judging unit and the first determination unit;
When second image includes first two dimensional image and first two dimensional image includes the cross sectional image, The contour line acquiring unit is used to obtain process object described in the cross sectional image automatically according to the cross sectional image First contour line, and obtain the second contour line of process object described in the first image;
The judging unit judges described first for be aligned comparing the first contour line with second contour line Whether contour line is overlapped with second contour line, if it is not, then calling the determination unit;
First determination unit is for determining that second image is the image for needing correcting process.
12. image processing system as claimed in claim 10, which is characterized in that the module that automatically determines includes probability calculation Unit and the second determination unit;
When second image includes first two dimensional image, the probability calculation unit for calculating described first automatically Two dimensional image needs the probability of correcting process;
Second determination unit is used to determine that the probability to be greater than first two dimensional image for setting probability threshold value as needs The image of correcting process.
13. image processing system as claimed in claim 10, which is characterized in that the module that automatically determines includes that abnormal point obtains Take unit, image acquisition unit and third determination unit;
When second image includes first 3-D image, the abnormal point acquiring unit is for obtaining the described 1st Tie up the abnormal point in image;
Described image acquiring unit is for obtaining first two dimension corresponding with abnormal point described in first 3-D image Image;
The third determination unit is for determining that first two dimensional image is the image for needing correcting process.
14. image processing system as claimed in claim 9, which is characterized in that described image processing system includes correction module With acquiring three-dimensional images module;
The correction module is used to obtain Corrections Division to needing the image of correcting process to be modified processing in second image The second two dimensional image after reason;
The acquiring three-dimensional images module be used for according to after correcting process the corresponding image data of second two dimensional image and The first image data that the image of correcting process is not needed in second image, obtain the two or three of the process object Tie up image.
15. image processing system as claimed in claim 14, which is characterized in that the correction module includes that correcting region obtains Unit and amending unit;
The correcting region acquiring unit obtains institute for comparing second image with corresponding the first image It states and needs modified correcting region in the second image;
The amending unit is used to delete the contour line in the correcting region, and repaints the corresponding wheel of the correcting region Profile obtains obtaining amendment until the contour line in the correcting region is overlapped with contour line corresponding in the first image Treated the second two dimensional image;Or,
The amending unit is used to adjust the profile curvature of a curve in the correcting region, until the profile in the correcting region Line is overlapped with contour line corresponding in the first image, obtains obtaining the second two dimensional image after correcting process.
16. the image processing system as described in any one of claim 9 to 15, which is characterized in that the process object includes Bone.
CN201810878707.0A 2018-08-03 2018-08-03 Image processing method and system Pending CN109118501A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810878707.0A CN109118501A (en) 2018-08-03 2018-08-03 Image processing method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810878707.0A CN109118501A (en) 2018-08-03 2018-08-03 Image processing method and system

Publications (1)

Publication Number Publication Date
CN109118501A true CN109118501A (en) 2019-01-01

Family

ID=64851784

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810878707.0A Pending CN109118501A (en) 2018-08-03 2018-08-03 Image processing method and system

Country Status (1)

Country Link
CN (1) CN109118501A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114511560A (en) * 2022-04-18 2022-05-17 武汉大学 Bile duct segmentation method and device for MRCP (Markov random Access control) image

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104091365A (en) * 2014-07-12 2014-10-08 大连理工大学 Acetabulum tissue model reconstruction method for serialization hip joint CT image
CN105912874A (en) * 2016-04-29 2016-08-31 青岛大学附属医院 Liver three-dimensional database system constructed on the basis of DICOM (Digital Imaging and Communications in Medicine) medical image
CN106447764A (en) * 2016-09-08 2017-02-22 福州大学 Three-dimensional visualization automatic measuring method for human body pelvis parameters
CN106469445A (en) * 2015-08-18 2017-03-01 青岛海信医疗设备股份有限公司 A kind of calibration steps of 3-D view, device and system
CN106846346A (en) * 2017-01-23 2017-06-13 天津大学 Sequence C T image pelvis profile rapid extracting methods based on key frame marker
CN107203989A (en) * 2017-04-01 2017-09-26 南京邮电大学 End-to-end chest CT image dividing method based on full convolutional neural networks
CN107292928A (en) * 2017-06-16 2017-10-24 沈阳东软医疗系统有限公司 A kind of method and device of blood vessel positioning
CN108197539A (en) * 2017-12-21 2018-06-22 西北大学 A kind of Diagnosis of Crania By Means identification method
CN108335304A (en) * 2018-02-07 2018-07-27 华侨大学 A kind of aortic aneurysm dividing method of abdominal CT scan sequence image

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104091365A (en) * 2014-07-12 2014-10-08 大连理工大学 Acetabulum tissue model reconstruction method for serialization hip joint CT image
CN106469445A (en) * 2015-08-18 2017-03-01 青岛海信医疗设备股份有限公司 A kind of calibration steps of 3-D view, device and system
CN105912874A (en) * 2016-04-29 2016-08-31 青岛大学附属医院 Liver three-dimensional database system constructed on the basis of DICOM (Digital Imaging and Communications in Medicine) medical image
CN106447764A (en) * 2016-09-08 2017-02-22 福州大学 Three-dimensional visualization automatic measuring method for human body pelvis parameters
CN106846346A (en) * 2017-01-23 2017-06-13 天津大学 Sequence C T image pelvis profile rapid extracting methods based on key frame marker
CN107203989A (en) * 2017-04-01 2017-09-26 南京邮电大学 End-to-end chest CT image dividing method based on full convolutional neural networks
CN107292928A (en) * 2017-06-16 2017-10-24 沈阳东软医疗系统有限公司 A kind of method and device of blood vessel positioning
CN108197539A (en) * 2017-12-21 2018-06-22 西北大学 A kind of Diagnosis of Crania By Means identification method
CN108335304A (en) * 2018-02-07 2018-07-27 华侨大学 A kind of aortic aneurysm dividing method of abdominal CT scan sequence image

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114511560A (en) * 2022-04-18 2022-05-17 武汉大学 Bile duct segmentation method and device for MRCP (Markov random Access control) image

Similar Documents

Publication Publication Date Title
US5771310A (en) Method and apparatus for recording three-dimensional topographies
US6201889B1 (en) Apparatus and computer program for generating a feedback code
US8384716B2 (en) Image processing method
CN103854270A (en) CT and MR inter-machine three dimensional image fusion registration method and system
Jones Facial Reconstruction Using Volumetric Data.
JPH06149991A (en) Method for converting image data into vector data
CN105096372A (en) Synchronized views of video data and three-dimensional model data
CN108711187A (en) The method that registration fusion CT and MRI signal establish human body lumbar vertebrae three-dimensional simulation model
CN107895364B (en) A kind of three-dimensional reconstruction system for the preoperative planning of virtual operation
CN103679810A (en) Method for three-dimensional reconstruction of liver computed tomography (CT) image
CN102254317A (en) Method for automatically extracting dental arch curved surface in dental implantation navigation
EP1214688A1 (en) Processing method and system for 3-d geometric modeling of the spine
JP2007523402A (en) Mesh model with internal individual elements
US11291423B2 (en) System and method of radiograph correction and visualization
AU2018301580B2 (en) Three-dimensional ultrasound image display method
Chougule et al. Clinical case study: spine modeling for minimum invasive spine surgeries (MISS) using rapid prototyping
Chougule et al. Development of patient specific implants for Minimum Invasive Spine Surgeries (MISS) from non-invasive imaging techniques by reverse engineering and additive manufacturing techniques
CN106991720B (en) Personalized acetabulum reconstruction steel plate pre-bending method based on finite element analysis and operation
US20060098008A1 (en) Method, device and computer program product for generating a three-dimensional model
CN106934810B (en) A kind of spine correcting device
Irwansyah et al. Algorithm for segmentation and reduction of fractured bones in computer-aided preoperative surgery
Ben-Hamadou et al. Teeth3ds: a benchmark for teeth segmentation and labeling from intra-oral 3d scans
CN109118501A (en) Image processing method and system
Qian et al. An automatic tooth reconstruction method based on multimodal data
JP2001118058A (en) Image processor and radiation medical treatment planning system

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20190101

RJ01 Rejection of invention patent application after publication