CN104299217B - A kind of region real time method for segmenting of medical image - Google Patents
A kind of region real time method for segmenting of medical image Download PDFInfo
- Publication number
- CN104299217B CN104299217B CN201310296542.3A CN201310296542A CN104299217B CN 104299217 B CN104299217 B CN 104299217B CN 201310296542 A CN201310296542 A CN 201310296542A CN 104299217 B CN104299217 B CN 104299217B
- Authority
- CN
- China
- Prior art keywords
- region
- real
- real time
- segmenting
- interest
- 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 OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
- G06F3/0484—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
- G06F3/0486—Drag-and-drop
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present invention provides a kind of region real time method for segmenting of medical image, including：Starting point is selected, positions area-of-interest；By controlling the position of terminating point to adjust the scope of partitioning parameters t and the area-of-interest in real time, and region Real-time segmentation is carried out in the area-of-interest.The region real time method for segmenting of medical image provided by the invention is by controlling terminating point while positioning adjusts the scope of area-of-interest, adjustment partitioning parameters t in real time, Real-time segmentation processing intuitively is carried out to image according to the demand of user, really real-time, accurate segmentation result is obtained, meets the real-time interactive demand of user.
Description
Technical field
The present invention relates to the region real time method for segmenting in medical information processing field, more particularly to a kind of medical image.
Background technology
Image Segmentation Technology is that digital picture is subdivided into the process of multiple images subregion, it is therefore intended that simplifies or changes
The representation of image so that image is more easy to be understood and analyzed, image split as digital picture important process means and
It is widely used in all trades and professions.In field of medical image processing, image Segmentation Technology understands form, the size of lump for doctor
Play the part of very important role Deng key property.
In the prior art, substantial amounts of medical image Real-time segmentation technology is limited only to the positioning of area-of-interest：(1) base
Handled in the Real-time segmentation of gradient, partitioning parameters t can not be controlled in real time, what is obtained in real time is the result of fixed segmentation, for side
The complicated lump segmentation effect of edge is simultaneously bad；(2) after selecting starting point in area-of-interest, the process for choosing terminating point is pulled
In, region growing is carried out from four direction based on starting point, only when four direction all reaches the border in region to be split
Final segmentation result is shown, real-time region segmentation result can not be obtained.
The content of the invention
Present invention solves the technical problem that it is to provide a kind of region real time method for segmenting of medical image, to solve user
During carrying out real-time region segmentation, fail to adjust partitioning parameters in real time while the positioning of area-of-interest is realized, make
Segmentation result can not be controlled in real time according to their needs by obtaining user.
In order to solve the above problems, the invention provides a kind of region segmentation method of medical image, comprise the following steps：
Starting point is selected, positions area-of-interest；
By controlling the position of terminating point to adjust the scope of partitioning parameters t and the area-of-interest in real time, and described
Region Real-time segmentation is carried out in area-of-interest；
Wherein, the partitioning parameters t be the starting point and ending point line vector on y direction component it is big
It is small.
Optionally, the method by controlling terminating point to carry out region Real-time segmentation comprises the following steps：
If the terminating point is located in region to be split, with the region growing of fixed threshold in the area-of-interest
Carry out Real-time segmentation；
If the terminating point is located at outside region to be split, sent out in the area-of-interest with adaptive region growing
Carry out Real-time segmentation.
Optionally, the bottom threshold of the region growing methods of the fixed threshold is that the starting point and the terminating point connect
Minimum gradation value on line.
Optionally, the bottom threshold of the adaptive region growing methods is：ThL=[(1-t) × (Tmax+Tmin)]/
2, wherein TmaxAnd TminRespectively current maximum gradation value and minimum gradation value in growth district.
Optionally, the area-of-interest be with the line vector of the starting point and ending point in X direction component
Size for radius circle composition region.
Optionally, projection of the line vector of the starting point and ending point in x-axis is located in the region to be split,
Then the terminating point is located in the region to be split, and otherwise the terminating point is located at outside region to be split.
Optionally, the terminating point is determined by pulling the method for mouse action or contact action.
Compared with prior art, the present invention has advantages below：
(1) present invention can be widely applied in the region Real-time segmentation processing of various images, and the degree of accuracy is high, and speed is fast,
It is directly perceived convenient；(2) the parameter t of partitioning algorithm can be adjusted in the present invention while area-of-interest scope is controlled so that point
Cut result and be rapidly and accurately presented to user in real time with area-of-interest and partitioning parameters t adjustment, obtain really real
When, accurate segmentation result.
Brief description of the drawings
Fig. 1 show the schematic flow sheet of the region real time method for segmenting of the medical image of one embodiment of the invention；
Fig. 2 is the result schematic diagram of the region real time method for segmenting of the medical image of one embodiment of the invention；
Fig. 3 a~3b are result schematic diagram of the one embodiment of the invention in the inside and outside Real-time segmentation in region to be split；
Fig. 4 a~4b are the result schematic diagram of the breast lump Real-time segmentation of one embodiment of the invention.
Embodiment
Many details are elaborated in the following description in order to fully understand the present invention.But the present invention can be with
Much it is different from other manner described here to implement, those skilled in the art can be in the situation without prejudice to intension of the present invention
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 the embodiment of the present invention is described in detail, for purposes of illustration only, institute
It is example to state schematic diagram, and it should not limit the scope of protection of the invention herein.
In the prior art, user often only defines the model of area-of-interest during the region Real-time segmentation of medical image
Enclose, but partitioning parameters can not be defined according to their needs so as to efficiently control Real-time segmentation result to solve above-mentioned technology
Problem, the invention provides a kind of region real time method for segmenting of medical image, including：Starting point is selected, positions region of interest
Domain；By controlling the position of terminating point to adjust partitioning parameters and the scope of the area-of-interest in real time, and described interested
Region Real-time segmentation is carried out in region.
The present invention is by the way that while area-of-interest is positioned on various types of medical images, ginseng is split in adjustment in real time
Number, so as to obtain accurate, real-time region segmentation result.
The present invention will be described in detail below in conjunction with the accompanying drawings.Such as the area of the medical image of Fig. 1 one embodiment of the invention
Shown in the schematic flow sheet of domain real time method for segmenting, methods described comprises the following steps：
First, step S10 is performed：Starting point 2 is chosen according to region 1 to be split on the original image, determined as seed point
Position area-of-interest 3.Then S20 is performed：User's control terminating point 4, it is real-time while 3 scope of area-of-interest is adjusted
Partitioning parameters t is adjusted, and region Real-time segmentation is carried out in the area-of-interest 3, obtains Real-time segmentation result 5.Wherein, such as
Shown in the result schematic diagram of the region real time method for segmenting of the medical image of Fig. 2 one embodiment of the invention, the area to be split
Domain 1 is the breast lump region in segmentation object region, such as mammary gland nuclear magnetic resonance image；The area-of-interest 3 is with institute
The line vector 6 of starting point 2 and terminating point 4 is stated in the region that the size of X direction component 7 forms for the circle of radius；Such as Fig. 3 a
~3b is the partitioning parameters shown in result schematic diagram of the one embodiment of the invention in 1 inside and outside Real-time segmentation of region to be split
T is the size of the component 8 on y direction of line vector 6 of starting point 2 and terminating point 4.The terminating point can pass through dragging
The determination of the method for mouse or contact action, the invention is not limited in this regard.
It should be noted that in a kind of region real time method for segmenting of medical image provided by the invention, the segmentation ginseng
Number t changes in real time with the change of terminating point 4, because will to partitioning parameters t value under different images or scene
Ask different, such as during the MRI progress Real-time segmentation of corresponding breast lump, suitable for sketch figure picture and strengthen image
Partitioning parameters t is typically different so that partitioning parameters t can be adjusted flexibly as needed by user, emerging in Perceived control in the prior art
While interesting region, partitioning parameters are not adjusted in real time so that accurate, real-time requirement that segmentation result can not meet, so this
Method can be applied in the region Real-time segmentation processing of the various images under different scenes.
Then step S30~S60 is performed：Divided in real time by controlling terminating point 4 to carry out region in the area-of-interest 3
Cut：If the terminating point 4 is located in region 1 to be split, with the region growing side of fixed threshold in the area-of-interest 3
Method carries out Real-time segmentation, obtains the Real-time segmentation result 5 in region 1 to be split, as shown in Figure 3 a；Treated if the terminating point 4 is located at
Outside cut zone 1, then Real-time segmentation is carried out with adaptive region growing forwarding method in the area-of-interest 3, treated
The Real-time segmentation result 5 of cut zone 1, as shown in Figure 3 b.It should be noted that whether the terminating point 4 is located at area to be split
The inside and outside standard in domain 1 is：If projection of the line vector 6 of the starting point 2 and the terminating point 4 in x-axis is treated positioned at described
In cut zone 1, then the terminating point 4 is located in the region 1 to be split, and otherwise the terminating point 4 is located at region 1 to be split
Outside, as shown in the inside and outside result schematic diagram in real time in the region to be split that Fig. 3 a~3b are one embodiment of the invention.
Further, the bottom threshold of the region growing methods of the fixed threshold is the starting point 2 and the termination
Minimum gradation value on 4 lines of point；The bottom threshold of the adaptive region growing methods is：ThL=[(1-t) × (Tmax+
Tmin)]/2, wherein TmaxAnd TminRespectively current maximum gradation value and minimum gradation value in growth district, according to above-mentioned threshold
It is worth lower limit ThLFormula understand, TmaxAnd TminThe continuous real-time update with the change of the partitioning parameters t, and with segmentation
Parameter t size is directly proportional, the bottom threshold increased by changing partitioning parameters t sizes come real-time adjustment region, according to formula,
Partitioning parameters t value is smaller, ThLBe worth it is smaller, so as to including pixel it is more, the scope of growth district is also bigger.
As Fig. 4 a~4b show the result schematic diagram of the breast lump Real-time segmentation of one embodiment of the invention.Wherein,
Seed point of the starting point 2 that user chooses as positioning area-of-interest, user pull what control was chosen by mouse or touch screen
Terminating point 4, partitioning parameters t and control area-of-interest size are adjusted by controlling terminating point 4, real-time mammary gland is obtained and swells
The segmentation result 9 of block, obtained as shown in Figure 4 a after the region growth of the fixed value threshold value in region to be split real-time
Breast lump segmentation result, the real-time breast obtained after increasing as shown in Figure 4 b for the adaptive region outside region to be split
Adenoncus block segmentation result 10.Compared with prior art, real time method for segmenting provided by the invention is adjusted by controlling terminating point 2
Feel the size and partitioning parameters in emerging region so that segmentation result is quick, accurate with the adjustment of area-of-interest and partitioning parameters
User really is presented in real time, Real-time segmentation processing intuitively is carried out to image according to the demand of user, meets that user's is real-time
Interactive demand.
Although the present invention is disclosed as above with preferred embodiment, it is not for limiting the present invention, any this area
Technical staff without departing from the spirit and scope of the present invention, may be by the methods and technical content of the disclosure above to this hair
Bright technical scheme makes possible variation and modification, therefore, every content without departing from technical solution of the present invention, according to the present invention
Any simple modifications, equivalents, and modifications made to above example of technical spirit, belong to technical solution of the present invention
Protection domain.
Claims (7)
1. the region real time method for segmenting of a kind of medical image, it is characterised in that comprise the following steps：
Starting point is selected, positions area-of-interest；
By controlling the position of terminating point to adjust the scope of partitioning parameters t and the area-of-interest in real time, and it is emerging in the sense
Region Real-time segmentation is carried out in interesting region；
Wherein, the partitioning parameters t is the size of vector component on y direction of the starting point and ending point line.
2. region real time method for segmenting as claimed in claim 1, it is characterised in that described by controlling terminating point to carry out region
The method of Real-time segmentation comprises the following steps：
If the terminating point is located in region to be split, with the region growing method of fixed threshold in the area-of-interest
Carry out Real-time segmentation；
If the terminating point is located at outside region to be split, entered in the area-of-interest with adaptive region growing method
Row Real-time segmentation.
3. region real time method for segmenting as claimed in claim 2, it is characterised in that the region growing methods of the fixed threshold
Bottom threshold be the minimum gradation value in the starting point and the terminating point line.
4. region real time method for segmenting as claimed in claim 2, it is characterised in that the adaptive region growing methods
Bottom threshold is：ThL=[(1-t) × (Tmax+Tmin)]/2, wherein TmaxAnd TminRespectively current maximum in growth district
Gray value and minimum gradation value.
5. the region real time method for segmenting as described in any one of Claims 1 to 4, it is characterised in that the area-of-interest is
The region that circle by radius of the size of line vector component in X direction of the starting point and ending point forms.
6. region real time method for segmenting as claimed in claim 2, it is characterised in that the line of the starting point and ending point to
The projection measured in x-axis is located in the region to be split, then the terminating point is located in the region to be split, otherwise described
Terminating point is located at outside region to be split.
7. region real time method for segmenting as claimed in claim 1, it is characterised in that the terminating point is by pulling mouse action
Or the method for contact action determines.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310296542.3A CN104299217B (en) | 2013-07-16 | 2013-07-16 | A kind of region real time method for segmenting of medical image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310296542.3A CN104299217B (en) | 2013-07-16 | 2013-07-16 | A kind of region real time method for segmenting of medical image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104299217A CN104299217A (en) | 2015-01-21 |
| CN104299217B true CN104299217B (en) | 2018-04-03 |
Family
ID=52318939
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310296542.3A Active CN104299217B (en) | 2013-07-16 | 2013-07-16 | A kind of region real time method for segmenting of medical image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104299217B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109509197B (en) * | 2018-09-26 | 2021-11-09 | 东软医疗系统股份有限公司 | Method, device and equipment for dividing region of interest and storage medium |
| CN115511894A (en) * | 2021-06-22 | 2022-12-23 | 上海微创卜算子医疗科技有限公司 | Blood vessel image segmentation method, electronic device and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102306373A (en) * | 2011-08-17 | 2012-01-04 | 深圳市旭东数字医学影像技术有限公司 | Method and system for dividing up three-dimensional medical image of abdominal organ |
| WO2012017375A2 (en) * | 2010-08-05 | 2012-02-09 | Koninklijke Philips Electronics N.V. | In-plane and interactive surface mesh adaptation |
| CN103049907A (en) * | 2012-12-11 | 2013-04-17 | 深圳市旭东数字医学影像技术有限公司 | Interactive image segmentation method |
-
2013
- 2013-07-16 CN CN201310296542.3A patent/CN104299217B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2012017375A2 (en) * | 2010-08-05 | 2012-02-09 | Koninklijke Philips Electronics N.V. | In-plane and interactive surface mesh adaptation |
| CN102306373A (en) * | 2011-08-17 | 2012-01-04 | 深圳市旭东数字医学影像技术有限公司 | Method and system for dividing up three-dimensional medical image of abdominal organ |
| CN103049907A (en) * | 2012-12-11 | 2013-04-17 | 深圳市旭东数字医学影像技术有限公司 | Interactive image segmentation method |
Non-Patent Citations (4)
| Title |
|---|
| Aitomated extraction of bronchns from 3D CT images of lung based on genetic algorithm and 3D region growing;Tsui Ying Law 等;《Medical Imaging 2000: Image Processing》;20000606;906-916 * |
| 一种基于改进随机游走的肺结节分割方法;依玉峰 等;《东北大学学报（自然科学版）》;20120315;第33卷(第3期);第319页左栏第2段，第320页右栏第3-4段 * |
| 一种改进的交互式医学图像序列分割方法;罗希平 等;《电子学报》;20030125(第1期);29-32 * |
| 面向生物医学图像的交互式分割算法;吴琳 等;《计算机工程》;20100820;第36卷(第16期);208-209 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104299217A (en) | 2015-01-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3745339B1 (en) | Method for implanting advertisements in video, and computer device | |
| CN103942755B (en) | Brightness of image adjusting method and device | |
| CN103049907B (en) | Interactive image segmentation method | |
| CN103489224B (en) | A kind of interactive three-dimensional point cloud color edit methods | |
| US9547410B2 (en) | Selection editing using a localized level set algorithm | |
| CN103456010A (en) | Human face cartoon generation method based on feature point localization | |
| TWI553590B (en) | Method and device for retargeting a 3d content | |
| TW201546758A (en) | Image segmentation device and image processing method | |
| JP6582472B2 (en) | Image processing apparatus, image processing system, and program | |
| CN102214306A (en) | Leaf disease spot identification method and device | |
| JP2015057698A (en) | Color video processing system and method, and corresponding computer program | |
| US20220309670A1 (en) | Method and system for visualizing information on gigapixels whole slide image | |
| CN104299217B (en) | A kind of region real time method for segmenting of medical image | |
| CN106228590B (en) | A kind of human body attitude edit methods in image | |
| JP2012104099A (en) | Method of enhancing contrast using bezier curve | |
| CN110009082B (en) | Three-dimensional code optimization method, medium, computer device and apparatus | |
| CN109117837A (en) | Method and device for determining region of interest | |
| CN112528991A (en) | Method, device, equipment and storage medium for labeling lumbar intervertebral disc | |
| CN103164843B (en) | A kind of medical image colorize method | |
| CN111010605B (en) | Method for displaying video picture-in-picture window | |
| CN110135315A (en) | Eye pupil replacement method and device based on human eye key point | |
| CN107918944B (en) | A kind of picture color filling method and device | |
| CN105427242A (en) | Interactive grid constraint deformation image self-adaptive scaling method based on content perception | |
| CN118714244A (en) | Film and television production system based on deep learning reverse relighting algorithm | |
| CN112307923A (en) | Partitioned expression migration method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 201815 No. 1180 Xingxian Road, Jiading Industrial Zone, Jiading District, Shanghai Patentee after: Shanghai Lianying Medical Technology Co., Ltd Address before: 201815 No. 1180 Xingxian Road, Jiading Industrial Zone, Jiading District, Shanghai Patentee before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd. |
|
| CP02 | Change in the address of a patent holder | ||
| CP02 | Change in the address of a patent holder |
Address after: 201807 2258 Chengbei Road, Jiading District, Shanghai Patentee after: Shanghai Lianying Medical Technology Co.,Ltd. Address before: 201815 No. 1180 Xingxian Road, Jiading Industrial Zone, Jiading District, Shanghai Patentee before: Shanghai Lianying Medical Technology Co.,Ltd. |