CN100543778C - The virtual grid of sub-volumes is aimed at - Google Patents
The virtual grid of sub-volumes is aimed at Download PDFInfo
- Publication number
- CN100543778C CN100543778C CNB2005800363220A CN200580036322A CN100543778C CN 100543778 C CN100543778 C CN 100543778C CN B2005800363220 A CNB2005800363220 A CN B2005800363220A CN 200580036322 A CN200580036322 A CN 200580036322A CN 100543778 C CN100543778 C CN 100543778C
- Authority
- CN
- China
- Prior art keywords
- volumes
- sub
- subvolumes
- implemented method
- computer implemented
- 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
Images
Landscapes
- Image Generation (AREA)
Abstract
The method that is used to feature extraction and aims at sub-volumes comprises: receive data volume, a plurality of parameters (401) of sub-volumes are provided, and determine the starting point (402) of each sub-volumes in two or more subvolumes according to a plurality of parameters.This method also comprises the amount of mis-alignment of determining these two or more subvolumes, the starting point that moves each sub-volumes is with the grid alignment (403) with these starting points and virtual volume, and the feature (407) in two or more subvolumes of determining to be aligned.
Description
The sequence number that the application requires on October 22nd, 2004 to submit to is No.60/621,198 U.S. Provisional Application No., and this application is introduced into as a reference in this integral body.
Technical field
The present invention relates to graphical analysis, and relate in particular to the system and method that is used for the volume sampling.
Background technology
Usually carry out data processing to determine eigenwert or to produce letter data that for example, three-dimensional computed tomography (CT) scans to sub-volumes rather than to whole data volume.Usually, from initial volume, extract sub-volumes by this way, make the sample frequency of sub-volumes and the sample frequency of initial volume not match.For example, some algorithms may need the sub-volumes isotropy, and initial volume may not be isotropic or under different resolution isotropy.In this case, from initial volume, extract sub-volumes by initial volume being sampled with given sample frequency.Because this sampling, voxel and the voxel in the sub-volumes in the initial volume do not have correspondence one by one.
When extracting a plurality of sub-volumes from initial volume, sub-volumes is overlapping usually.Fig. 1 illustrates an overlapping example.Initial volume is shown as border 101.From this volume, extract two subvolumes 102 and 103.The algorithm of supposing to handle sub-volumes is extracting the feature on the area-of-interest (ROI) 104.This area-of-interest 104 can have for two overlapping sub-volumes 102 and 103, as shown in fig. 1.Ideally, if two or more subvolumes within it portion comprise the same section of ROI, then the eigenwert of determining of two or more subvolumes should be identical.Similarly, when the same section of the ROI of antithetical phrase internal volume carried out video picture, they should seem identical.
In fact, may be different to two subvolumes institute calculated feature values, and the video picture of two subvolumes may seem different.The reason of this difference is as follows.When having extracted first sub-volumes, initial volume is sampled to obtain the desirable sample frequency of sub-volumes.When having extracted second sub-volumes, the repeated sampling process.The sample position of two subvolumes by observing initial volume inside can determine that the sample position of two subvolumes does not match in their overlapping areas.For example, referring to Fig. 2, wherein first and second sub-volumes 202 and 203 sample position are respectively equidistant, but do not match.As a result of, the voxel value difference in the total zone of two subvolumes.This species diversity of voxel value causes the difference of determined eigenwert.
Fig. 2 illustrates the sample position of two overlapping sub-volumes.The position of line 201 expression original samples, the sample position of line 202 expressions first sub-volumes, and the sample position of line 203 expressions second sub-volumes.Can observe, two subvolumes 202 and 203 sample position do not match.This causes eigenwert is different for two subvolumes.
Therefore, need a kind of system and method that be used to make the sample-synchronous of a plurality of sub-volumes, wherein sample position is identical in the sub-volumes overlapping areas.
Summary of the invention
Embodiment according to present disclosure, a kind of computer implemented method that is used for aiming at for feature extraction sub-volumes comprises: receives data volume, a plurality of parameters of sub-volumes is provided, and the starting point of determining each sub-volumes in two or more subvolumes according to a plurality of parameters.This method also comprises the amount of mis-alignment of determining two or more subvolumes, and the starting point that moves each sub-volumes is so that the grid alignment of these starting points and virtual volume, and the feature in definite two or more subvolumes that have been aligned.
Provide a plurality of parameters to comprise to select the central point of each sub-volumes in two or more subvolumes, select the size and dimension of these two or more subvolumes, and select the spacing between the voxel in these two or more subvolumes.This method also comprises and converts voxel spacing to volume unit.This method utilizes the voxel spacing of starting point and volume unit form to extract two or more subvolumes from volume before being included in and extracting feature.
Carry out the starting point of determining these two or more subvolumes according to central point, sub-volumes and voxel spacing.
Grid with respect to virtual volume is determined amount of mis-alignment, and wherein the spacing of virtual volume grid is the spacing between the voxel of sub-volumes, and the initial point of the grid of virtual volume conforms to the initial point of world coordinate system (world coordinate system).
This method comprises that the starting point that will be moved converts volume coordinates to, and wherein (0,0,0) position of each sub-volumes is as initial point.These two or more subvolumes can have different resolution, and data volume can comprise two or more data volume of representing different modalities.
Only carry out mobile starting point with respect to one or two axle of sub-volumes.Sub-volumes can be two dimension and on two-dimensional sub-volumes, determine video picture and feature.Three-dimensional or more determine the aligning of starting point on the multidimensional.Under the situation of multidimensional volume, this method comprises the selection one or more dimensions, wherein determines the aligning of starting point on selected dimension.Under the situation of multi-modal volume, the aligning of using starting point is to cross over the different modalities that has been write down in advance.
According to the embodiment of present disclosure, a kind of program storage device is set to machine readable, comprises effectively to be used for aiming at for feature extraction the method step of sub-volumes with execution by the instruction repertorie of this machine execution.This method step comprises: receives data volume, a plurality of parameters of sub-volumes is provided, and the starting point of determining each sub-volumes in two or more subvolumes according to a plurality of parameters.This method also comprises: determine the amount of mis-alignment of these two or more subvolumes, the starting point that moves each sub-volumes is with the grid alignment with these starting points and virtual volume, and the feature in two or more subvolumes of determining to be aligned.
Embodiment according to present disclosure, a kind of computer implemented method that is used for aiming at for video picture sub-volumes comprises: receive data volume, a plurality of parameters of sub-volumes are provided, and determine the starting point of each sub-volumes in two or more subvolumes according to a plurality of parameters.This method comprises: the amount of mis-alignment of determining these two or more subvolumes, the starting point that moves each sub-volumes is with the grid alignment with these starting points and virtual volume, and show the sub-volumes with the starting point that is aligned, wherein the area-of-interest in each sub-volumes has identical voxel value.
Provide a plurality of parameters to comprise: to select the central point of each sub-volumes in two or more subvolumes, select the size and dimension of these two or more subvolumes, and select the spacing between the voxel in these two or more subvolumes.This method comprises and converts voxel spacing to volume unit.This method utilizes the voxel spacing of starting point and volume unit form to extract two or more subvolumes from volume before being included in and showing sub-volumes.
According to the embodiment of present disclosure, a kind of program storage device is set to machine readable, comprises the method step that can be used for aiming at for video picture sub-volumes by the instruction repertorie that machine is carried out with execution effectively.The method that is used for aiming at for video picture sub-volumes comprises: receives data volume, a plurality of parameters of sub-volumes is provided, and the starting point of determining each sub-volumes in two or more subvolumes according to a plurality of parameters.This method comprises: the amount of mis-alignment of determining two or more subvolumes, the starting point that moves each sub-volumes is with the grid alignment with these starting points and virtual volume, and show sub-volumes with the starting point that is aligned, wherein the area-of-interest in each sub-volumes has identical voxel value.
Description of drawings
The preferred embodiments of the present invention are described below with reference to the accompanying drawings in more detail:
Fig. 1 is the diagram that the overlapping sub-volumes that extracts from volume is shown;
Fig. 2 is the diagram of the sample position of two overlapping sub-volumes;
Fig. 3 is the diagram according to the input volume embodiment of present disclosure, on virtual grid;
Fig. 4 be according to the embodiment of present disclosure, in two subvolumes overlapping areas the diagram of identical sub-volumes sample position;
Fig. 5 is the process flow diagram according to the method for the embodiment of present disclosure; And
Fig. 6 is the block diagram according to the system of the embodiment of present disclosure.
Embodiment
Embodiment according to present disclosure, the system and method hypothesis overlapping region that is used to extract the sub-volumes with substantially the same eigenwert of determining for a plurality of sub-volumes with overlapping region covers whole area-of-interest (ROI), for example tubercle, is wherein just determining feature on this area-of-interest.This system and method makes the sample-synchronous of a plurality of sub-volumes, and wherein sample position is substantially the same in the sub-volumes overlapping areas.
With reference to figure 3, created virtual volume 301 by on virtual grid 303, original input volume 32 being sampled.The initial point of virtual grid 303 (0,0,0) conforms to the initial point of world coordinate system 304.In addition, the required mesh spacing with sub-volumes is identical basically for mesh spacing.From virtual volume 301, extract required sub-volumes.Owing to further virtual volume 301 is not sampled, so the voxel in the sub-volumes overlapping areas is substantially the same.
Do not need to create virtual volume, so that save memory resource.Can from initial volume, directly extract the required sub-volumes of aiming at virtual volume grid.Can use if define enough storeies, can create virtual volume so and can from virtual volume, extract sub-volumes.By using the virtual volume whole initial volume is not carried out resampling, allow to be used for the high sample frequency of sub-volumes according to the method for the embodiment of present disclosure.This can improve efficient and the precision of handling sub-volumes.
With reference to figure 4, comprise that the input of a plurality of parameters of the sub-volumes that data volume and definition will be extracted is provided 401.These parameters for example comprise with central point (C coordinate representation, in the world coordinate system
W), required subvolume size (D
W) and the voxel spacing (VS of sub-volumes
W).These parameters can be provided or can be determined in advance by the user.Utilize central point, subvolume size and voxel spacing, determine the sub-volumes S in the world coordinate system
WStarting point 402:
S
W＝C
W-D
W＊VS
W
The starting point of sub-volumes is moved with the grid alignment 403 with itself and virtual volume, for example, and following expression:
Δ＝S
W?mod?VS
W
If (Δ〉VS
W/ 2)
S
W＝S
W+(VS
W-Δ)
Otherwise
S
W＝S
W-VS
W
The described grid alignment of determining the starting point and the virtual volume of sub-volumes.The spacing VS of virtual volume grid
WIdentical with the mesh spacing of required sub-volumes.Δ is the amount of mis-alignment of sub-volumes with respect to virtual volume grid.In order to reduce the amount of movement of starting point, with Δ and (VS
W/ 2) compare.This restriction from original start point to+/-VS
W/ 2 move.
Starting point is converted into volume coordinates 404, and wherein (0,0,0) position of initial volume is as initial point.
Required voxel spacing is converted into original volume units 405 from millimeter.
The voxel spacing of utilization determined starting point and volume unit form in module 404 and 405 is extracted sub-volumes 406 from initial volume.Utilize (for example aiming at) sub-volumes that is extracted on data volume, to determine feature.
Fig. 5 is the identical diagram of sample position that is illustrated in sub-volumes in the two subvolumes overlapping areas.For the sake of simplicity, only show vertical sample line.In Fig. 5, show the result based on the situation shown in Fig. 2 according to present disclosure.Can observe, two subvolumes 202 are identical with sample position in 203 the initial volume.As a result of, the voxel in the overlapping region is shared identical between should the sub-volumes in zone.
The sub-volumes of given aligning, for sub-volumes on shared ROI 104 determined any feature all with identical.This feature unchangeability depends on the position of ROI and based on such hypothesis, promptly the voxel in shared ROI does not help feature calculation and value, and between sub-volumes shared whole ROI.
Should be understood that the present invention can implement with the various forms of hardware, software, firmware, dedicated processor or their combination.In one embodiment, the present invention can be implemented as the application program that is included in conscientiously on the program storage device with software.This application program can be uploaded on the machine that comprises any suitable construction and by this machine and carry out.
With reference to figure 6, according to the embodiment of present disclosure, the computer system 601 that is used to carry out the method for the sample-synchronous that is used to make a plurality of sub-volumes especially comprises CPU (central processing unit) (CPU) 602, storer 603 and I/O (I/O) interface 604.This computer system 601 is coupled on display 605 and the various input equipments 606 such as mouse and keyboard by I/O interface 604 usually.This display 605 can show the view of virtual volume and the image that is write down.Auxiliary circuit can comprise such as cache memory, power supply, clock circuit and communication bus.Storer 603 can comprise random-access memory (ram), ROM (read-only memory) (ROM), disc driver, tape drive or the like or their combination.The present invention can be implemented as program 607, and it is stored in the storer 603 and by CPU 602 and carries out to handle the signal from signal source 608.Equally, computer system 601 is general-purpose computing systems, and when carrying out program 607 of the present invention, this computer system becomes dedicated computer system.
Should further be appreciated that because the parts and the method step of some construction systems of being described in the accompanying drawings can be implemented with software, so the mode that the actual connection between the system unit (perhaps method step) can be programmed according to the present invention and difference.Give to fix on the instruction of the present invention that this provides, those of ordinary skill in the related art can be susceptible to of the present invention these with similar embodiment or configuration.
System and method described herein is suitable for wide range of applications preferably, for example, handles sub-volumes so that determine the application of feature.Other exemplary application comprises computer-aided diagnosis (CAD) application that is used for various organs in the human body.
System and method described herein is implemented in two kinds of application: the measurement of colonic pathological change; And the feature detection among the colon C AD.
In the measurement of colonic pathological change, the user selects click-point, and this click-point is with acting on the central point that sub-volumes extracts.Handle this sub-volumes and detect pathology in the sub-volumes automatically.When detecting pathology, measure the size of pathology and to user report.Because the user can freely click pathology Anywhere on every side, so the click-point that is input in this application changes.As a result of, a plurality of click-point produce a plurality of sub-volumes.According to the embodiment of present disclosure, a plurality of sub-volumes will return the identical measurement of pathology separately.
Similarly, for colon C AD, intracolic different position candidate are extracted sub-volumes.At each sub-volumes, determine feature.According to the embodiment of present disclosure, determined eigenwert does not change along with the trickle change of position candidate.This causes more sane eigenwert.
Can use different mode to aim at the sub-volumes that has been determined: CT, magnetic resonance (MR), ultrasonic, single photon emission CT (SPECT) and PET (positron emission tomography) (PET).The sub-volumes that can implement to aim at has the feature of substantially the same eigenwert with extraction, thereby for example reduces the precision of false positive or raising characteristic detection method in lesion detection approach.According to the embodiment of present disclosure, the sub-volumes that can implement to aim at be used to cut apart and handle various types of, for example have data of different resolution.
Described the embodiment of the system and method for the sample-synchronous that is used to make a plurality of sub-volumes, should be noted that those skilled in the art can make amendment and change according to instruction.Thereby it should be understood that in disclosed specific embodiments of the present invention and can change, described embodiment as by appended claim within the defined scope and spirit of the present invention.Thereby at length and especially the present invention has been described according to the requirement of Patent Law, the content that requires and wish to obtain patent certificate to protect is illustrated in the appended claims.
Claims (19)
1. computer implemented method that is used for aiming at for feature extraction sub-volumes comprises:
Receive data volume;
A plurality of parameters of sub-volumes are provided;
Determine the starting point of each sub-volumes in two or more subvolumes according to described a plurality of parameters;
Determine the amount of mis-alignment of described two or more subvolumes;
The starting point that moves each sub-volumes is with the grid alignment with these starting points and virtual volume; And
Feature in two or more subvolumes of determining to be aligned.
2. according to the computer implemented method of claim 1, wherein provide a plurality of parameters to comprise:
Select the central point of each sub-volumes in two or more subvolumes;
Select the size and dimension of described two or more subvolumes; And
Select the spacing between the voxel in described two or more subvolumes.
3. according to the computer implemented method of claim 2, also comprise converting voxel spacing to volume unit.
4. according to the computer implemented method of claim 3, also be included in the extraction feature and utilize the voxel spacing of described starting point and volume unit form from volume, to extract two or more subvolumes before.
5. according to the computer implemented method of claim 2, wherein,, carry out the determining of starting point of each sub-volumes in two or more subvolumes according to central point, sub-volumes and voxel spacing.
6. according to the computer implemented method of claim 1, wherein, grid with respect to virtual volume is determined amount of mis-alignment, and wherein the spacing of virtual volume grid is the spacing between the voxel of sub-volumes, and the initial point of the grid of virtual volume conforms to the initial point of world coordinate system.
7. according to the computer implemented method of claim 1, comprise that also the starting point that will be moved converts volume coordinates to, wherein (0,0,0) position of each sub-volumes is as initial point.
8. according to the computer implemented method of claim 1, wherein, described two or more subvolumes have different resolution.
9. according to the computer implemented method of claim 1, wherein, described data volume comprises two or more data volume of representing different modalities.
10. according to the computer implemented method of claim 1, wherein, only carry out moving of starting point with respect to one or two of sub-volumes.
11. according to the computer implemented method of claim 1, wherein, described sub-volumes is two-dimentional.
12., wherein, on two-dimensional sub-volumes, determine feature according to the computer implemented method of claim 11.
13. according to the computer implemented method of claim 1, wherein, three-dimensional or more determine the aligning of starting point on the multidimensional.
14. according to the computer implemented method of claim 1, also comprise the selection one or more dimensions, wherein on selected dimension, determine the aligning of starting point.
15. according to the computer implemented method of claim 1, wherein use the aligning of starting point, to have crossed over the different modalities of record in advance.
16. a computer implemented method that is used for aiming at for video picture sub-volumes comprises:
Receive data volume;
A plurality of parameters of sub-volumes are provided;
Determine the starting point of each sub-volumes in two or more subvolumes according to described a plurality of parameters;
Determine the amount of mis-alignment of described two or more subvolumes;
The starting point that moves each sub-volumes is with the grid alignment with these starting points and virtual volume; And
Show the sub-volumes with the starting point that is aligned, wherein the area-of-interest in each sub-volumes has identical voxel value.
17., wherein provide a plurality of parameters to comprise according to the computer implemented method of claim 16:
Select the central point of each sub-volumes in two or more subvolumes;
Select the size and dimension of described two or more subvolumes; And
Select the spacing between the voxel in described two or more subvolumes.
18., also comprise converting voxel spacing to volume unit according to the computer implemented method of claim 17.
19., also be included in the demonstration sub-volumes and utilize the voxel spacing of described starting point and volume unit form from volume, to extract two or more subvolumes before according to the computer implemented method of claim 18.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US62119804P | 2004-10-22 | 2004-10-22 | |
| US60/621,198 | 2004-10-22 | ||
| US11/255,263 | 2005-10-21 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101048804A CN101048804A (en) | 2007-10-03 |
| CN100543778C true CN100543778C (en) | 2009-09-23 |
Family
ID=38772257
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005800363220A Active CN100543778C (en) | 2004-10-22 | 2005-10-24 | The virtual grid of sub-volumes is aimed at |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100543778C (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107391885B (en) * | 2017-08-29 | 2020-09-22 | 西北工业大学 | Shearing sliding moving grid method based on finite volume method |
-
2005
- 2005-10-24 CN CNB2005800363220A patent/CN100543778C/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN101048804A (en) | 2007-10-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109493328B (en) | Medical image display method, viewing device and computer device | |
| CN103054602B (en) | The method and system that vector detects is referred to for rib in computed tomography volume | |
| EP3210163B1 (en) | Gaze-tracking driven region of interest segmentation | |
| US10362941B2 (en) | Method and apparatus for performing registration of medical images | |
| US7813535B2 (en) | System and method for fused PET-CT visualization for heart unfolding | |
| US8160320B2 (en) | Medical image display apparatus, method and program, and recording medium for the program | |
| RU2479012C2 (en) | Three-dimensional shadow mouse pointer | |
| JP5226974B2 (en) | Image diagnosis support apparatus, method and program | |
| WO2017114366A1 (en) | Data rendering method and system thereof | |
| US20080267499A1 (en) | Method and system for automatic detection of objects in an image | |
| US10210667B2 (en) | Displaying 3D image with a plurality of surface images at depths of interest | |
| US20080287796A1 (en) | Method and system for spine visualization in 3D medical images | |
| CN104424647A (en) | Method and apparatus for registering medical images | |
| CN110969623B (en) | Lung CT multi-symptom automatic detection method, system, terminal and storage medium | |
| GB2496255A (en) | Generating an enhanced image from medical imaging data | |
| US10088992B2 (en) | Enabling a user to study image data | |
| CN107480673B (en) | Method and device for determining interest region in medical image and image editing system | |
| CN111310793A (en) | Medical image classification method and device, mobile terminal and medium | |
| GB2496246A (en) | Identifying regions of interest in medical imaging data | |
| EP3843038B1 (en) | Image processing method and system | |
| CN101802877B (en) | Path proximity rendering | |
| CN100543778C (en) | The virtual grid of sub-volumes is aimed at | |
| US7609876B2 (en) | Virtual grid alignment of sub-volumes | |
| CN1836258B (en) | Method and system for using structure tensors to detect lung nodules and colon polyps | |
| US8326007B2 (en) | Methods and apparatus for combined 4D presentation of quantitative regional measurements and morphology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200902 Address after: Erlangen Patentee after: Siemens Healthcare GmbH Address before: Pennsylvania, USA Patentee before: Siemens Medical Solutions USA, Inc. |