CN104224212B - CT system, its Scan orientation method and calibration method - Google Patents
CT system, its Scan orientation method and calibration method Download PDFInfo
- Publication number
- CN104224212B CN104224212B CN201310236459.7A CN201310236459A CN104224212B CN 104224212 B CN104224212 B CN 104224212B CN 201310236459 A CN201310236459 A CN 201310236459A CN 104224212 B CN104224212 B CN 104224212B
- Authority
- CN
- China
- Prior art keywords
- space coordinates
- data
- scan
- video camera
- scanned
- 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
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The present invention relates to CT system, its Scan orientation method and calibration methods.The CT system includes: hospital bed, is used to support patient to be scanned disposed thereon；CT scan and imaging device, for carrying out CT scan and imaging to patient to be scanned；3D video camera, for shooting the 3D rendering for the patient to be scanned being located on hospital bed；And processing unit, for the human anatomic structure information of patient to be scanned being obtained according to the 3D rendering of 3D shot by camera with determination range to be scanned, and the range to be scanned under the space coordinates of 3D video camera is transformed under the space coordinates of CT scan and imaging device, wherein the CT scan and imaging device are scanned and are imaged to patient to be scanned according to the range to be scanned under the space coordinates of the CT scan and imaging device converted.By the invention it is possible to improve the accuracy of scanning range, shorten sweep time and reduce dose of radiation.
Description
Technical field
The present invention relates to CT technical field of imaging, more particularly to CT system, its Scan orientation method and calibration method.
Background technique
When being scanned imaging to patient to be scanned using CT system, generally include two stages: first stage is
The Scan orientation stage determines the stage of range to be scanned, be used to determine initial position, end position and the rack of scanning
Tilt angle；Second stage is the stage that real scanning imagery is carried out according to determining range to be scanned.
Currently, being all to determine range to be scanned using x-ray imaging when being scanned positioning in first stage.Though
The X-ray radiation dosage of right this x-ray imaging further decreases Scan orientation in the process to patient within safe range
Dose of radiation be still the pursuit of CT technical field of imaging all the time.
In addition, existing positioning scanning is positioned by laser lamp, take a long time, it is troublesome in poeration.
Summary of the invention
According to an aspect of the present invention, a kind of CT system is provided comprising:
Hospital bed is used to support patient to be scanned disposed thereon；
CT scan and imaging device, for carrying out CT scan and imaging to patient to be scanned；
3D video camera, for shooting the 3D rendering for the patient to be scanned being located on hospital bed；And
Processing unit, the human anatomic structure for obtaining patient to be scanned according to the 3D rendering of 3D shot by camera are believed
Range to be scanned under the space coordinates of 3D video camera is transformed into CT scan and imaging is set with determination range to be scanned by breath
Under standby space coordinates, wherein the CT scan and imaging device are sat according to the space for the CT scan and imaging device converted
Range to be scanned under mark system is scanned and is imaged to patient to be scanned.
According to one embodiment of present invention, wherein the processing unit further comprises:
Pattern recognition device, for being obtained by image recognition technology wait sweep according to the 3D rendering of 3D shot by camera
Retouch the human anatomic structure information of patient.
According to one embodiment of present invention, wherein the processing unit further comprises:
Coordinate system calibrating installation, the space coordinate for space coordinates and CT scan and imaging device to 3D video camera
System is calibrated to obtain transformational relation between the two；And
Coordinate system conversion equipment, for according to by calibrating the transformational relation obtained for the space coordinate of 3D video camera
Range to be scanned under system is transformed under the space coordinates of CT scan and imaging device.
According to one embodiment of present invention, wherein the transformational relation is the space coordinates and CT scan of 3D video camera
Affine transform matrix between the space coordinates of imaging device.
According to one embodiment of present invention, wherein the coordinate system calibrating installation according to from 3D video camera about school
The 3D video data of quasi- object and the CT scan data about the caliberator from CT scan and imaging device image to obtain 3D
Transformational relation between the space coordinates and CT scan of machine and the space coordinates of imaging device.
According to another aspect of the present invention, a kind of Scan orientation method for CT system is provided comprising:
The 3D rendering of the patient to be scanned on the hospital bed of CT is obtained by 3D video camera；
The obtained 3D rendering is identified using image recognition technology, to obtain the human body solution of patient to be scanned
Cut open structural information；
Range to be scanned is determined on the 3D rendering based on the human anatomic structure information；And
The range to be scanned under the space coordinates of the 3D video camera is transformed into CT scan and imaging device
Under space coordinates.
According to one embodiment of present invention, wherein will be described to be scanned under the space coordinates of the 3D video camera
Range is transformed under the space coordinates of CT scan and imaging device
The space coordinates of space coordinates and CT scan and imaging device to 3D video camera are calibrated to obtain two
Transformational relation between person；And
The range to be scanned under the space coordinates of 3D video camera is turned according to the transformational relation obtained by calibration
It changes under the space coordinates of CT scan and imaging device.
According to one embodiment of present invention, wherein the transformational relation is that the space coordinates of 3D video camera are swept with CT
Retouch the affine transform matrix between the space coordinates of imaging device.
According to one embodiment of present invention, wherein the calibration is realized by following processing: being imaged according to from 3D
The 3D video data about caliberator of machine and the CT scan data about the caliberator from CT scan and imaging device come
Obtain the transformational relation between the space coordinates and CT scan and the space coordinates of imaging device of 3D video camera.
According to another aspect of the invention, a kind of calibration side for the space coordinates conversion in detection system is provided
Method, the detection system include for carrying the examination couch of caliberator, the 3D video camera in the space being mounted on above examination couch
And detection device, the calibration method include:
A) caliberator is placed on examination couch；
B) height of bed of examination couch is set；
C) caliberator on examination couch is detected using detection device to obtain the data of the detection image of caliberator；
D) caliberator on examination couch is shot using 3D video camera to obtain the data of the shooting image of caliberator；
E) caliberator is subjected to horizontal-shift relative to 3D video camera and detection device；
F) step d) and e) is repeated, the bat of the caliberator of enough horizontal extents until obtaining covering examination couch
Take the photograph the data of image；
G) change the examination couch height of bed；
H) step c) is repeated to g), and the bat of the caliberator of enough vertical ranges of examination couch is covered until obtaining
Take the photograph the data of image and the data of detection image；And
I) it is obtained using the data for the detection image for obtaining the caliberator by the detection device and by 3D video camera
The caliberator shooting image data, to calculate the space coordinates of the 3D video camera and the sky of the detection device
Between relationship between coordinate system.
According to one embodiment of present invention, wherein the detection system is CT system, and the examination couch is removable disease
Bed, the detection device is CT scan and imaging device, and the data of the detection image are the data of CT scan image.
According to one embodiment of present invention, wherein step i) further comprises:
I-1) identify the specified point on the caliberator in the sky of the detection device according to the data of the detection image
Between coordinate data under coordinate system；
I-2 the coordinate data of the specified point obtained) is added into each horizontal-shift respectively, described in obtaining
Specified point is sat in the horizontal extent of covering examination couch and the space of the vertical range in the space of the detection device
Coordinate data under mark system；
I-3 the coordinate data of the specified point on the caliberator) is identified according to the obtained captured image data,
It is imaged with obtaining the specified point in the horizontal extent of covering examination couch and the space of the vertical range in the 3D
Coordinate data under the space coordinates of machine；And
I-4) the coordinate data according to the specified point under the space coordinates of the detection device and in the 3D
Coordinate data under the space coordinates of video camera calculates the space coordinates of the 3D video camera and the sky of the detection device
Between relationship between coordinate system.
According to one embodiment of present invention, wherein the step i-1) further comprise:
Using Feature Extraction Technology, the pixel region of the specified point is determined according to the detection data；
To the method that center of gravity is sought in each identified pixel region application, with the determination specified point comprising described specific
Coordinate in the detection image of point；And
According to the coordinate of the parameter of the detection image comprising the specified point and the specified point in the detection image
Acquire three-dimensional coordinate data of each specified point under the space coordinates of the detection device.
According to one embodiment of present invention, wherein the parameter of the detection image includes examining when obtaining the detection image
Look into the height of bed of bed.
According to one embodiment of present invention, wherein the step i-3) further comprise:
Each of extraction and each examination couch height of bed and caliberator from the captured image data that the 3D video camera is shot
The corresponding RGB image of horizontal-shift and point cloud chart picture；
Identify the 2-d index of each specified point in RGB image；And
The specified point is acquired in the 3D in corresponding point cloud chart picture according to the 2-d index of each specified point
Three-dimensional coordinate data under the space coordinates of video camera.
According to one embodiment of present invention, wherein being extracted and each examination couch height of bed and school from captured image data
When the corresponding RGB image of each horizontal-shift of quasi- object, obtained by being carried out to the RGB image within the scope of a period of time averagely
RGB image as the corresponding RGB image, wherein a period of time range refers to shoot corresponding RGB figure
As at the time of centered on a period of time.
According to one embodiment of present invention, wherein being extracted and each examination couch height of bed and school from captured image data
The corresponding point cloud chart of each horizontal-shift of quasi- object as when, the point cloud chart picture within the scope of a period of time will be asked obtained by intermediate value
Point cloud chart picture as the corresponding point cloud chart picture, wherein a period of time range refers to shoot the corresponding cloud
A period of time centered at the time of image.
According to one embodiment of present invention, wherein the sky of the space coordinates of the 3D video camera and the detection device
Between relationship between coordinate system be affine transform matrix between them, wherein acquiring the affine transform square by following equation
Battle array:
M=Y/X,
Wherein, M is the affine transform matrix, and X is the specified point under the space coordinates of the 3D video camera
The matrix of coordinate data composition, and Y is coordinate data composition of the specified point under the space coordinates of the detection device
Matrix.
According to one embodiment of present invention, wherein the step i-4) include:
It is selected in the horizontal extent of covering examination couch and the space of the vertical range spaced apart more
A selected coordinate position；
Coordinate position is selected for each of the multiple selected coordinate position:
Solve the local affine invariant between the space coordinates of the 3D video camera and the space coordinates of the detection device
Transition matrix Mlocal=Ylocal/Xlocal, wherein XlocalCoordinate position is selected at a distance of pre- spacing with this for falling in for the specified point
From spatial dimension under the space coordinates of the 3D video camera coordinate data composition matrix, YlocalFor the spy
Fall in and this of fixed point are selected space in the detection device of the coordinate position in the spatial dimension of the preset distance and are sat
The matrix of coordinate data composition under mark system.
According to one embodiment of present invention, the calibration method further include: to obtained and the multiple selected seat
The corresponding multiple local affine invariant transition matrixes of cursor position are fitted, to obtain affine transform matrix about arbitrary coordinate position
The function F setM(X,Y,Z)。
According to one embodiment of present invention, wherein before step i) further include: to the number of obtained detection image
Validation verification is carried out according to the data with shooting image, and rejects and does not pass through the abnormal data of validation verification.
According to one embodiment of present invention, the calibration method further include: by the sky of the obtained 3D video camera
Between relationship between coordinate system and the space coordinates of the detection device be stored in the detection system.
By the CT system of embodiment according to the present invention, its Scan orientation method or for the calibration side of its Scan orientation
The accuracy of scanning range can be improved in method, shortens sweep time and reduces dose of radiation.
Detailed description of the invention
In order to thoroughly understand content of this disclosure, below with reference to the following description carried out in conjunction with attached drawing, in attached drawing
In:
Fig. 1 shows the composition schematic diagram of one embodiment of CT system according to the present invention；
Fig. 2 shows the flow diagrams of the Scan orientation method of the CT system for Fig. 1；
Fig. 3 shows according to the present invention for obtaining one of the calibration method of the transformational relation between two coordinate systems
The flow diagram of embodiment；
Fig. 4, which is shown, according to an embodiment of the invention finds out specified point on caliberator in CT from CT image data
The flow diagram of coordinate data under the space coordinates of scanning and imaging device；And
Fig. 5 shows according to an embodiment of the invention from 3D(three-dimensional) image data of shot by camera finds out
The flow diagram of coordinate data of the specified point under the space coordinates of 3D video camera on caliberator.
Specific embodiment
It is described in detail below with reference to the accompanying drawings for realizing specific embodiments of the present invention.But it is to be understood that this
Invention is not limited to following specific embodiments.
Fig. 1 shows the composition schematic diagram of one embodiment of CT system according to the present invention.In this embodiment, CT system
System includes 3D video camera 2, hospital bed 3, CT scan and imaging device 4 and processing unit (not shown in FIG.).Hospital bed 3 is the height of bed
Adjustable and horizontally movable removable hospital bed, is used to support patient to be scanned disposed thereon.3D video camera 2 is for shooting position
In the image of the patient to be scanned on hospital bed 3.3D video camera mentioned herein is to refer to acquire RGB data and depth simultaneously
The video camera of degree evidence.Generally, 3D video camera 2 is located in the space of 3 top of hospital bed, for example, being mounted on the day above hospital bed
On card 1 (as shown in Figure 1).The acquisition range of 3D video camera 2 has certain subtended angle.It can be big according to the subtended angle of video camera 2
The length of small and hospital bed 3 adjusts position and setting angle θ of the video camera 2 in CT system, to cover entire hospital bed range.
CT scan and imaging device 4 are used to carry out CT scan and imaging to patient to be scanned.Processing unit and 3D video camera 2
It is communicated with CT scan and imaging device 4, the human body solution of patient to be scanned is obtained according to image captured by 3D video camera 2
Structural information is cutd open with determination range to be scanned, and coordinate system is carried out to the range to be scanned under the space coordinates of 3D video camera and is turned
It changes, so that the range to be scanned under the space coordinates of CT scan and imaging device is obtained, so that CT scan and imaging device 4 make
With.CT scan and imaging device 4 are according to the range pair to be scanned under the space coordinates of CT scan and imaging device after conversion
Patient to be scanned is scanned and is imaged.
In order to determine range to be scanned according to human body image captured by video camera 2, processing unit includes for taking the photograph
Image application image identification technology captured by camera 2 is to obtain the image recognition of the human anatomic structure information of patient to be scanned
Device.After image recognition, processing unit is determined to be scanned based on the human anatomic structure information of the patient to be scanned shown
Range, to obtain the range to be scanned under the space coordinates of 3D video camera.Later, processing unit is needed the model to be scanned
It encloses and is transformed under the space coordinates of CT scan and imaging device.For this purpose, processing unit further comprises: coordinate system calibration cartridge
Both it sets, the space coordinates for space coordinates and CT scan and imaging device to 3D video camera are calibrated to obtain
Between transformational relation；And coordinate system conversion equipment, for 3D to be imaged according to the transformational relation obtained by calibration
Range to be scanned under the space coordinates of machine is transformed under the space coordinates of CT scan and imaging device.
In one embodiment, coordinate system calibrating installation obtains the space coordinates and CT scan of 3D video camera by calibration
Affine transform matrix between the space coordinates of imaging device, as the transformational relation between them.Optionally, coordinate system
Calibrating installation according to from 3D video camera about caliberator 3D rendering data with from CT scan and imaging device about
The CT scan data of the caliberator calculates the affine transform matrix between two kinds of coordinate systems.
Fig. 2 shows the flow diagrams of one embodiment of the Scan orientation method for CT system shown in FIG. 1.?
In step 201, the image of the patient to be scanned on the hospital bed of CT is obtained by 3D video camera.Later, pass through image recognition
Technology obtains the human anatomic structure information (step 202) of patient to be scanned from captured image.In step 203, according to
Human anatomic structure information determines range to be scanned on captured patient image to be scanned, to obtain the space of video camera
Range to be scanned under coordinate system.Then, coordinate system conversion is carried out to obtain under the space coordinates of CT scan and imaging device
Range (step 204) to be scanned.In this way, CT scan and imaging device can be according to being switched under its space coordinates
Range to be scanned is scanned and is imaged to patient to be scanned.
In the embodiment of CT system and its Scan orientation method described above, with it is traditional by CT prescan come into
The method of row Scan orientation (determining range to be scanned) is different, and CT system and its Scan orientation method of the invention is to pass through camera shooting
Machine shooting patient image simultaneously carries out image recognition to determine range to be scanned, and carries out later to identified range to be scanned
Space coordinates are converted to obtain the range to be scanned under the space coordinates of CT scan and imaging device.Therefore, pass through basis
CT system of the invention and/or its Scan orientation method, can be improved the accuracy of scanning range, shorten entire sweep time simultaneously
Reduce the dose of radiation to patient.
In the above description, it is noted that from the space coordinates of video camera to the space coordinate of CT scan and imaging device
The conversion of system.Before carrying out the conversion, the transformational relation between both space coordinates should be first determined, such as calculate this
Affine transform matrix between two kinds of space coordinates.The determination of transformational relation is calibration method according to the present invention to realize
's.Therefore, calibration method according to the present invention is the important step for realizing Scan orientation method according to the present invention.Pass through basis
The available accurate space coordinates transformational relation of calibration method of the invention, so as to obtain ideal Scan orientation effect
Fruit.The precision that the calibration process can reach will directly affect the precision of CT system Scan orientation.
Although according to the present invention it should be understood that be illustrated by taking CT system as an example in the following description
Calibration method can be used for other detection systems, and the coordinate system conversion established between 3D video camera and other detection devices is closed
System solves the space coordinates interaction problems between 3D video camera and a variety of detection devices.
Fig. 3 shows the conversion according to the present invention for obtaining between two or more coordinate systems in detection system and closes
The flow diagram of one embodiment of the calibration method of system.In this embodiment, calibration method is used for CT system, the CT system
System includes hospital bed, the 3D video camera being mounted in the space above hospital bed and CT scan and imaging device.Calibration needs to acquire
Such as DICOM data that two class data, i.e. CT scan obtain CT scan data and video camera acquisition such as video data with
And the captured image data of depth data.So-called depth the distance between refers to from the camera of 3D video camera to measured point.
Generally, actual use scope is 1m to 2.6m.The depth of each pixel and rgb value are to correspond in captured data
's.In order to enable the transformational relation of such as affine transform matrix acquired to be applicable in biggish spatial dimension, need in disease
It is acquired under the different height of bed and horizontal-shift of bed.
It is assembling CT system, after the installation site including adjusting 3D video camera and setting angle, is starting to calibrate.It is first
First, in step 301, caliberator is placed on hospital bed.At least one specified point is disposed on caliberator, the specified point was both
It can be identified, and can be formed by CT scan and imaging device in the image that 3D video camera is shot by image recognition technology
It is identified in CT image.
Then, in step 302, hospital bed is set in certain height of bed.Then, using CT scan and imaging device to disease
Caliberator on bed carries out CT scan (step 303), to obtain the CT image of caliberator.Also using 3D video camera on hospital bed
Caliberator is shot (step 304), to obtain the shooting image of caliberator.Although step 304 is shown as in step in Fig. 3
After rapid 303, but it is understood that, their sequence is not fixation, can also first carry out step 304 and execute step again
303, or both carry out simultaneously.
Enough data in order to obtain set the caliberator on hospital bed relative to 3D video camera and CT scan and imaging
It is standby to carry out horizontal-shift (step 305).In one embodiment, by moving horizontally hospital bed rather than school on Mobile sickbed
Quasi- object realizes the caliberator relative to 3D video camera and the horizontal-shift of CT scan and imaging device.It ensure that operating
The displacement in other directions will not be introduced in journey.Then, the shooting image of the caliberator after horizontal-shift is acquired using 3D video camera
(step 304).In this way, repeating step 305 and 304 to obtain enough data.Horizontal-shift operation (step is being carried out every time
305) and/or after or before shooting operation (step 304), it can first judge obtained data whether enough (step 306), i.e.,
Whether the photographed data of the caliberator of the enough horizontal extent of covering hospital bed is obtained, for example, the level of hospital bed is partially
All horizontal mobile ranges for whether having spread hospital bed moved.If data deficiencies is enough, continue horizontal-shift operation and shooting
Operation.If data are enough, enter next step 307.Optionally, the shooting of 3D video camera can be continued for, i.e.,
Make also not stop during the relative level offset of adjustment caliberator.In Fig. 3, inquiry step 306 is shown as in step 304
After 305, but as described above, it can also be executed after step 304 or step 305, or in step 304 or step
It is executed before 305.
In step 307, to obtain enough data of covering vertical direction, change the height of bed of hospital bed.Then it repeats
Step 303-306 has obtained the CT image data and shooting figure of enough vertical ranges of covering hospital bed until being judged as
As data (step 308).Equally, step 308 is shown as after step 307 in Fig. 3, but it is understood that,
It can be executed before step 307.
By above data collection process, caliberator can be obtained and covering sufficiently large horizontal extent and vertical range
Corresponding two kinds of data in space, i.e. captured image data and CT image data.According to both correspondence numbers collected
According to the transformational relation between the space coordinates of 3D video camera and CT scan and the space coordinates of imaging device can be calculated
(step 309).
In one embodiment of the invention, by the space of the space coordinates of 3D video camera and CT scan and imaging device
Transformational relation between coordinate system is solved as affine transform matrix.It is available at caliberator by data collected
When each spatial position (each spatial position corresponds to certain the hospital bed height of bed and hospital bed horizontal offset positions) on caliberator
Corresponding coordinate data of the specified point under two spaces coordinate system.It is horizontal partially for corresponding to certain hospital bed height of bed and hospital bed
All there is coordinate data of the specified point under the space coordinates of 3D video camera and corresponding in each spatial position of pan position
The coordinate data under the space coordinates of CT scan and imaging device.
If homogeneous coordinates of i-th of specified point under the space coordinates of 3D video camera are as follows:
If homogeneous coordinates of the specified point under the space coordinates of CT scan and imaging device are as follows:
Since the space coordinates of 3D video camera can be sat by the space of rotationally and/or translationally CT scan and imaging device
Mark system obtains (vice versa), therefore there is the affine transform matrixes of a 4*4, so that the multiple groups specified point collected
Coordinate meets following relationship:
Wherein, affine transform matrix of the M between Two coordinate system.
For each spatial position for corresponding to certain the hospital bed height of bed and hospital bed horizontal offset positions, specified point is taken the photograph in 3D
Coordinate data under the space coordinates of camera can be directly obtained from the image data of shot by camera by image recognition.
But a CT image is only acquired due to being directed to each hospital bed height of bed as described above, spy can only be obtained from CT image data
Fixed-point needle (is the horizontal-shift of hospital bed in the embodiment show in figure 3 to a horizontal position of each hospital bed height of bed and hospital bed
The horizontal position for being zero) CT imaging and scanning device space coordinates under coordinate data.In order to obtain specified point needle
To the coordinate data under the space coordinates of CT scan and imaging device of each hospital bed height of bed and hospital bed horizontal offset positions,
It also needs obtained for each hospital bed height of bed and the CT coordinate data of hospital bed zero level deviation post plus each level
Offset data.
Fig. 4, which is shown, according to an embodiment of the invention finds out specified point on caliberator in CT from CT image data
The flow diagram of coordinate data under the space coordinates of scanning and imaging device.
In step 401, using Feature Extraction Technology, the pixel region of each specified point is determined from CT scan image data
Domain.As described above, specified point is can be identified in CT image.This is because for example, the material for preparing specified point is selected
The density of adjacent material on caliberator is markedly different from for density.In one embodiment, can by the setting of gray threshold,
The pixel region with specified point identical material is found in CT image first.Then, then by Feature Extraction Technology, other objects are excluded
The interference of matter, so that it is determined that the pixel region of specified point.
In step 402, the method for seeking center of gravity to each identified pixel region application, is existed with each specified point of determination
Coordinate (x in CT scan image comprising the specified pointi',yi').
In step 403, for each specified point, according to the parameter of the CT scan image comprising the specified point and the spy
Pinpoint the coordinate (x in the CT scan imagei',yi') specified point is acquired in the space coordinates of CT scan and imaging device
Under three-dimensional coordinate data (xi',yi',zi').Due to including the parameter of the image in each CT image for collecting,
Such as height of bed of hospital bed etc..Therefore it can use the coordinate position of the parameter and specified point of CT image in CT image, calculate
Three-dimensional coordinate data of the specified point under CT coordinate system (space coordinates of CT scan and imaging device).
As described above, the three-dimensional coordinate data is specified point in the case where certain hospital bed height of bed and hospital bed zero level deviate
Data.In order to obtain for the coordinate data of each horizontal-shift of hospital bed, need plus corresponding horizontal-shift (step
404).Assuming that the direction that hospital bed moves horizontally is the Z-direction of CT coordinate system, then specified point is directed to each horizontal-shift of hospital bed
Coordinate data be (xi',yi',zi'+Δzi).Wherein, Δ ziFor the horizontal-shift vector of hospital bed, value is according to the movement of hospital bed
Direction is identical as Z axis positive direction or mutually takes just or takes instead and is negative.
Fig. 5 shows the image data according to an embodiment of the invention from 3D shot by camera and finds out caliberator
On coordinate data of the specified point under the space coordinates of 3D video camera flow diagram.In the present embodiment, 3D is imaged
Image data captured by machine is video data.It is understood that 3D video camera can also shoot with each hospital bed height of bed and
The corresponding still image of hospital bed horizontal offset positions (position for being zero including horizontal-shift).It is as shown in Figure 5:
In step 501, it is extracted from the video data that 3D video camera is shot inclined with each hospital bed height of bed and each level
Move (position for being zero including horizontal-shift) corresponding RGB image and point cloud chart picture.Video camera is during acquiring data
It will receive extraneous interference.To guarantee that the reliability of data is being extracted from video data according to one embodiment of present invention
When RGB image corresponding with each hospital bed height of bed and horizontal-shift, it will be averaged to the RGB image within the scope of a period of time
Obtained RGB image is as the corresponding RGB image.A period of time range refers to shoot the corresponding RGB figure
A period of time centered at the time of as frame.In addition, according to one embodiment of present invention, in the extraction from video data and often
A hospital bed height of bed and the corresponding point cloud chart of each horizontal-shift as when, intermediate value will be sought to the point cloud chart picture within the scope of a period of time
Obtained point cloud chart picture is as the corresponding point cloud chart picture, wherein a period of time range refers to shoot this accordingly
Point cloud chart as at the time of centered on a period of time.
In step 502, the 2-d index of each specified point in each RGB image is identified.In one embodiment, it utilizes
Findchesscorner () function in OpenCV finds 2-d index (x of each specified point in each RGB imagei,
yi).OpenCV is the database of 3D video camera, and full name is open source computer vision library (Open Source Computer
Vision Library), be one based on (open source) issue cross-platform computer vision library, may operate in Linux,
In the operating systems such as Windows and Mac OS.It is understood that the scope of the present invention be not limited to it is listed above
OpenCV database and library function findchesscorner (), those skilled in the art will appreciate that can be used any suitable
Algorithmic tool and/or library function find 2-d index of each specified point in each RGB image.
In step 503, the 2-d index according to each specified point in each RGB image is in corresponding point cloud chart picture
In acquire three-dimensional coordinate data (x of the specified point under the space coordinates of the 3D video camerai,yi,zi)。
Pass through the above-described reality that coordinate data of the specified point under two coordinate systems is found out for each spatial position
Example is applied, for each spatial position that specified point is traversed, a pair of of the coordinate being located under two coordinate systems can be obtained
Data.We are by these coordinate datas to being named as " point coordinate pair ".
In order to accurately acquire the space coordinates and CT scan and imaging of the 3D video camera suitable for larger space range
Affine transform matrix between the space coordinates of equipment, acquires a large amount of data in a calibration process.Due to data volume compared with
Greatly, some mistakes of artificial introducing are likely to during seeking specified point coordinate in data sampling and processing, to influence imitative
Penetrate the final result of transition matrix.It therefore, optionally can be to these numbers before solving affine transform matrix using these data
According to progress validation verification.
Verifying main foundation be: between specified point and specified point, corresponding different level deviate video data between with
And there is certain spatial relations in the X, Y, Z direction between the data of the corresponding different height of bed.Three sides are drawn respectively
Upward coordinate value will obtain the coordinate curve there are certain rule.By discovery curve present in exception coordinate points simultaneously and
When rejected, it can be ensured that the accuracy of the calculated result of affine transform matrix.
According to one embodiment of present invention, by following equation acquire 3D video camera space coordinates and CT scan and
Affine transform matrix between the space coordinates of imaging device:
M=Y/X,
Wherein, M is the affine transform matrix, and X is by the space coordinates in 3D video camera in all the points coordinate pair
Under coordinate data composition matrix, and Y be by the space coordinates in CT scan and imaging device in all the points coordinate pair
Under coordinate data composition matrix.
Available affine transform matrix M through the foregoing embodiment.Whether camera coordinates can be realized for matrix M
It is the demand whether correct conversion and its precision to CT coordinate system can satisfy practical application, certain side can be executed
Method is tested.
In one embodiment, the data collected can be divided into two groups, one group of data is used for the solution of M, and another
One group of data is used for the validation verification of calibration result.Specified point coordinates matrix X and Y are obtained from later group data.Utilize basis
The Metzler matrix that previous group data acquire converts X to Y, then transformed error are as follows:
Δ=M·X-Y
Using the methods of histogram, error of the Δ on tri- directions X, Y, Z is analyzed respectively, can be respectively obtained
The conversion accuracy of M in three directions.So as to judge, whether M meets the requirement of practical application.
In fact, the resolution ratio of 3D video camera is not constant.Generally, with the increase of detecting distance, resolution ratio
It can worse and worse.This has very bad influence to the accuracy of the output of 3D video camera.Therefore, it is found out by above formula
Affine transform matrix is only applicable in certain spatial dimension, and the size of spatial dimension is determined by the ability of camera.
In order to obtain more accurate transition matrix, in one embodiment of the invention, local affine invariant conversion square is defined
Battle array Mlocal.It in this embodiment, is including the sky for all corresponding to the spatial position of each hospital bed height of bed and hospital bed horizontal-shift
Between select multiple selected coordinate positions spaced apart in range.For each selected coordinate position, pass through following formula
To find out the local affine invariant transition matrix M for selecting coordinate position corresponding to thislocal:
Mlocal=Ylocal/Xlocal,
Wherein XlocalThe spatial dimension of coordinate position at a predetermined distance is selected with this to be fallen in by a coordinate center position
The matrix of the interior coordinate data composition under the space coordinates of 3D video camera, YlocalFor by a coordinate center position fall in
The selected coordinate position in the spatial dimension of the preset distance under the space coordinates of CT scan and imaging device
The matrix of coordinate data composition.
A local affine invariant transition matrix can be acquired corresponding to each selected coordinate position, the local affine invariant transition matrix is suitable
For selecting all coordinate positions of the coordinate position in the spatial dimension of the preset distance with this.In this way, just obtain with
The corresponding multiple local affine invariant transition matrixes of the multiple selected coordinate position.
It in one embodiment of the invention, can be quasi- to obtained multiple local affine invariant transition matrix further progresss
It closes, to obtain function F of the affine transform matrix about arbitrary coordinate positionM(X,Y,Z)。
It can be by obtained such as affine transform matrix M or affine transform matrix function FMThe coordinate system of (X, Y, Z) turns
The relationship of changing is stored in CT system, in case turning during the Scan orientation stage of CT system for the coordinate system of range to be scanned
It changes.
By the CT system of any embodiment according to the present invention, its Scan orientation method or for the calibration of its Scan orientation
The accuracy of scanning range can be improved in method, shortens sweep time, reduces equipment operation step and reduces dose of radiation.
Although the above-mentioned attached drawing that has been combined describes specific embodiments of the present invention, those skilled in the art are not taking off
In the case where from the spirit and scope of the present invention, various changes, modification and equivalent substitution can be carried out to the present invention.These change
Become, modification and equivalent substitution all mean and fall within spirit and scope defined by appended claims.
Claims (21)
1. a kind of CT system, comprising:
Hospital bed is used to support patient to be scanned disposed thereon；
CT scan and imaging device, for carrying out CT scan and imaging to patient to be scanned；
3D video camera, for shooting the 3D rendering for the patient to be scanned being located on hospital bed；And
Processing unit, for obtained according to the 3D rendering of 3D shot by camera the human anatomic structure information of patient to be scanned with
It determines range to be scanned, and the range to be scanned under the space coordinates of 3D video camera is transformed into CT scan and imaging device
Under space coordinates, wherein the CT scan and imaging device are according to the CT scan converted and the space coordinates of imaging device
Under range to be scanned patient to be scanned is scanned and is imaged, and the processing unit is also used to the sky based on 3D video camera
Between the space coordinates of coordinate system and CT scan and imaging device obtain it is horizontal partially relative to certain hospital bed height of bed and hospital bed
The data of each spatial position of pan position calibrate described two coordinate systems to obtain transformational relation between the two.
2. CT system according to claim 1, wherein the processing unit further comprises:
Pattern recognition device, for obtaining trouble to be scanned by image recognition technology according to the 3D rendering of 3D shot by camera
The human anatomic structure information of person.
3. CT system according to claim 1, wherein the processing unit further comprises:
Coordinate system calibrating installation, for space coordinates and CT scan and imaging device to 3D video camera space coordinates into
Row calibration is to obtain transformational relation between the two；And
Coordinate system conversion equipment, will be under the space coordinates of 3D video camera by the transformational relation that calibration obtains for basis
Range to be scanned be transformed under the space coordinates of CT scan and imaging device.
4. CT system according to claim 3, wherein the space coordinates that the transformational relation is 3D video camera are swept with CT
Retouch the affine transform matrix between the space coordinates of imaging device.
5. CT system according to claim 3, wherein the coordinate system calibrating installation according to from 3D video camera about
The 3D video data of caliberator is taken the photograph with the CT scan data about the caliberator from CT scan and imaging device to obtain 3D
Transformational relation between the space coordinates and CT scan of camera and the space coordinates of imaging device.
6. a kind of Scan orientation method for CT system, comprising:
The 3D rendering of the patient to be scanned on the hospital bed of CT is obtained by 3D video camera；
The obtained 3D rendering is identified using image recognition technology, to obtain the human dissection knot of patient to be scanned
Structure information；
Range to be scanned is determined on the 3D rendering based on the human anatomic structure information；
Space coordinates based on 3D video camera and the space coordinates of CT scan and imaging device obtain relative to certain
The data of each spatial position of the hospital bed height of bed and hospital bed horizontal offset positions calibrate to obtain described two coordinate systems
Transformational relation between the two；And
The range to be scanned under the space coordinates of the 3D video camera is transformed into CT scan according to the transformational relation
Under the space coordinates of imaging device.
7. Scan orientation method according to claim 6, wherein the transformational relation is the space coordinates of 3D video camera
Affine transform matrix between CT scan and the space coordinates of imaging device.
8. Scan orientation method according to claim 6, wherein the calibration is realized by following processing: according to next
It is swept from the 3D video data about caliberator of 3D video camera with the CT about the caliberator from CT scan and imaging device
Data are retouched to obtain the transformational relation between the space coordinates of 3D video camera and CT scan and the space coordinates of imaging device.
9. a kind of calibration method for the space coordinates conversion in detection system, the detection system includes for carrying school
The examination couch of quasi- object, the 3D video camera being mounted in the space above examination couch and detection device, the calibration method include:
A) caliberator is placed on examination couch；
B) height of bed of examination couch is set；
C) caliberator on examination couch is detected using detection device to obtain the data of the detection image of caliberator；
D) caliberator on examination couch is shot using 3D video camera to obtain the data of the shooting image of caliberator；
E) caliberator is subjected to horizontal-shift relative to 3D video camera and detection device；
F) step d) and e) is repeated, the shooting figure of the caliberator of enough horizontal extents until obtaining covering examination couch
The data of picture；
G) change the examination couch height of bed；
H) step c) is repeated to g), and the shooting figure of the caliberator of enough vertical ranges of examination couch is covered until obtaining
The data of picture and the data of detection image；And
I) data for the detection image that the caliberator is obtained by the detection device and the institute obtained by 3D video camera are utilized
The data of the shooting image of caliberator are stated, to calculate the space coordinates of the 3D video camera and the space seat of the detection device
Relationship between mark system.
10. calibration method according to claim 9, wherein the detection system is CT system, and the examination couch is removable
Dynamic hospital bed, the detection device is CT scan and imaging device, and the data of the detection image are the number of CT scan image
According to.
11. calibration method according to claim 9, wherein step i) further comprises:
I-1) identify that the specified point on the caliberator is sat in the space of the detection device according to the data of the detection image
Coordinate data under mark system；
I-2 the coordinate data of the specified point obtained) is added into each horizontal-shift respectively, it is described specific to obtain
Point is in the horizontal extent of covering examination couch and the space of the vertical range in the space coordinates of the detection device
Under coordinate data；
I-3 the coordinate data of the specified point on the caliberator is identified, according to the obtained captured image data) to obtain
The specified point is obtained in the horizontal extent of covering examination couch and the space of the vertical range in the 3D video camera
Coordinate data under space coordinates；And
I-4) coordinate data according to the specified point under the space coordinates of the detection device and the 3D image
The space of coordinate data under the space coordinates of machine, the space coordinates and the detection device that calculate the 3D video camera is sat
Relationship between mark system.
12. calibration method according to claim 11, wherein the step i-1) further comprise:
Using Feature Extraction Technology, the pixel region of the specified point is determined according to the data of the detection image；
To the method that center of gravity is sought in each identified pixel region application, the specified point is being included with the determination specified point
Coordinate in detection image；And
It is acquired according to the coordinate of the parameter of the detection image comprising the specified point and the specified point in the detection image
Three-dimensional coordinate data of each specified point under the space coordinates of the detection device.
13. calibration method according to claim 12, wherein the parameter of the detection image includes obtaining the detection figure
As when examination couch the height of bed.
14. calibration method according to claim 11, wherein the step i-3) further comprise:
Each level with each examination couch height of bed and caliberator is extracted from the captured image data that the 3D video camera is shot
Deviate corresponding RGB image and point cloud chart picture；
Identify the 2-d index of each specified point in RGB image；And
The specified point is acquired in corresponding point cloud chart picture according to the 2-d index of each specified point to image in the 3D
Three-dimensional coordinate data under the space coordinates of machine.
15. calibration method according to claim 14, wherein being extracted and each examination couch bed from captured image data
When high RGB image corresponding with each horizontal-shift of caliberator, it will be averaged to the RGB image within the scope of a period of time
Obtained RGB image as the corresponding RGB image, wherein a period of time range refer to it is described opposite to shoot
A period of time centered at the time of the RGB image answered.
16. calibration method according to claim 14, wherein being extracted and each examination couch bed from captured image data
High point cloud chart corresponding with each horizontal-shift of caliberator as when, intermediate value will be sought to the point cloud chart picture within the scope of a period of time
Obtained point cloud chart picture is as the corresponding point cloud chart picture, wherein a period of time range refers to shoot the phase
Corresponding point cloud chart as at the time of centered on a period of time.
17. calibration method according to claim 11, wherein the space coordinates of the 3D video camera are set with the detection
Relationship between standby space coordinates is the affine transform matrix between them, wherein being acquired by following equation described affine
Transition matrix:
M=Y/X,
Wherein, M is the affine transform matrix, and X is coordinate of the specified point under the space coordinates of the 3D video camera
The matrix of data composition, and Y is the square of coordinate data composition of the specified point under the space coordinates of the detection device
Battle array.
18. calibration method according to claim 11, wherein the step i-4) include:
Multiple choosings spaced apart are selected in the horizontal extent of covering examination couch and the space of the vertical range
Position fixing position；
Coordinate position is selected for each of the multiple selected coordinate position:
Solve the local affine invariant conversion between the space coordinates of the 3D video camera and the space coordinates of the detection device
Matrix Mlocal=Ylocal/Xlocal, wherein XlocalCoordinate position is selected at a predetermined distance with this for falling in for the specified point
The matrix of the coordinate data composition under the space coordinates of the 3D video camera in spatial dimension, YlocalFor the specified point
Fall in and select space coordinates in the detection device of the coordinate position in the spatial dimension of the preset distance with this
Under coordinate data composition matrix.
19. calibration method according to claim 18, further includes: to obtained with the multiple selected coordinate position phase
Corresponding multiple local affine invariant transition matrixes are fitted, to obtain function of the affine transform matrix about arbitrary coordinate position
FM(X, Y, Z).
20. calibration method according to claim 9, wherein before step i) further include: to obtained detection image
Data and shooting image data carry out validation verification, and reject do not pass through the abnormal data of validation verification.
21. calibration method according to claim 9, further includes: by the space coordinates of the obtained 3D video camera
Relationship between the space coordinates of the detection device is stored in the detection system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310236459.7A CN104224212B (en) | 2013-06-14 | 2013-06-14 | CT system, its Scan orientation method and calibration method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310236459.7A CN104224212B (en) | 2013-06-14 | 2013-06-14 | CT system, its Scan orientation method and calibration method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104224212A CN104224212A (en) | 2014-12-24 |
| CN104224212B true CN104224212B (en) | 2019-07-23 |
Family
ID=52213752
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310236459.7A Active CN104224212B (en) | 2013-06-14 | 2013-06-14 | CT system, its Scan orientation method and calibration method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104224212B (en) |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11576645B2 (en) | 2015-03-02 | 2023-02-14 | Shanghai United Imaging Healthcare Co., Ltd. | Systems and methods for scanning a patient in an imaging system |
| CN104644205A (en) | 2015-03-02 | 2015-05-27 | 上海联影医疗科技有限公司 | Method and system for positioning patient during diagnostic imaging |
| US11576578B2 (en) | 2015-03-02 | 2023-02-14 | Shanghai United Imaging Healthcare Co., Ltd. | Systems and methods for scanning a patient in an imaging system |
| JP6446348B2 (en) * | 2015-09-30 | 2018-12-26 | ジーイー・メディカル・システムズ・グローバル・テクノロジー・カンパニー・エルエルシー | Radiation tomography apparatus and program |
| CN105852893A (en) * | 2016-04-14 | 2016-08-17 | 杭州美诺瓦医疗科技股份有限公司 | Human body parameter detector for medical imaging system and intelligent digital X-ray machine |
| US10182771B2 (en) * | 2017-02-10 | 2019-01-22 | Siemens Healthcare Gmbh | Method and system for dose-optimized computed tomography scanning of a target organ |
| WO2018209633A1 (en) * | 2017-05-18 | 2018-11-22 | 深圳配天智能技术研究院有限公司 | Automatic correction method and device based on and vibration displacement of vision system, and vision system |
| CN107320124A (en) * | 2017-06-28 | 2017-11-07 | 上海联影医疗科技有限公司 | The method and medical image system of spacer scanning are set in medical image system |
| US9958559B1 (en) * | 2017-07-11 | 2018-05-01 | Uih America, Inc. | Method and apparatus for automatic detection and correction of patient bed shift using intrinsic scintillation crystal radiations |
| CN107334487A (en) * | 2017-08-11 | 2017-11-10 | 上海联影医疗科技有限公司 | A kind of medical image system and its scan method |
| CN107789001B (en) * | 2017-10-31 | 2021-08-31 | 上海联影医疗科技股份有限公司 | Positioning method and system for imaging scanning |
| CN108670286A (en) * | 2018-06-13 | 2018-10-19 | 上海联影医疗科技有限公司 | A kind of CT system and CT scan method |
| CN109598717A (en) * | 2018-12-06 | 2019-04-09 | 宁波耀通管阀科技有限公司 | Scan pattern switching system |
| CN109464155B (en) * | 2018-12-29 | 2022-12-20 | 上海联影医疗科技股份有限公司 | Medical scanning positioning method |
| CN112006711A (en) * | 2019-05-31 | 2020-12-01 | 通用电气精准医疗有限责任公司 | 3D perspective indicator and generation method and application thereof |
| CN112085797A (en) * | 2019-06-12 | 2020-12-15 | 通用电气精准医疗有限责任公司 | 3D camera-medical imaging device coordinate system calibration system and method and application thereof |
| CN111297383A (en) * | 2019-12-24 | 2020-06-19 | 上海联影医疗科技有限公司 | Correction method, device and storage medium for X-ray imaging |
| CN111053570B (en) * | 2020-03-02 | 2020-11-24 | 明峰医疗系统股份有限公司 | Non-contact CT scanning positioning method |
| CN111631744B (en) * | 2020-05-09 | 2023-04-14 | 赛诺威盛科技(北京)股份有限公司 | Method, device and system for CT scanning positioning |
| CN112113502B (en) * | 2020-09-10 | 2022-08-05 | 杭州三坛医疗科技有限公司 | Fracture block positioning method and device |
| CN112450954A (en) * | 2020-11-27 | 2021-03-09 | 上海优医基医疗影像设备有限公司 | CT imaging field of view adjusting method, CT imaging method and CT imaging system |
| CN113180709B (en) * | 2021-07-01 | 2021-09-07 | 晓智未来(成都)科技有限公司 | Human body to-be-detected part posture recognition method based on photogrammetry |
| CN113440156A (en) * | 2021-07-14 | 2021-09-28 | 南京安科医疗科技有限公司 | Mobile CT intelligent scanning positioning system, positioning method and storage medium |
| CN114748086B (en) * | 2021-12-21 | 2023-08-08 | 首都医科大学附属北京友谊医院 | CT scanning method and system, electronic device and computer readable storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1737810A (en) * | 2004-05-18 | 2006-02-22 | 爱克发-格法特公司 | Method for automatically mapping of geometric objects in digital medical images |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7869562B2 (en) * | 2008-05-19 | 2011-01-11 | Siemens Aktiengesellschaft | Automatic patient positioning system |
-
2013
- 2013-06-14 CN CN201310236459.7A patent/CN104224212B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1737810A (en) * | 2004-05-18 | 2006-02-22 | 爱克发-格法特公司 | Method for automatically mapping of geometric objects in digital medical images |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104224212A (en) | 2014-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104224212B (en) | CT system, its Scan orientation method and calibration method | |
| CN104132613B (en) | Noncontact optical volume measurement method for complex-surface and irregular objects | |
| CN103181775B (en) | For detecting the method and system of patient body's cursor position | |
| CN109598762A (en) | A kind of high-precision binocular camera scaling method | |
| US9456200B2 (en) | Method and apparatus for calibration of stereo-optical three-dimensional surface-mapping system | |
| CN105258710B (en) | A kind of high-precision principal point for camera scaling method | |
| CN107886547B (en) | Fisheye camera calibration method and system | |
| WO2018196303A1 (en) | Projector calibration method and apparatus based on multi-directional projection | |
| CN104240262B (en) | Calibration device and calibration method for outer parameters of camera for photogrammetry | |
| CN106447733B (en) | Method, system and device for determining cervical vertebra mobility and moving axis position | |
| CN101292255A (en) | Artifact mitigation in three-dimensional imaging | |
| CN106662734A (en) | Determining the position of an object in the beam path of an optical device | |
| CN110375648A (en) | The spatial point three-dimensional coordinate measurement method that the single camera of gridiron pattern target auxiliary is realized | |
| WO2004044522A1 (en) | Three-dimensional shape measuring method and its device | |
| TW201043949A (en) | Inspection method | |
| CN110986757B (en) | Three-dimensional human body scanning method, device and system | |
| CN106169076B (en) | A kind of angle license plate image library building method based on perspective transform | |
| CN104616348A (en) | Method for reconstructing fabric appearance based on multi-view stereo vision | |
| CN112132908A (en) | Camera external parameter calibration method and device based on intelligent detection technology | |
| CN110958450A (en) | Imaging system space testing device and contrast and frequency testing method | |
| CN112509060B (en) | CT secondary scanning positioning method and system based on image depth learning | |
| CN112258593A (en) | CT or PET-CT intelligent positioning scanning method under monocular camera | |
| CN111105466A (en) | Calibration method of camera in CT system | |
| CN109171789A (en) | A kind of calibration method and calibration system for diagnostic imaging equipment | |
| CN111179335A (en) | Standing tree measuring method based on binocular vision |
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 |