CN106889999B

CN106889999B - The method and apparatus of multi-modal detection system image co-registration

Info

Publication number: CN106889999B
Application number: CN201611239157.5A
Authority: CN
Inventors: 姜丽丽; 李楠; 于庆泽
Original assignee: Sino United Medical Technology (beijing) Co Ltd
Current assignee: Sino United Medical Technology (beijing) Co Ltd
Priority date: 2016-12-28
Filing date: 2016-12-28
Publication date: 2019-12-03
Anticipated expiration: 2036-12-28
Also published as: CN106889999A

Abstract

The invention discloses a kind of method and apparatus of multi-modal detection system image co-registration, belong to medical imaging technology field.Method includes: to detect target object by multi-modal detection system, obtains the first image and the second image, and the FOV coordinate system of the first image is the first coordinate system, and the FOV coordinate system of the second image is the second coordinate system；The first transition matrix is obtained, the first transition matrix is used to coordinate of the first pixel of each of first image in the first coordinate system being converted to coordinate in the second coordinate system；According to the first coordinate of the first transition matrix and each first pixel in the first coordinate system, the second coordinate of each first pixel in the second coordinate system is determined；The first image and the second image co-registration are obtained into multi-modal detection image according to the third coordinate of the second pixel of each of the second coordinate of each first pixel and the second image in the second coordinate system.The present invention improves the accuracy of the first image and the second image co-registration.

Description

The method and apparatus of multi-modal detection system image co-registration

Technical field

The present invention relates to medical imaging technology field, in particular to a kind of method of multi-modal detection system image co-registration and Device.

Background technique

Include two kinds of detectors in multi-modal detection system, same target object is visited by both detectors It surveys, obtains two images；Two images are merged, multi-modal detection image is obtained, by multi-modal detection image into Row analysis, analyzes target object with realizing.For example, PET-CT (Positron Emission Tomography- Computed Tomography, positron emission tomography-computed tomography) it include pet detector in detection system With CT detector, the PET image of human body inner disease foci is detected by pet detector, and human organ tissue is detected by CT detector CT image, PET image and CT image are merged, PET-CT image is obtained；Doctor is according to the PET-CT image, energy in this way Enough intracorporal lesions of finder, and lesions position is accurately positioned.

For the ease of merging PET image and CT image, PET image is being detected by pet detector, and pass through Before CT detector detects CT image, engineer manually adjusts pet detector and CT detector, by the FOV of pet detector (Field-of-View, the visual field) coordinate system is consistent with the adjusting of the FOV coordinate system of CT detector；Namely the FOV of pet detector is sat Mark system origin and CT detector FOV coordinate system origin be overlapped, the X, Y, Z axis of the FOV coordinate system of pet detector respectively with The X, Y, Z axis of the FOV coordinate system of CT detector is parallel.

When being merged PET image and CT image, the controlling terminal in PET-CT detection system is according in PET image Each of the second pixel of the first coordinate and each of CT image of the first pixel third coordinate, by PET image and CT Image is merged, and PET-CT image is obtained, and the first coordinate of each first pixel is each first pixel Coordinate in the FOV coordinate system of the pet detector, the third coordinate of each second pixel are described each second Coordinate of the pixel in the FOV coordinate system of the CT detector.

In the implementation of the present invention, the inventor finds that the existing technology has at least the following problems:

Due to the limitation of mechanical adjustment precision, between the FOV coordinate system of pet detector and the FOV coordinate system of CT detector Usually there is error, the accuracy so as to cause above-mentioned image co-registration is poor.

Summary of the invention

In order to solve problems in the prior art, the present invention provides a kind of method of multi-modal detection system image co-registration and Device.Technical solution is as follows:

In a first aspect, the embodiment of the invention provides a kind of method of multi-modal detection system image co-registration, the method Include:

Target object is detected by multi-modal detection system, obtains the first image and the second image, the first image Visual field FOV coordinate system is the first coordinate system, and the FOV coordinate system of second image is the second coordinate system；

The first transition matrix is obtained, first transition matrix is used for the first pixel of each of the first image The coordinate in second coordinate system is converted in the coordinate in first coordinate system；

According to the first coordinate of first transition matrix and each first pixel, each first picture is determined Second coordinate of vegetarian refreshments, the first coordinate of each first pixel are each first pixel in first coordinate Coordinate in system, the second coordinate of each first pixel are each first pixel in second coordinate system Coordinate；

According to the of the second pixel of each of the second coordinate of each first pixel and second image Three coordinates merge the first image and second image, obtain multi-modal detection image, each second picture The third coordinate of vegetarian refreshments is coordinate of each second pixel in second coordinate system.

In a kind of possible design, the first transition matrix of the acquisition, comprising:

Coordinate registration die body is detected by the multi-modal detection system, obtains third image and the 4th image, the seat Standard configuration quasi-mode body includes n point source, and the n point source is the object that can be detected by the multi-modal detection system, described Object can be the F for being mixed with medical lipiodol¹⁸- FDG solution or solid-state point source, the FOV coordinate system of the third image are the One coordinate system, the FOV coordinate system of the 4th image are the second coordinate system, and n is the integer more than or equal to 4；

It determines the 4-coordinate of each point source in the n point source in the third image, and determines described the The Five Axis of each point source in four images, the 4-coordinate of each point source are each point source described the Coordinate in one coordinate system, the Five Axis of each point source are coordinate of each point source in second coordinate system；

According to the 4-coordinate and Five Axis of each point source, first transition matrix is determined.

In a kind of possible design, each point source in the n point source in the determination third image 4-coordinate, comprising:

For each point source, the sphere where the point source is determined in the third image, obtains the sphere The pixel value and the 6th coordinate of interior each third pixel, the 6th coordinate of each third pixel are each third Coordinate of the pixel in first coordinate system；

The point is determined by following formula one (1) according to the pixel value and the 6th coordinate of each third pixel The 4-coordinate in source:

Wherein, (x₄,y₄,z₄) be the point source 4-coordinate, (x₆,y₆,z₆) it is the of each third pixel Six coordinates, PixelValue_pFor the pixel value of each third pixel.

In a kind of possible design, the Five Axis of each point source in determination the 4th image, packet It includes:

For each point source, the sphere where the point source is determined in the 4th image, obtains the sphere The pixel value and the 9th coordinate of interior each 4th pixel, the 9th coordinate of each 4th pixel are the described each 4th Coordinate of the pixel in second coordinate system；

The point is determined by following formula one (2) according to the pixel value and the 9th coordinate of each 4th pixel The Five Axis in source:

Wherein, (x₅,y₅,z₅) be the point source Five Axis, (x₉,y₉,z₉) it is the of each 4th pixel Nine coordinates, PixelValue_qFor the pixel value of each 4th pixel.

In a kind of possible design, the 4-coordinate and Five Axis according to each point source, determine described in First transition matrix, comprising:

According to the 4-coordinate of each point source, the transformed representation of Five Axis and first transition matrix, really The coordinate difference expression formula of fixed each point source；

According to the coordinate difference expression formula of each point source, total coordinate difference expression formula of the multiple point source is determined；

According to total coordinate difference expression formula, first transition matrix for meeting preset condition is determined, it is described default Condition is that the value of total coordinate difference expression formula is minimum value.

In a kind of possible design, 4-coordinate, the Five Axis and described first according to each point source The transformed representation of transition matrix determines the coordinate difference expression formula of each point source, comprising:

It is determined described every according to the 4-coordinate and the transformed representation of each point source by following formula two 7th coordinate of a point source, the 7th coordinate of each point source are seat of each point source in second coordinate system Mark:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

Each point source is determined by following formula three according to the Five Axis and the 7th coordinate of each point source Coordinate difference expression formula:

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

Wherein, (x₇,y₇,z₇) be each point source the 7th coordinate, (x₅,y₅,z₅) it is the 5th of each point source Coordinate, (x₄,y₄,z₄) be each point source 4-coordinate, M be the transformed representation, Δ x be the coordinate difference table Up to formula.

In a kind of possible design, the multi-modal detection system is positron emission tomography-computerized tomography PET-CT detection system is scanned, the first image is the PET image of pet detector detection, and second image is CT detection The CT image of device detection, first coordinate according to first transition matrix and each first pixel determine institute Before the second coordinate for stating each first pixel, the method also includes:

According to first transition matrix, the second transition matrix is obtained, second transition matrix is used for described second The second pixel of each of image is converted to the coordinate in first coordinate system in the coordinate in second coordinate system；

According to the third coordinate of second transition matrix and each second pixel, each second picture is determined 8th coordinate of vegetarian refreshments, the 8th coordinate of each second pixel are each second pixel in first coordinate Coordinate in system；

According to the first coordinate of the 8th coordinate of each second pixel and each first pixel, from described The attenuation coefficient of each first pixel is determined in second image；

According to the attenuation coefficient of each first pixel, the first image is corrected.

Second aspect, the present invention provide a kind of device of multi-modal detection system image co-registration, described device in implementing Include:

Detecting module obtains the first image and the second image, institute for detecting target object by multi-modal detection system The visual field FOV coordinate system for stating the first image is the first coordinate system, and the FOV coordinate system of second image is the second coordinate system；

First obtains module, and for obtaining the first transition matrix, first transition matrix is used for the first image Each of the first pixel be converted to the coordinate in second coordinate system in the coordinate in first coordinate system；

First determining module, for the first coordinate according to first transition matrix and each first pixel, Determine that the second coordinate of each first pixel, the first coordinate of each first pixel are each first picture Coordinate of the vegetarian refreshments in first coordinate system, the second coordinate of each first pixel are each first pixel Coordinate in second coordinate system；

Fusion Module, for according to each of the second coordinate of each first pixel and described second image The third coordinate of two pixels merges the first image and second image, obtains multi-modal detection image, institute The third coordinate for stating each second pixel is coordinate of each second pixel in second coordinate system.

In a kind of possible design, described first obtains module, comprising:

Probe unit, for by the multi-modal detection system detection coordinate registration die body, obtaining third image and the Four images, the coordinate registration die body include n point source, and the n point source is that can be detected by the multi-modal detection system The object arrived, the object can be the F for being mixed with medical lipiodol¹⁸- FDG solution or solid-state point source, the third image FOV coordinate system is the first coordinate system, and the FOV coordinate system of the 4th image is the second coordinate system, and n is whole more than or equal to 4 Number；

Determination unit, for determining the 4-coordinate of each point source in the n point source in the third image, with And determining the Five Axis of each point source in the 4th image, the 4-coordinate of each point source is described each Coordinate of the point source in first coordinate system, the Five Axis of each point source are that each point source is sat described second Mark the coordinate of system；According to the 4-coordinate and Five Axis of each point source, first transition matrix is determined.

In a kind of possible design, the determination unit, comprising:

First determines subelement, is used for for each point source, in the third image where the determining point source Sphere, obtain the pixel value and the 6th coordinate of each third pixel in the sphere, the of each third pixel Six coordinates are coordinate of each third pixel in first coordinate system；

Second determines subelement, for the pixel value and the 6th coordinate according to each third pixel, by following Formula one (1) determines the 4-coordinate of the point source:

In a kind of possible design, the determination unit, comprising:

Third determines subelement, is used for for each point source, in the 4th image where the determining point source Sphere, obtain the pixel value and the 9th coordinate of each 4th pixel in the sphere, the of each 4th pixel Nine coordinates are coordinate of each 4th pixel in second coordinate system；

4th determines subelement, for the pixel value and the 9th coordinate according to each 4th pixel, by following Formula one (2) determines the Five Axis of the point source:

In a kind of possible design, the determination unit, comprising:

5th determines subelement, for according to the 4-coordinate of each point source, Five Axis and first conversion The transformed representation of matrix determines the coordinate difference expression formula of each point source；

6th determines that subelement determines the multiple point source for the coordinate difference expression formula according to each point source Total coordinate difference expression formula；

7th determines subelement, meets described the of preset condition for determining according to total coordinate difference expression formula One transition matrix, the preset condition are that the value of total coordinate difference expression formula is minimum value.

In a kind of possible design, the described 5th determines subelement, is also used to sit according to the 4th of each point source the Mark and the transformed representation determine the 7th coordinate of each point source by following formula two, the of each point source Seven coordinates are coordinate of each point source in second coordinate system:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

It is described 5th determine subelement, be also used to the Five Axis and the 7th coordinate according to each point source, by with Lower formula three determines the coordinate difference expression formula of each point source:

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

In a kind of possible design, the multi-modal detection system is positron emission tomography-computerized tomography PET-CT detection system is scanned, the first image is the PET image of pet detector detection, and second image is CT detection The CT image of device detection, described device further include:

Second obtains module, for obtaining the second transition matrix, the second conversion square according to first transition matrix Battle array is for being converted to coordinate of the second pixel of each of second image in second coordinate system described the Coordinate in one coordinate system；

Second determining module, for the third coordinate according to second transition matrix and each second pixel, Determine that the 8th coordinate of each second pixel, the 8th coordinate of each second pixel are each second picture Coordinate of the vegetarian refreshments in first coordinate system；

Third determining module, for the 8th coordinate and each first pixel according to each second pixel The first coordinate, the attenuation coefficient of each first pixel is determined from second image；

Correction module is corrected the first image for the attenuation coefficient according to each first pixel.

In the embodiment of the present invention, after controlling terminal detects the first image and the second image of target object, first is obtained Transition matrix, and according to the first coordinate of the first pixel each in first transition matrix and the first image, by this each First coordinate of one pixel is converted to second coordinate of each first pixel in the second coordinate system, and then, control is eventually The third coordinate further according to the second pixel of each of the second coordinate of each first pixel and second image is held, it will First image and second image co-registration are multi-modal detection image, due to first transition matrix be will be in the first image First coordinate of each first pixel is converted to the conversion of second coordinate of each first pixel in the second coordinate system Matrix, it therefore reduces the error between the first coordinate system of the first image and the second coordinate system of the second image, further mentions It is high by the first image and the second image co-registration when accuracy.

Detailed description of the invention

Fig. 1 is a kind of method flow diagram of multi-modal detection system image co-registration provided in an embodiment of the present invention；

Fig. 2 is a kind of method flow diagram of multi-modal detection system image co-registration provided in an embodiment of the present invention；

Fig. 3 is a kind of coordinate registration die body structure chart provided in an embodiment of the present invention；

Fig. 4 is a kind of coordinate registration die body top view provided in an embodiment of the present invention；

Fig. 5 is a kind of method flow diagram of multi-modal detection system image co-registration provided in an embodiment of the present invention；

Fig. 6 is a kind of apparatus structure schematic diagram of multi-modal detection system image co-registration provided in an embodiment of the present invention；

Fig. 7 is a kind of apparatus structure schematic diagram (control of multi-modal detection system image co-registration provided in an embodiment of the present invention The general structure of terminal processed).

Specific embodiment

To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention Formula is described in further detail.

The embodiment of the invention provides a kind of method of multi-modal detection system image co-registration, this method is applied to any more In mode detection system, which can be PET-CT detector system or SPECT-MRI (Single- Photon Emission Computed Tomography-Magnetic Resonance Imaging, single photon emission calculate Machine tomography-magnetic resonance imaging) detector system.In embodiments of the present invention, can be with multi-modal detection system It is illustrated for PET-CT detector system, but specific limit is not made to multi-modal detection system.

PET-CT detection system includes nuclear imaging detector, anatomy detector and controlling terminal；The nuclear imaging detector It can be pet detector, anatomy detector can be CT detector.Pet detector is used to detect the PET figure of target object Picture, CT detector are used to detect the CT image of target object, and controlling terminal is used for PET image and CT image synthesis PET-CT figure Picture；Executing subject of the invention can be controlling terminal.

It should be noted that the controlling terminal of PET-CT detection system can pass through pet detector in the embodiment of the present invention Target object is detected with CT detector, obtains the first image and the second image of the target object；The control of PET-CT detection system Terminal processed can detect coordinate registration die body by pet detector and CT detector, obtain the third figure of the coordinate registration die body Picture and the 4th image.

Referring to Fig. 1, this method comprises:

Step 101: target object being detected by multi-modal detection system, obtains the first image and the second image, this first The visual field FOV coordinate system of image is the first coordinate system, and the FOV coordinate system of second image is the second coordinate system.

Step 102: obtaining the first transition matrix, which is used for the first pixel of each of first image Coordinate o'clock in the first coordinate system is converted to coordinate in the second coordinate system.

Step 103: according to first coordinate of first transition matrix and each first pixel, determining each first picture Second coordinate of vegetarian refreshments, the first coordinate of each first pixel are the seat of each first pixel in the first coordinate system Mark, the second coordinate of each first pixel are the coordinate of each first pixel in the second coordinate system.

Step 104: according to the second pixel of each of the second coordinate of each first pixel and second image Third coordinate, first image and second image are merged, obtain multi-modal detection image, each second pixel The third coordinate of point is the coordinate of each second pixel in second coordinate system.

In a kind of possible design, first transition matrix of acquisition, comprising:

Coordinate registration die body is detected by the multi-modal detection system, obtains third image and the 4th image, which matches Quasi-mode body includes n point source, which is the object that can be detected by the multi-modal detection system, which can be It is mixed with the F of medical lipiodol¹⁸- FDG solution or solid-state point source, the FOV coordinate system of the third image are the first coordinate system, this The FOV coordinate system of four images is the second coordinate system, and n is the integer more than or equal to 4；

It determines the 4-coordinate of each point source in n point source in the third image, and determines in the 4th image Each point source Five Axis, the 4-coordinate of each point source is the coordinate of each point source in the first coordinate system, should The Five Axis of each point source is coordinate of each point source in the second coordinate system；

According to the 4-coordinate and Five Axis of each point source, the first transition matrix is determined.

In a kind of possible design, the 4-coordinate of each point source in n point source in the determination third image, Include:

For each point source, the sphere where the point source is determined in the third image, obtains each third in the sphere The pixel value of pixel and the 6th coordinate, the 6th coordinate of each third pixel are that each third pixel is sat first Coordinate in mark system；

The point source is determined by following formula one (1) according to the pixel value of each third pixel and the 6th coordinate 4-coordinate:

Wherein, (x₄,y₄,z₄) be the point source 4-coordinate, (x₆,y₆,z₆) sat for the 6th of each third pixel Mark, PixelValue_pFor the pixel value of each third pixel.

In a kind of possible design, the Five Axis of each point source in the 4th image of determination, comprising:

For each point source, the sphere where the point source is determined in the 4th image, is obtained each the in the sphere The pixel value of four pixels and the 9th coordinate, the 9th coordinate of each 4th pixel be each 4th pixel this Coordinate in two coordinate systems；

The point source is determined by following formula one (2) according to the pixel value and the 9th coordinate of each 4th pixel Five Axis:

Wherein, (x₅,y₅,z₅) be the point source Five Axis, (x₉,y₉,z₉) sat for the 9th of each 4th pixel Mark, PixelValue_qFor the pixel value of each 4th pixel.

In a kind of possible design, which determines this first turn Change matrix, comprising:

According to the 4-coordinate of each point source, the transformed representation of Five Axis and first transition matrix, determine each The coordinate difference expression formula of point source；

According to the coordinate difference expression formula of each point source, total coordinate difference expression formula of multiple point sources is determined；

According to total coordinate difference expression formula, first transition matrix for meeting preset condition is determined, which is The value of total coordinate difference expression formula is minimum value.

In a kind of possible design, 4-coordinate, Five Axis and first transition matrix according to each point source Transformed representation, determine the coordinate difference expression formula of each point source, comprising:

The of each point source is determined by following formula two according to the 4-coordinate of each point source and the transformed representation 7th coordinate of seven coordinates, each point source is the coordinate of each point source in the second coordinate system:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

The coordinate of each point source is determined by following formula three according to the Five Axis and the 7th coordinate of each point source Differential expression formula:

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

Wherein, (x₇,y₇,z₇) be each point source the 7th coordinate, (x₅,y₅,z₅) be each point source Five Axis, (x₄,y₄,z₄) be each point source 4-coordinate, M be the transformed representation, Δ x be the coordinate difference expression formula.

In a kind of possible design, which is that positron emission tomography-computerized tomography is swept PET-CT detection system is retouched, which is the PET image of pet detector detection, which is the detection of CT detector CT image, which determines each first pixel The second coordinate before, this method further include:

According to first transition matrix, the second transition matrix is obtained, which is used for will be in second image Each of the coordinate of the second pixel in the second coordinate system be converted to the coordinate in the first coordinate system；

According to the third coordinate of second transition matrix and each second pixel, the 8th of each second pixel is determined Coordinate, the 8th coordinate of each second pixel are coordinate of each second pixel in the first coordinate system；

According to the first coordinate of the 8th coordinate of each second pixel and each first pixel, from the second image The attenuation coefficient of middle each first pixel of determination；

Since there are errors between the FOV coordinate system of pet detector and the FOV coordinate system of CT detector, so as to cause control Terminal processed is poor by the accuracy of PET image and CT image co-registration；In embodiments of the present invention, controlling terminal obtains a conversion square Battle array, the transition matrix are used to PET image being converted to the image in the FOV coordinate system of CT detector, will by the transition matrix PET image is converted to the image in the FOV coordinate system of CT detector, and the PET image after CT image and conversion is merged, is obtained To PET-CT image.Since the coordinate system of PET image after CT image and conversion is all the FOV coordinate system of CT detector, reduce Error between the coordinate system of PET image and the coordinate system of CT image, therefore improve the accuracy of image co-registration.

Alternatively, controlling terminal obtains a transition matrix, which is used to be converted to CT image under PET detection CT image is converted to the image in the FOV coordinate system of pet detector by the transition matrix by the image in FOV coordinate system, will CT image after PET image and conversion is merged, and PET-CT image is obtained.Due to the seat of CT image after PET image and conversion Mark system is all the FOV coordinate system of pet detector, reduces the error between the coordinate system of PET image and the coordinate system of CT image, Therefore the accuracy of image co-registration is improved, also, improves the precision of correction for attenuation.

Referring to fig. 2, the embodiment of the present invention obtains transition matrix by following steps 201-203, comprising:

Step 201: controlling terminal detects coordinate registration die body by the multi-modal detection system, obtains third image and the Four images, the coordinate registration die body include n point source, which is pair that can be detected by the multi-modal detection system As the object can be the F for being mixed with medical lipiodol¹⁸The FOV coordinate system of-FDG solution or solid-state point source, the third image is First coordinate system, the FOV coordinate system of the 4th image are the second coordinate system, and n is the integer more than or equal to 4.

In the embodiment of the present invention, PET-CT equipment may include pet detector and CT detector, and therefore, this step can be with It is realized by following first way or the second way.

For the first implementation, this step can be with are as follows:

Controlling terminal detects the coordinate registration die body by pet detector, obtains the third image of the coordinate registration die body, Controlling terminal detects the coordinate registration die body by CT detector, obtains the 4th image of the coordinate registration die body.

Wherein, using the FOV coordinate system of third image as the first coordinate system, the i.e. corresponding FOV coordinate system of pet detector, Using the FOV coordinate system of the 4th image as the second coordinate system, the i.e. corresponding FOV coordinate system of CT detector.

In this step, when controlling terminal detects the coordinate registration die body by pet detector, have in pet detector multiple Detector cells pair, pet detector can by multiple detector cells to detect the coordinate registration die body in three-dimensional space In the photon pair launched, form the third image of the coordinate registration die body；Controlling terminal detects the coordinate by CT detector When being registrated die body, CT detector detects the coordinate registration die body real-time perfoming, obtains multiple CT figure of the coordinate registration die body Multiple CT images are formed the 4th image by picture.

It include multiple point sources in the coordinate registration die body, controlling terminal is to the coordinate registration die body in the embodiment of the present invention In point source detected, the point source can be mixed with medical lipiodol by the solution of isotope labeling or solid-state point source, should A pair of contrary photon pair can be launched by the solution of isotope labeling, wherein the solution by isotope labeling can be F¹⁸- FDG (2-Fluorine-18-Fluoro-2-deeoxy-D-glucose, 2- Value linear-fluoro- 2-deoxy-D-glucose) is molten Liquid, or other objects that can be detected simultaneously by multi-modal detection system, the embodiment of the present invention are not made this specifically It limits.

Therefore, controlling terminal detects the coordinate registration die body by pet detector, obtains the third of the coordinate registration die body The step of image, can be with are as follows: controlling terminal detects the photon pair of coordinate registration die body sending by pet detector, and will detection The detection data of the photon pair arrived forms third image.

Wherein, which may include position and number of the photon pair that the pet detector detects etc.；Due to appointing Meaning four can not uniquely determine a three-dimensional space in conplane point, and therefore, which includes at least four It is a not in conplane point source, i.e. n is the integer more than or equal to 4.

In the embodiment of the present invention, to be illustrated for including the coordinate registration die body of six point sources, as shown in figure 3, should The structure of coordinate registration die body can be ladder configuration, include six point sources in the coordinate registration die body, six point sources are along vertical It directly in three groups of direction average mark of ground, is placed on every layer of plate of the coordinate registration die body, the coordinate registration die body midpoint The position that source is placed is without departing from the FOV range of the PET-CT equipment, and the projection of multiple point source in vertical direction will not be mutual Interference will not overlap.

Since each point source in the coordinate registration die body can both be detected by pet detector, can also be detected by CT Device detects.The point source can be the F for being mixed with medical lipiodol¹⁸- FDG solution, then it includes six points that third image, which is one, The 3-D image in source；Multiple CT images composition the 4th image can also be counted as one include six point sources three-dimensional figure Picture.

For second of implementation, this step can be with are as follows:

Controlling terminal can detect the coordinate registration die body by CT detector, obtain the third figure of the coordinate registration die body Picture, controlling terminal can detect the coordinate registration die body by pet detector, obtain the 4th image of the coordinate registration die body. At this point it is possible to using the FOV coordinate system of third image as the first coordinate system, the i.e. corresponding FOV coordinate system of CT detector, by the 4th The FOV coordinate system of image is as the second coordinate system, the i.e. corresponding FOV coordinate system of pet detector.

In the embodiment of the present invention, after controlling terminal determines the third image and the 4th image of coordinate registration die body, by with Lower step 202 determines position of each point source in third image and the 4th image in coordinate registration die body.

Step 202: controlling terminal determines the 4-coordinate of each point source in the n point source in the third image, with And determining the Five Axis of each point source in the 4th image, the 4-coordinate of each point source is each point source at this Coordinate in first coordinate system, the Five Axis of each point source are coordinate of each point source in second coordinate system.

In this step, determine the point source in third image and the 4th figure by the barycentric coodinates of each point source of determination Position as in, therefore, this step can be realized by following steps 2021-2024.

Step 2021: for each point source, controlling terminal determines the sphere where the point source in third image, and obtaining should The pixel value and the 6th coordinate of each third pixel in sphere, the 6th coordinate of each third pixel are each third Coordinate of the pixel in first coordinate system.

During the present invention is implemented, for multiple point sources that coordinate registration die body includes, controlling terminal passes through pet detector and CT Detector separately detects the coordinate registration die body, and each point source in the coordinate registration die body is detected by pet detector In the CT image that the point source is detected by CT detector in obtained position in PET image and the coordinate registration die body Position corresponds.

In order to accurately determine in coordinate registration die body each point source respectively the 4-coordinate in the first coordinate system with And the corresponding Five Axis in the second coordinate system of the point source, it need to be by default time of each point source in the coordinate registration die body Sequence successively determines the 4-coordinate and Five Axis of point source, therefore, in this step, the point source institute is determined in third image Sphere the step of can be with are as follows:

The order in multiple point sources for obtaining the point source, according to the order, in third image where the determining point source Sphere.

Wherein, which can need to be arranged and change according to user, and the embodiment of the present invention does not limit this specifically It is fixed.

It should be noted that controlling terminal determines time of the sphere in multiple point sources where each point source in third image Sequence determines that the order of sphere in multiple point sources where each point source is consistent with controlling terminal in the 4th image.

For example, when the preset order can detect the coordinate registration die body for controlling terminal, it is more in the coordinate registration die body A point source along the axial positive direction of the PET-CT equipment order.

Fig. 3 is the structure chart of the coordinate registration die body, as shown in figure 3, by taking coordinate registration die body includes six point sources as an example It is illustrated, which includes the first transverse slat, the second transverse slat, third transverse slat and for fixing the first transverse slat, the Point source 1 and point source 2 are arranged on the first transverse slat, is arranged on the second transverse slat for two transverse slats, the starting stave of third transverse slat and the second riser Point source 5 and point source 6 are arranged on third transverse slat for point source 3 and point source 4.

Fig. 4 is the top view of the coordinate registration die body, and each point source is perpendicular to ground direction in the coordinate registration die body Projective distribution it is as shown in Figure 4.

In the embodiment of the present invention, can by PET-CT equipment perpendicular to the direction definition of the plane where detector annulus For axial direction, if by from the first transverse slat to the direction definition of third transverse slat being axial positive direction, the coordinate registration in top view In the projection of die body in vertical direction, the projection of multiple point sources is non-interference, and each point source is along the PET- in multiple point sources The axial positive direction of CT equipment is successively are as follows: 2 → 1 → 3 → 4 → 6 → 5；Therefore, controlling terminal is in third image and the 4th image Sphere sequence in the middle multiple point sources of determination where each point source can be equal are as follows: 2 → 1 → 3 → 4 → 6 → 5.

In this step, in third image, first determine that the centre of sphere determines the practical body than the point source according to the centre of sphere The bigger sphere of product, therefore, the step of controlling terminal determines the sphere where the point source in third image, can be with are as follows: control is eventually End obtains the pixel value of each third pixel in third image, determines the maximum point of pixel value in multiple third pixels, with The point is the centre of sphere, determines that a volume is greater than the sphere of the point source actual volume, and using the sphere as the ball where the point source Body.

Wherein, controlling terminal determine a volume be greater than the point source actual volume sphere when, can by following steps, Determine a sphere more bigger than the point source actual volume: controlling terminal determine the centre of sphere to the maximum distance in point source boundary, According to the maximum distance, the pre-determined distance for being greater than the maximum distance is determined, which is determined as the half of the sphere Diameter, and then determine a sphere more bigger than the actual volume of the point source.

Wherein, which can need to be arranged and change according to user, and the embodiment of the present invention does not do specific limit to this It is fixed.For example, the pre-determined distance can increase the distance of unit length for the maximum distance, for example, if the maximum distance is 3 millis Rice, the pre-determined distance can be 3.1 millimeters.It should be noted that controlling terminal is in third image where the determining point source In the step of sphere, the implementation that one volume of the determination is greater than the sphere of the point source actual volume can be according to user's needs Be arranged and change, the embodiment of the present invention to one volume of the determination be greater than the point source actual volume sphere implementation not Make specific limit.

In this step, controlling terminal is obtained in the sphere after the pixel value of each third pixel, determines each third Coordinate of the pixel in the first coordinate system, using the coordinate of the third pixel in the first coordinate system as the 6th coordinate.

Step 2022: controlling terminal passes through following formula according to the pixel value and the 6th coordinate of each third pixel One (1) determines the 4-coordinate of the point source:

It should be noted that the 4-coordinate of the point source is the barycentric coodinates of sphere where the point source.

Step 2023: for each point source, controlling terminal determines the sphere where the point source in the 4th image, and obtaining should The pixel value and the 9th coordinate of each 4th pixel in sphere, the 9th coordinate of each 4th pixel be this each 4th Coordinate of the pixel in second coordinate system.

This step realizes that process is similar with step 2021, and details are not described herein.

Step 2024: controlling terminal passes through following formula according to the pixel value and the 9th coordinate of each 4th pixel One (2) determine the Five Axis of the point source:

This step realizes that process is similar with step 2022, and details are not described herein.

In turn, controlling terminal passes through following steps according to the 4-coordinate of third image and the Five Axis of the 4th image 203, determine the first transition matrix, which can be used for the 4-coordinate by third image in the first coordinate system It is converted into the 7th coordinate of third image in the second coordinate system.

Step 203: controlling terminal determines the first conversion square according to the 4-coordinate and Five Axis of each point source Battle array.

If third image is the PET image of pet detector detection, the 4th image is the CT image of CT detector detection, Then the first transition matrix is the corresponding transition matrix of image that PET image is converted to the FOV coordinate system of CT detector.

If third image is the CT image of CT detector detection, the 4th image is the PET image of pet detector detection, Then the first transition matrix is the corresponding transition matrix of image that CT image is converted to the FOV coordinate system of pet detector.

It is third figure by coordinate transformation of the third image in the first coordinate system by the first transition matrix in this step When as coordinate in the second coordinate system, for each point source in coordinate registration die body, after controlling terminal is by determining conversion The third image Five Axis of the 7th coordinate and the 4th image in the second coordinate system in the second coordinate system coordinate difference It is different, further determine that the smallest first transition matrix of total coordinate difference so that multiple point sources.

Therefore, this step can be realized by following steps 2031-2033.

Step 2031: controlling terminal is according to the 4-coordinate of each point source, Five Axis and first transition matrix Transformed representation determines the coordinate difference expression formula of each point source.

In the embodiment of the present invention, for each point source in coordinate registration die body, controlling terminal has been pre-defined third Image is converted to the transformed representation of the 7th coordinate of third image in the second coordinate system in the 4-coordinate of the first coordinate system, Then, by the transformed representation, the corresponding Five Axis of the point source in the 7th coordinate and the 4th image is further determined that Coordinate difference.Therefore, this step can be realized by following steps 2031a-2031b.

Step 2031a: controlling terminal passes through following formula according to the 4-coordinate and the transformed representation of each point source Two, determine that the 7th coordinate of each point source, the 7th coordinate of each point source are each point source in second coordinate system Coordinate:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

Wherein, (x₇,y₇,z₇) be each point source the 7th coordinate, (x₄,y₄,z₄) be each point source 4-coordinate, M is the transformed representation.

In the embodiment of the present invention, for each point source in coordinate registration die body, controlling terminal turns according in formula two Expression formula is changed, 4-coordinate of the third image in the first coordinate system is converted into the 7th of third image in the second coordinate system Coordinate.

When the corresponding conversion square of image that the first transition matrix is the FOV coordinate system that CT image is converted to pet detector When battle array, which can be with are as follows: (x₇,y₇,z₇)=M_CT→PET×(x₄,y₄,z₄), the M_CT→PETIt indicates to detect CT image in CT When coordinate in the FOV coordinate system of device is converted to coordinate of the CT image in the FOV coordinate system of pet detector, the first conversion square The corresponding transformed representation of battle array；

When the corresponding conversion square of image that the first transition matrix is the FOV coordinate system that PET image is converted to CT detector When battle array, which can be with are as follows: (x₇,y₇,z₇)=M_PET→CT×(x₄,y₄,z₄), the M_PET→CTIt indicates to visit PET image in PET When the coordinate surveyed in the FOV coordinate system of device is converted to coordinate of the PET image in the FOV coordinate system of CT detector, the first conversion The corresponding transformed representation of matrix.

In this step, 4-coordinate of the third image in the first coordinate system can be converted to second by rotation and translation The 7th coordinate in coordinate system.

If each coordinate for including by the first coordinate system is rotating around x, y, z using the second coordinate system as reference frame Axis rotates a certain angle, so that the x, y, z axis of the first coordinate system is parallel with the x, y, z axis of reference frame respectively, it then, will Each coordinate that first coordinate system includes translates a certain distance along x, y, z axis direction respectively, so that the coordinate of the first coordinate system Origin is Chong Die with the coordinate origin of reference frame, finally, so that the first coordinate system and the second coordinate system are consistent.

Wherein, each coordinate that the first coordinate system includes is denoted as respectively rotating around the angle of the rotation of x, y, z axisθ、 Each coordinate that first coordinate system includes is denoted as x along the distance that x, y, z axis direction translates respectively by ψ₀、y₀、z₀, in this way, obtaining 4-coordinate of the third image in the first coordinate system is converted to the conversion parameter of the 7th coordinate in the second coordinate system, this turn Changing parameter includes rotation parameterθ, ψ and translation parameters x₀、y₀、z₀；Then by third image in the first coordinate system the 4th The transformed representation M that coordinate is converted to the 7th coordinate in the second coordinate system can be indicated with the conversion parameter are as follows:

Due to the factors on test data such as experimental error and pet detector and CT detector resolution generate influence, Cause actual experiment data that can not obtain the accurate solution for the multiple parameters that the matrix includes.Therefore, present invention implementation cannot root The accurate solution of multiple parameter is directly obtained according to the formula two.

In the embodiment of the present invention, using the algorithm from wound, pass through following steps 2031b, there is shown the after conversion the 4th sits Existing coordinate difference between Five Axis in mark and the second coordinate system.

Step 2031b: controlling terminal is according to the Five Axis and the 7th coordinate of each point source, by following formula three, Determine the coordinate difference expression formula of each point source:

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

Wherein, (x₇,y₇,z₇) be each point source the 7th coordinate, (x₅,y₅,z₅) be each point source Five Axis, Δ x is the coordinate difference expression formula.

By step 2031a it is found that the relationship between the 7th coordinate and 4-coordinate can be indicated by following formula two are as follows:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄),

Therefore, formula two is substituted into formula three, obtains following formula four, i.e., the coordinate difference expression formula of each point source can be with It indicates are as follows:

Formula four: Δ x=(x₅,y₅,z₅)-M×(x₄,y₄,z₄)。

When the corresponding conversion square of image that the first transition matrix is the FOV coordinate system that CT image is converted to pet detector When battle array, which can be with are as follows: Δ x_CT→PET=(x₅,y₅,z₅)-M_CT→PET×(x₄,y₄,z₄), Δ x_CT→PETFor by CT image When the coordinate in the FOV coordinate system of CT detector is converted to coordinate of the CT image in the FOV coordinate system of pet detector, the The corresponding coordinate difference expression formula of one transition matrix.

When the corresponding conversion square of image that the first transition matrix is the FOV coordinate system that PET image is converted to CT detector When battle array, which can be with are as follows: Δ x_PET→CT=(x₅,y₅,z₅)-M_PET→CT×(x₄,y₄,z₄), Δ x_PET→CTFor by PET image When the coordinate in the FOV coordinate system of pet detector is converted to coordinate of the PET image in the FOV coordinate system of CT detector, the The corresponding coordinate difference expression formula of one transition matrix.

Step 2032: controlling terminal determines total seat of multiple point source according to the coordinate difference expression formula of each point source Mark differential expression formula.

In the embodiment of the present invention, by following formula five, i.e., total coordinate difference expression formula, further by coordinate registration die body In total coordinate difference of multiple point sources be expressed as:

Wherein, n indicates the n point source that the coordinate registration die body includes, the dimension of j indicates coordinate axis, i.e., for indicating this Three dimensions of the coordinate x, y, z axis in the first coordinate system and the second coordinate system respectively, Error indicate total coordinate difference table Up to formula.

Therefore, by the formula five, define in coordinate registration die body the Five Axis of multiple point sources and the 7th coordinate it Between difference.

Wherein, when the first transition matrix is corresponding turn of image of FOV coordinate system that CT image is converted to pet detector When changing matrix, which can be with are as follows:The Error_CT→PETFor by CT image It is right when the coordinate in the FOV coordinate system of CT detector is converted to coordinate of the CT image in the FOV coordinate system of pet detector The total coordinate difference expression formula answered.

When the corresponding conversion square of image that the first transition matrix is the FOV coordinate system that PET image is converted to CT detector When battle array, which can be with are as follows:The Error_PET→CTFor PET image is existed It is corresponding when coordinate in the FOV coordinate system of pet detector is converted to coordinate of the PET image in the FOV coordinate system of CT detector Total coordinate difference expression formula.

Step 2033: controlling terminal determines first conversion for meeting preset condition according to total coordinate difference expression formula Matrix, the preset condition are that the value of total coordinate difference expression formula is minimum value.

In this step, according to total coordinate difference expression formula formula five, the coordinate difference expression formula of each point source is substituted into Total coordinate difference expression formula, that is, substitute into formula four in formula five, then, by each point source determined in step 202 in third The 4-coordinate and the point source of image successively substitute into the formula five in the Five Axis of the 4th image, determine that one makes this total The smallest transition matrix of value of coordinate difference expression formula determines the smallest conversion of value so that total coordinate difference expression formula Each conversion parameter in matrix, in turn, using the transition matrix as the first transition matrix.

Specific implementation process are as follows: by each point source in the 4-coordinate of third image and the point source the of the 4th image Five Axis successively substitutes into the formula five, reasonable model on the basis of the first transition matrix initial value, required by physical process In enclosing, suitable step-length is chosen, the value of total coordinate difference expression formula is successively calculated, therefrom chosen so that coordinate difference expression formula Value corresponding transition matrix, optimal solution as the first transition matrix when minimum.

Wherein, the first transition matrix initial value can be with are as follows:

In this way, controlling terminal can be according to first transition matrix, by fourth seat of the third image in the first coordinate system Mark is converted to the 7th coordinate of the third image in the second coordinate system.

In the embodiment of the present invention, controlling terminal stores the numerical value of each of the first transition matrix conversion parameter, is convenient for It is subsequent using controlling terminal when, according to first transition matrix, by PET-CT the first image and the second image merge.

In a kind of possible design of the embodiment of the present invention, due to can store target in the CT image of CT detector detection The attenuation coefficient of object, controlling terminal first can carry out correction for attenuation to PET image by the attenuation coefficient, then, further according to Revised PET image is converted to the image in the FOV coordinate system of CT detector by the first transition matrix, and then will be after amendment PET image and CT image merged, further improve the accuracy of fused PET-CT image.

Therefore, when the first transition matrix is corresponding turn of image of FOV coordinate system that PET image is converted to CT detector When changing matrix, also need to determine that the image for the FOV coordinate system that CT image is converted to pet detector is corresponding by following steps a Transition matrix, i.e. the second transition matrix.

Step a: controlling terminal obtains the second transition matrix according to first transition matrix, which is used for The 4th pixel of each of 4th image is converted in first coordinate system in the coordinate in second coordinate system Coordinate.

In this step, if first transition matrix can be M_PET→CT, at this point, due to needing according to the 4th after conversion The image data of the 4th pixel of each of image, determines attenuation coefficient, therefore, also needs to obtain the second transition matrix M_CT→PET, Will the 4th pixel of each of the 4th image be converted in first coordinate system in the coordinate in second coordinate system Coordinate transition matrix.

By formula two: (x₇,y₇,z₇)=M_PET→CT×(x₄,y₄,z₄) it is found that first transition matrix and the second transition matrix With orthogonality, the orthogonality of first transition matrix and the second transition matrix can be indicated by following formula six are as follows:

Formula six: M_CT→PET=M_PET→CT ^-1,

Therefore, controlling terminal determines the second transition matrix M directly according to the formula six_CT→PETIn each parameter value be It can.

In the embodiment of the present invention, controlling terminal stores the numerical value of each of the second transition matrix conversion parameter, is convenient for When subsequent use PET-CT equipment, according to second transition matrix, CT image is converted in the FOV coordinate system of pet detector Image, and according to the CT image after the conversion to PET image carry out correction for attenuation.

In the embodiment of the present invention, after controlling terminal detection coordinate registration die body obtains third image and the 4th image, pass through The Five Axis and transformed representation of the 4-coordinate of third image, the 4th image define each point source in multiple point sources and sit Differential expression formula is marked, and then determines that total coordinate difference expression formula of multiple point sources is determined according to total coordinate difference expression formula So that the smallest first transition matrix of the value of total coordinate difference expression formula；Since the embodiment of the present invention is not only according to each The conversion expression and 4-coordinate, Five Axis of point source directly obtain the first transition matrix, but have pre-defined multiple Total coordinate difference expression formula of point source, substitutes into total coordinate difference expression formula for 4-coordinate, Five Axis, determines the first conversion Matrix, moreover, first transition matrix makes the value of total coordinate difference expression formula minimum, therefore, the embodiment of the present invention makes Coordinate difference is minimum after conversion, to improve image co-registration precision.In PET/CT system, not only improves and improve lesion localization Accuracy is also beneficial to the precision that PET image carries out correction for attenuation using CT image.

The embodiment of the invention provides a kind of method of multi-modal detection system image co-registration, this method utilizes above-mentioned implementation The first transition matrix that example obtains, merges image.The executing subject of this method can be that controlling terminal should referring to Fig. 5 Method includes:

Step 301: controlling terminal detects target object by multi-modal detection system, obtains the first image and the second figure Picture, the visual field FOV coordinate system of first image are the first coordinate system, and the FOV coordinate system of second image is the second coordinate system.

In this step, controlling terminal may include pet detector and CT detector, and therefore, this step can be by following First way or the second way are realized.

For the first implementation, this step can be with are as follows:

Controlling terminal detects the target object by pet detector, obtains the first image of the target object, controlling terminal The target object is detected by CT detector, obtains the second image of the target object.

Wherein, using the FOV coordinate system of the first image as the first coordinate system, the i.e. corresponding FOV coordinate system of pet detector, Using the FOV coordinate system of the second image as the second coordinate system, the i.e. corresponding FOV coordinate system of CT detector.

It should be noted that controlling terminal detects the target object by pet detector, the first of the target object is obtained Image, controlling terminal detect the target object by CT detector, and the implementation for obtaining the second image of the target object is same The implementation of step 201 is consistent, no longer repeats one by one herein.

For second of implementation, this step can be with are as follows:

The step of controlling terminal detects target object by multi-modal detection system, obtains the first image and the second image is also It can be with are as follows: controlling terminal detects the target object by CT detector, obtains the first image of the target object, and controlling terminal is logical It crosses pet detector and detects the target object, obtain the second image of the target object.

At this point, using the FOV coordinate system of the first image as the first coordinate system, the i.e. corresponding FOV coordinate system of CT detector, it will The FOV coordinate system of second image is as the second coordinate system, the i.e. corresponding FOV coordinate system of pet detector.

Step 302: controlling terminal obtains the first transition matrix, which is used for will be every in first image A first pixel is converted to the coordinate in second coordinate system in the coordinate in first coordinate system.

In the embodiment of the present invention, controlling terminal first passes through step 201-203 in advance, it is determined that the first conversion square of controlling terminal Battle array, and each conversion parameter in first transition matrix is stored, therefore, in this step, PET-CT directly acquires this first turn Change matrix.

It should be noted that first transition matrix can convert 4-coordinate of the third image in the first coordinate system Corresponding for the 7th coordinate of the third image in the second coordinate system, in this step, which is used for the Coordinate of one image in the first coordinate system is converted to the coordinate of the first image in the second coordinate system.

In a kind of possible design of the embodiment of the present invention, after controlling terminal obtains the first transition matrix, can directly it hold First coordinate of the first pixel each in the first image is converted to each of first image first by row following steps 303 The second coordinate of pixel in the second coordinate system.

In a kind of possible design of the embodiment of the present invention, which can pass through pet detector for controlling terminal The PET image of detection, then the second image is the CT image that controlling terminal is detected by CT detector, at this point, due in CT image The attenuation coefficient of target object is stored, controlling terminal first can carry out correction for attenuation to PET image by the attenuation coefficient, Then, then pass through following steps 303-304, PET image is converted into the image in the FOV coordinate system of CT detector, and then will PET image and CT image after conversion are merged.

Therefore, before controlling terminal executes step 303, following steps b-c is first passed through, the first image is corrected.

Step b: controlling terminal determines that this is every according to the third coordinate of second transition matrix and each second pixel 8th coordinate of a second pixel, the 8th coordinate of each second pixel are each second pixel in first seat Coordinate in mark system.

In this step, if first transition matrix can be M_PET→CT, at this point, due to needing according to second after conversion The image data of the second pixel of each of image, determines attenuation coefficient, therefore, also needs to obtain the second transition matrix M_CT→PET, Second transition matrix is for being converted to coordinate of the second pixel of each of second image in second coordinate system Coordinate in first coordinate system.

In this step, since controlling terminal passes through step a, it is determined that the second transition matrix, therefore, controlling terminal directly obtains The second transition matrix is taken, according to second transformed matrix M_CT→PET, the conversion parameter in second transition matrix is determined, according to this Conversion parameter in second transition matrix, respectively by the third coordinate of the second pixel of each of second image be converted to this Eightth coordinate of two pixels in the first coordinate system.

Step c: controlling terminal is according to the first of the 8th coordinate of each second pixel and each first pixel Coordinate determines the attenuation coefficient of each first pixel from second image；Controlling terminal is according to each first pixel The attenuation coefficient of point, is corrected first image.

In the embodiment of the present invention, when detecting target object by CT detector due to controlling terminal, which can be with Target object is scanned by X-ray beam, therefore, declining for X-ray can be obtained from the image data of second image Subtract coefficient, the i.e. attenuation coefficient of the first pixel；Due to by step b, by the of the second pixel of each of second image After three coordinates are converted to the 8th coordinate of second pixel in the first coordinate system, sat according to the 8th of second pixel the Mark, can determine attenuation coefficient of the target object in the first image at corresponding second pixel position, i.e. the first image In the first pixel attenuation coefficient.

Correspondingly, for each second pixel, controlling terminal is according to the 8th coordinate of second pixel, from the second figure Determine that the 8th coordinate position corresponds to attenuation coefficient as in.

Wherein, controlling terminal is according to the attenuation coefficient of each first pixel, to the corrected step of the first image It suddenly can be with are as follows:

Controlling terminal is according to the attenuation coefficient of the first pixel each in first image, in first image each The image data of one pixel is corrected, by first after the image data composition correction of the first pixel each of after correction Image.

Step 303: for controlling terminal according to first coordinate of first transition matrix and each first pixel, determining should Second coordinate of each first pixel, the first coordinate of each first pixel be each first pixel this first Coordinate in coordinate system, the second coordinate of each first pixel are each first pixel in second coordinate system Coordinate.

In the embodiment of the present invention, the first pixel is the pixel in the first image, and the second pixel is each second figure Pixel as in, first pixel corresponding coordinate in the first coordinate system is the first coordinate.

In this step, controlling terminal makees each conversion parameter according to each conversion parameter in the first transition matrix For first coordinate of each first pixel in the first coordinate system, each first pixel is obtained in the second coordinate system The second coordinate.

Step 304: controlling terminal is according to each of the second coordinate of each first pixel and second image First image and second image are merged, obtain multi-modal detection image by the third coordinate of two pixels, this is each The third coordinate of second pixel is the coordinate of each second pixel in second coordinate system.

In the embodiment of the present invention, corresponding coordinate is each of the second image includes the second pixel in the second coordinate system Third coordinate.

This step can be with are as follows: controlling terminal is according to the second coordinate of the first pixel of each of first image, from The image data that each first pixel is determined in one image, sits according to the third of the second pixel of each of second image Mark determines the image data of each second pixel from the second image, according to the second coordinate of each first pixel and The third coordinate of the second pixel of each of second image, by the image data of each first pixel and each second picture The image data of vegetarian refreshments is merged, and mode detection image, such as PET-CT image are obtained.

In a kind of possible design of the embodiment of the present invention, if controlling terminal carries out the first image by step b-c Correction, that is, when the multi-modal detection system is PET-CT detection system, which is PET image, which is CT Image, then this step can be with are as follows:

Controlling terminal is according to the second pixel of each of the second coordinate of each first pixel and second image Third coordinate, the first image after second image and the correction is merged, multi-modal detection image is obtained, this is each The third coordinate of second pixel is the coordinate of each second pixel in second coordinate system.

Controlling terminal is according to the second coordinate of the first pixel of each of the first image after the correction, after correction The image data that each first pixel is determined in first image, according to the third of the second pixel of each of second image Coordinate determines the image data of each second pixel, according to the second coordinate of each first pixel from the second image With the third coordinate of the second pixel of each of second image, by the image data of each first pixel and each second The image data of pixel is merged, and mode detection image namely PET-CT image are obtained.

The present invention provides a kind of device of multi-modal detection system image co-registration in implementing, referring to Fig. 6, the device packet It includes:

Detecting module 401 obtains the first image and the second figure for detecting target object by multi-modal detection system Picture, the visual field FOV coordinate system of first image are the first coordinate system, and the FOV coordinate system of second image is the second coordinate system；

First obtains module 402, and for obtaining the first transition matrix, which is used for will be in first image Each of the first pixel be converted to the coordinate in second coordinate system in the coordinate in first coordinate system；

First determining module 403, for the first coordinate according to first transition matrix and each first pixel, really Second coordinate of fixed each first pixel, the first coordinate of each first pixel are each first pixel at this Coordinate in first coordinate system, the second coordinate of each first pixel are each first pixel in second coordinate system In coordinate；

Fusion Module 404, for according to each of the second coordinate of each first pixel and second image First image and second image are merged, obtain multi-modal detection image by the third coordinate of two pixels, this is each The third coordinate of second pixel is the coordinate of each second pixel in second coordinate system.

In a kind of possible design, the first acquisition module 402, comprising:

Probe unit obtains third image and the 4th for detecting coordinate registration die body by the multi-modal detection system Image, the coordinate registration die body include n point source, which is the object that can be detected by the multi-modal detection system, The object can be the F for being mixed with medical lipiodol¹⁸- FDG solution or solid-state point source, the FOV coordinate system of the third image are the One coordinate system, the FOV coordinate system of the 4th image are the second coordinate system, and n is the integer more than or equal to 4；

Determination unit, for determining the 4-coordinate of each point source in the n point source in the third image, and really The Five Axis of each point source in fixed 4th image, the 4-coordinate of each point source be each point source this first Coordinate in coordinate system, the Five Axis of each point source are coordinate of each point source in second coordinate system；It is every according to this The 4-coordinate and Five Axis of a point source determine first transition matrix.

In a kind of possible design, the determination unit, comprising:

First determines subelement, for determining the sphere where the point source in the third image for each point source, The pixel value and the 6th coordinate of each third pixel in the sphere are obtained, the 6th coordinate of each third pixel is that this is every Coordinate of a third pixel in first coordinate system；

Second determines that subelement passes through following public affairs for the pixel value and the 6th coordinate according to each third pixel Formula one (1) determines the 4-coordinate of the point source:

In a kind of possible design, the determination unit, comprising:

Third determines subelement, for determining the sphere where the point source in the 4th image for each point source, The pixel value and the 9th coordinate of each 4th pixel in the sphere are obtained, the 9th coordinate of each 4th pixel is that this is every Coordinate of a 4th pixel in second coordinate system；

4th determines that subelement passes through following public affairs for the pixel value and the 9th coordinate according to each 4th pixel Formula one (2) determines the Five Axis of the point source:

In a kind of possible design, the determination unit, comprising:

5th determines subelement, for the 4-coordinate, Five Axis and first transition matrix according to each point source Transformed representation, determine the coordinate difference expression formula of each point source；

6th determines that subelement determines the total of multiple point source for the coordinate difference expression formula according to each point source Coordinate difference expression formula；

7th determines subelement, for determining this first turn for meeting preset condition according to total coordinate difference expression formula Matrix is changed, which is that the value of total coordinate difference expression formula is minimum value.

In a kind of possible design, the 5th determines subelement, be also used to according to the 4-coordinate of each point source and The transformed representation determines the 7th coordinate of each point source, the 7th coordinate of each point source is should by following formula two Coordinate of each point source in second coordinate system:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

5th determines subelement, is also used to the Five Axis and the 7th coordinate according to each point source, passes through following public affairs Formula three determines the coordinate difference expression formula of each point source:

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

In a kind of possible design, which is that positron emission tomography-computerized tomography is swept PET-CT detection system is retouched, which is the PET image of pet detector detection, which is the detection of CT detector CT image, the device further include:

Second obtains module, and for obtaining the second transition matrix according to first transition matrix, which is used In being converted to the second pixel of each of second image in first coordinate system in the coordinate in second coordinate system Coordinate；

Second determining module is determined for the third coordinate according to second transition matrix and each second pixel 8th coordinate of each second pixel, the 8th coordinate of each second pixel be each second pixel this Coordinate in one coordinate system；

Third determining module, for according to the of the 8th coordinate of each second pixel and each first pixel One coordinate determines the attenuation coefficient of each first pixel from second image；

Correction module is corrected first image for the attenuation coefficient according to each first pixel.

It should be understood that the device of multi-modal detection system image co-registration provided by the above embodiment is in multi-modal detection When system image co-registration, only the example of the division of the above functional modules, in practical application, can according to need and Above-mentioned function distribution is completed by different functional modules, i.e., the internal structure of device is divided into different functional modules, with Complete all or part of function described above.In addition, multi-modal detection system image co-registration provided by the above embodiment The embodiment of the method for device and multi-modal detection system image co-registration belongs to same design, and specific implementation process is detailed in method reality Example is applied, which is not described herein again.

Referring to Fig. 7, the embodiment of the invention provides a kind of controlling terminals 500.The processing terminal 500 is for implementing above-mentioned reality The method that the multi-modal detection system image co-registration provided in example is provided.Specifically:

Processing terminal 500 may include processor 510, transceiver 520, memory 530, input unit 550, display unit 550, the components such as voicefrequency circuit 560 and power supply 570, as shown in fig. 7, it will be understood by those skilled in the art that shown in Fig. 7 The restriction of terminal structure not structure paired terminal, may include than illustrating more or fewer components or the certain components of combination, Or different component layout.Wherein:

Processor 510 can be the control centre of terminal, utilize each of various interfaces and the entire terminal device of connection A part, such as transceiver 520 and memory 530, by run or execute the software program being stored in memory 530 and/ Or module, and the data being stored in memory 530 are called, the various functions and processing data of processing terminal 500 are executed, from And integral monitoring is carried out to processing terminal 500.Optionally, processor 510 may include one or more processing cores.In the present invention In, processor 510 is determined for the relevant treatment of gate-control signal.Transceiver 520 can be used for sending and receiving data, eventually End can send and receive data by transceiver 520, and terminal can be net by internet sending and receiving data, transceiver Card.

Memory 530 can be used for storing software program and module, and processor 510 is stored in memory 530 by operation Software program and module, thereby executing various function application and data processing.Memory 530 can mainly include storage journey Sequence area and storage data area, wherein storing program area can the (ratio of application program needed for storage program area, at least one function Such as determine gate-control signal function) etc.；Storage data area, which can be stored, uses created data (such as the vanishing point that falls into oblivion according to terminal Location information etc.) etc..In addition, memory 530 may include high-speed random access memory, it can also include non-volatile memories Device, for example, at least a disk memory, flush memory device or other volatile solid-state parts.Input unit 540 can be with Number for receiving input or character information, and generate keyboard related with user setting and function control, mouse, behaviour Make bar, optics or trackball signal input.Display unit 550 is displayed for information input by user or is supplied to use The information at family and the various graphical user interface of terminal, these graphical user interface can be by figure, text, icon, videos It is constituted with any combination thereof.Display unit 550 may include display panel 551, optionally, can use LCD (Liquid Crystal Display, liquid crystal display), OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode) Etc. forms configure display panel 551.Voicefrequency circuit 560, loudspeaker 561, microphone 562 can provide between user and terminal The audio data received can be converted to electric signal by audio interface, voicefrequency circuit 560.Power supply 570 can pass through power management System and processor 510 are logically contiguous, to realize management charging, electric discharge and power managed etc. by power-supply management system Function.Power supply 570 can also include one or more direct current or AC power source, recharging system, power failure monitor The random components such as circuit, power adapter or inverter, power supply status indicator.

Specifically in embodiments of the present invention, processing terminal 500 further includes having memory and one or more than one Program, perhaps more than one program is stored in memory and is configured to by one or more than one processing for one of them Device executes.Said one or more than one program include the instruction for performing the following operation:

Target object is detected by multi-modal detection system, obtains the first image and the second image, the view of first image Wild FOV coordinate system is the first coordinate system, and the FOV coordinate system of second image is the second coordinate system.

The first transition matrix is obtained, which is used for the first pixel of each of first image first Coordinate in coordinate system is converted to coordinate in the second coordinate system.

According to first coordinate of first transition matrix and each first pixel, the of each first pixel is determined Two coordinates, the first coordinate of each first pixel are the coordinate of each first pixel in the first coordinate system, this is every Second coordinate of a first pixel is the coordinate of each first pixel in the second coordinate system.

It is sat according to the third of the second pixel of each of the second coordinate of each first pixel and second image First image and second image are merged, obtain multi-modal detection image, the third of each second pixel by mark Coordinate is the coordinate of each second pixel in second coordinate system.

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware It completes, relevant hardware can also be instructed to complete by program, the program can store in a kind of computer-readable In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..

The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.

Claims

1. a kind of method of multi-modal detection system image co-registration, which is characterized in that the described method includes:

Target object is detected by multi-modal detection system, obtains the first image and the second image, the visual field of the first image FOV coordinate system is the first coordinate system, and the FOV coordinate system of second image is the second coordinate system；

The first transition matrix is obtained, first transition matrix is used for the first pixel of each of the first image in institute It states the coordinate in the first coordinate system and is converted to the coordinate in second coordinate system；

According to the first coordinate of first transition matrix and each first pixel, each first pixel is determined The second coordinate, the first coordinate of each first pixel is each first pixel in first coordinate system Coordinate, the second coordinate of each first pixel is seat of each first pixel in second coordinate system Mark；

It is sat according to the third of the second pixel of each of the second coordinate of each first pixel and second image Mark, the first image and second image are merged, multi-modal detection image, each second pixel are obtained Third coordinate be coordinate of each second pixel in second coordinate system；

The first transition matrix of the acquisition, comprising:

Coordinate registration die body is detected by the multi-modal detection system, obtains third image and the 4th image, the coordinate is matched Quasi-mode body includes n point source, and the n point source is the object that can be detected by the multi-modal detection system, the object It can be the F for being mixed with medical lipiodol¹⁸- FDG solution or solid-state point source, the FOV coordinate system of the third image are the first seat Mark system, the FOV coordinate system of the 4th image are the second coordinate system, and n is the integer more than or equal to 4；

It determines the 4-coordinate of each point source in the n point source in the third image, and determines the 4th figure The Five Axis of each point source as in, the 4-coordinate of each point source are that each point source is sat described first Coordinate in mark system, the Five Axis of each point source are coordinate of each point source in second coordinate system；

According to the 4-coordinate and Five Axis of each point source, first transition matrix is determined；

The 4-coordinate of each point source in the n point source in the determination third image, comprising:

For each point source, the sphere where the point source is determined in the third image, is obtained every in the sphere The pixel value and the 6th coordinate of a third pixel, the 6th coordinate of each third pixel are each third pixel Coordinate of the point in first coordinate system；

The point source is determined by following formula one (1) according to the pixel value and the 6th coordinate of each third pixel 4-coordinate:

Formula one (1):

2. the method according to claim 1, wherein each point source in the determination the 4th image Five Axis, comprising:

For each point source, the sphere where the point source is determined in the 4th image, is obtained every in the sphere The pixel value and the 9th coordinate of a 4th pixel, the 9th coordinate of each 4th pixel are each 4th pixel Coordinate of the point in second coordinate system；

Formula one (2):

3. the method according to claim 1, wherein the 4-coordinate and the 5th according to each point source Coordinate determines first transition matrix, comprising:

According to the 4-coordinate of each point source, the transformed representation of Five Axis and first transition matrix, institute is determined State the coordinate difference expression formula of each point source；

According to the coordinate difference expression formula of each point source, total coordinate difference expression formula of the n point source is determined；

According to total coordinate difference expression formula, first transition matrix for meeting preset condition, the preset condition are determined Value for total coordinate difference expression formula is minimum value.

4. according to the method described in claim 3, it is characterized in that, the 4-coordinate according to each point source, the 5th The transformed representation of coordinate and first transition matrix determines the coordinate difference expression formula of each point source, comprising:

Each point is determined by following formula two according to the 4-coordinate and the transformed representation of each point source 7th coordinate of the 7th coordinate in source, each point source is coordinate of each point source in second coordinate system:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

The seat of each point source is determined by following formula three according to the Five Axis and the 7th coordinate of each point source Mark differential expression formula:

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

5. the method according to claim 1, wherein the multi-modal detection system is aobvious for positron emission fault Picture-computed tomography PET-CT detection system, the PET image that the first image detects for pet detector, described second Image is the CT image of CT detector detection, described according to the first of first transition matrix and each first pixel Coordinate, before the second coordinate for determining each first pixel, the method also includes:

According to first transition matrix, the second transition matrix is obtained, second transition matrix is used for second image Each of the second pixel be converted to the coordinate in first coordinate system in the coordinate in second coordinate system；

According to the third coordinate of second transition matrix and each second pixel, each second pixel is determined The 8th coordinate, the 8th coordinate of each second pixel is each second pixel in first coordinate system Coordinate；

According to the first coordinate of the 8th coordinate of each second pixel and each first pixel, from described second The attenuation coefficient of each first pixel is determined in image；

6. a kind of device of multi-modal detection system image co-registration, which is characterized in that described device includes:

Detecting module, for obtaining the first image and the second image by multi-modal detection system detection target object, described the The visual field FOV coordinate system of one image is the first coordinate system, and the FOV coordinate system of second image is the second coordinate system；

First obtains module, and for obtaining the first transition matrix, first transition matrix is used for will be in the first image Each first pixel is converted to the coordinate in second coordinate system in the coordinate in first coordinate system；

First determining module is determined for the first coordinate according to first transition matrix and each first pixel Second coordinate of each first pixel, the first coordinate of each first pixel are each first pixel Coordinate in first coordinate system, the second coordinate of each first pixel are each first pixel in institute State the coordinate in the second coordinate system；

Fusion Module, for according to the second picture of each of the second coordinate of each first pixel and second image The third coordinate of vegetarian refreshments merges the first image and second image, obtains multi-modal detection image, described every The third coordinate of a second pixel is coordinate of each second pixel in second coordinate system；

Described first obtains module, comprising:

Probe unit obtains third image and the 4th figure for detecting coordinate registration die body by the multi-modal detection system Picture, the coordinate registration die body include n point source, and the n point source can be detected by the multi-modal detection system Object, the object can be the F for being mixed with medical lipiodol¹⁸The FOV of-FDG solution or solid-state point source, the third image is sat Mark system is the first coordinate system, and the FOV coordinate system of the 4th image is the second coordinate system, and n is the integer more than or equal to 4；

Determination unit, for determining the 4-coordinate of each point source in the n point source in the third image, and really The Five Axis of each point source in fixed 4th image, the 4-coordinate of each point source are each point source Coordinate in first coordinate system, the Five Axis of each point source are each point source in second coordinate system Coordinate；According to the 4-coordinate and Five Axis of each point source, first transition matrix is determined；

The determination unit, comprising:

First determines subelement, for determining the ball where the point source in the third image for each point source Body, obtains the pixel value and the 6th coordinate of each third pixel in the sphere, and the 6th of each third pixel sits It is designated as coordinate of each third pixel in first coordinate system；

Second determines that subelement passes through following formula for the pixel value and the 6th coordinate according to each third pixel One, determine the 4-coordinate of the point source:

Formula one:

7. device according to claim 6, which is characterized in that the determination unit, comprising:

Third determines subelement, for determining the ball where the point source in the 4th image for each point source Body, obtains the pixel value and the 9th coordinate of each 4th pixel in the sphere, and the 9th of each 4th pixel sits It is designated as coordinate of each 4th pixel in second coordinate system；

4th determines that subelement passes through following formula for the pixel value and the 9th coordinate according to each 4th pixel One (2) determine the Five Axis of the point source:

Formula one (2):

8. device according to claim 6, which is characterized in that the determination unit, comprising:

5th determines subelement, for 4-coordinate, Five Axis and first transition matrix according to each point source Transformed representation, determine the coordinate difference expression formula of each point source；

6th determines that subelement determines total seat of the n point source for the coordinate difference expression formula according to each point source Mark differential expression formula；

7th determines subelement, for determining described first turn for meeting preset condition according to total coordinate difference expression formula Matrix is changed, the preset condition is that the value of total coordinate difference expression formula is minimum value.

9. device according to claim 8, which is characterized in that the described 5th determines subelement, is also used to according to described every The 4-coordinate and the transformed representation of a point source determine the 7th coordinate of each point source, institute by following formula two The 7th coordinate for stating each point source is coordinate of each point source in second coordinate system:

Formula two: (x₇,y₇,z₇)=M × (x₄,y₄,z₄)

Described 5th determines subelement, is also used to pass through following public affairs according to the Five Axis and the 7th coordinate of each point source Formula three determines the coordinate difference expression formula of each point source:

Formula three: Δ x=(x₅,y₅,z₅)-(x₇,y₇,z₇)

10. device according to claim 6, which is characterized in that the multi-modal detection system is positron emission fault Imaging-computed tomography PET-CT detection system, the first image are the PET image of pet detector detection, described the Two images are the CT image of CT detector detection, described device further include:

Second obtains module, and for obtaining the second transition matrix according to first transition matrix, second transition matrix is used It is sat in being converted to coordinate of the second pixel of each of second image in second coordinate system described first Coordinate in mark system；

Second determining module is determined for the third coordinate according to second transition matrix and each second pixel 8th coordinate of each second pixel, the 8th coordinate of each second pixel are each second pixel Coordinate in first coordinate system；