CN101833784A - Method and device for synthesizing volume data - Google Patents
Method and device for synthesizing volume data Download PDFInfo
- Publication number
- CN101833784A CN101833784A CN200910119408A CN200910119408A CN101833784A CN 101833784 A CN101833784 A CN 101833784A CN 200910119408 A CN200910119408 A CN 200910119408A CN 200910119408 A CN200910119408 A CN 200910119408A CN 101833784 A CN101833784 A CN 101833784A
- Authority
- CN
- China
- Prior art keywords
- volume data
- resampled
- data
- pixel
- subunit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000002194 synthesizing effect Effects 0.000 title claims abstract description 31
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000012952 Resampling Methods 0.000 claims description 50
- 230000015572 biosynthetic process Effects 0.000 claims description 31
- 238000003786 synthesis reaction Methods 0.000 claims description 31
- 238000001308 synthesis method Methods 0.000 claims description 20
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 abstract 1
- 210000001015 abdomen Anatomy 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 4
- 210000001367 artery Anatomy 0.000 description 3
- 230000002792 vascular Effects 0.000 description 3
- 238000002583 angiography Methods 0.000 description 2
- 230000036770 blood supply Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000002595 magnetic resonance imaging Methods 0.000 description 2
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 230000007012 clinical effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 210000002345 respiratory system Anatomy 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a method for synthesizing volume data, which comprises the following steps of: acquiring a basic slice direction of a synthesized image; acquiring an outer bounding box of each input volume data based on the acquired basic slice direction of the synthesized image; re-sampling each input volume data in the outer bounding box of each input volume data; determining that an overlapping region between the re-sampled volume data has a translation parameter of maximum similarity; and synthesizing the re-sampled volume data into one volume data according to the determined translation parameter to acquire a synthesized image. The invention discloses a device for synthesizing the volume data at the same time. The method and the device can synthesize the volume data with random slice direction.
Description
Technical Field
The present invention relates to magnetic resonance imaging technology, and in particular, to a method and an apparatus for synthesizing volume data in a magnetic resonance imaging system.
Background
Due to the hardware limitation of the magnetic resonance system, in practical applications, when the magnetic resonance system is used for scanning imaging, the scanning image of the whole body cannot be acquired at one time, so that a plurality of volume data with overlapping regions are acquired for many times, and then the volume data are synthesized into one volume data in a certain way, so as to obtain a Virtual Large Field of View (Virtual Large Field of View) for diagnosing some systemic diseases, such as respiratory system and vascular system.
Currently, the synthesis technology of volume data with different slice directions is widely applied in clinic. One typical application is for examining the ability of arteries to supply blood to the legs, abdomen, and upper body. Firstly, respectively acquiring volume data corresponding to parts such as legs, abdomen, upper part of the body and the like, wherein overlapping regions exist among the parts; then, through a volume data synthesis technique, an angiography image covering the whole body of the patient for the complete vascular system examination is acquired, so that the doctor can perform the subsequent diagnosis of the vascular system diseases. The slice direction referred to herein means the direction of the normal vector of the slice plane.
Although the volume data synthesis technology has important clinical effects, the technology has great limitations in practical use.
For example, when the slice direction between the acquired plurality of volume data has a smaller tilt angle, a more ideal synthesis result can be obtained by using the existing volume data synthesis method, for example, a feature-based three-dimensional (3D) Magnetic Resonance (MR) angiography image synthesis method proposed in the prior art can be used. This method warps volume data based on an angle (an angle between an actual slice direction and a standard/specified direction), and after warping volume data each having an inclination angle to volume data having a parallel slice direction (i.e., the slice directions of the volume data coincide), determines each volume data arrangement such that an overlapping region between the volume data has the maximum similarity by using rigid body matching.
The arrangement referred to herein includes the positional relationship of each volume data and the manner of overlap between two adjacent volume data. Since the positional relationship between the respective volume data is known in advance, for example, if the blood supply capability of the artery to the leg, the abdomen, and the upper body is to be checked, it is assumed that three corresponding volume data of the leg, the abdomen, and the upper body are acquired respectively. It is apparent that there is an overlapping region between two volume data corresponding to the leg and the abdomen, and an overlapping region between two volume data corresponding to the abdomen and the upper part of the body. The essence of volume data synthesis is to find the overlapping manner between adjacent volume data after the three volume data are synthesized into one volume data in the order of the upper part of the body, the abdomen and the legs.
For example, the following steps are carried out: as shown in fig. 1, fig. 1 is a schematic diagram of a conventional volume data warping process. It can be seen that there is a certain inclination angle between the volume data 101 and 102, where the oblique lines in the volume data 101 and 102 shown in fig. 1 indicate slices, and the hatched areas at the corners between the volume data 101 and 102 indicate image-related information. The warped volume data 103 and 104 are obtained by warping the volume data 101 and 102 by shifting different values in the respective slices (shift in the slice plane direction, no interpolation), and the area of the warped volume data 103 and 104 having no data can be filled with 0.
The warping process described above can be decomposed into a rotation plus a translation for each slice in the vertical direction, i.e. the column direction of the slice. If the tilt angle between the slice directions of the volume data is relatively small, the warped volume data can be regarded as being similar to the original volume data, and accordingly, the resultant synthesized image is relatively real and reliable.
However, in practical applications, due to various reasons, such as the natural curvature of the body of the patient, a large tilt angle exists between the slice directions of the acquired volume data. In this case, the above-mentioned volume data synthesis method is no longer applicable because: as the tilt angle increases, more and more distortion is introduced into the warped volume data, which in turn causes the phenomenon of mismatch of the overlapping regions in the volume data synthesis based on the warped volume data. Fig. 2 is a synthesis result obtained by synthesizing volume data having a large inclination angle between three slice directions by using the above-described volume data synthesis method suitable for a small inclination angle between slice directions. As can be seen from fig. 2, at the volume data overlap regions 201 and 202, there are cases of severe blood vessel mismatch and intra-vascular intrinsic strike inconsistency, both caused by distorting the volume data in order to cope with the inclination angle.
It can be seen that the above-mentioned volume data synthesis method is not suitable for the case of large tilt angle between the volume data, and the prior art has not proposed a feasible solution to this problem.
Disclosure of Invention
In view of the above, it is a primary object of the present invention to provide a volume data synthesis method that can synthesize volume data having an arbitrary slice direction.
Another object of the present invention is to provide a volume data synthesis device capable of synthesizing volume data having an arbitrary slice direction.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
a method of volumetric data synthesis, the method comprising:
acquiring a basic slice direction of the synthesized image; acquiring an outer bounding box of each input volume data based on the basic slice direction of the synthesized image;
resampling each input volume data in the bounding box of each input volume data;
determining a translation parameter which enables an overlapping area between the resampled volume data to have the maximum similarity;
and synthesizing the volume data after the resampling into one volume data according to the determined translation parameters to obtain a synthesized image.
Wherein the acquiring of the base slice direction of the synthesized image comprises: calculating an average value of all input volume data in the slice direction, and taking the calculated average value as a basic slice direction of the synthesized image; alternatively, the designated direction is set as the basic slice direction of the synthesized image.
The resampling each input volume data in each input volume data's bounding box comprises: resampling each input volume data in the bounding box of each input volume data according to the same slice thickness and the same pixel interval; the same slice thickness and the same pixel interval mean that the slice thickness and the pixel interval adopted in each outer bounding box and between the outer bounding boxes are the same.
After resampling each input volume data in each input volume data's bounding box, further comprising: and filling the area without data in the outer surrounding box after resampling with data zero.
The determining the translation parameter that enables the overlap region between the resampled volume data to have the maximum similarity includes:
and according to the position relation between the input volume data which can be known in advance, searching for the translation parameter which enables the overlap area between the two resampled volume data to have the maximum similarity between every two resampled volume data which have the overlap area.
Preferably, the finding the translation parameter between the resampled volume data at each of the two mutual positions where there is an overlap area includes:
keeping one of the two resampled volume data fixed, moving the other volume data on the fixed volume data, and searching a maximum similarity metric value between the two volume data when the translation parameters of the moving volume data are different values; and determining the translation parameter corresponding to the searched maximum similarity metric value as the translation parameter to be searched.
Wherein the similarity metric is a normalized cross-correlation coefficient:
t is a translation parameter; vfA pixel average value of an overlapping region in the fixed volume data; vtA pixel average value which is an overlapping region in moving body data; p is the pixel position of the overlapping region in the fixed volume data; vf(p) is the pixel value at p point in the fixed volume data; vt(p + t) is the pixel value of a pixel point corresponding to the p point in the fixed volume data after the moving volume data is translated by t; o isvAll pixels whose pixel values are not 0 in both volume data in the overlapping region of the fixed volume data and the moving volume data.
Preferably, the searching for the maximum similarity metric value is an adaptive searching, including:
firstly, coarse searching is carried out according to a preset larger step length to obtain a temporary maximum similarity metric value; then, fine search with smaller step length is carried out near the area corresponding to the searched temporary maximum similarity metric value, and the maximum similarity metric value searched in the area is taken as the final required maximum similarity metric value.
The synthesizing of the resampled volume data into one volume data includes: determining the optimal overlapping mode between the resampled volume data according to the found translation parameters between every two resampled volume data with the overlapping area; and synthesizing the resampled volume data into a data volume according to the determined optimal overlapping mode.
The synthesizing of the resampled volume data into one volume data includes: setting the pixel value of each pixel point in the overlapped area of every two synthesized volume data as the pixel average value of the corresponding pixel points in the two volume data; or setting the pixel value of each pixel point in the overlapped region of every two synthesized volume data as the pixel value of the corresponding pixel point in any volume data of the two volume data.
A volume data synthesis apparatus, the apparatus comprising: the device comprises an acquisition unit, a resampling unit, a searching unit and a synthesizing unit; wherein,
the acquisition unit is used for acquiring the basic slice direction of the synthesized image and acquiring the bounding box of each input volume data based on the basic slice direction of the synthesized image;
the resampling unit is used for resampling each input volume data in the surrounding box of each input volume data;
the search unit is used for determining a translation parameter which enables the overlapped area between the resampled volume data to have the maximum similarity;
and the synthesis unit is used for synthesizing the volume data after resampling into one volume data according to the determined translation parameter to obtain a synthesized image.
Wherein the acquisition unit includes: a first acquisition subunit and a second acquisition subunit;
the first acquisition subunit is configured to calculate an average value of all input volume data in the slice direction, and use the calculated average value as a basic slice direction of the synthesized image; or, the received designated direction is taken as the basic slice direction of the synthesized image;
the second acquiring subunit is configured to acquire an outer bounding box of each input volume data based on the basic slice direction of the synthesized image.
The resampling unit includes: a resampling subunit and a padding subunit;
the resampling subunit is configured to resample each input volume data in the bounding box of each input volume data according to the same slice thickness and the same pixel interval; the same slice thickness and the same pixel interval mean that the slice thickness and the pixel interval adopted in each outer enclosure box and between the outer enclosure boxes are the same;
and the filling subunit is used for filling the area without data in the outer enclosure box after resampling by using zeros.
The search unit includes: a first search subunit and a second search subunit;
the first searching subunit is configured to, according to a position relationship between input volume data that is known in advance, fix one of two resampled volume data having an overlap region between them, move the other volume data on the fixed volume data, and perform a coarse search according to a preset large step size to obtain a temporary maximum similarity metric value between the two volume data when a translation parameter of the moving volume data is a different value;
the second searching subunit is configured to perform fine searching with a smaller step size near an area corresponding to the searched temporary maximum similarity metric value, and determine a translation parameter corresponding to the maximum similarity metric value searched in the area as a required translation parameter.
The synthesis unit includes: determining a subunit and synthesizing a subunit;
the determining subunit is configured to determine an optimal overlap manner between the resampled volume data according to the determined translation parameter between each two resampled volume data having an overlap region therebetween;
and the synthesis subunit is used for synthesizing the resampled volume data into a data volume according to the determined optimal overlapping mode to obtain a synthesized image.
Therefore, by adopting the technical scheme of the invention, the volume data with the same slice direction, the same slice thickness and the same pixel interval can be obtained by resampling each input volume data in the surrounding box of each input volume data, so that the limitation that the input volume data has smaller inclination angle is avoided, and the application range of the volume data synthesis method is expanded; in addition, after the resampling is used for replacing the distortion in the prior art, the distortion introduced by the distorted volume data is avoided, and a more real and reliable synthesized image can be obtained.
Drawings
The foregoing and other features and advantages of the invention will become more apparent to those skilled in the art to which the invention relates upon consideration of the following detailed description of a preferred embodiment of the invention with reference to the accompanying drawings, in which:
FIG. 1 is a schematic diagram of a conventional volume data warping process;
FIG. 2 is a schematic diagram of a synthetic result obtained by synthesizing volume data with a large inclination angle between three slice directions by using a conventional volume data synthesis method suitable for a small inclination angle between slice directions;
FIG. 3 is a flow chart of an embodiment of a volume data synthesis method of the present invention;
FIG. 4 is a schematic diagram of a resampling process in an embodiment of the invention;
FIG. 5 is a diagram illustrating resampling results according to an embodiment of the invention;
fig. 6 is a schematic structural diagram of a volume data synthesis apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In order to solve the problems in the prior art, the embodiment of the present invention provides a new volume data synthesis method: acquiring a basic slice direction of the synthesized image, and acquiring an outer bounding box of each input volume data based on the basic slice direction of the synthesized image; secondly, resampling each input volume data in an outer bounding box of each input volume data, and determining a translation parameter which enables an overlapping area between the resampled volume data to have the maximum similarity; and finally, synthesizing the volume data after resampling into one volume data according to the determined translation parameters to obtain a synthesized image. By adopting the volume data synthesis mode, volume data with the same slice direction, the same slice thickness and the same pixel interval are obtained by resampling the volume data, so that the limitation that the input volume data has smaller inclination angle is avoided.
The volume data synthesis method according to the present invention will be described in further detail below with reference to specific examples.
FIG. 3 is a flow chart of an embodiment of a volume data synthesis method according to the invention. As shown in fig. 3, the method comprises the steps of:
step 301: the basic slice direction of the synthesized image is acquired.
In the present embodiment, two ways of acquiring the basic slice direction of the synthesized image are provided, including:
1) the average of the slice directions of all the input volume data is calculated.
In practical applications, The slicing direction of each input volume data is known in advance, and is described in a Digital image and Communications standard (DICOM) file obtained during scanning. Therefore, in this step, the average value of the slice directions of the input volume data is directly calculated, and the calculated average value is used as the basic slice direction of the synthesized image.
2) The designated direction is taken as the basic slice direction of the synthesized image.
As necessary, a certain direction is designated as a basic slice direction of the synthesized image.
Step 302: based on the acquired basic slice direction of the synthesized image, an outer bounding box of each input volume data is acquired.
The outer bounding box referred to herein is a concept well known in the art and refers to a cuboid that circumscribes volumetric data. It is required in the present invention that one of the faces of the cuboid is parallel to the base slice plane of the synthesized image. As shown in fig. 4, fig. 4 is a schematic diagram of a resampling process in the embodiment of the present invention. Where 403 is input volume data and 401 is the determined basic slice direction of the synthesized image; 402 is an bounding box of input volume data 403 acquired in the basic slice direction of the synthesized image shown in 401.
Step 303: each input volume data is resampled in the bounding box of each input volume data.
How to perform resampling is the prior art, and is not described herein again. However, in this embodiment, it is necessary to resample each input volume data according to the same slice thickness and the same pixel interval (the slice thickness and the pixel interval adopted in each outer bounding box and between each outer bounding box are the same), so as to ensure that each volume data obtained after resampling has the same slice direction (the basic slice direction is the same), the same slice thickness, and the same pixel interval. The specific resampling according to how large slice thickness and how large pixel interval can be set according to actual needs.
In addition, after resampling each input volume data, the area of the outer bounding box corresponding to the volume data, that is, the area indicated by 405 in fig. 4, needs to be filled with 0. The region surrounded by 405 is the resampled volume data. It should be noted that the step of filling the no-data area in the outer bounding box with 0 may also be performed in step 302, that is, the no-data area in the outer bounding box is filled with 0 before resampling, as shown in 404 in fig. 4.
Fig. 5 is a diagram illustrating a resampling result in an embodiment of the invention. As shown in fig. 5, where 501 and 502 are two input volume data, it can be seen that there is a large tilt angle between the slice directions of the two volume data. After resampling the volume data 501 and 502, the resampled volume data shown at 503 and 504, respectively, can be obtained according to the present invention. It can be seen that the method of the present invention avoids warping the volume data and can obtain volume data with the same slice direction, the same slice thickness, and the same pixel spacing.
Step 304: translation parameters are determined such that the overlap region between the resampled volume data has the greatest similarity.
The specific implementation of the step is as follows: and according to the position relation between the input volume data which can be known in advance, searching for the translation parameter which enables the overlap area between the two resampled volume data to have the maximum similarity between every two resampled volume data which have the overlap area.
The clinical application of examining the blood supply capacity of arteries to the legs, abdomen and upper body is still exemplified. Assume that three volume data corresponding to the leg, abdomen, and upper body are acquired. The purpose of this step is to determine the overlapping manner between the adjacent volume data after the three volume data are combined into one volume data in the order of the upper part of the body, the abdomen, and the legs. The overlap mode may be determined by finding a translation parameter that gives the maximum similarity to the overlap region between the resampled volume data.
The specific implementation mode is as follows: for every two resampled volume data with an overlapping area between the two volume data, keeping one of the volume data fixed, moving the other volume data on the fixed volume data, and searching the maximum similarity metric value between the two volume data when the translation parameters of the moving body data are different values; and determining the translation parameter corresponding to the searched maximum similarity metric value as the translation parameter to be searched.
The similarity metric used in this embodiment is a normalized cross-correlation coefficient:
t is a translation parameter;
Vfa pixel average value of an overlapping region in the fixed volume data;
Vta pixel average value which is an overlapping region in moving body data;
p is the pixel position of the overlapping region in the fixed volume data;
Vf(p) is the pixel value at p point in the fixed volume data;
Vt(p + t) is the pixel value of a pixel point corresponding to the p point in the fixed volume data after the moving volume data is translated by t;
Ovall pixels of which pixel values are not 0 in both volume data in an overlapping region of the fixed volume data and the moving volume data;
wherein,
n is the number of non-0 pixels in the overlap region.
In the present embodiment, since it is known in advance that the two volumes of data are substantially overlapped, for example, the upper portion of the moving body data and the lower portion of the fixed body data are overlapped, the search can be performed from the middle position, thereby saving the search time. However, even so, it is known to those skilled in the art that searching in the above manner still takes a long time. Therefore, in order to save time, an adaptive search mode can be adopted, namely: firstly, coarse searching is carried out according to a preset larger step length to obtain a temporary maximum similarity metric value (which is distinguished from a subsequent maximum similarity metric value), namely a most possible matching area is found; then, a fine search with a smaller step size is performed near the found most likely matching region, and the searched maximum similarity metric value is taken as the final maximum similarity metric value.
The above-mentioned calculation mode of the normalized cross correlation coefficient and the adaptive search mode are both the prior art, and are not described again.
Step 305: and synthesizing the volume data after the resampling into one volume data based on the determined translation parameters to obtain a synthesized image.
In the step, according to the found translation parameter between every two resampled volume data with the overlapping area, determining the optimal overlapping mode between the resampled volume data; and synthesizing the resampled volume data into a data volume according to the determined optimal overlapping mode.
When synthesizing, the pixel value of each pixel point in the overlapping region of two adjacent volume data can be set in the following two ways, namely: setting the pixel average value of corresponding pixel points in the two-volume data; or, the pixel values of the corresponding pixel points in any one of the two volume data are set. In practical applications, the former arrangement is usually adopted to obtain a smoother synthesized image.
Based on the above method, fig. 6 is a schematic diagram of a composition structure of an embodiment of the volume data synthesis apparatus according to the present invention. As shown in fig. 6, the apparatus includes: an acquisition unit 601, a resampling unit 602, a search unit 603, and a synthesis unit 604; wherein:
an acquisition unit 601 configured to acquire a basic slice direction of the synthesized image, and acquire an bounding box of each input volume data based on the basic slice direction of the synthesized image;
a resampling unit 602 is configured to resample each input volume data in an outer bounding box of each input volume data;
the search unit 603 is configured to determine a translation parameter that enables an overlap area between the resampled volume data to have a maximum similarity;
the synthesizing unit 604 is configured to synthesize the resampled volume data into a volume data according to the determined translation parameter, and obtain a synthesized image.
Wherein, the obtaining unit 601 includes: first acquisition subunit 6011 and second acquisition subunit 6012:
the first acquisition subunit 6011 is configured to calculate an average value of the slice directions of all input volume data, and use the calculated average value as a basic slice direction of the synthesized image; or, the received appointed direction is taken as the basic slice direction of the synthesized image; the second acquiring subunit 6012 is configured to acquire an outer bounding box of each input volume data based on the basic slice direction of the synthesized image.
The resampling unit 602 includes: resampling subunit 6021 and padding subunit 6022:
a resampling sub-unit 6021 for resampling each input volume data in the bounding box of each input volume data with the same slice thickness and the same pixel interval; the same slice thickness and the same pixel interval mentioned herein mean that the slice thickness and the pixel interval adopted in each outer bounding box and between the outer bounding boxes are the same; the padding subunit 6022 is used to pad the area without data in the resampled outer bounding box with zeros.
The search unit 603 includes: first search subunit 6031 and second search subunit 6032:
the first search subunit 6031 is configured to, according to a position relationship between input volume data that is known in advance, fix one of two resampled volume data having an overlap region between them, move the other volume data on the fixed volume data, and perform a coarse search according to a preset large step length to obtain a temporary maximum similarity metric value between the two volume data when a translation parameter of the moving volume data is a different value; the second searching subunit 6032 is configured to perform fine searching with a smaller step size near the area corresponding to the searched temporary maximum similarity metric value, and determine the translation parameter corresponding to the maximum similarity metric value searched in the area as the required translation parameter.
Wherein the similarity measureThe values are normalized cross-correlation coefficients:t is a translation parameter; vfA pixel average value of an overlapping region in the fixed volume data; vtA pixel average value which is an overlapping region in moving body data; p is the pixel position of the overlapping region in the fixed volume data; vf(p) is the pixel value at p point in the fixed volume data; vt(p + t) is the pixel value of a pixel point corresponding to the p point in the fixed volume data after the moving volume data is translated by t; o isvAll pixels whose pixel values are not 0 in both volume data in the overlapping region of the fixed volume data and the moving volume data.
Further, the synthesizing unit 604 includes: determination subunit 6041 and synthesis subunit 6042:
a determining subunit 6041 is configured to determine an optimal overlap manner between the resampled volume data according to the determined translation parameter between each two pieces of volume data having an overlap region therebetween; the synthesizing sub-unit 6042 is configured to synthesize the resampled volume data into one data volume in the determined optimal superimposition manner, and obtain a synthesized image.
For a specific work flow of the apparatus embodiment shown in fig. 6, please refer to the related description in the method embodiment shown in fig. 3, which is not repeated herein.
In short, by adopting the technical scheme of the invention, the volume data with the same slice direction, the same slice thickness and the same pixel interval is obtained by resampling the input volume data, so that the application range of the volume data synthesis method is greatly expanded, the limitation that the input volume data must have the same slice direction or only have a smaller inclination angle is avoided, and the volume data synthesis method can process the synthesis of the volume data with any slice direction, different resolutions and different slice thicknesses; in addition, after the resampling is used for replacing the distortion in the existing volume data synthesis, the distortion caused by the distorted volume data is avoided, and a more real and reliable synthesized image can be obtained; in addition, in order to better search, the invention adopts a self-adaptive search mode, thereby accelerating the speed of correctly finding the needed translation parameters.
It should be noted that the above embodiments are only for illustration and are not used to limit the technical solutions of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (15)
1. A method of volumetric data synthesis, the method comprising:
acquiring a basic slice direction of the synthesized image; acquiring an outer bounding box of each input volume data based on the basic slice direction of the synthesized image;
resampling each input volume data in the bounding box of each input volume data;
determining a translation parameter which enables an overlapping area between the resampled volume data to have the maximum similarity;
and synthesizing the volume data after the resampling into one volume data according to the determined translation parameters to obtain a synthesized image.
2. The volume data synthesis method according to claim 1, wherein the acquiring of the basic slice direction of the synthesized image includes:
calculating an average value of all input volume data in the slice direction, and taking the calculated average value as a basic slice direction of the synthesized image; alternatively, the designated direction is set as the basic slice direction of the synthesized image.
3. The method of synthesizing volumetric data according to claim 1, wherein the resampling each input volumetric data in the bounding box of each input volumetric data comprises:
resampling each input volume data in the bounding box of each input volume data according to the same slice thickness and the same pixel interval;
the same slice thickness and the same pixel interval mean that the slice thickness and the pixel interval adopted in each outer bounding box and between the outer bounding boxes are the same.
4. The volume data synthesizing method according to claim 1 or 3, wherein after resampling each input volume data in the bounding box of each input volume data, further comprising:
and filling the area without data in the outer surrounding box after resampling with data zero.
5. The volume data synthesis method according to claim 1, wherein the determining of the translation parameter such that the overlap region between the resampled volume data has the maximum similarity includes:
and according to the position relation between the input volume data which can be known in advance, searching for the translation parameter which enables the overlap area between the two resampled volume data to have the maximum similarity between every two resampled volume data which have the overlap area.
6. The volume data synthesis method according to claim 5, wherein the finding of the translation parameter between the resampled volume data at each of the two mutual positions where the overlap area exists enables the overlap area between the two resampled volume data to have the maximum similarity comprises:
keeping one of the two resampled volume data fixed, moving the other volume data on the fixed volume data, and searching a maximum similarity metric value between the two volume data when the translation parameters of the moving volume data are different values;
and determining the translation parameter corresponding to the searched maximum similarity metric value as the translation parameter to be searched.
7. The volumetric data synthesis method of claim 6, wherein the similarity metric is a normalized cross-correlation coefficient:
wherein,
t is a translation parameter; vfA pixel average value of an overlapping region in the fixed volume data; vtA pixel average value which is an overlapping region in moving body data; p is the pixel position of the overlapping region in the fixed volume data; vf(p) is the pixel value at p point in the fixed volume data; vt(p + t) is the pixel value of a pixel point corresponding to the p point in the fixed volume data after the moving volume data is translated by t; o isvAll pixels whose pixel values are not 0 in both volume data in the overlapping region of the fixed volume data and the moving volume data.
8. The volume data synthesis method according to claim 6 or 7, wherein the searching for the maximum similarity metric value is performed in an adaptive search mode, including:
firstly, coarse searching is carried out according to a preset larger step length to obtain a temporary maximum similarity metric value;
and performing fine search with a smaller step length near the area corresponding to the searched temporary maximum similarity metric value, and taking the maximum similarity metric value searched in the area as the finally required maximum similarity metric value.
9. The method according to claim 5, wherein the synthesizing of the resampled volume data into one volume data includes:
determining the optimal overlapping mode between the resampled volume data according to the found translation parameters between every two resampled volume data with the overlapping area;
and synthesizing the resampled volume data into a data volume according to the determined optimal overlapping mode.
10. The method according to claim 9, wherein the synthesizing of the resampled volume data into one volume data includes:
setting the pixel value of each pixel point in the overlapped area of every two synthesized volume data as the pixel average value of the corresponding pixel points in the two volume data; or,
and setting the pixel value of each pixel point in the overlapped region of every two synthesized volume data as the pixel value of the corresponding pixel point in any volume data of the two volume data.
11. A volume data synthesis apparatus, the apparatus comprising: an acquisition unit (601), a resampling unit 602), a search unit 603), and a synthesis unit (604),
the acquisition unit (601) is used for acquiring the basic slice direction of the synthesized image and acquiring the outer bounding box of each input volume data based on the basic slice direction of the synthesized image;
the resampling unit 602 is configured to resample each input volume data in the bounding box of each input volume data;
the search unit (603) is configured to determine a translation parameter such that an overlapping region between the resampled volume data has a maximum similarity;
and the synthesis unit (604) is used for synthesizing the resampled volume data into one volume data according to the determined translation parameter to obtain a synthesized image.
12. The volume data synthesis apparatus according to claim 11, wherein the acquisition unit (601) includes: a first acquisition subunit (6011) and a second acquisition subunit (6012);
the first acquisition subunit (6011) is configured to calculate an average value of slice directions of all input volume data, and use the calculated average value as a basic slice direction of the synthesized image; or, the received designated direction is taken as the basic slice direction of the synthesized image;
the second acquisition subunit (6012) is configured to acquire an outer bounding box of each input volume data based on a basic slice direction of the synthesized image.
13. The volume data synthesis apparatus according to claim 11, wherein the resampling unit (602) comprises: a resampling subunit (6021) and a padding subunit (6022);
the resampling subunit (6021) is configured to resample each input volume data in the bounding box of each input volume data at the same slice thickness and the same pixel interval; the same slice thickness and the same pixel interval mean that the slice thickness and the pixel interval adopted in each outer enclosure box and between the outer enclosure boxes are the same;
the padding subunit (6022) is configured to pad the area without data in the outer bounding box after resampling with zeros.
14. The volume data synthesis apparatus according to claim 11, wherein the search unit (603) includes: a first search subunit (6031) and a second search subunit (6032);
the first searching subunit (6031) is configured to, according to a known positional relationship between input volume data, fix one of two resampled volume data in which an overlap region exists between the two volume data, move the other volume data on the fixed volume data, and perform a coarse search according to a preset larger step size to obtain a temporary maximum similarity metric value between the two volume data when a translation parameter of the moving volume data is a different value;
the second searching subunit (6032) is configured to perform a fine search with a smaller step size near an area corresponding to the searched temporary maximum similarity metric value, and determine a translation parameter corresponding to the maximum similarity metric value searched in the area as a required translation parameter.
15. The volume data synthesis apparatus according to claim 14, wherein the synthesis unit (604) includes: a determination subunit (6041) and a synthesis subunit (6042);
the determining subunit (6041) is configured to determine an optimal overlapping manner between the resampled volume data according to the determined translation parameter between every two resampled volume data having an overlapping area;
and the synthesis subunit (6042) is used for synthesizing the resampled volume data into a data volume according to the determined optimal overlapping mode to obtain a synthesized image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910119408A CN101833784A (en) | 2009-03-12 | 2009-03-12 | Method and device for synthesizing volume data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910119408A CN101833784A (en) | 2009-03-12 | 2009-03-12 | Method and device for synthesizing volume data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101833784A true CN101833784A (en) | 2010-09-15 |
Family
ID=42717845
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910119408A Pending CN101833784A (en) | 2009-03-12 | 2009-03-12 | Method and device for synthesizing volume data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101833784A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112364421A (en) * | 2020-11-12 | 2021-02-12 | 天河超级计算淮海分中心 | Rendering method and device of building information model, computer equipment and storage medium |
-
2009
- 2009-03-12 CN CN200910119408A patent/CN101833784A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| 陈秦玉: "人体三维重建的实践和技术研究", 《中国优秀博硕士学位论文全文数据库（博士）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112364421A (en) * | 2020-11-12 | 2021-02-12 | 天河超级计算淮海分中心 | Rendering method and device of building information model, computer equipment and storage medium |
| CN112364421B (en) * | 2020-11-12 | 2023-10-27 | 天河超级计算淮海分中心 | Rendering method and device of building information model, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Porter et al. | Three-dimensional registration and fusion of ultrasound and MRI using major vessels as fiducial markers | |
| US10242450B2 (en) | Coupled segmentation in 3D conventional ultrasound and contrast-enhanced ultrasound images | |
| US6500123B1 (en) | Methods and systems for aligning views of image data | |
| US20190066309A1 (en) | Single- and multi-modality alignment of medical images in the presence of non-rigid deformations using phase correlation | |
| US8537159B2 (en) | Visualization of voxel data | |
| US7945117B2 (en) | Methods and systems for registration of images | |
| US5937083A (en) | Image registration using closest corresponding voxels with an iterative registration process | |
| US6159152A (en) | Medical diagnostic ultrasound system and method for multiple image registration | |
| US4989142A (en) | Three-dimensional images obtained from tomographic slices with gantry tilt | |
| US20120121150A1 (en) | Ultrasonic image processing apparatus | |
| JP5815722B2 (en) | Magnetic resonance imaging apparatus and magnetic resonance imaging method | |
| WO2013018575A1 (en) | Image diagnostic device and image correction method | |
| CA2354616A1 (en) | Methods and systems for registering image data | |
| WO1997041532A9 (en) | Iterative image registration process using closest corresponding voxels | |
| US20120121149A1 (en) | Ultrasonic image processing apparatus | |
| Poon et al. | Three-dimensional extended field-of-view ultrasound | |
| JP2009050544A (en) | Magnetic resonance imaging device and navigator data analytic method | |
| Dai et al. | Real-time visualized freehand 3D ultrasound reconstruction based on GPU | |
| JP5194138B2 (en) | Image diagnosis support apparatus, operation method thereof, and image diagnosis support program | |
| US20080212863A1 (en) | Image processing apparatus, image processing method, and magnetic resonance imaging apparatus | |
| CN111009032A (en) | Blood vessel three-dimensional reconstruction method based on improved epipolar line constraint matching | |
| JP2008200097A (en) | Ultrasonograph and data processing method | |
| JPH09198490A (en) | Three-dimensional discrete data projector | |
| JPH021082A (en) | Apparatus and method for generating definition of three-dimensional surface | |
| US20130072782A1 (en) | System and method for automatic magnetic resonance volume composition and normalization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20100915 |