WO2017020580A1 - 一种医学组织的切片图像三维重建的方法及设备 - Google Patents

一种医学组织的切片图像三维重建的方法及设备 Download PDF

Info

Publication number
WO2017020580A1
WO2017020580A1 PCT/CN2016/074260 CN2016074260W WO2017020580A1 WO 2017020580 A1 WO2017020580 A1 WO 2017020580A1 CN 2016074260 W CN2016074260 W CN 2016074260W WO 2017020580 A1 WO2017020580 A1 WO 2017020580A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
slice
slice image
display
medical tissue
Prior art date
Application number
PCT/CN2016/074260
Other languages
English (en)
French (fr)
Inventor
王佳
刘静静
田广野
陈永健
Original Assignee
青岛海信医疗设备股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 青岛海信医疗设备股份有限公司 filed Critical 青岛海信医疗设备股份有限公司
Publication of WO2017020580A1 publication Critical patent/WO2017020580A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects

Definitions

  • the invention relates to the technical field of display of medical images, in particular to a method and a device for three-dimensional reconstruction of slice images of medical tissues.
  • 3D reconstruction refers to the establishment of a mathematical model suitable for computer representation and processing of three-dimensional objects, that is, the basis for processing, manipulating and analyzing the properties of three-dimensional objects in a computer environment, and also the key technology for establishing virtual reality in the objective world. .
  • the inferior vena cava and non-homogeneous tumors are two very important tissues, and doctors knowing the details of these two tissues is extremely important for doctors to plan their surgery.
  • the doctor needs to know in detail the erosion of the inferior vena cava by the diseased tissue; for example, for non-homogeneous tumors, doctors need to obtain detailed tumor volume, as well as tumors and surrounding tissues such as liver, blood vessels, etc. relationship.
  • Traditional three-dimensional reconstruction methods include: mathematical morphology methods, region generation methods, level set methods, and Hessian Matrix methods.
  • the mathematical morphology method is used to analyze the intrinsic geometry of the image; the region generation method can be extended by a single feature point to divide the entire region with similar characteristics; the level set method can realize the active contour model; the black matrix method Can be used to achieve a prominent target area. These methods are specific to the question Sometimes the problem is not used alone.
  • the inferior vena cava due to human metabolism and CT imaging principles, the inferior vena cava can only determine its spatial position in continuous CT images, and according to different periods of CT images, the inferior vena cava is on the image.
  • the brightness of the display is different, and even the contrast agent is not in the scanning position at the scanning time, so that the blood vessel cannot be observed, which is caused by the faster flow velocity of the inferior vena cava and the larger diameter of the vena cava of the inferior vena cava.
  • the situation is more common in CT images.
  • the shape of black and white plaques will appear in the sliced images.
  • the traditional 3D reconstruction method can not automatically distinguish the edges of non-homogeneous tumors by using existing algorithms. Tumors with blood vessels, cysts, etc. inside the tumor are also common, so it is difficult to distinguish them from other tissues quickly and accurately.
  • Embodiments of the present invention provide a method and a device for three-dimensional reconstruction of a slice image of a medical tissue, which are difficult to quickly and accurately distinguish it from other tissues when performing three-dimensional reconstruction on intermittently displayed tissue and plaque tissue. The problem that comes.
  • an embodiment of the present invention provides a method for three-dimensional reconstruction of a sliced image of a medical tissue, including:
  • the isosurfaces of the generated display patches are spliced to obtain a three-dimensional image of the medical tissue.
  • drawing the medical tissue edge in the first slice image includes:
  • each of the two adjacent points of the at least three points are respectively connected by an arc to form a closed curve to form a medical tissue edge in the first slice image.
  • drawing the medical tissue edge in the first slice image includes:
  • a circle is drawn with a distance between the first point and the second point to form a medical tissue edge in the first slice image.
  • the first slice image and the second slice image that are drawn are extracted for the spatially adjacent first slice image and the second slice image.
  • the isosurface, the generated display patch includes:
  • the rendered isoforms of the first slice image and the second slice image are extracted by the Marching Cubes algorithm to generate a display patch.
  • the first slice image that is drawn is extracted for the spatially adjacent first slice image and the second slice image And generating the display patch by the isosurface of the second slice image further includes:
  • the image operation instruction is executed on the display patch according to the image operation instruction, and the image operation instruction includes selection, scaling, translation, and rotation of the medical image.
  • the method before the splicing the isosurface of the generated display patch to obtain the three-dimensional image of the medical tissue, the method further The method includes: streamlining and smoothing the display patch.
  • the medical tissue is an inferior vena cava or a non-homogeneous tumor.
  • an embodiment of the present invention provides an image processing apparatus, including:
  • a drawing unit for drawing an edge of the medical tissue in the edge-sharp slice image of the set of slice images of the medical tissue
  • a filling unit for filling each of the drawn slice images with an edge enclosed by the medical tissue
  • a generating unit configured to extract an isosurface of two slice images adjacent to each two spaces in the filled slice image, respectively, to generate a display patch
  • a display unit configured to splicing the isosurfaces of the generated display patches to obtain a three-dimensional image of the medical tissue.
  • the drawing unit includes:
  • a first receiving module configured to receive at least three points input by a user
  • the first processing module respectively connects each two adjacent points of the at least three points with an arc to form a closed curve to form a medical tissue edge in the first slice image.
  • the drawing unit includes:
  • the drawing unit includes:
  • a second receiving module configured to receive first and second points input by the user
  • a second processing module configured to use a first point of the user input as a center, and draw a circle with a distance between the first point and the second point to form a medical organization in the first slice image edge.
  • the generating unit is specifically configured to extract, by using a Marching Cubes algorithm, an isosurface of the first slice image and the second slice image that are drawn, Generate a display patch.
  • the generating unit further includes:
  • a third receiving module configured to receive an image operation instruction input by the user on the display patch
  • an execution module configured to execute the image operation instruction on the display patch according to the image operation instruction, where the image operation instruction includes selecting, scaling, panning, and rotating the medical image.
  • the device before the display unit, the device further includes:
  • a processing unit configured to streamline and smooth the display patch.
  • a method for three-dimensional reconstruction of a slice image of a medical tissue by drawing a side of a medical tissue in a slice image of a sharp slice in a set of slice images of a medical tissue Filling the image of the edge of the medical tissue with each slice image, by drawing the edge of the medical tissue, the medical tissue to be reconstructed can be identified from the slice image for subsequent extraction, etc.
  • the value plane generates a display patch and improves the accuracy of the three-dimensional reconstruction, extracts the isosurfaces of the two slice images adjacent to each of the two spatially drawn slice images, respectively generates a display patch, and the generated display patch
  • the isosurface is spliced to obtain a three-dimensional image of the medical tissue, so that by constructing a two-dimensional slice image, the edge is drawn, and the two-dimensional slice image is connected by the three-dimensional reconstruction method by extracting the isosurface of the two-dimensional slice image, which can be accurately Quickly presenting the three-dimensional visualization result of the actual structural information and spatial information of the sliced image in the human body, and solving the traditional three-dimensional reconstruction method cannot quickly and accurately display the human tissue from the sliced image showing intermittent display or plaque Bottlenecks for effective processing, for doctors to learn more about the structural information of such organizations and Weave relations with neighboring organizations provide important help.
  • FIG. 1 is a hardware structural diagram of an image processing device according to an embodiment of the present invention.
  • FIG. 2 is a schematic flowchart 1 of a method for three-dimensional reconstruction of a sliced image of a medical tissue according to Embodiment 1 of the present invention
  • FIG. 3 is a schematic flow chart 2 of a method for three-dimensional reconstruction of a sliced image of a medical tissue according to Embodiment 1 of the present invention
  • FIG. 4 is a schematic flowchart 3 of a method for three-dimensional reconstruction of a sliced image of a medical tissue according to a first embodiment of the present invention
  • FIG. 5 is a schematic flow chart of a method for three-dimensional reconstruction of a lower vena cava by using a method for three-dimensional reconstruction of a sliced image of a medical tissue according to a second embodiment of the present invention
  • 6a-6e are schematic diagrams of a three-dimensional reconstruction method of a sliced image of a medical tissue for three-dimensional reconstruction of a lower vena cava according to a second embodiment of the present invention
  • FIG. 7 is a schematic flow chart of a three-dimensional reconstruction method of a sliced image of a medical tissue for three-dimensional reconstruction of a non-homogeneous tumor according to a third embodiment of the present invention
  • FIGS. 8a-8b are schematic diagrams of a three-dimensional reconstruction method of a sliced image of a medical tissue for three-dimensional reconstruction of a non-homogeneous tumor according to a third embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram 1 of an image processing apparatus according to Embodiment 4 of the present invention.
  • FIG. 10 is a schematic structural diagram 2 of an image processing apparatus according to Embodiment 4 of the present invention.
  • FIG. 11 is a schematic structural diagram 3 of an image processing apparatus according to Embodiment 4 of the present invention.
  • the words “first”, “second” and the like are used to distinguish the same or similar items whose functions and functions are substantially the same, in the field.
  • the skilled person can understand the words “first”, “second” and the like and limit the number and execution order.
  • FIG. 1 shows a hardware diagram of an image processing apparatus including one or more (only one shown in the figure) processing.
  • the device 101 the memory 102, the user interface 103, the communication bus 104, and the display screen 105.
  • FIG. 1 is merely illustrative and does not limit the structure of the image processing apparatus 10.
  • the image processing apparatus 10 may further include more or less components than those shown in FIG. 1, or have a different configuration than that shown in FIG.
  • the communication bus 104 is used for communication between the various components in the image processing apparatus 10.
  • the user interface 103 is used to plug in external devices, such as a mouse and a keyboard, to receive information input by the user.
  • the display screen 105 is used to display the slice image, the medical tissue after the three-dimensional reconstruction of the slice image, and the image and image changes involved in the intermediate processing.
  • the memory 102 can be used to store software programs as well as modules, databases, as in the present invention The determination in the embodiment requires drawing a method of drawing the medical organization and a program instruction/module corresponding to the device.
  • Memory 102 can include high speed random access memory and can also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid state memory.
  • memory 102 can further include memory remotely located relative to processor 101, which can be coupled to image processing device 10 over a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
  • the processor 101 executes various functional applications and data processing by running software programs and modules stored in the memory 102, for example, by calling the application in the memory 102 that determines that the drawing method of the medical organization needs to be drawn, Achieving a fast and accurate determination requires drawing a test case for a medical organization.
  • An embodiment of the present invention provides a method for three-dimensional reconstruction of a slice image of a medical tissue. As shown in FIG. 2, the method for three-dimensional reconstruction includes:
  • the slice image refers to the use of a computer imaging device to obtain a two-dimensional sequence digital tomographic image in the human body and its internal organs, and provides a diagnosis and treatment to the doctor through a clear human body structure and a detailed two-dimensional sequence digital tomographic image of the case information, for example,
  • the slice image may include a CT (Computed Tomography) image or an MRI (Magnetic Resonance Imaging) image, which is not limited in the present invention.
  • the slice image selected by the embodiment of the present invention is a CT image.
  • medical organization refers to an organic cell population having a certain function and structure.
  • the medical tissue in the embodiment of the invention is an inferior vena cava or a non-homogeneous tumor.
  • the edge of medical organization usually means that the two sides of the medical organization belong to two regions, and the characteristics inside each region are relatively uniform and uniform, and there are certain differences in the characteristics between the two regions.
  • the slice image with clear edges in the slice image refers to a medical tissue edge that can be segmented in a group of slice images and other medical organizations. Separate slice images.
  • the processed object is only for the slice image with sharp edges.
  • the slice image with clear edges may be selected from a group of slice images by various methods. For example, the user may select according to his own judgment or may select according to a certain algorithm, and the process does not affect the realization of the object of the present invention. This embodiment of the present invention does not limit this.
  • the edge of the medical tissue is drawn in the slice image displayed on the display, with the specific lines identifying the edges of the medical tissue to distinguish it from other tissues.
  • the edge of the medical tissue can be drawn in various ways, and no specific limitation is imposed herein.
  • the processing principle and the process for each slice image are the same, here, only the first slice image is taken as an example, and the first slice image refers to any one of the selected slice images with sharp edges. Does not have a special meaning.
  • step S101 specifically includes a1-a2:
  • A1 receiving at least three points input by the user
  • each of two adjacent points of at least three points are respectively connected by an arc to form a closed curve to form a medical tissue edge in the first slice image.
  • the first slice image may be displayed on a display screen, and at least three point sampling points input by the user are obtained through a user interface, and the processor receives at least three user input according to a three-spline interpolation algorithm (Spline).
  • the curves between the points are connected by a three-spline interpolation to form a closed curve and displayed in the display, that is, the drawing of the medical tissue edge in the first slice image is completed.
  • the first slice image after the drawing of the tissue edge is completed may also be saved. For example, you can save after taking a double-click operation entered by the user.
  • the closed curve is adjusted by the user interface to obtain the user dragging at least three sampling points.
  • step S101 specifically includes a3-a4:
  • A3 receiving the first point and the second point input by the user
  • A4 taking a first point input by the user as a center, and drawing a circle with a distance between the first point and the second point as a radius to form a medical tissue edge in the first slice image.
  • the first slice image may be displayed on the display screen, and the first point and the second point input by the user are sequentially obtained through the user interface, and the processor obtains the first point as the center, the first point and the second point.
  • the distance between the points is a radius circled to form a medical tissue edge in the first slice image and displayed in the display screen, ie the rendering of the medical tissue edge in the first slice image is completed.
  • the first slice image after the drawing of the tissue edge is completed may also be saved. For example, you can save after taking a double-click operation entered by the user.
  • the input operation of the user dragging the circular curve is obtained through the user interface to adjust.
  • the medical tissue to be reconstructed can be identified from the slice image, thereby facilitating subsequent extraction of the isosurface to generate the display patch and improving the accuracy of the three-dimensional reconstruction, for example, for non- Uniform tumors, because of the appearance of black and white plaques in two-dimensional slice images, tumors that appear in the non-homogeneous tumors such as blood vessels, cysts, etc. are also more common, by drawing the edge of the medical tissue Quickly and accurately distinguish it from other organizations, improve the extraction of isosurfaces, and generate accurate display patches, thus improving the accuracy of subsequent 3D reconstruction.
  • spatial proximity means that the two slice images are spatially closest and the features are similar.
  • the isosurface refers to the volume data as a sampling set of a certain physical property in a certain spatial region, and the value at the non-sampling point is estimated by interpolation of the sampling values at the adjacent sampling points, then the space A collection of all points in a region that have a certain value of the same value will define one or more surfaces.
  • step S103 uses the Marching Cubes algorithm (a surface rendering processing algorithm commonly used in image processing three-dimensional display)
  • the drawn isosurfaces of the first slice image and the second slice image are extracted to generate a display patch.
  • the display patch Since the obtained display patch has a spatial position deviation and scaling from the position of the real tissue, the display patch is adjusted according to the spatial position and the separation distance of the tissue image (the original CT image) generated by the display patch.
  • the display patch is adjusted, and specifically includes the following steps a5-a6:
  • A5. Receive an image operation instruction input by the user on the display patch
  • A6 Perform the image operation instruction on the display patch according to an image operation instruction.
  • the image operation instruction may include selection, scaling, translation, and rotation of the medical image.
  • the Marching Cubes algorithm when generating a display patch, simply fits the space between each two adjacent spatially sliced images, in order to obtain a better-performing display patch, and reduce the display surface. There is spatial position deviation and scaling between the slice and the real tissue.
  • the display patch that needs to be adjusted is selected by obtaining the user input selection command.
  • the translation instruction input by the user is obtained according to the original
  • the spatial position of the CT image spatially moves the display patch.
  • the three-dimensional reconstruction method further includes: performing thinning processing and smoothing processing on the display patch.
  • the streamlining process refers to the use of the face reduction algorithm (Triangle Reduction) to merge some of the triangles, reducing the number of triangles and improving the display effect.
  • the face reduction algorithm Triangle Reduction
  • the smoothing process improves the visual effect of the display patch by removing noise.
  • the thinning process uses a vertex merging method to combine several small display patches located in a rectangular block of length, width, and height into one large display surface.
  • a piece, the length, width, and height parameters of the rectangular block are adjustable, and the display patches in the rectangular block are sequentially read, and then judged one by one, if only one vertex of the three vertices of the display patch is in the rectangular block, The display patch is retained; if two of the three vertices of the display patch are within the rectangular block, one edge between the two vertices is retained; if the three vertices of the display patch are all in a rectangle Inside the block, the three vertices of the display patch are reduced to one point, and all the rectangular blocks are combined with the triangular patch;
  • the smoothing process adopts Laplacian smoothing technology, and sets a certain number of iterations, and reduces the surface noise by adjusting the position of the point, and the surface drawing is smoother. Calculate the normal of each point on the grid, and select the Gauther coloring algorithm to realize the illumination smoothing of the triangle patch.
  • the higher the number of iterations of the smoothing operation the better the smoothing effect and the slower the speed.
  • the present invention does not limit this. In actual operation, the corresponding number of iterations can be set according to the smoothing effect that needs to be obtained.
  • steps S103 and S104 of the present invention may be performed after all the slice images are drawn in step S101, or may be executed after performing the step S103 after each two spatially adjacent slice images are drawn in step S101.
  • S104 the present invention does not limit this, and the user can perform according to actual needs.
  • a set of slice images with sharp edges is obtained, and the slice images are recorded as a slice image 1, a slice image 2, a slice image N, ... in the order in which the slice images are spatially adjacent.
  • the medical tissue edge of the slice image 1 and the slice image 2 may be first drawn, and then the area enclosed by the edge of the medical tissue of the sliced image 1 and the slice image 2 is filled, and the filled slice image 1 and the slice image 2 are extracted.
  • the isosurfaces of the slice images adjacent to the middle space respectively generate a display patch; then the medical tissue edges of the image three are sliced; the image of the edge of the medical tissue surrounded by the post-image image is filled, and the filled slice image is extracted.
  • the isosurfaces of the spatially adjacent slice images in the second and third slice images respectively generate a display patch; the display patches generated according to the slice image one and the slice image two are generated according to the slice image two and the third slice image
  • the isosurface of the display patch is spliced to obtain a three-dimensional image of the medical tissue, and the generation of the three-dimensional image of the slice image four to the slice image N is similar to the above method, and the present invention will not be described herein.
  • a method for three-dimensional reconstruction of a slice image of a medical tissue by drawing an edge of a medical tissue in each slice image of a set of slice images of a medical tissue, filling each slice image drawn The area enclosed by the edge of the medical organization; the medical tissue to be reconstructed can be identified from the slice image, which facilitates subsequent extraction of the isosurface in the slice image, generates a display patch and improves the accuracy of the three-dimensional reconstruction, and extracts the drawn An isosurface of two slice images adjacent to each other in the slice image respectively generates a display patch, and splices the isosurfaces of the generated display patch to obtain a three-dimensional image of the medical tissue, thereby establishing The two-dimensional slice image is drawn on the edge of the tissue, and the two-dimensional slice image is spatially connected by the three-dimensional reconstruction method by extracting the isosurface of the two-dimensional slice image, which can accurately and quickly represent the actual structure information and spatial information of the slice image organization in the human body.
  • the following vena cava is taken as an example, and a method for three-dimensional reconstruction of a slice image of a medical tissue is specifically described.
  • the original set of inferior vena cava has about 300 CT images, and fifteen CT images that can clearly distinguish the tissue margin of the inferior vena cava are obtained, and the CT images of the fifteen inferior vena cava will be obtained.
  • the spatial adjacent order it is recorded as the inferior vena cava CT image, the inferior vena cava CT image II, the inferior vena cava CT image three... the inferior vena cava CT image fifteen.
  • the specific steps of the three-dimensional reconstruction method of the sliced image of the medical tissue for the three-dimensional reconstruction of the inferior vena cava provided by the embodiment of the present invention are as follows:
  • the inferior vena cava CT image 1 can be displayed on the display screen.
  • the processor interpolates according to the three-type spline.
  • the algorithm (Spline) connects the curves between the at least three points of the received user input by three-spline interpolation to form a closed curve, as shown in FIG. 6c, and the closed curve is displayed in the display after being drawn. That is, the drawing of the medical tissue edge in the image of the inferior vena cava CT image is completed.
  • the user enters the "display” operation, the curve is saved, the curve disappears on the display screen, and then the inferior vena cava CT image is drawn: of course, in order to facilitate the subsequent processing and the user
  • the view can also save the first slice image after the drawing of the tissue edge is completed.
  • S205 splicing the isosurfaces of the generated display patches to obtain a three-dimensional image of the inferior vena cava, as shown in FIG. 6e;
  • the present invention can use the steps S201 and S202 to generate the inferior vena cava CT image three display patch, and then the inferior vena cava CT.
  • the image 1 and the inferior vena cava CT image are generated to display the isosurface of the patch to be spliced, and the isosurface of the inferior vena cava CT image generating display patch and the inferior vena cava CT image 3 are spliced.
  • the three-dimensional image of the inferior vena cava is obtained, and the principle and method for generating the three-dimensional image of the inferior vena cava CT image four to the inferior vena cava CT image are similar, and the present invention will not be repeated herein. It should be noted that when the isosurface is spliced, the display patch generated by splicing each of the inferior vena cava CT images and the display patch generated by the adjacent inferior vena cava CT image are used to acquire the inferior vena cava. 3D image.
  • a method for three-dimensional reconstruction of a slice image of a medical tissue provided by an embodiment of the present invention, Filling the edge of the inferior vena cava with each CT image drawn by drawing the edge of the inferior vena cava in each CT image in a set of CT images of the inferior vena cava; extracting the drawn CT image
  • the isosurfaces of two CT images adjacent to each of the two spaces respectively generate a display patch, and splicing the isosurfaces of the generated display patches to obtain a three-dimensional image of the inferior vena cava due to human metabolism and CT imaging
  • the inferior vena cava is difficult to distinguish from the surrounding tissue.
  • the spatial position can only be judged, and according to the different periods of the CT image, the display of the inferior vena cava on the image is different, and there may even be When the tissue is occluded, the blood vessels cannot be observed.
  • tissue edge rendering by extracting the isosurface of the two-dimensional slice image, the two-dimensional slice image is spatially connected by the three-dimensional reconstruction method, and the slice image can be accurately and quickly presented.
  • the three-dimensional visualization results of the actual structural information and spatial information in the human body are solved, and the traditional three-dimensional reconstruction method cannot be solved quickly.
  • accurately presented its intermittent display from human tissue or plaque effectively address the bottleneck of slice images, information for doctors to learn more about the structure and organization of this type of relationship with the surrounding tissue organizations provide important help.
  • a method for three-dimensional reconstruction of a slice image of a medical tissue provided by an embodiment of the present invention is applied to three-dimensional reconstruction of a non-homogeneous tumor as an example.
  • a non-homogeneous tumor is taken as an example, and a method for three-dimensional reconstruction of a sliced image of a medical tissue is specifically described.
  • the original CT image of a group of non-homogeneous tumors is about 200 sheets, and ten CT images of the tissue margins of the non-homogeneous tumors are clearly selected, and the CT images of the ten non-homogeneous tumors are obtained according to the space.
  • the adjacent sequences are respectively recorded as non-homogeneous tumor CT images, non-homogeneous tumor CT images, non-homogeneous tumor CT images three... non-homogeneous tumor CT images ten.
  • FIG. 8 provides an experimental example for three-dimensional reconstruction of a non-homogeneous tumor by using a method for three-dimensional reconstruction of a sliced image of a medical tissue according to an embodiment of the present invention.
  • FIG. 8 is a block diagram of FIG.
  • the display interface of the homogeneous tumor CT image in the image processing apparatus it can be seen from Fig. 8a that the spatial three-dimensional structure of the non-homogeneous tumor is similar to a relatively smooth sphere, and the non-homogeneous tumor CT image shown in Fig. 8a is located.
  • the tumor is located in the middle of the tumor, and the liver is located in the lower left of the tumor.
  • 8b is a three-dimensional reconstructed display interface of a set of non-homogeneous tumor CT images in an image processing apparatus, and the image located in the lower right corner of the image is a three-dimensionally reconstructed tissue image of the present invention, and the generated tissue image is located at the upper left.
  • the two-dimensional tissue image comparison of the non-homogeneous tumor of the angle shows that the tissue image obtained by the three-dimensional reconstruction of the present invention fits closely with the tumor edge in the two-dimensional tissue image recognized by the naked eye.
  • a method for three-dimensional reconstruction of a slice image of a medical tissue by drawing an edge of a non-homogeneous tumor in each CT image with a sharp edge in a set of CT images of a non-homogeneous tumor, filling each of the drawn images CT image image enclosing the edge of the non-homogeneous tumor; extracting the isosurfaces of two CT images adjacent to each of the spatially located CT images, respectively generating a display patch, and the generated display patch
  • the isosurface is spliced to obtain a three-dimensional image of the non-homogeneous tumor, so that the two-dimensional CT image is constructed by edge-forming, and the two-dimensional slice image is spatially connected by the three-dimensional reconstruction method by extracting the isosurface of the two-dimensional slice image.
  • the three-dimensional visualization result corresponding to the actual structural information and the spatial information of the sliced image tissue in the human body can be accurately and quickly presented, and the traditional three-dimensional reconstruction method cannot quickly and accurately display the sliced image from the intermittent display or the plaque.
  • the bottleneck of effective treatment of human tissue, for doctors to understand the structural information of such organizations and the relationship between tissues and surrounding organizations Provided important help.
  • An embodiment of the present invention provides an image processing apparatus, as shown in FIG.
  • the built equipment includes:
  • a drawing unit 401 configured to draw an edge of the medical tissue in each slice image with a sharp edge in a set of slice images of the medical tissue
  • a filling unit 402 configured to fill an area enclosed by the edge of the medical tissue of each slice image drawn;
  • a generating unit 403 configured to extract an isosurface of two slice images adjacent to each of two spaces in the filled slice image, and respectively generate a display patch
  • the display unit 404 is configured to splicing the isosurfaces of the generated display patches to obtain a three-dimensional image of the medical tissue.
  • the drawing unit 401 includes:
  • the first receiving module 401A1 is configured to receive at least three points input by the user;
  • the first processing module 401A2 according to the three-spline interpolation algorithm, respectively connecting each adjacent two points of the at least three points with an arc to form a closed curve, forming a medical tissue edge in the first slice image .
  • the drawing unit 401 includes:
  • the second receiving module 401B1 is configured to receive the first point and the second point that the user inputs in succession;
  • a second processing module 401B2 configured to use a first point input by the user as a center, and draw a circle with a distance between the first point and the second point to form a medicine in the first slice image Organizational edge.
  • the generating unit 403 is specifically configured to extract the isosurfaces of the first slice image and the second slice image that are drawn by using a Marching Cubes algorithm to generate a display patch.
  • the generating unit 403 further includes:
  • the third receiving module 4031 is configured to receive an image operation instruction input by the user on the display patch
  • the executing module 4032 is configured to execute the image operation instruction on the display patch according to the image operation instruction, where the image operation instruction includes selecting, scaling, panning, and rotating the medical image.
  • the three-dimensional reconstruction device 40 further includes:
  • the processing unit 405 is configured to streamline and smooth the display patch.
  • the drawing unit is used to draw the edge of the medical tissue
  • the filling unit fills the area enclosed by the edge of the drawn medical tissue
  • the generating unit extracts two slice images of each two adjacent spaces in the filled slice image.
  • the isosurfaces respectively generate display patches
  • the display unit is configured to splicing the isosurfaces of the generated display patches to obtain a three-dimensional image of the medical tissue.
  • the 3D visualization result corresponding to the spatial information solves the bottleneck that the traditional 3D reconstruction method cannot quickly and accurately process the human tissue from the intermittently displayed or plaque sliced image.
  • the doctor understands the structure of such tissue in detail. Information and the relationship between the organization and surrounding organizations provide important assistance.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Graphics (AREA)
  • Geometry (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

本发明的实施例提供一种医学组织的切片图像三维重建的方法及设备，用以解决对断续显示的组织及斑块化的组织进行三维重建时难以准确、快速的将其与其他组织区分开来的问题。该医学组织的切片图像三维重建的方法包括：绘制所述医学组织的一组切片图像中边缘清晰的切片图像中的所述医学组织的边缘；填充绘制的切片图像所述医学组织的边缘围合的区域；提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片；将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。本发明适用于医学影像的显示技术领域。

Description

一种医学组织的切片图像三维重建的方法及设备
本申请要求于2015年08月03日提交中国专利局、申请号为201510481588.1、发明名称为“一种医学组织的切片图像三维重建的方法及设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及医学影像的显示技术领域，尤其涉及一种医学组织的切片图像三维重建的方法及设备。
背景技术
传统的CT(Computer Tomography，简称CT,电子计算机断层扫描)阅片限于二维图像展示，且信息量的表达很大程度上取决于医生的经验水平。随着计算机技术的发展以及数字医学的推广，基于图像的三维重建(英文：3D Reconstruction)被广泛应用于外科手术中，如术前规划、手术模拟以及术前风险评估等。三维重建是指对三维物体建立适合计算机表示和处理的数学模型,即在计算机环境下对三维物体进行处理、操作和分析其性质的基础,也是在计算机中建立表达客观世界的虚拟现实的关键技术。
下腔静脉和非匀质肿瘤是两种非常重要的组织，医生了解这两种组织的详细信息对于医生进行手术规划是极其重要的。例如：对于下腔静脉，医生需要详细了解下腔静脉受到病变组织的侵蚀情况；再例如，对于非匀质肿瘤，医生更需要详细获取肿瘤的体积、以及肿瘤与周边组织如肝脏、血管等的关系。
传统的三维重建方法包括：数学形态学方法、区域生成方法、水平集(Level Set)方法及黑塞矩阵(Hessian Matrix)方法。数学形态学方法用来分析图像内在几何结构；区域生成方法可以由单个特征点延伸，从而将与之相邻特性相似的整个区域划分出来；水平集方法能够实现主动轮廓线模型；黑塞矩阵方法可以用以实现突出目标区域。这些方法在针对具体问 题时有时并不会单独使用。
但是，利用上述传统的三维重建方法对下腔静脉和非匀质肿瘤进行重建时，不能准确、快速将其与其他组织区分开来。具体的：对于下腔静脉，由于人体代谢以及CT成像原理等原因，下腔静脉在连续的CT图像中只能大约判断其空间位置，并且根据CT图像不同的时期，下腔静脉在图像上的显示明亮程度不同，甚至会有造影剂在扫描时刻不在扫描位置的情况使得无法观察到血管，造成该情况的原因是下腔静脉流速较快、下腔静脉的脉管直径较大，并且此种情况在CT图像中出现的较为普遍；对于非匀质肿瘤，在切片图像中会呈现黑白斑块的形状，传统的三维重建方法无法利用现有算法自动区分非匀质肿瘤的边缘，非匀质肿瘤内部出现血管、囊肿等情况的肿瘤也较为普遍，所以难以快速准确将其与其他组织区分开来。
发明内容
本发明的实施例提供一种医学组织的切片图像三维重建的方法及设备，用以解决对断续显示的组织及斑块化的组织进行三维重建时难以快速、准确将其与其他组织区分开来的问题。
为达到上述目的，本发明的实施例采用如下技术方案：
第一方面，本发明实施例提供了一种医学组织的切片图像三维重建的方法，包括：
绘制所述医学组织的一组切片图像中边缘清晰的切片图像中的所述医学组织的边缘；
填充绘制的每张切片图像所述医学组织的边缘围合的区域；
提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片；
将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。
结合第一方面，在本发明实施例的第一种可能实现方式中，对于第一切片图像，绘制所述第一切片图像中的医学组织边缘包括：
接收用户输入的至少三个点，
根据三式样条插值算法，用弧线分别连接所述至少三个点中每两个相邻的点绘制成闭合曲线，形成所述第一切片图像中的医学组织边缘。
结合第一方面，在本发明实施例的第二种可能实现方式中，对于第一切片图像，绘制所述第一切片图像中的医学组织边缘包括：
接收用户先后输入的第一点和第二点；
以所述用户输入的第一点为圆心，以所述第一点和第二点之间的距离为半径绘圆，形成所述第一切片图像中的医学组织边缘。
结合第一方面，在本发明实施例的第三种可能实现方式中，对于空间相邻的第一切片图像和第二切片图像，提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片包括：
利用Marching Cubes算法提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片。
结合第三种可能实现方式，在本发明实施例的第四种可能实现方式中，所述对于空间相邻的第一切片图像和第二切片图像，提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片还包括：
接收用户输入的对所述显示面片的图像操作指令；
根据所述图像操作指令，对所述显示面片执行所述图像操作指令，所述图像操作指令包括对所述医学影像的选择、缩放、平移、旋转。
结合第一方面，在本发明实施例提供的第五种可能实现方式中，在所述将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像之前，所述方法还包括：对所述显示面片进行精简以及平滑处理。
结合第一方面，在本发明实施例提供的第六种可能实现方式中，所述医学组织为下腔静脉或者非匀质肿瘤。
第二方面，本发明实施例提供了一种图像处理设备，包括：
绘制单元，用于绘制所述医学组织的一组切片图像中边缘清晰的切片图像中的所述医学组织的边缘；
填充单元，用于填充绘制的每张切片图像所述医学组织的边缘围合的区域；
生成单元，用于提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片；
显示单元，用于将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。
结合第二方面，在本发明实施例的第一种可能实现方式中，所述绘制单元包括：
第一接收模块，用于接收用户输入的至少三个点；
第一处理模块，根据三式样条插值算法，用弧线分别连接所述至少三个点中每两个相邻的点绘制成闭合曲线，形成所述第一切片图像中的医学组织边缘。
结合第二方面，在本发明实施例的第二种可能实现方式中，所述绘制单元包括：
所述绘制单元包括：
第二接收模块，用于接收用户先后输入的第一点和第二点；
第二处理模块，用于以所述用户输入的第一点为圆心，以所述第一点和第二点之间的距离为半径绘圆，形成所述第一切片图像中的医学组织边缘。
结合第二方面，在本发明实施例的第三种可能实现方式中，所述生成单元具体用于利用Marching Cubes算法提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片。
结合本发明实施例的第三种可能实现方式，在本发明实施例的第四种可能实现方式中，所述生成单元还包括：
第三接收模块，用于接收用户输入的对所述显示面片的图像操作指令；
执行模块，用于根据所述图像操作指令，对所述显示面片执行所述图像操作指令，所述图像操作指令包括对所述医学影像的选择、缩放、平移、旋转。
结合第一方面，在本发明实施例的第五种种可能实现方式中，在所述显示单元之前，所述设备还包括：
处理单元，用于对所述显示面片进行精简以及平滑处理。
基于本发明实施例提供的医学组织的切片图像三维重建的方法，通过绘制医学组织的一组切片图像中边缘清晰的切片图像中的医学组织的边 缘，填充绘制的每张切片图像所述医学组织的边缘围合的区域，通过绘制所述医学组织的边缘围合区域，能够将所要重建的医学组织从切片图像中标识出来，便于后续提取等值面生成显示面片且提高了三维重建的精确度，提取绘制的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片，将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像，这样通过建立二维切片图像组织边缘绘制，通过提取二维切片图像的等值面将二维切片图像通过三维重建方法空间连接，可以准确、快速呈现该切片图像组织在人体中实际的结构信息与空间信息对应的三维可视化结果，解决了传统的三维重建方法无法快速准确将其从呈现断续显示或者斑块化的切片图像中人体组织进行有效处理的瓶颈，对于医生详细了解这类组织的结构信息以及组织与周边组织的关系提供了重要的帮助。
附图说明
为了更清楚地说明本发明实施例的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本发明实施例提供的一种图像处理设备的硬件结构图；
图2为本发明实施例一提供的一种医学组织的切片图像三维重建的方法的流程示意图一；
图3为本发明实施例一提供的一种医学组织的切片图像三维重建的方法的流程示意图二；
图4为本发明实施例一提供的一种医学组织的切片图像三维重建的方法的流程示意图三；
图5为本发明实施例二提供的一种医学组织的切片图像三维重建的方法用于三维重建下腔静脉的流程示意图；
图6a-图6e为本发明实施例二提供的一种医学组织的切片图像三维重建方法用于三维重建下腔静脉的示意图；
图7为本发明实施例三提供的一种医学组织的切片图像三维重建方法用于三维重建非匀质肿瘤的流程示意图；
图8a-图8b为本发明实施例三提供的一种医学组织的切片图像三维重建方法用于三维重建非匀质肿瘤的示意图；
图9为本发明实施例四提供的一种图像处理设备的结构示意图一；
图10为本发明实施例四提供的提供一种图像处理设备的结构示意图二；
图11为本发明实施例四提供的提供一种图像处理设备的结构示意图三。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
为了便于清楚描述本发明实施例的技术方案，在本发明的实施例中，采用了“第一”、“第二”等字样对功能和作用基本相同的相同项或相似项进行区分，本领域技术人员可以理解“第一”、“第二”等字样并对不对数量和执行次序进行限定。
本发明实施例的方法可以由图像处理设备执行，参见图1，图1示出了一种图像处理设备的硬件示意图，该图像处理设备10包括一个或多个(图中仅示出一个)处理器101、存储器102、用户接口103、通信总线104以及显示屏105。本领域普通技术人员可以理解，图1所示的结构仅为示意，其并不对图像处理设备10的结构造成限定。例如，图像处理设备10还可包括比图1中所示更多或者更少的组件，或者具有与图1所示不同的配置。
通信总线104用于图像处理设备10中各组成部件之间的通信。用户接口103用于插接外部设备，例如鼠标及键盘等，以接收用户输入的信息。
显示屏105用于对切片图像、切片图像三维重建后的医学组织以及中间的处理过程涉及的图像以及图像变化等进行显示。
存储器102可用于存储软件程序以及模块，数据库，如本发明实 施例中的确定需要绘制医学组织的绘制方法及装置对应的程序指令/模块。存储器102可包括高速随机存储器，还可包括非易失性存储器，如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。在一些实例中，存储器102可进一步包括相对于处理器101远程设置的存储器，这些远程存储器可以通过网络连接至图像处理设备10。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
处理器101通过运行存储在存储器102内的软件程序以及模块，从而执行各种功能应用以及数据处理，例如，处理器101通过调用存储器102中的确定需要绘制医学组织的绘制方法的应用程序，以实现快速而准确的确定需要进行绘制医学组织的绘用试例。
实施例一
本发明的实施例提供一种医学组织的切片图像三维重建的方法，如图2所示，该三维重建的方法包括：
S101、绘制医学组织的一组切片图像中边缘清晰的切片图像中的所述医学组织的边缘。
其中，切片图像是指利用计算机成像设备得到人体及其内部器官内的二维序列数字断层图像，通过明晰的人体结构及详细病例信息二维序列数字断层图像给医生提供辅助诊断和治疗，例如，切片图像可以是包括CT(Computed Tomography，电子计算机断层扫描)图像或MRI(Magnetic Resonance Imaging，磁共振成像)图像，本发明并不限定。仅是示例性的，本发明的实施例选取的切片图像为CT图像。
本发明实施例中，医学组织是指具有某种机能和构造的有机的细胞群。优选的，本发明实施例中的医学组织为下腔静脉或者非匀质肿瘤。
其中，医学组织边缘通常是指医学组织的两侧分属于两个区域，每个区域内部的特性相对比较均匀一致，而两个区域之间在特性上则存在一定差异。
本发明实施例中，切片图像中边缘清晰的切片图像是指在一组切片图像中可以较为清晰将需要分割的医学组织边缘与其他医学组织 区分开的切片图像。
通常，在一组切片图像中，对于一种组织，有的切片图像的组织边缘清晰，有的组织边缘则不清晰。在S101中，处理的对象仅针对边缘清晰的切片图像。其中，可以通过多种方法从一组切片图像中选择边缘清晰的切片图像，例如，可以由用户根据自己的判断选择，也可以根据一定的算法进行选择，该过程不影响本发明目的的实现，本发明实施例对此不进行限制。
绘制医学组织的边缘是在显示屏上显示的切片图像中，用特定的线条标识医学组织的边缘，以将其与其他组织区分开来。
本发明实施例中，可以通过多种方式绘制医学组织的边缘，在此不进行具体限制。其中，因为对于每张切片图像的处理原理和过程相同，在此，仅以第一切片图像为例进行说明，该第一切片图像指选出的边缘清晰的切片图像中的任意一张，并不具有特殊的指示含义。
例如，一种可能的实现方式中，如图3所示，步骤S101具体包括a1-a2：
a1、接收用户输入的至少三个点；
a2、根据三式样条插值算法，用弧线分别连接至少三个点中每两个相邻的点绘制成闭合曲线，形成所述第一切片图像中的医学组织边缘。
示例的，可以在显示屏上显示该第一切片图像，通过用户接口获取用户输入的至少三个点取样点，处理器根据三式样条插值算法(Spline)将接收到的用户输入的至少三个点之间的曲线会以三式样条插值进行连接，形成闭合曲线，并在显示屏中显示，即完成对该第一切片图像中的医学组织边缘的绘制。当然，为方便后续处理过程的进行以及用户的查看，还可以保存组织边缘的绘制完成后的第一切片图像。例如，可以在获取用户输入的双击操作后进行保存。
同时为了获取效果更好的绘制图像，通过用户接口获取用户拖动至少三个点取样点的操作对闭合曲线进行调整。
另一种可能的实现方式中，如图4所示，步骤S101具体包括a3-a4：
a3、接收用户先后输入的第一点和第二点；
a4、以所述用户输入的第一点为圆心，以所述第一点和第二点之间的距离为半径绘圆，形成所述第一切片图像中的医学组织边缘。
示例的，可以在显示屏上显示该第一切片图像，通过用户接口获取用户先后输入的第一点和第二点，处理器以获取的第一点为圆心，以第一点和第二点之间的距离为半径绘圆，形成第一切片图像中的医学组织边缘，并在显示屏中显示，即完成对该第一切片图像中的医学组织边缘的绘制。当然，为方便后续处理过程的进行以及用户的查看，还可以保存组织边缘的绘制完成后的第一切片图像。例如，可以在获取用户输入的双击操作后进行保存。
同时为了获取效果更好的绘制图像，通过用户接口获取用户拖动该圆形曲线的输入操作进行调整。
S102、填充绘制的每张切片图像所述医学组织的边缘围合的区域。
其中，通过绘制所述医学组织的边缘围合区域，能够将所要重建的医学组织从切片图像中标识出来，便于后续提取等值面生成显示面片且提高三维重建的准确度，例如，对于非匀质肿瘤，由于在二维切片图像中会呈现黑白斑块的形状，在非匀质肿瘤内部出现血管、囊肿等情况的肿瘤也较为普遍，通过绘制其医学组织的边缘围合的区域，能够快速准确将其与其他组织区分开来，提高了提取等值面，生成显示面片的准确，从而提高了后续三维重建的精确度。
S103、提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片。
其中，空间相邻是指两张切片图像在空间上距离最近，特征相似。
其中，等值面是指将体数据看成是某个空间区域内关于某种物理属性的采样集合，非采样点上的值以其邻近采样点上的采样值的插值来估计，则该空间区域内所有具有某一个相同值的点的集合将定义一个或多个曲面。
进一步的，对于步骤S103的一种可能的实现方式为：利用Marching Cubes算法(图像处理三维显示中常用的面绘制处理算法) 提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片。
由于得到的显示面片与真实组织所在位置存在空间位置偏差与缩放，依据显示面片生成的组织图像(原CT图像的)的空间位置以及间隔距离，对显示面片进行调整。
进一步的，对显示面片进行调整，具体包括以下步骤a5-a6：
a5、接收用户输入的对所述显示面片的图像操作指令；
a6、根据图像操作指令，对所述显示面片执行所述图像操作指令。
其中，图像操作指令可以包括对所述医学影像的选择、缩放、平移、旋转。
示例的，在实际应用中，由于生成显示面片时，Marching Cubes算法只是将每两张空间相邻的切片图像之间做了空间拟合，为了获取效果更佳的显示面片，减少显示面片与真实组织所在位置存在空间位置偏差与缩放，通过获取用户输入的选择指令选取需要调整的显示面片，为了使得到的显示面片显示更加真实，依据原CT图像在所在空间位置的实际距离，获取用户输入的缩放指令对该显示面片进行缩放使该显示面片贴合原CT图像实际的空间分布，或为了使显示面片符合所在空间的实际位置，获取用户输入的平移指令根据原CT图像所在空间位置对该显示面片进行空间移动。
S104、将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。
进一步的，为了提高三维重建的效率以及最终三维图像的显示效果，在步骤S104之前，该三维重建方法，还包括：对所述显示面片进行精简处理以及平滑处理。
其中，精简处理是指利用面片精简算法(Triangle Reduction)合并部分三角片，减少三角片数量，提高显示效果。
其中，平滑(Smoothing)处理是通过去除噪声改善显示面片的视觉效果。
一种可能的实现方式中，该精简处理采用顶点合并法将位于长、宽、高固定的矩形块内的几个小的显示面片合并为一个大的显示面 片，该矩形块的长、宽、高参数可调,并依次读取该矩形块内的显示面片，然后逐一进行判断，如果该显示面片的三个顶点只有一个顶点在矩形块内，则保留该显示面片；如果该显示面片的三个顶点中的两个顶点在矩形块内，则保留该两个顶点之间的一条边；如果该显示面片的三个顶点全部在矩形块内部，则将显示面片的三个顶点缩成一个点，所有的矩形块内都进行三角面片合并的操作；
一种可能的实现方式中，该平滑处理采用拉普拉斯平滑技术，并设置一定迭代次数，通过调整点的位置减少表面噪点，表面绘制更加平滑。计算网格上每个点的法线，选取高洛德着色算法实现对三角面片的光照平滑处理。
需要注意的是，平滑运算的迭代次数越高，平滑效果越好，也越慢，本发明对此不作限定，在实际操作中可以根据实际需要获取的平滑效果设置相应的迭代次数。
需要说明的是，在实际应用中，本发明的步骤S103和S104可以在步骤S101绘制完所有的切片图像进行，也可以在步骤S101绘制每两张空间相邻的切片图像后执行步骤S103再执行S104，本发明对此不进行限制，用户可以根据实际需要执行。
在实际应用中获取边缘清晰的一组切片图像，按照该切片图像空间相邻的顺序，将切片图像记为切片图像一、切片图像二……切片图像N……。
例如，可以先绘制切片图像一和切片图像二的医学组织边缘，然后填充绘制后切片图像一和切片图像二所述医学组织的边缘围合的区域，提取填充后的切片图像一和切片图像二中空间相邻的切片图像的等值面，分别生成显示面片；然后切片图像三的医学组织边缘；填充绘制后切片图像三所述医学组织的边缘围合的区域，提取填充后的切片图像二和第三切片图像中空间相邻的切片图像的等值面，分别生成显示面片；将根据切片图像一和切片图像二生成的显示面片与根据切片图像二和第三切片图像生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像，对于切片图像四至切片图像N的三维图像的生成与上述方法类似，本发明在此不再赘述。
或者，绘制所述医学组织的一组切片图像中所有边缘清晰的每张切片图像中的所述医学组织的边缘；然后填充绘制的所有切片图像所述医学组织的边缘围合的区域，提取填充后的切片图像中空间相邻的切片图像的等值面，分别生成显示面片；将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。
需要说明的是，上述只是两种可实现的方式，并不对本发明目的实现构成影响，所以，也不对本发明实施例构成任何限制。
基于本发明实施例提供的医学组织的切片图像三维重建的方法，通过绘制医学组织的一组切片图像中边缘清晰的每张切片图像中的医学组织的边缘，填充绘制的每张切片图像所述医学组织的边缘围合的区域；能够将所要重建的医学组织从切片图像中标识出来，便于后续提取切片图像中的等值面，生成显示面片且提高了三维重建的精确度，提取绘制的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片，将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像，这样通过建立二维切片图像组织边缘绘制，通过提取二维切片图像的等值面将二维切片图像通过三维重建方法空间连接，可以准确、快速呈现该切片图像组织在人体中实际的结构信息与空间信息对应的三维可视化结果，解决了传统的三维重建方法无法快速、准确将其从呈现断续显示或者斑块化的切片图像中人体组织进行有效处理的瓶颈，对于医生详细了解这类组织的结构信息以及组织与周边组织的关系提供了重要的帮助。
实施例二
本实施例以下腔静脉为例，对医学组织的切片图像三维重建的方法进行具体说明。本实施例中，原始的一组下腔静脉的CT图像约为300张，从中选取十五张可以清晰分辨下腔静脉的组织边缘的CT图像，将得到的十五张下腔静脉的CT图像按照空间相邻顺序分别记为下腔静脉CT图像一，下腔静脉CT图像二，下腔静脉CT图像三……下腔静脉CT图像十五。
如图5所示，本发明实施例提供的一种医学组织的切片图像三维重建的方法用于三维重建下腔静脉的具体步骤如下：
S201、绘制下腔静脉CT图像一和下腔静脉CT图像二中的所述下腔静脉的边缘；
示例的，如图6a所示，可以在显示屏上显示该下腔静脉CT图像一，如图6b所示，通过用户接口获取用户输入的至少三个点取样点，处理器根据三式样条插值算法(Spline)将接收到的用户输入的至少三个点之间的曲线会以三式样条插值进行连接，形成闭合曲线，如图6c所示，该闭合曲线绘制完成后在显示屏中显示，即完成对该下腔静脉CT图像一图像中的医学组织边缘的绘制。如图6d所示，曲线绘制完成之后获取用户输入的“显示”操作，绘制曲线保存，曲线在显示屏上消失，然后绘制下腔静脉CT图像二：当然，为方便后续处理过程的进行以及用户的查看，还可以保存组织边缘的绘制完成后的第一切片图像。
S202、填充绘制后下腔静脉CT图像一和下腔静脉CT图像二中所述下腔静脉的边缘围合的区域。
S203、提取填充后的下腔静脉CT图像一和下腔静脉CT图像二中的等值面，分别生成显示面片。
S204、对所述显示面片进行精简处理以及平滑处理。
S205、将生成的显示面片的等值面进行拼接，获取所述下腔静脉的三维图像，如图6e所示；
需要说明的是，下腔静脉CT图像一和下腔静脉CT图像二生成显示面片后，本发明可以采用步骤S201及S202生成下腔静脉CT图像三的显示面片，然后将下腔静脉CT图像一和下腔静脉CT图像二生成显示面片的等值面进行拼接，并将下腔静脉CT图像二生成显示面片及下腔静脉CT图像三的显示面片的等值面进行拼接，获取所述下腔静脉的三维图像，下腔静脉CT图像四至下腔静脉CT图像十五的三维图像生成的原理和方法类似，本发明在此不再赘述。需要说明的是在进行等值面进行拼接时，拼接每一张下腔静脉CT图像生成的显示面片与其之前相邻的下腔静脉CT图像生成的显示面片用以获取所述下腔静脉的三维图像。
基于本发明实施例提供的医学组织的切片图像三维重建的方法， 通过绘制下腔静脉的一组CT图像中边缘清晰的每张CT图像中的下腔静脉的边缘，填充绘制的每张CT图像所述下腔静脉的边缘围合的区域；提取绘制的CT图像中每两个空间相邻的两张CT图像的等值面，分别生成显示面片，将生成的显示面片的等值面进行拼接，获取下腔静脉的三维图像，由于人体代谢以及CT成像原理等原因，下腔静脉难以与周边组织区分，在连续的CT图像中只能大约判断其空间位置，并且根据CT图像不同的时期，下腔静脉在图像上的显示明亮程度不同，甚至会有组织遮挡的情况无法观察到血管，这样通过建立二维CT图像组织边缘绘制，通过提取二维切片图像的等值面将二维切片图像通过三维重建方法空间连接，可以准确、快速呈现该切片图像组织在人体中实际的结构信息与空间信息对应的三维可视化结果，解决了传统的三维重建方法无法快速、准确将其从呈现断续显示或者斑块化的切片图像中人体组织进行有效处理的瓶颈，对于医生详细了解这类组织的结构信息以及组织与周边组织的关系提供了重要的帮助。
实施例三
如图7所示，以本发明实施例提供的一种医学组织的切片图像三维重建的方法应用于三维重建非匀质肿瘤为例进行说明。
具体步骤如下：
本实施例以非匀质肿瘤为例，对医学组织的切片图像三维重建的方法进行具体说明。本实施例中，原始的一组非匀质肿瘤CT图像约为200张，从中选取十张可以清晰分辨非匀质肿瘤的组织边缘的CT图像，将得到的十张非匀质肿瘤的CT图像按照空间相邻顺序分别记为非匀质肿瘤CT图像一，非匀质肿瘤CT图像二，非匀质肿瘤CT图像三……非匀质肿瘤CT图像十。
S301、绘制非匀质肿瘤的一组CT图像中边缘清晰的CT图像中的所述非匀质肿瘤的边缘；
S302、填充绘制的每张CT图像所述非匀质肿瘤的边缘围合的区域。
S303、提取填充后的CT图像中每两个空间相邻的两张CT图像 的等值面，分别生成显示面片。
S304、对所述显示面片进行精简处理以及平滑处理。
S305、将生成的显示面片的等值面进行拼接，获取所述非匀质肿瘤的三维图像。
示例性的，图8提供了本发明实施例一种医学组织的切片图像三维重建的方法用于三维重建非匀质肿瘤的实验实例，现以图8为例进行说明，图8a为一组非匀质肿瘤CT图像在图像处理设备中的显示界面，从图8a中可以看出非匀质肿瘤的空间三维结构类似于一种较为光滑的球体，图8a显示的非匀质肿瘤CT图像中位于中间偏上的为肿瘤，位于肿瘤左下方的为肝脏，可以看出该肿瘤虽然整体灰度低于位于其左下方的肝脏，但是肿瘤内部明暗交错，肿瘤边缘能够通过肉眼判断。图8b为一组非匀质肿瘤CT图像在图像处理设备中的三维重建后显示界面，位于该图像右下角的图像为经过本发明三维重建后的组织图像，将该生成的组织图像与位于左上角的该非匀质肿瘤的二维组织图像对比，可以发现经过本发明三维重建后的组织图像与肉眼识别的二维组织图像中的肿瘤边缘十分贴合。
基于本发明实施例提供的医学组织的切片图像三维重建的方法，通过绘制非匀质肿瘤的一组CT图像中边缘清晰的每张CT图像中的非匀质肿瘤的边缘，填充绘制的每张CT图像所述非匀质肿瘤的边缘围合的区域；提取绘制的CT图像中每两个空间相邻的两张CT图像的等值面，分别生成显示面片，将生成的显示面片的等值面进行拼接，获取所述非匀质肿瘤的三维图像，这样通过建立二维CT图像组织边缘绘制，通过提取二维切片图像的等值面将二维切片图像通过三维重建方法空间连接，可以准确、快速呈现该切片图像组织在人体中实际的结构信息与空间信息对应的三维可视化结果，解决了传统的三维重建方法无法快速、准确将其从呈现断续显示或者斑块化的切片图像中人体组织进行有效处理的瓶颈，对于医生详细了解这类组织的结构信息以及组织与周边组织的关系提供了重要的帮助。
实施例四
本发明的实施例提供一种图像处理设备，如图9所示，该三维重 建设备包括：
绘制单元401，用于绘制所述医学组织的一组切片图像中边缘清晰的每张切片图像中的所述医学组织的边缘；
填充单元402，用于填充绘制的每张切片图像所述医学组织的边缘围合的区域；
生成单元403，用于提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片；
显示单元404，用于将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。
可选的，如图10所示，绘制单元401包括：
第一接收模块401A1，用于接收用户输入的至少三个点；
第一处理模块401A2，根据三式样条插值算法，用弧线分别连接所述至少三个点中每两个相邻的点绘制成闭合曲线，形成所述第一切片图像中的医学组织边缘。
可选的，如图10所示，绘制单元401包括：
第二接收模块401B1，用于接收用户先后输入的第一点和第二点；
第二处理模块401B2，用于以所述用户输入的第一点为圆心，以所述第一点和第二点之间的距离为半径绘圆，形成所述第一切片图像中的医学组织边缘。
可选的，生成单元403具体用于利用Marching Cubes算法提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片。
可选的，如图10所示，生成单元403还包括：
第三接收模块4031，用于接收用户输入的对所述显示面片的图像操作指令；
执行模块4032，用于根据所述图像操作指令，对所述显示面片执行所述图像操作指令，所述图像操作指令包括对所述医学影像的选择、缩放、平移、旋转。
可选的，如图11所示，该三维重建设备40还包括：
处理单元405，用于对显示面片进行精简以及平滑处理。
这样一来，绘制单元用于绘制医学组织的边缘，填充单元填充绘制后的医学组织的边缘围合的区域，生成单元提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片，显示单元，用于将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。这样通过建立二维切片图像组织边缘绘制，通过提取二维切片图像的等值面将二维切片图像通过三维重建方法空间连接，可以准确、快速呈现该切片图像组织在人体中实际的结构信息与空间信息对应的三维可视化结果，解决了传统的三维重建方法无法快速准确将其从呈现断续显示或者斑块化的切片图像中人体组织进行有效处理的瓶颈，对于医生详细了解这类组织的结构信息以及组织与周边组织的关系提供了重要的帮助。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以所述权利要求的保护范围为准。

Claims (11)

  1. 一种医学组织的切片图像三维重建的方法，其特征在于，包括：
    绘制所述医学组织的一组切片图像中边缘清晰的切片图像中的所述医学组织的边缘；
    填充绘制的每张切片图像所述医学组织的边缘围合的区域；
    提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片；
    将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。
  2. 根据权利要求1所述的方法，其特征在于，对于第一切片图像，绘制所述第一切片图像中的医学组织边缘包括：
    接收用户输入的至少三个点，
    根据三式样条插值算法，用弧线分别连接所述至少三个点中每两个相邻的点绘制成闭合曲线，形成所述第一切片图像中的医学组织边缘。
  3. 根据权利要求1所述的方法，其特征在于，对于空间相邻的第一切片图像和第二切片图像，提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片包括：
    利用Marching Cubes算法提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片。
  4. 根据权利要求3所述的方法，其特征在于，所述对于空间相邻的第一切片图像和第二切片图像，提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片还包括：
    接收用户输入的对所述显示面片的图像操作指令；
    根据所述图像操作指令，对所述显示面片执行所述图像操作指令，所述图像操作指令包括对所述医学影像的选择、缩放、平移、旋转。
  5. 根据权利要求1所述的方法，其特征在于，在所述将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像之前，所述方法还包括：对所述显示面片进行精简以及平滑处理。
  6. 根据权利要求1所述的方法，其特征在于，所述医学组织为下腔静脉或者非匀质肿瘤。
  7. 一种图像处理设备，其特征在于，包括：
    绘制单元，用于绘制所述医学组织的一组切片图像中边缘清晰的切 片图像中的所述医学组织的边缘；
    填充单元，用于填充绘制的每张切片图像所述医学组织的边缘围合的区域；
    生成单元，用于提取填充后的切片图像中每两个空间相邻的两张切片图像的等值面，分别生成显示面片；
    显示单元，用于将生成的显示面片的等值面进行拼接，获取所述医学组织的三维图像。
  8. 根据权利要求7所述的设备，其特征在于，所述绘制单元包括：
    第一接收模块，用于接收用户输入的至少三个点；
    第一处理模块，根据三式样条插值算法，用弧线分别连接所述至少三个点中每两个相邻的点绘制成闭合曲线，形成所述第一切片图像中的医学组织边缘。
  9. 根据权利要求7所述的设备，其特征在于，所述生成单元具体用于利用Marching Cubes算法提取绘制的所述第一切片图像和第二切片图像的等值面，生成显示面片。
  10. 根据权利要求9所述的设备，其特征在于，所述生成单元还包括：
    第三接收模块，用于接收用户输入的对所述显示面片的图像操作指令；
    执行模块，用于根据所述图像操作指令，对所述显示面片执行所述图像操作指令，所述图像操作指令包括对所述医学影像的选择、缩放、平移、旋转。
  11. 根据权利要求7所述的设备，其特征在于，在所述显示单元之前，所述设备还包括：
    处理单元，用于对所述显示面片进行精简以及平滑处理。
PCT/CN2016/074260 2015-08-03 2016-02-22 一种医学组织的切片图像三维重建的方法及设备 WO2017020580A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510481588.1A CN106408648A (zh) 2015-08-03 2015-08-03 一种医学组织的切片图像三维重建的方法及设备
CN201510481588.1 2015-08-03

Publications (1)

Publication Number Publication Date
WO2017020580A1 true WO2017020580A1 (zh) 2017-02-09

Family

ID=57942348

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/074260 WO2017020580A1 (zh) 2015-08-03 2016-02-22 一种医学组织的切片图像三维重建的方法及设备

Country Status (2)

Country Link
CN (1) CN106408648A (zh)
WO (1) WO2017020580A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109859304A (zh) * 2018-11-16 2019-06-07 华中科技大学同济医学院附属同济医院 三维打印技术在角膜缘组织体外建立三维结构数字化模型中的应用
CN112396609A (zh) * 2019-07-31 2021-02-23 上海正雅齿科科技股份有限公司 一种牙列分割方法、牙齿分割方法、装置及电子设备

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107895364B (zh) * 2017-10-31 2019-06-14 哈尔滨理工大学 一种用于虚拟手术术前规划的三维重建***
CN108510443A (zh) * 2018-03-30 2018-09-07 河北北方学院 一种医学图像离线重建定位方法
CN108597038B (zh) * 2018-04-16 2022-05-27 北京市神经外科研究所 一种三维表面建模方法及装置、计算机存储介质
CN109242964B (zh) * 2018-11-01 2023-04-11 青岛海信医疗设备股份有限公司 三维医学模型的处理方法和装置
CN109727240B (zh) * 2018-12-27 2021-01-19 深圳开立生物医疗科技股份有限公司 一种三维超声图像的遮挡组织剥离方法及相关装置
CN109887091A (zh) * 2019-02-27 2019-06-14 华侨大学 一种用于将下肢和胸腰椎筋膜重建为3d模型的方法
CN111991015B (zh) * 2020-08-13 2024-04-26 上海联影医疗科技股份有限公司 三维图像拼接方法、装置、设备、***和存储介质
CN112113937A (zh) * 2020-07-20 2020-12-22 浙江大学 一种基于连续切片、多色荧光和三维重建的组织器官立体成像与分析方法
CN113160417B (zh) * 2021-04-13 2022-08-05 同济大学 一种基于泌尿***的多器官三维重建控制方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090171627A1 (en) * 2007-12-30 2009-07-02 Olson Eric S System and Method for Surface Reconstruction from an Unstructured Point Set
CN102592311A (zh) * 2011-03-10 2012-07-18 上海大学 利用vtk进行肠道三维重建的方法
CN104504760A (zh) * 2014-12-09 2015-04-08 北京畅游天下网络技术有限公司 实时更新三维图像的方法和***

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100381104C (zh) * 2005-03-22 2008-04-16 东软飞利浦医疗设备***有限责任公司 一种自动消减边缘伪影的x－射线计算机层析成像机
CN201029876Y (zh) * 2007-01-15 2008-03-05 杭州市萧山区中医院 骨科手术导航***
CN100581447C (zh) * 2007-01-15 2010-01-20 杭州市萧山区中医院 骨科手术导航***
CN103732151B (zh) * 2011-08-19 2016-08-17 株式会社日立制作所 医用图像装置以及医用图像构成方法
CN102920477B (zh) * 2012-03-05 2015-05-20 杭州弘恩医疗科技有限公司 医学影像的目标区域边界确定装置和方法
CN103679810B (zh) * 2013-12-26 2017-03-08 海信集团有限公司 肝部ct图像的三维重建方法
CN104463853A (zh) * 2014-11-22 2015-03-25 四川大学 一种基于图像分割的阴影检测与去除算法
CN104809736B (zh) * 2015-05-12 2017-09-12 河海大学常州校区 基于先验知识的医学断层图像闭合骨骼轮廓计算方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090171627A1 (en) * 2007-12-30 2009-07-02 Olson Eric S System and Method for Surface Reconstruction from an Unstructured Point Set
CN102592311A (zh) * 2011-03-10 2012-07-18 上海大学 利用vtk进行肠道三维重建的方法
CN104504760A (zh) * 2014-12-09 2015-04-08 北京畅游天下网络技术有限公司 实时更新三维图像的方法和***

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109859304A (zh) * 2018-11-16 2019-06-07 华中科技大学同济医学院附属同济医院 三维打印技术在角膜缘组织体外建立三维结构数字化模型中的应用
CN109859304B (zh) * 2018-11-16 2023-08-04 华中科技大学同济医学院附属同济医院 三维打印技术在角膜缘组织体外建立三维结构数字化模型
CN112396609A (zh) * 2019-07-31 2021-02-23 上海正雅齿科科技股份有限公司 一种牙列分割方法、牙齿分割方法、装置及电子设备
CN112396609B (zh) * 2019-07-31 2023-08-08 正雅齿科科技(上海)有限公司 一种牙列分割方法、牙齿分割方法、装置及电子设备

Also Published As

Publication number Publication date
CN106408648A (zh) 2017-02-15

Similar Documents

Publication Publication Date Title
WO2017020580A1 (zh) 一种医学组织的切片图像三维重建的方法及设备
Correa et al. Feature aligned volume manipulation for illustration and visualization
CN104794758B (zh) 一种三维图像的裁剪方法
Beichel et al. Liver segmentation in contrast enhanced CT data using graph cuts and interactive 3D segmentation refinement methods
CN108242077A (zh) 快速渲染二次曲面
Huang et al. Visualizing industrial CT volume data for nondestructive testing applications
AU2019430369B2 (en) VRDS 4D medical image-based vein Ai endoscopic analysis method and product
Huang et al. 3D reconstruction and visualization from 2D CT images
CN107507212A (zh) 数字脑可视化方法、装置、计算设备及存储介质
Wang et al. Three-dimensional reconstruction of jaw and dentition CBCT images based on improved marching cubes algorithm
US20110090222A1 (en) Visualization of scaring on cardiac surface
CN108510580A (zh) 一种椎骨ct图像三维可视化方法
US20190096118A1 (en) Imaging system and method
WO2021030995A1 (zh) 基于vrds ai下腔静脉影像的分析方法及产品
Tan et al. Computer assisted system for precise lung surgery based on medical image computing and mixed reality
JP2012085833A (ja) ３次元医用画像データの画像処理システム、その画像処理方法及びプログラム
Dai et al. Volume‐Rendering‐Based Interactive 3D Measurement for Quantitative Analysis of 3D Medical Images
WO2021081846A1 (zh) 静脉血管肿瘤影像处理方法及相关产品
WO2020168695A1 (zh) 基于VRDS 4D医学影像的肿瘤与血管Ai处理方法及产品
CN109147035B (zh) 三维模型的显示方法及***
Liu et al. Optimization of reconstruction of 2D medical images based on computer 3D reconstruction technology
WO2020168694A1 (zh) 基于VRDS 4D医学影像的肿瘤Ai处理方法及产品
Adeshina et al. CAHECA: computer aided hepatocellular carcinoma therapy planning
Lv et al. Interactive curved planar reformation based on snake model
Kirmizibayrak et al. Interactive focus+ context medical data exploration and editing

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16832069

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16832069

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 16832069

Country of ref document: EP

Kind code of ref document: A1