WO2017028516A1 - 一种三维图像的校准方法、装置和系统 - Google Patents

一种三维图像的校准方法、装置和系统 Download PDF

Info

Publication number
WO2017028516A1
WO2017028516A1 PCT/CN2016/074291 CN2016074291W WO2017028516A1 WO 2017028516 A1 WO2017028516 A1 WO 2017028516A1 CN 2016074291 W CN2016074291 W CN 2016074291W WO 2017028516 A1 WO2017028516 A1 WO 2017028516A1
Authority
WO
WIPO (PCT)
Prior art keywords
dimensional
dimensional image
image
mapping
images
Prior art date
Application number
PCT/CN2016/074291
Other languages
English (en)
French (fr)
Inventor
刘丽丽
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 WO2017028516A1 publication Critical patent/WO2017028516A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30056—Liver; Hepatic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular

Definitions

  • the present invention relates to the field of image processing technologies, and in particular, to a method, device and system for calibrating a three-dimensional image.
  • the prior art adopts the conventional abdominal CT (English full name: Computed Tomography) to enhance the three-phase dynamic scanning.
  • Three phases of arterial phase, portal phase and balance phase, and three-dimensional reconstruction techniques were used to image the hepatic artery, portal vein and hepatic vein, and the distribution and variation of the three in the liver were analyzed.
  • the resection has important guiding significance.
  • the quality of the CT image is not high. Therefore, when the three-dimensional reconstruction technique is used to image the hepatic artery, portal vein and hepatic vein, the obtained image is obtained.
  • Three-dimensional vascular images often have vascular tissue, such as overlap of hepatic artery, portal vein and hepatic vein, and other non-vascular tissue confounding, so that the three-dimensional model is not accurate enough, and the medical reference value is reduced.
  • Embodiments of the present invention provide a method, an apparatus, and a system for calibrating a three-dimensional image, which are used to solve the problem that the three-dimensional reconstruction object in the existing three-dimensional image is overlapped and mixed, resulting in an inaccurate three-dimensional model establishment.
  • an embodiment of the present invention provides a method for calibrating a three-dimensional image, including:
  • mapping contour to a position of a tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image
  • the second three-dimensional image is obtained by three-dimensional reconstruction according to the two-dimensional data included in the adjusted map contour.
  • an embodiment of the present invention provides a calibration apparatus for a three-dimensional image, including:
  • An acquiring unit configured to map the first three-dimensional image onto a set of two-dimensional images, and acquire a mapping contour line of the first three-dimensional image on the first two-dimensional image in the set of two-dimensional images;
  • a determining unit configured to determine whether a mapping contour of the first three-dimensional image on the first two-dimensional image and a tissue edge of the three-dimensional reconstructed object in the corresponding first two-dimensional image are coincident;
  • Adjusting unit if not overlapping, adjusting the mapping contour to a position of a tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image
  • the three-dimensional reconstruction unit is configured to obtain a second three-dimensional image by three-dimensional reconstruction according to the two-dimensional data included in the adjusted mapping contour.
  • an embodiment of the present invention provides a calibration system for a three-dimensional image, including: a calibration device and a display for a three-dimensional image according to the second aspect;
  • the display includes a first display area for displaying the first three-dimensional image or the second three-dimensional image, and a second display area for displaying the A mapping contour of the first two-dimensional image and the first three-dimensional image on the first two-dimensional image.
  • mapping the first three-dimensional image onto the first two-dimensional image in a set of two-dimensional images by mapping the first three-dimensional image onto a set of two-dimensional images a contour line; determining whether a mapping contour of the first three-dimensional image on the first two-dimensional image coincides with a tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image; if not overlapping, adjusting the mapping contour to a corresponding one
  • the position of the tissue edge of the three-dimensional reconstruction object in the first two-dimensional image changes the two-dimensional data category used by the three-dimensional reconstruction model, and then the second-dimensional reconstruction is obtained according to the two-dimensional data included in the adjusted mapping contour.
  • the three-dimensional image realizes the correction of the first three-dimensional image, so that the second three-dimensional image after the three-dimensional reconstruction is more Accurate, in order to meet the actual modeling requirements, solves the problem that the three-dimensional reconstruction objects overlap in the three-dimensional images in the prior art, resulting in the inaccuracy of the three-dimensional model establishment.
  • FIG. 1 is a schematic flowchart 1 of a method for calibrating a three-dimensional image according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of a coordinate axis according to an embodiment of the present invention.
  • FIG. 3 is a schematic flowchart 2 of a method for calibrating a three-dimensional image according to an embodiment of the present invention
  • FIG. 4 is a schematic diagram 1 of a three-dimensional image calibration apparatus according to an embodiment of the present invention.
  • FIG. 5 is a second schematic diagram of a three-dimensional image calibration apparatus according to an embodiment 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 will understand that the words “first”, “second” and the like do not limit the number and order of execution.
  • An embodiment of the present invention provides a method for calibrating a three-dimensional image. Referring to FIG. 1, the method includes the following steps:
  • Step 101 Mapping the first three-dimensional image onto a set of two-dimensional images, and acquiring the first three-dimensional image A mapping outline on a first two-dimensional image in a set of two-dimensional images.
  • a set of two-dimensional images includes two-dimensional images of the same period and the same aspect type, and the same period includes an arterial phase, a portal phase or a balanced phase, and the type of the facet includes a shaft surface, a coronal plane or a sagittal plane.
  • the arterial phase, the portal phase and the equilibrium phase are terminology used in MRI scans, which specifically refers to the three periods of contrast agent development after the contrast agent portal vein is injected into the blood vessel.
  • the arterial phase is characterized by the contrast agent flowing into the artery.
  • the portal phase is characterized by blood passing through the portal vein at about 40 s after the arterial phase. The equilibrium period is shown after these two periods.
  • the coronal plane or the sagittal plane more clearly, as shown in Fig. 2, when the human body is in the lying position, the X axis is positively directed to the left side of the human body, and the Y axis is positively directed to the back of the human body. The Z axis is pointing forward to the head of the human body.
  • the axial plane (also referred to as a cross section) is a plane formed by cutting the human body vertically in a direction from the head to the foot when the human body is in a lying position, that is, the axial plane is with the Z a plurality of planes perpendicular to the axis, the axial direction of the axial plane is Z;
  • the coronal plane is a layer of the surface of the human body which is cut horizontally from the upper surface to the lower surface when the human body is in the lying position, that is, the coronal plane It is a plurality of faces perpendicular to the Y-axis, and the axial direction of the coronal plane is the Y-axis direction;
  • the sagittal plane is when the human body is in a lying position, and the human body is cut in layers from right to left along both sides of the human body.
  • the face that is, the sagittal plane
  • the face is a plurality of faces perpendicular to the X-axis, and the axial direction of the sagittal plane is the X-axis direction.
  • Two-dimensional images of the same period and the same aspect type can be marked with different identifiers in order to facilitate subsequent use.
  • step 101 specifically includes:
  • the tissue edge line of the three-dimensional reconstructed object in the first three-dimensional image is obtained according to the number of layers, and the tissue edge line and the number of layers of the layer are stored, and the coordinate values of the points constituting the edge lines of the tissues are obtained, according to the points.
  • the coordinate values are plotted on the corresponding two-dimensional image to obtain a mapping contour of the first three-dimensional image.
  • Step 102 Determine whether a mapping contour of the first three-dimensional image on the first two-dimensional image coincides with a tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image.
  • the tissue edge of the three-dimensional reconstruction object in the first two-dimensional image can be calculated by the boundary region algorithm. Whether the coincidence is determined by comparing the projected contour of the projection with the edge of the tissue calculated by the algorithm. If not, step 103 is performed.
  • Step 103 If not coincident, adjust the mapping contour to a position where the tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image is located.
  • step 103 specifically includes:
  • the calibration instruction may be that the user (such as a medical staff) drags and stretches the first blood vessel mapping contour by clicking a mouse, and then aligns the mapping contour to the corresponding first two-dimensional image according to the calibration instruction.
  • the position of the tissue edge of the three-dimensional reconstruction object is to align the mapping contour to coincide with the tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image. Since jitter occurs when artificially calibrated, an algorithm with anti-shake can be employed to align the mapping contour as much as possible to coincide with the tissue edge of the three-dimensional reconstructed object in the corresponding first two-dimensional image.
  • the calibration instruction may also be performed when the device (such as the CPU) detects that the mapping contour of the first three-dimensional image on the first two-dimensional image does not coincide with the tissue edge of the three-dimensional reconstructed object in the corresponding first two-dimensional image. And calibrating the first blood vessel mapping contour to a position corresponding to the original contour of the blood vessel in the corresponding two-dimensional blood vessel image, that is, calibrating the mapping contour to the corresponding first two-dimensional image according to the calibration instruction The tissue edges of the 3D reconstruction object coincide. It should be noted that, in the process of adjusting, the mapping contour may be calibrated to the tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image according to the pixel value comparison method. Hehe.
  • Step 104 Perform three-dimensional reconstruction to obtain a second three-dimensional image according to the two-dimensional data included in the adjusted mapping contour.
  • mapping contour Since the mapping contour is adjusted, the two-dimensional data category used in the three-dimensional reconstruction model is changed, and then the second three-dimensional image is reconstructed according to the two-dimensional data included in the adjusted mapping contour, and the first is realized.
  • the correction of the three-dimensional image makes the second three-dimensional image reconstructed again more accurately, in order to meet the actual modeling requirements, and solves the problem that the three-dimensional reconstruction object overlaps in the three-dimensional image in the prior art, which leads to the inaccurate establishment of the three-dimensional model.
  • the three-dimensional blood vessel image of the liver organ will be calibrated according to the calibration method of the three-dimensional image shown in FIG. 1 as an example.
  • 3D reconstruction image is reconstructed based on all CT scan images of a certain period, that is, 3D reconstruction can select all CT scan images of the arterial phase, 3D reconstruction can obtain 3D blood vessel images of hepatic artery; also select all the portal phase CT scan images, three-dimensional reconstruction to obtain three-dimensional blood vessel images of the three portal veins; all CT scan images of the equilibrium period can also be selected, and three-dimensional blood vessel images of the hepatic veins can be obtained by three-dimensional reconstruction.
  • the CT scan image of the hepatic artery, portal phase and balance region can be segmented by the region growing algorithm (English name: region growing) to obtain the blood vessel segmentation sequence.
  • the region growing algorithm (English name: region growing) is to collect pixels with similar properties to form a region.
  • each two-dimensional blood vessel image can be divided into a plurality of regions, and thus, a blood vessel segmentation sequence of each two-dimensional blood vessel image is obtained.
  • the medical staff can confirm the anatomical structure of the blood vessel in the liver and the shape of the blood vessel, and confirm the obvious non-vascular tissue.
  • the medical staff can draw a closed The curve defines the non-vascular tissue, and after clearing and subtracting the circled region, a relatively accurate first three-dimensional blood vessel image is obtained, and the first accurate three-dimensional blood vessel image is mapped to the method shown in FIG.
  • the mapping contour of the first three-dimensional blood vessel image on each CT scan image is compared with the corresponding Whether the tissue edges of the blood vessels overlap in the CT scan image, if not coincident, manually drag and stretch the mapping contour, until the mapping contour is adjusted to the tissue edge of the blood vessel in the corresponding CT scan image.
  • the medical staff can classify the second three-dimensional blood vessel image. Before the classification, the medical staff can identify the incorrect connection position with a sphere or other shape according to the artificial anatomy, obtain the spatial coordinates of the identified position, and store the data of the second three-dimensional blood vessel image data.
  • the coordinate values of this part are emptied, thus breaking the incorrect connection position, thereby classifying the blood vessels in the second three-dimensional blood vessel image, and the user can determine the spatial position of the blood vessel segment in the second three-dimensional blood vessel image, and The actual vascular system anatomical location of the internal organs determines whether the vascular branch is an artery, a vein or a branch of the portal vein, and is identified by a different color after confirmation, which is convenient for the user to observe. According to the user's choice, the user can operate the space, which can improve the accuracy of the classification, so as to provide accurate reference for subsequent medical operations, and improve the medical reference value.
  • An embodiment of the present invention provides a method for calibrating a three-dimensional image, by mapping a first three-dimensional image onto a set of two-dimensional images, and acquiring a mapping contour of the first three-dimensional image on the first two-dimensional image in a set of two-dimensional images.
  • the position of the tissue edge of the 3D reconstructed object in a 2D image changes the 2D data category used by the 3D reconstruction model, and then the 2D data is reconstructed according to the 2D data contained in the adjusted map contour.
  • the image realizes the correction of the first three-dimensional image, so that the second three-dimensional image after the three-dimensional reconstruction is more accurate, in order to meet the actual modeling requirements, and solves the overlapping of the three-dimensional reconstruction objects in the three-dimensional image in the prior art, resulting in three-dimensional The model is not built accurately enough.
  • the embodiment of the present invention provides a calibration apparatus for a three-dimensional image, and the respective functional units of the calibration apparatus correspond to the calibration method of the three-dimensional image in the first embodiment.
  • the three-dimensional image calibration apparatus 40 includes:
  • the acquiring unit 401 is configured to map the first three-dimensional image onto a set of two-dimensional images, and acquire a mapping contour line of the first three-dimensional image on the first two-dimensional image in the set of two-dimensional images;
  • a determining unit 402 configured to determine whether a mapping contour of the first three-dimensional image on the first two-dimensional image and a tissue edge of the three-dimensional reconstructed object in the corresponding first two-dimensional image are coincident;
  • the adjusting unit 403 is configured to adjust the mapping contour to a position where the tissue edge of the three-dimensional reconstruction object is located in the corresponding first two-dimensional image if not overlapping;
  • the three-dimensional reconstruction unit 404 is configured to perform three-dimensional reconstruction to obtain a second three-dimensional image according to the two-dimensional data included in the adjusted mapping contour.
  • a set of two-dimensional images includes a two-dimensional image of the same period and the same slice type, and the same period includes an arterial phase, a portal phase, or an equilibrium phase, and the slice type includes a shaft surface, a coronal plane, or a sagittal plane.
  • the obtaining unit 401 includes:
  • a first obtaining sub-unit 401a configured to acquire a tissue edge of the three-dimensional reconstructed object in the first three-dimensional image when the number of layers is different, and the number of layers corresponds to the number of two-dimensional images in the two-dimensional image;
  • a second obtaining subunit 401b configured to acquire a group of three-dimensional reconstructed objects in the first three-dimensional image Position information of the woven edge
  • the drawing sub-unit 401c is configured to draw a mapping contour of the first three-dimensional image on the corresponding first two-dimensional image according to the position information acquired by the second acquiring sub-unit 401b.
  • the adjusting unit 403 includes:
  • a third obtaining subunit 403a configured to acquire a calibration instruction
  • the adjustment sub-unit 403b is configured to adjust the mapping contour to the position of the tissue edge of the three-dimensional reconstruction object in the corresponding first two-dimensional image according to the calibration instruction acquired by the third acquisition sub-unit 403a.
  • An embodiment of the present invention provides a calibration apparatus for a three-dimensional image, by mapping a first three-dimensional image onto a set of two-dimensional images, and acquiring a mapping contour of the first three-dimensional image on the first two-dimensional image in a set of two-dimensional images.
  • the position of the tissue edge of the 3D reconstructed object in a 2D image changes the 2D data category used by the 3D reconstruction model, and then the 2D data is reconstructed according to the 2D data contained in the adjusted map contour.
  • the image realizes the correction of the first three-dimensional image, so that the second three-dimensional image after the three-dimensional reconstruction is more accurate, in order to meet the actual modeling requirements, and solves the overlapping of the three-dimensional reconstruction objects in the three-dimensional image in the prior art, resulting in three-dimensional The model is not built accurately enough.
  • the embodiment of the present invention further provides a calibration system for a three-dimensional image, comprising: the calibration device and the display of the three-dimensional image described in the second embodiment; wherein the display comprises a first display area and a second display area, the first display area For displaying the first three-dimensional image or the second three-dimensional image, the second display area is configured to display a mapping contour of the first two-dimensional image and the first three-dimensional image on the first two-dimensional image.
  • the two-dimensional image required for display modeling and the first three-dimensional image and the second three-dimensional image obtained by three-dimensional reconstruction are displayed for easy visual comparison by the user.
  • the foregoing program may be stored in a computer readable storage medium, and the program is executed when executed. Including the steps of the above method embodiments;
  • the foregoing storage medium includes various media that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

本发明实施例提供了一种三维图像的校准方法、装置和系统，涉及图像处理技术领域，用以解决现有三维图像中三维重建对象出现重叠混杂，导致三维模型建立不够准确的问题。该三维图像的校准方法包括：将第一三维图像映射到一组二维图像上，获取第一三维图像在一组二维图像中第一二维图像上的映射轮廓线；确定第一三维图像在第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；若未重合，则将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置；根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像。本发明应用于对医疗中的三维图像处理。

Description

一种三维图像的校准方法、装置和系统
本申请要求于2015年08月18日提交中国专利局、申请号为201510509147.8、发明名称为“一种三维图像的校准方法、装置和系统”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
本发明涉及图像处理技术领域，尤其涉及一种三维图像的校准方法、装置和系统。
背景技术
随着医疗技术水平的不断提高，为了更加精准的获取病人的病变位置，现有技术中采用对人体进行常规腹部CT(英文全称：Computed Tomography中文：电子计算机断层扫描)增强三期动态扫描，获得动脉期、门脉期及平衡期的三期图像，进而采用三维重建技术对肝动脉、门脉和肝静脉血管成像，分析三者在肝内的分布结构以及变异，对于肝脏分段、肝脏肿瘤的切除术具有重要的指导意义。
但是，由于医学成像自身以及病人个体肝脏肿瘤的位置和血管的变异等因素，导致CT图像质量并不高，因此，使用三维重建技术对肝动脉、门脉和肝静脉血管进行成像时，得到的三维血管图像经常会出现血管组织，如肝动脉、门脉和肝静脉的重叠，以及其他非血管组织混杂的现象，从而三维模型建立不够准确，医用参考价值降低。
发明内容
本发明的实施例提供一种三维图像的校准方法、装置和系统，用以解决现有三维图像中三维重建对象出现重叠混杂，导致三维模型建立不够准确的问题。
第一方面，本发明实施例提供了一种三维图像的校准方法，包括：
将第一三维图像映射到一组二维图像上，获取所述第一三维图像在所述一组二维图像中第一二维图像上的映射轮廓线；
确定所述第一三维图像在所述第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；
若未重合，则将所述映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置；
根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像。
第二方面，本发明实施例提供了一种三维图像的校准装置，包括：
获取单元，用于将第一三维图像映射到一组二维图像上，获取所述第一三维图像在所述一组二维图像中第一二维图像上的映射轮廓线；
确定单元，用于确定所述第一三维图像在所述第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；
调整单元，用于若未重合，则将所述映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置；
三维重建单元，用于根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像。
第三方面，本发明实施例提供了一种三维图像的校准系统，包括：第二方面所述的三维图像的校准装置和显示器；
其中，所述显示器包括第一显示区域和第二显示区域，所述第一显示区域用于显示所述第一三维图像或所述第二三维图像，所述第二显示区域用于显示所述第一二维图像和所述第一三维图像在所述第一二维图像上的映射轮廓线。
基于上述所提供的三维图像的校准方法、装置和系统，通过将第一三维图像映射到一组二维图像上，获取第一三维图像在一组二维图像中第一二维图像上的映射轮廓线；确定第一三维图像在第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；若未重合，则将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置，从而改变了三维重建模型所使用的二维数据范畴，进而根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像，实现了对第一三维图像的校正，使得再次三维重建后的第二三维图像更加 准确，以符合实际建模需求，解决了现有技术中三维图像中出现三维重建对象重叠混杂，导致三维模型建立不够准确的问题。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，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三维图像的校准方法的流程示意图一；
图2为本发明实施例提供的坐标轴示意图；
图3为本发明实施例提供的一种三维图像的校准方法的流程示意图二；
图4为本发明实施例提供的一种三维图像的校准装置的示意图一；
图5为本发明实施例提供的一种三维图像的校准装置的示意图二。
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做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”等字样并不对数量和执行次序进行限定。
实施例一
本发明的实施例提供一种三维图像的校准方法，参照图1所示，该方法包括以下步骤：
步骤101、将第一三维图像映射到一组二维图像上，获取第一三维图 像在一组二维图像中第一二维图像上的映射轮廓线。
其中，一组二维图像包括同一期、同一切面类型的二维图像，同一期包括动脉期、门脉期或平衡期，切面类型包括轴状面、冠状面或矢状面。
需要说明的是，动脉期、门脉期和平衡期(或称为延迟期)是核磁共振扫描时用的专业术语，它具体指将造影剂门脉注入血管后，造影剂显影的3个时段。动脉期指造影剂流入动脉后所表现出来的特点，门脉期是在动脉期后40s左右血液经过肝脏门门脉时所表现出的特点，平衡期是在这两个时期之后所表现出的特点。
另外，为了更加清楚的对轴状面、冠状面或矢状面进行解释，如图2所示，人体处于躺位时，X轴正向指向人体的左侧，Y轴正向指向人体的背部，Z轴正向指向人体的头部。其中，轴状面(也可称之为横截面)就是人体处于躺位时，沿着从头到脚这样的方向，垂直的将人体切开形成的一个个面，也即轴状面是与Z轴垂直的多个面，轴状面的轴位方向为Z；冠状面就是人体处于躺位时，从上表面到下表面将人体沿着水平切开的一层层的面，也即冠状面是与Y轴垂直的多个面，冠状面的轴位方向为Y轴方向；矢状面就是人体处于躺位时，沿着人体两侧从右到左的方向，将人体一层层切开的面，也即矢状面是于X轴垂直的多个面，矢状面的轴位方向为X轴方向。对于同一期、同一切面类型的二维图像可以依次采用不同的标识进行标记，以方便后续使用。
由于第一三维图像是由一组二维血管图像进行三维重建得到的，而在三维重建的过程中，由于二维图像的图像质量以及算法等方面的原因，第一三维图像中会出现三维重建对象混杂，导致三维建模不准确的问题，因此，需要对第一三维图像进行映射校准。其中，在本发明的一个优选的实施例中，如图3所示，步骤101具体包括：
101a、获取在不同层数时第一三维图像中三维重建对象的组织边缘，其中，层数与一组二维图像中二维图像的张数相对应。
101b、获取第一三维图像中三维重建对象的组织边缘的位置信息。
101c、根据位置信息，在对应的第一二维图像上绘制得到第一三维图像的映射轮廓线。
示例的，根据层数获取第一三维图像中三维重建对象的组织边缘线，并把该层的组织边缘线和层数进行存储，进而获取组成这些组织边缘线的点的坐标值，根据这些点的坐标值在对应的二维图像上绘制得到第一三维图像的映射轮廓线。
步骤102、确定第一三维图像在第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合。
其中，对于第一二维图像中三维重建对象的组织边缘可以通过边界区域算法计算得到。通过将投影得到的映射轮廓线和算法计算得到的组织边缘进行比对，确定是否重合。若未重合，则执行步骤103。
步骤103、若未重合，则将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置。
在本发明的一个优选的实施例中，如图3所示，步骤103具体包括：
103a、获取校准指令。
103b、根据校准指令，将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置。
其中，校准指令可以是用户(如医护人员)通过点击鼠标对第一血管映射轮廓线进行拖动、拉伸等操作指令，进而根据校准指令，将映射轮廓线校准至对应的第一二维图像中三维重建对象的组织边缘所在的位置，即将映射轮廓线校准至与对应的第一二维图像中三维重建对象的组织边缘重合。由于人为的校准时会出现抖动的情况，因此，可以采用具有防抖动的算法，以尽可能的将映射轮廓线校准至与对应的第一二维图像中三维重建对象的组织边缘重合。
另外，校准指令也可以是设备(如CPU)检测到第一三维图像在第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘未重合时，发出的校准指令，进而根据该校准指令，将第一血管映射轮廓线校准至与对应的二维血管图像中血管的原始轮廓线所在的位置，即将映射轮廓线校准至与对应的第一二维图像中三维重建对象的组织边缘重合。需要说明的是，在进行调整的过程中，可以根据像素值相近比较方法，将映射轮廓线校准至与对应的第一二维图像中三维重建对象的组织边缘重 合。
步骤104、根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像。
由于对映射轮廓线进行调整，即改变了三维重建模型所使用的二维数据范畴，进而根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像，实现了对第一三维图像的校正，使得再次三维重建后的第二三维图像更加准确，以符合实际建模需求，解决了现有技术中三维图像中出现三维重建对象重叠混杂，导致三维模型建立不够准确的问题。
下面，根据图1所示的三维图像的校准方法，以对肝脏器官的三维血管图像进行校准为例进行具体说明。
具体的，在对人体进行常规腹部CT增强三期动态扫描时，获得动脉期、门脉期和平衡期三期图像，每一期中包含有多张CT扫描图像，这些CT扫描图像都是二维图像，三维重建图像是基于某一期的所有CT扫描图像进行重建，也就是三维重建可以选择动脉期的所有CT扫描图像，三维重建得到肝动脉的三维血管图像；也可以选择门脉期的所有CT扫描图像，三维重建得到三门脉的三维血管图像；也可以选择平衡期的所有CT扫描图像，三维重建得到肝静脉的三维血管图像。
对于如何从CT扫描图像到得到肝脏血管的三维图像，可以采用区域生长算法(英文全称：region growing)对肝脏动脉期、门脉期和平衡区的CT扫描图像进行血管分割，得到血管分割序列，然后再采用移动立方体(英文全称：Marching Cubes，简称MC)算法对同一期所有二维血管图像的血管分割序列进行三维重建，得到第一三维血管图像。其中，区域生长算法(英文全称：region growing)是将具有相似性质的像素集合起来构成区域。首先在需要分割的区域中选取一个种子点(x，y)作为生长点，然后根据预先定义的规则将种子点周围领域中的相似像素合并到种子点像素所属的区域中，将这些新像素作为新的种子像素继续进行上述过程，直到没有满足条件的像素点被包括进来，这样一个区域就生成完成。通过区域生长算法对每个二维血管图像进行血管分割后，可以将每个二维血管图像分割成多个区域，因而，得到每个二维血管图像的血管分割序列。
但是，由于在对每个二维血管图像根据区域生长算法进行血管分割时，有时会把不属于血管区域的相似像素点也归类到血管区域，导致得到的第一三维血管图像中或出现肝动脉、门静脉和肝静脉重叠的问题，以及其他非血管组织的混杂，导致第一三维血管图像不够准确，医用参考价值低。因此，需要将第一三维血管图像进行校准，以符合实际的内脏器官的血管系统。
示例的，在将第一三维血管图像进行映射之前，医护人员可以根基肝脏内血管的解剖结构以及血管的走形，确认出明显的非血管组织，对于上述非血管组织，医护人员可以绘制闭合的曲线将该非血管组织圈定出来，对于圈定出的区域，进行清空消减后，得到较为准确的第一三维血管图像，进而参照图1所示的方法，将较为准确的第一三维血管图像映射到动脉期的所有CT扫描图像上，若第一三维血管图像在每张CT扫描图像上均有映射轮廓线，那么，比较第一三维血管图像在每张CT扫描图像上的映射轮廓线与对应的CT扫描图像中血管的组织边缘是否重合，若未重合，则手动对映射轮廓线进行拖动、拉伸等编辑操作，直至将映射轮廓线调整至对应的CT扫描图像中血管的组织边缘所在的位置，对映射轮廓线进行调整之后，再次使用移动立方体算法进行三维血管重建，得到更符合实际肝脏血管系统的第二三维血管图像。
由于进行校准后得到的第二三维血管图像更加符合实际的肝脏血管系统，因此，医护人员可以对第二三维血管图像进行分类。在进行分类之前，医护人员可以根据人工解剖学知识，对于不正确的连接位置，用球体或者其他形状标识出，取得标识出的位置的空间坐标，在存储第二三维血管图像数据的数据集中把这部分的坐标值清空，这样就断开了不正确的连接位置，进而对第二三维血管图像中的血管进行分类标识，用户可以根据血管支段在第二三维血管图像中的空间位置、以及实际的内脏器官的血管系统解剖学位置，确定该血管支段是动脉、静脉还是门脉的支段，确定之后用不同的颜色标识出来，便于用户观察。根据用户的选择进行分类，给予用户可操作的空间，能够提高分类的准确性，从而为后续的医学手术等提供准确的参考，提高医用参考价值。
本发明实施例提供了一种三维图像的校准方法，通过将第一三维图像映射到一组二维图像上，获取第一三维图像在一组二维图像中第一二维图像上的映射轮廓线；确定第一三维图像在第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；若未重合，则将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置，从而改变了三维重建模型所使用的二维数据范畴，进而根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像，实现了对第一三维图像的校正，使得再次三维重建后的第二三维图像更加准确，以符合实际建模需求，解决了现有技术中三维图像中出现三维重建对象重叠混杂，导致三维模型建立不够准确的问题。
实施例二
本发明实施例提供了一种三维图像的校准装置，该校准装置中各个功能单元与实施例一中三维图像的校准方法相对应，具体可以参考实施例一中的描述，在此不再赘述。如图4所示，该三维图像的校准装置40包括：
获取单元401，用于将第一三维图像映射到一组二维图像上，获取第一三维图像在一组二维图像中第一二维图像上的映射轮廓线；
确定单元402，用于确定第一三维图像在第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；
调整单元403，用于若未重合，则将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置；
三维重建单元404，用于根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像。
可选的，一组二维图像中包括同一期、同一切面类型的二维图像，同一期包括动脉期、门脉期或平衡期，切面类型包括轴状面、冠状面或矢状面。
可选的，如图5所示，获取单元401包括：
第一获取子单元401a，用于获取在不同层数时第一三维图像中三维重建对象的组织边缘，层数与一组二维图像中二维图像的张数相对应；
第二获取子单元401b，用于获取第一三维图像中三维重建对象的组 织边缘的位置信息；
绘制子单元401c，用于根据第二获取子单元401b获取的位置信息，在对应的第一二维图像上绘制得到第一三维图像的映射轮廓线。
可选的，如图5所示，调整单元403包括：
第三获取子单元403a，用于获取校准指令；
调整子单元403b，用于根据第三获取子单元403a获取的校准指令，将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置。
本发明实施例提供了一种三维图像的校准装置，通过将第一三维图像映射到一组二维图像上，获取第一三维图像在一组二维图像中第一二维图像上的映射轮廓线；确定第一三维图像在第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；若未重合，则将映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置，从而改变了三维重建模型所使用的二维数据范畴，进而根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像，实现了对第一三维图像的校正，使得再次三维重建后的第二三维图像更加准确，以符合实际建模需求，解决了现有技术中三维图像中出现三维重建对象重叠混杂，导致三维模型建立不够准确的问题。
本发明实施例还提供了一种三维图像的校准系统，包括：实施例二中所述的三维图像的校准装置和显示器；其中，显示器包括第一显示区域和第二显示区域，第一显示区域用于显示第一三维图像或第二三维图像，第二显示区域用于显示第一二维图像和第一三维图像在第一二维图像上的映射轮廓线。这样，通过显示器建模所需的二维图像和三维重建得到的第一三维图像和第二三维图像进行显示，便于用户直观的进行比对。需要说明的是，对于系统实施例而言，由于其与方法实施例基本相似，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤； 而前述的存储介质包括：ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
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 (9)

  1. 一种三维图像的校准方法，其特征在于，包括：
    将第一三维图像映射到一组二维图像上，获取所述第一三维图像在所述一组二维图像中第一二维图像上的映射轮廓线；
    确定所述第一三维图像在所述第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；
    若未重合，则将所述映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置；
    根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像。
  2. 根据权利要求1所述的方法，其特征在于，所述一组二维图像中包括同一期、同一切面类型的二维图像，所述同一期包括动脉期、门脉期或平衡期，所述切面类型包括轴状面、冠状面或矢状面。
  3. 根据权利要求1或2所述的方法，其特征在于，所述将第一三维图像映射到一组二维图像上，获取所述第一三维图像在所述一组二维图像中第一二维图像上的映射轮廓线包括：
    获取在不同层数时所述第一三维图像中所述三维重建对象的组织边缘，所述层数与所述一组二维图像中二维图像的张数相对应；
    获取所述第一三维图像中所述三维重建对象的组织边缘的位置信息；
    根据所述位置信息，在对应的所述第一二维图像上绘制得到所述第一三维图像的映射轮廓线。
  4. 根据权利要求1所述的方法，其特征在于，所述将所述映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置包括：
    获取校准指令；
    根据所述校准指令，将所述映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置。
  5. 一种三维图像的校准装置，其特征在于，包括：
    获取单元，用于将第一三维图像映射到一组二维图像上，获取所述第 一三维图像在所述一组二维图像中第一二维图像上的映射轮廓线；
    确定单元，用于确定所述第一三维图像在所述第一二维图像上的映射轮廓线与对应的第一二维图像中三维重建对象的组织边缘是否重合；
    调整单元，用于若未重合，则将所述映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置；
    三维重建单元，用于根据调整后的映射轮廓线所包含的二维数据，三维重建得到第二三维图像。
  6. 根据权利要求5所述的装置，其特征在于，所述一组二维图像中包括同一期、同一切面类型的二维图像，所述同一期包括动脉期、门脉期或平衡期，所述切面类型包括轴状面、冠状面或矢状面。
  7. 根据权利要求5或6所述的装置，其特征在于，所述获取单元包括：
    第一获取子单元，用于获取在不同层数时所述第一三维图像中所述三维重建对象的组织边缘，所述层数与所述一组二维图像中二维图像的张数相对应；
    第二获取子单元，用于获取所述第一三维图像中所述三维重建对象的组织边缘的位置信息；
    绘制子单元，用于根据所述第二获取子单元获取的位置信息，在对应的所述第一二维图像上绘制得到所述第一三维图像的映射轮廓线。
  8. 根据权利要求5所述的装置，其特征在于，所述调整单元包括：
    第三获取子单元，用于获取校准指令；
    调整子单元，用于根据所述第三获取子单元获取的校准指令，将所述映射轮廓线调整至对应的第一二维图像中三维重建对象的组织边缘所在的位置。
  9. 一种三维图像的校准系统，其特征在于，包括：权利要求5-8任一项所述的三维图像的校准装置和显示器；
    其中，所述显示器包括第一显示区域和第二显示区域，所述第一显示区域用于显示所述第一三维图像或所述第二三维图像，所述第二显示区域用于显示所述第一二维图像和所述第一三维图像在所述第一二维图像上的 映射轮廓线。
PCT/CN2016/074291 2015-08-18 2016-02-22 一种三维图像的校准方法、装置和系统 WO2017028516A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510509147.8A CN106469445A (zh) 2015-08-18 2015-08-18 一种三维图像的校准方法、装置和系统
CN201510509147.8 2015-08-18

Publications (1)

Publication Number Publication Date
WO2017028516A1 true WO2017028516A1 (zh) 2017-02-23

Family

ID=58051909

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/074291 WO2017028516A1 (zh) 2015-08-18 2016-02-22 一种三维图像的校准方法、装置和系统

Country Status (2)

Country Link
CN (1) CN106469445A (zh)
WO (1) WO2017028516A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109658501A (zh) * 2018-12-21 2019-04-19 Oppo广东移动通信有限公司 一种图像处理方法、图像处理装置及终端设备
CN114663592A (zh) * 2022-03-24 2022-06-24 河北彩居科技有限公司 基于医学图像三维重建的模型制作方法及装置
CN116542977A (zh) * 2023-07-06 2023-08-04 宁德时代新能源科技股份有限公司 图像处理方法、装置、设备、存储介质和程序产品

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109118501A (zh) * 2018-08-03 2019-01-01 上海电气集团股份有限公司 图像处理方法及系统
CN114820731A (zh) * 2022-03-10 2022-07-29 青岛海信医疗设备股份有限公司 Ct影像和三维体表图像的配准方法及相关装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102411794A (zh) * 2011-07-29 2012-04-11 南京大学 一种基于球谐变换的三维模型的二维投影的输出方法
CN103035009A (zh) * 2012-12-19 2013-04-10 长春工业大学 一种基于ct影像的肺结节边缘重建与分割方法
CN103679805A (zh) * 2013-12-19 2014-03-26 苏州大学 基于三维主动轮廓模型和三维哈弗变换的三维肾皮质定位方法
CN103747740A (zh) * 2011-08-17 2014-04-23 日立阿洛卡医疗株式会社 超声波数据处理设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8155411B2 (en) * 2008-07-22 2012-04-10 Pie Medical Imaging B.V. Method, apparatus and computer program for quantitative bifurcation analysis in 3D using multiple 2D angiographic images
TW201015491A (en) * 2008-10-03 2010-04-16 Himax Tech Ltd 3D depth generation by vanishing line detection
US9070019B2 (en) * 2012-01-17 2015-06-30 Leap Motion, Inc. Systems and methods for capturing motion in three-dimensional space
CN104657984B (zh) * 2015-01-28 2018-10-16 复旦大学 三维超声乳腺全容积图像感兴趣区域的自动提取方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102411794A (zh) * 2011-07-29 2012-04-11 南京大学 一种基于球谐变换的三维模型的二维投影的输出方法
CN103747740A (zh) * 2011-08-17 2014-04-23 日立阿洛卡医疗株式会社 超声波数据处理设备
CN103035009A (zh) * 2012-12-19 2013-04-10 长春工业大学 一种基于ct影像的肺结节边缘重建与分割方法
CN103679805A (zh) * 2013-12-19 2014-03-26 苏州大学 基于三维主动轮廓模型和三维哈弗变换的三维肾皮质定位方法

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109658501A (zh) * 2018-12-21 2019-04-19 Oppo广东移动通信有限公司 一种图像处理方法、图像处理装置及终端设备
CN114663592A (zh) * 2022-03-24 2022-06-24 河北彩居科技有限公司 基于医学图像三维重建的模型制作方法及装置
CN116542977A (zh) * 2023-07-06 2023-08-04 宁德时代新能源科技股份有限公司 图像处理方法、装置、设备、存储介质和程序产品
CN116542977B (zh) * 2023-07-06 2024-02-06 宁德时代新能源科技股份有限公司 图像处理方法、装置、设备、存储介质和程序产品

Also Published As

Publication number Publication date
CN106469445A (zh) 2017-03-01

Similar Documents

Publication Publication Date Title
JP6851359B2 (ja) ボリューム測定位相コントラストｍｒｉによる総合的心血管解析
US20240013474A1 (en) Treatment procedure planning system and method
CN104622495B (zh) 医用图像处理装置以及医用图像处理方法
CN105427311B (zh) 用于交互式肝脏血管和胆管系统评估的方法、系统、装置和计算机程序产品
EP2157905B1 (en) A method for tracking 3d anatomical and pathological changes in tubular-shaped anatomical structures
JP5129480B2 (ja) 管状臓器の３次元再構成を行うシステム及び血管撮像装置の作動方法
CA2905050C (en) Method and system to facilitate intraoperative positioning and guidance
JP6080249B2 (ja) ３次元画像表示装置および方法並びにプログラム
WO2017028516A1 (zh) 一种三维图像的校准方法、装置和系统
US8611989B2 (en) Multi-planar reconstruction lumen imaging method and apparatus
US11132801B2 (en) Segmentation of three-dimensional images containing anatomic structures
WO2017028519A1 (zh) 一种肝脏血管的分类方法
US11017531B2 (en) Shell-constrained localization of vasculature
JP2010528750A (ja) 管状構造の検査
CN112862833A (zh) 血管分段方法、电子装置和存储介质
US20120224752A1 (en) Image-based diagnosis assistance apparatus, its operation method and program
JP4122463B2 (ja) 医療用可視画像の生成方法
Ballocca et al. Validation of quantitative 3-dimensional transesophageal echocardiography mitral valve analysis using stereoscopic display
US20120169735A1 (en) Improvements to curved planar reformation
US8938107B2 (en) System and method for automatic segmentation of organs on MR images using a combined organ and bone atlas
CN113645896A (zh) 手术计划、手术导航和成像用系统
CN114677436A (zh) 基于模型配准的脑部cta图像中的脑血管自动定位方法
Sørensen et al. A new virtual reality approach for planning of cardiac interventions
CN108510506A (zh) 一种管状结构图像分割方法
KR20140120236A (ko) 심근 및 심혈관 정보의 통합 분석 방법

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: 16836381

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: 16836381

Country of ref document: EP

Kind code of ref document: A1