WO2020133236A1 - 一种脊柱的成像方法以及超声成像*** - Google Patents

一种脊柱的成像方法以及超声成像*** Download PDF

Info

Publication number
WO2020133236A1
WO2020133236A1 PCT/CN2018/124950 CN2018124950W WO2020133236A1 WO 2020133236 A1 WO2020133236 A1 WO 2020133236A1 CN 2018124950 W CN2018124950 W CN 2018124950W WO 2020133236 A1 WO2020133236 A1 WO 2020133236A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
image
volume data
anatomical structure
spine
Prior art date
Application number
PCT/CN2018/124950
Other languages
English (en)
French (fr)
Inventor
贾洪飞
梁天柱
林穆清
邹耀贤
陈志杰
Original Assignee
深圳迈瑞生物医疗电子股份有限公司
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 深圳迈瑞生物医疗电子股份有限公司 filed Critical 深圳迈瑞生物医疗电子股份有限公司
Priority to PCT/CN2018/124950 priority Critical patent/WO2020133236A1/zh
Priority to CN201880097198.6A priority patent/CN112654301A/zh
Publication of WO2020133236A1 publication Critical patent/WO2020133236A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/13—Tomography
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis

Definitions

  • the present application relates to the field of medical devices, in particular to a spinal imaging method and an ultrasound imaging system.
  • ultrasound examination has a wide range of applications in clinical examination and has become one of the main auxiliary methods for doctors to diagnose diseases.
  • doctors can observe the internal tissue structure of the human body for clinically assisted diagnosis.
  • the spine is a very important structure in fetal development and an important part of prenatal examination.
  • three-dimensional ultrasound has been widely used in fetal spine examination. Its advantage is that it can acquire three-dimensional volume data of the region of interest in one scan, and can display any cut plane in the three-dimensional volume data, and the image is intuitive.
  • the three-dimensional volume data includes three-dimensional volume data of the spine, which can help the doctor locate the abnormal segment of the spine more accurately.
  • the doctor needs to manually rotate and translate the three-dimensional volume data of the spine frequently to achieve a proper observation angle. Then, according to the anatomical structure of the vertebral arch, vertebral body, etc., manually adjust the size and position of the volume of interest (VOI), or use the curved multi-planar reconstruction (CMPR) to manually select the vertebral arch and vertebrae For the body area, obtain standard vertebral arch, vertebral body volume (VR) image or spinal sagittal image. This entire process requires a certain amount of experience from the doctor, which is time-consuming and laborious.
  • VOI volume of interest
  • CMPR curved multi-planar reconstruction
  • Embodiments of the present application provide a spine imaging method and an ultrasound imaging system, which are used to automatically generate and display a target image of a target spine without manual selection, which improves the efficiency and accuracy of ultrasound imaging.
  • An aspect of an embodiment of the present application provides a spinal imaging method, which includes: acquiring three-dimensional volume data of a fetus; identifying a fetal spine image area from the three-dimensional volume data based on the characteristics of the fetal spine Acquiring a target image of the fetal spine according to the image area of the identified fetal spine, wherein the target image includes at least one of a stereoscopic VR image, a two-dimensional slice image, and a multi-surface reconstruction CMPR image; displaying the target image.
  • An aspect of an embodiment of the present application provides a spine imaging method, including: acquiring three-dimensional volume data of a scanning target; determining a predetermined anatomical structure of the target spine from the three-dimensional volume data, wherein the target spine
  • the preset anatomical structure is a partial anatomical structure or all anatomical structures of the target spine; acquiring a target image of the preset anatomical structure, wherein the target image includes a stereoscopic VR image, a two-dimensional slice image, and a multi-surface reconstruction CMPR At least one of the figures; displaying the target image.
  • a spinal imaging method includes: acquiring three-dimensional volume data of a fetus; identifying a spinal cone area from the three-dimensional volume data of the fetus based on the characteristics of the fetal spinal cone; According to the identified spinal cord cone area, the position of the spinal cord cone area is determined; the position of the spinal cord cone area is displayed.
  • An aspect of an embodiment of the present application provides an ultrasound imaging system, including: an ultrasound probe that transmits ultrasound waves to a scanning target and receives ultrasound echoes to obtain ultrasound echo signals; a processor, the processor according to Obtaining the three-dimensional volume data of the target spine by using the ultrasonic echo signal, and determining the preset anatomical structure of the target spine from the three-dimensional volume data, wherein the preset anatomical structure of the target spine is part of the target spine Anatomical structure or all anatomical structures; acquiring a target image of the preset anatomical structure, wherein the target image includes at least one of a stereoscopic VR image, a two-dimensional slice image, and a multi-curved reconstruction CMPR image; a display, the display The target image is displayed.
  • An aspect of an embodiment of the present application provides an ultrasound imaging system, which includes: an ultrasound probe that transmits ultrasound waves to a fetus and receives ultrasound echoes to obtain ultrasound echo signals; a processor, the processing The device obtains the three-dimensional volume data of the fetus according to the ultrasonic echo signal, identifies the spinal cone area from the three-dimensional volume data of the fetus based on the characteristics of the fetal spinal cone, and determines the spinal cone according to the identified spinal cone area The position of the area; a display that displays the position of the spinal cord cone area.
  • An aspect of an embodiment of the present application provides a computer-readable storage medium having instructions stored therein, which when run on a computer, causes the computer to perform the imaging of the spine provided in the first aspect method.
  • the ultrasound imaging system acquires the three-dimensional volume data of the target spine, and determines the preset anatomical structure of the target spine from the three-dimensional volume data, where the preset anatomical structure may be the target spine Part of the anatomical structure may also be the entire structural structure of the target spine.
  • the predetermined anatomical structure may be a vertebral arch or vertebral body in the target spine.
  • the ultrasound imaging system acquires a target image of the preset anatomical structure and displays the target image, where the target image may be a VR image, a two-dimensional standard cross-sectional image, or a CMPR image.
  • the ultrasound imaging system can automatically obtain the target image of the preset anatomical structure of the target spine, and it can be visually displayed without manual selection by the user, improving the efficiency and accuracy of ultrasound imaging, effectively helping the doctor to assist in the diagnosis of the disease, and improving the work effectiveness.
  • FIG. 1 is a schematic structural block diagram of a possible ultrasound imaging system in an embodiment of the present application.
  • FIG. 2 is a schematic diagram of an embodiment of a spinal imaging method in an embodiment of the present application.
  • FIG. 3 is a schematic diagram of an interface display of three-dimensional volume data of a target spine in an embodiment of the present application
  • FIG. 4 is a schematic diagram of an interface display for determining a target spine in three-dimensional volume data in an embodiment of the present application
  • FIG. 5 is a VR diagram of the target spine in the embodiment of the present application.
  • FIG. 6 is a two-dimensional cut-away view of the target spine in an embodiment of the present application.
  • FIG. 7a is a CMPR diagram of the target spine in the embodiment of the present application.
  • FIG. 7b is a CMPR diagram of the vertebral arch in the embodiment of the present application.
  • 7c is a CMPR diagram of the vertebral body in the embodiment of the present application.
  • FIG. 8 is a schematic diagram of the conical region of the spinal cord in the embodiment of the present application.
  • FIG. 1 is a schematic structural block diagram of an ultrasound imaging system 10 in an embodiment of the present application.
  • the ultrasound imaging system 10 may include a probe 100, wherein the probe 100 may be an ultrasound probe, a transmission/reception selection switch 101, a transmission/reception sequence controller 102, a processor 103, and a display 104.
  • the transmission/reception sequence controller 102 can excite the ultrasound probe 100 to transmit ultrasound waves to the target spine, and can also control the ultrasound probe 100 to receive ultrasound echoes returned from the target spine, thereby obtaining ultrasound echo signals/data.
  • the processor 103 processes the ultrasound echo signal/data to obtain tissue-related parameters and ultrasound images of the target spine.
  • the ultrasound images obtained by the processor 103 may be stored in the memory 105, and these ultrasound images may be displayed on the display 104.
  • the ultrasound imaging system 10 may not include the probe 100, the transmission/reception selection switch 101, and the transmission/reception sequence controller 102, and only needs to include the processor 103 and the display 104. That is, the ultrasound image or related parameters of the target spine are acquired directly from the other device through the processor 103 and displayed on the display 104, which is not specifically limited here.
  • the display 104 of the foregoing ultrasound imaging system 10 may be a touch display screen, a liquid crystal display screen, etc., or an independent display device independent of the ultrasound imaging system 10, such as a liquid crystal display, a television, or the like. It is a display screen on electronic devices such as mobile phones and tablet computers.
  • the probe 100 may be a three-dimensional (3-dimension, 3D) ultrasound probe, which may also be called a volume probe, and may receive ultrasound echo data returned from different angles of the target spine to obtain the target spine 3D volume data.
  • 3D three-dimensional
  • the probe 100 can also be a two-dimensional ultrasound probe, and the three-dimensional volume data of the target spine can be obtained by the two-dimensional ultrasound probe.
  • the specific implementation method is to control the two-dimensional ultrasound manually or by a motor or a support arm. The probe moves to obtain three-dimensional data of the target spine.
  • the ultrasound imaging system 10 may further include a mechanical scanning device (not shown in FIG. 1).
  • the mechanical scanning device can drive the probe 100 to move, so that the probe 100 can receive ultrasound echo data returned from different angles of the target spine to obtain three-dimensional volume data of the target spine.
  • the probe 100 may exist independently, or may be provided on a mechanical scanning device, and the mechanical scanning device drives the probe 100 to move.
  • the acoustic head portion of the probe 100 may be an array composed of a plurality of array elements, and the plurality is two or more.
  • the array element can be used to convert electrical signals into ultrasonic waves, send ultrasonic waves, and receive the returned ultrasonic echoes, and convert the ultrasonic echoes into electrical signals to obtain ultrasonic echo data/signals.
  • the shape of the array may be a linear arrangement, a fan arrangement, etc., which may be adjusted according to actual application scenarios.
  • Each array element performs the transmission of ultrasonic waves or the reception of ultrasonic echoes by receiving the transmission signal of the transmission circuit and the reception signal sent by the reception circuit.
  • the foregoing memory 105 of the ultrasound imaging system 10 may be a flash memory card, a solid-state memory, a hard disk, or the like.
  • a computer-readable storage medium stores a plurality of program instructions. After the plurality of program instructions are called and executed by the processor 103, the present application can be executed. In some embodiments, some or all of the steps in the ultrasound imaging method of the spine or any combination thereof.
  • the computer-readable storage medium may be the memory 105, which may be a non-volatile storage medium such as a flash memory card, solid state memory, or hard disk.
  • the aforementioned processor 103 of the ultrasound imaging system 10 may be implemented by software, hardware, firmware, or a combination thereof, and may use circuits, single or multiple application specific integrated circuits (application specific integrated circuits, ASIC ), single or multiple general-purpose integrated circuits, single or multiple microprocessors, single or multiple programmable logic devices, or a combination of the aforementioned circuits or devices, or other suitable circuits or devices, so that the processor 103 can Perform the corresponding steps of the ultrasound imaging method of the spine in various embodiments of the present application.
  • ASIC application specific integrated circuits
  • ASIC application specific integrated circuits
  • microprocessors single or multiple programmable logic devices
  • a combination of the aforementioned circuits or devices or other suitable circuits or devices
  • three-dimensional visualization information usually includes cut-plane (or called section, Multiple Planner Rendering, MPR) image display and volume (Rendering, VR) image display.
  • MPR Multiple Planner Rendering
  • VR volume
  • Stereoscopic image refers to the three-dimensional The image obtained by volume data rendering and the cross-sectional image display a plane where the current orientation is located in the three-dimensional volume data.
  • the clinical fetal spine examination needs to observe the standard spine VR image or the spine standard two-dimensional slice view.
  • the doctor needs to adjust the orientation of the three-dimensional volume data of the fetal spine by adjusting the translation and rotation of X, Y, and Z, so that the target spine can be better displayed in this orientation; also, because The VR map is used to render the volume of interest (VOI).
  • VOI volume of interest
  • the doctor also needs to adjust the size and position of the VOI. Therefore, the process of using three-dimensional ultrasound to check the spine often requires a doctor to have a deep understanding of the target spine and three-dimensional ultrasound adjustment. This greatly depends on the experience of the doctor, consumes clinical examination time, and reduces the efficiency of the doctor.
  • the present application provides a method of ultrasonic self-imaging of the spine, which can effectively help doctors diagnose the fetal spine and improve work efficiency.
  • the ultrasound imaging method of the spine in the present application is described in detail below. Please refer to FIG. 2.
  • An ultrasound imaging method of the spine provided by an embodiment of the present application is applied to the ultrasound imaging system 10 shown in FIG.
  • Ultrasound imaging system 10 including a touch display screen that is, using a touch touch display screen to perform input touch screen operations
  • other ultrasound imaging system 10 including a display screen that is, using a mouse, trackball, etc. for input operations, here No specific restrictions.
  • the ultrasound imaging system 10 can generate three-dimensional volume data using ultrasound echo data.
  • the embodiments of the ultrasound imaging method of the spine in the present application include:
  • the ultrasound imaging system can obtain the three-dimensional volume data of the target spine in real time, or can obtain the three-dimensional volume data of the target spine from a local memory or a cloud storage, where the target spine can be any fetus, newborn Waiting for the detection of the spine of the human body, not specifically limited here.
  • acquiring the three-dimensional volume data of the target spine may include: sending ultrasonic waves to the target spine, receiving the ultrasonic echo returned by the target spine, and determining the three-dimensional volume data of the target spine according to the ultrasonic echo.
  • the transmit/receive sequence controller 102 sends a set of delayed focused pulses to the probe 100.
  • the probe 100 transmits ultrasonic waves to the tissue of the body under test (including the target spine), and receives tissue from the body under test after a certain delay. (Including the target spine) Ultrasound echoes with tissue information (including target spine information) reflected back, and this ultrasonic echoes are re-converted into electrical signals.
  • the transmission/reception sequence controller 102 receives these electrical signals and sends these ultrasonic echo signals to the processor 103.
  • the ultrasonic echo signal can obtain the three-dimensional volume data of the target spine after completing the focus delay, weighting and channel summation, and then through signal processing and then through three-dimensional imaging processing.
  • the three-dimensional volume data of the target spine may be obtained by the ultrasound imaging system in FIG. 1 described above, where the probe sends ultrasonic waves to the target spine and receives the ultrasonic echoes returned from the target spine.
  • the probe in the ultrasound imaging system may be a three-dimensional probe including three-dimensional array elements, the three-dimensional probe may directly acquire the ultrasonic echo signal returned from the target spine to obtain three-dimensional volume data of the target spine.
  • the ultrasound imaging system may further include a mechanical scanning device that drives the probe to move, receives ultrasound echo signals returned from the target spine from different angles, and obtains three-dimensional volume data of the target spine.
  • the three-dimensional volume data of the target spine may also be obtained from the memory.
  • the three-dimensional volume data may be that within a preset time period, an ultrasound imaging system or other three-dimensional ultrasound probe in an ultrasound imaging device is used to send ultrasound waves to the target spine and receive ultrasound echoes returned from the target spine to obtain the three-dimensional volume of the target spine.
  • the three-dimensional volume data of the target spine is stored in the memory. Therefore, in the embodiment of the present application, the three-dimensional volume data of the target spine can be read from the memory.
  • the storage may be a local storage, a cloud storage, or a storage in other manners, which is not specifically limited in the embodiments of the present application.
  • the three-dimensional volume data of the target spine can also be obtained by copying from other ultrasound imaging systems.
  • the A ultrasound imaging system obtains the three-dimensional volume data of the target spine from the memory in the B ultrasound imaging system.
  • the three-dimensional volume data may be detected and obtained by the B ultrasound imaging system in real time or obtained and stored by other means, which is not specifically limited here.
  • the ultrasound imaging system After acquiring the three-dimensional volume data of the target spine, the ultrasound imaging system further determines a preset anatomical structure from the three-dimensional volume data, where the preset anatomical structure is a partial anatomical structure or the entire anatomical structure of the target spine.
  • the predetermined anatomical structure may be a partial anatomical structure of the target spine such as a vertebral arch, vertebral body, or spinal cord.
  • the predetermined anatomical structure may also be all anatomical structures corresponding to the target spine, that is, including the vertebral arch, vertebral body, and The entire anatomical structure of the target spine such as the spinal cord. As shown in FIG.
  • the target spine in three-dimensional volume data, wherein the preset anatomical structure is the entire target spine, that is, including all anatomical structures of the vertebral arch, vertebral body, and spinal cord in the target spine.
  • the target spine can be marked by a frame shape. Of course, it can also be marked by one or more combinations of boundary lines, points, colors, and the like, which is not specifically limited here.
  • the method for the ultrasound imaging system to determine the preset anatomical structure from the three-dimensional volume data may be manual or automatic, including but not limited to the following methods:
  • responding to the input operation to the three-dimensional volume data to determine the region of interest may include: receiving an input operation to draw a target frame of the three-dimensional volume data, determining the region of interest according to the drawn target frame; or, receiving the three-dimensional volume The input operation of the points or lines drawn by the data determines the region of interest based on the points or lines drawn.
  • manually determining the preset anatomical structure can refer to the user using a keyboard, mouse, trackball and other tools, through a certain workflow to point and draw lines on the three-dimensional volume data, and the ultrasound imaging system responds to the user's input operation to determine the preset The direction and position of the anatomical structure in the 3D volume data.
  • the ultrasound imaging system can determine the characteristic information of the preset anatomical structure in advance and store it locally, or can obtain the characteristic information of the preset anatomical structure from other cloud storage or other ultrasound imaging systems.
  • the preset information can also be obtained in real time.
  • the characteristic information of the anatomical structure wherein the characteristic information indicates the key differentiating characteristics of the preset anatomical structure, for example, taking the preset anatomical structure as a vertebral arch as an example, the shape of the vertebral arch is an arch, and the echo is a strong echo
  • ultrasound The imaging system can identify the vertebral arch from the three-dimensional volume data through the characteristic information of the vertebral arch.
  • the ultrasound imaging system determines the preset anatomical structure from the three-dimensional volume data according to the characteristic information of the preset anatomical structure.
  • the following examples illustrate several possible implementation methods:
  • a preset learning algorithm may be used to detect the partial anatomy of the vertebral arch or vertebral body or spinal cord in the target spine in the three-dimensional volume data of the target spine.
  • certain data can be collected in advance
  • a number of vertebral arch or vertebral body images called positive samples
  • a certain number of non-vertebral arch or vertebral body images called negative samples
  • design an artificial neural network to determine the vertebral arch or Vertebral body detection model. That is, the detection model of the vertebral arch or vertebral body is used to automatically learn the features that can distinguish the positive sample and the negative sample.
  • the area is calculated as a positive sample.
  • the corresponding anatomical structure of the entire target spine can also be detected from the three-dimensional volume data in the same or similar manner, for example, the key anatomical structures such as the vertebral arch, vertebral body, and spinal cord in the target spine are detected, and then based on All areas corresponding to the vertebral arch, vertebral body and spinal cord determine the target spine.
  • Adaboost algorithm Support Vector Machine (SVM)
  • neural network algorithm convolutional neural network algorithm (Convolutional Neural Networks, CNN), and recurrent neural network algorithm (Recurrent Neural Network Algorithm) Network, RNN), FastRCNN, target detection method (Signal Shot Multiple Detector, SSD), etc.
  • the image segmentation method uses the image segmentation method to perform image segmentation on the three-dimensional volume data to obtain several candidate regions; according to the feature information of the preset anatomical structure and the image features of several candidate regions, determine the probability that several candidate regions are regions of interest, Wherein, the region of interest includes a preset anatomical structure; the preset anatomical structure is determined from the candidate regions whose probability is greater than a preset threshold.
  • the cone segment or the vertebral body or the spinal cord in the target spine can be determined according to the three-dimensional volume data of the target spine through an image segmentation method.
  • the vertebral arch in the target spine usually shows a strong echo-shaped arch structure, and the vertebral arch in the three-dimensional volume data of the target spine can be segmented by image segmentation.
  • the three-dimensional volume data is binary segmented, that is, after performing some necessary morphological operations, many candidate regions can be obtained, and then the image features of each candidate region are determined, and then according to the characteristics of each candidate region, the candidate region is determined as The probability of the vertebral arch is determined from the area with a higher probability.
  • the key structure of the vertebral body or other target spine or the entire target spine can also be determined from the three-dimensional volume data through the image segmentation method.
  • the specific determination process is similar to the process of determining the vertebral arch. For details, please refer to the above description. I will not repeat them here.
  • image segmentation methods can also be used, such as Level Set, Graph Cut, Snake, Random Walker, and some other image segmentation methods in deep learning, such as full Convolutional networks (Fully Convolutional Networks, FCN), unified networks (Unity Networking, UNet), etc., will not repeat them one by one here.
  • full Convolutional networks FCN
  • unified networks UNet
  • the template matching method is used to match the three-dimensional volume data with the preset anatomical structure template; according to the characteristic information of the preset anatomical structure, the region with the highest similarity is determined from the three-dimensional volume data; the region with the highest similarity is determined Preset anatomy.
  • the template matching method may also be used to determine the preset anatomical structure from the three-dimensional volume data of the target spine.
  • the vertebral body in the target spine is usually a short cylinder. You can collect some vertebral body data in the target spine in advance to create a vertebral body template.
  • determining the vertebral body in the three-dimensional volume data of the target spine It is possible to traverse all possible regions in the three-dimensional volume data of the target spine, and perform similarity matching with the preset vertebral body template, and select the region with the highest similarity as the region of interest, and the region of interest includes the vertebral body.
  • the key structure of the vertebral arch or other target spine or the entire target spine can also be determined from the three-dimensional volume data through the template matching method.
  • the specific determination process is similar to the process of determining the vertebral body. For details, please refer to the above description. I will not repeat them here.
  • the target spine can also be determined in other ways, which is not specifically limited here.
  • the position of the vertebral body can also be determined by presetting the spatial distance between the vertebral arch and the vertebral body.
  • the position of the vertebral arch can also be determined by presetting the spatial distance between the vertebral body and the vertebral arch.
  • the position of the entire spine is determined within a preset range through the position of the vertebral arch and vertebral body. There are many specific implementation methods, which will not be repeated here.
  • the target spine determined according to the three-dimensional volume data of the target spine may include the direction and position of the target spine in the three-dimensional volume data, or may be the direction and position in a certain two-dimensional slice, which is not done here Specific restrictions.
  • the manually determined or automatically determined options may be displayed on the display screen of the ultrasound imaging system, and the user may select according to actual needs, for example, through manual options To support the manual determination of the preset anatomical structure, and the automatic option to support the ultrasound imaging system to automatically determine the preset anatomical structure.
  • the target image includes at least one of a three-dimensional VR image, a two-dimensional cross-sectional image, and a multi-planar reconstruction (CMPR) image.
  • CMPR multi-planar reconstruction
  • acquiring the target image of the preset anatomical structure may include: rotating the preset anatomical structure to the target orientation in the three-dimensional volume data; determining the size of the region of interest corresponding to the preset anatomical structure And position; adjust the size and position of the region of interest so that the region of interest surrounds the preset anatomical structure; render the region of interest to obtain a VR image of the preset anatomical structure.
  • a method for determining a VR map of a preset anatomical structure the VR map renders the area within the VOI frame, that is, the area where the preset anatomical structure is located.
  • the VR map renders the area within the VOI frame, that is, the area where the preset anatomical structure is located.
  • it is a VR image of the target spine, that is, taking the preset anatomical structure as the entire target spine as an example, and rendering the entire target spine by highlighting the color to obtain a VR image of the target spine, so that The area where the target spine is located is clearly different from the area other than the target spine.
  • acquiring the target image of the preset anatomical structure may include: selecting at least three target pixels from the preset anatomical structure; generating two according to the positions of the at least three target pixels Dimensional plane; determine a two-dimensional slice corresponding to the two-dimensional plane from the three-dimensional volume data, and determine the two-dimensional slice corresponding to the two-dimensional plane as the two-dimensional slice of the preset anatomical structure.
  • the two-dimensional cross-sectional view of the preset anatomical structure is obtained here, usually referring to the standard two-dimensional cross-sectional view of the preset anatomical structure, referred to as the standard cross-section, the standard cross-section can be the median sagittal plane, the coronal plane, Cross-sectional diagrams and other high-standard cross-sectional diagrams.
  • a preset plane can be generated through the preset anatomical structure detected in the three-dimensional volume data in the previous step, and the plane can be obtained by solving equations or fitting.
  • the plane contains part of the anatomical structure of the target spine such as the vertebral arch, vertebral body or spinal cord.
  • the plane may also contain the entire anatomical structure of the target spine such as the vertebral arch, vertebral body and spinal cord.
  • the mathematical theorem of a plane based on three non-collinear points in space and solve the plane equation; for example, if you know The positions of more than three preset anatomical structures in the standard section can be fitted to a plane equation using the fitting method. There are many fitting methods, such as least square estimation and Hough transform. After the plane equation is obtained, the grayscale image corresponding to the plane can be taken from the three-dimensional volume data, so as to obtain the standard cut plane of the preset anatomical structure.
  • the two-dimensional cutting plane adopts the expression of plane equations, or other equivalent expressions can also be used.
  • a point in space plus a normal vector can also represent a two-dimensional cutting plane.
  • FIG. 6 it is a two-dimensional view of the target spine, that is, the entire anatomical structure of the target spine is determined from the three-dimensional volume data, and the corresponding two-dimensional view is determined from the three-dimensional volume data for the entire anatomical view .
  • acquiring the target image of the preset anatomical structure may include: determining a curve path of the preset anatomical structure from the three-dimensional volume data, and determining the curve from the three-dimensional volume data according to a preset thickness Multiple curved surfaces corresponding to the path, and reconstructing the multiple curved surfaces to obtain a CMPR diagram of the preset anatomical structure.
  • the ultrasound imaging system reconstructs the plurality of curved surfaces to obtain the CMPR map of the preset anatomical structure may include: acquiring pixel values of target pixels on the plurality of curved surfaces, where the target pixels are the multiple curved surfaces respectively For all or part of the pixels corresponding to the target position, determine the target surface of the preset anatomical structure according to the pixel values of the target pixels on the multiple surfaces, and straighten the target surface according to the preset direction to obtain the preset anatomical structure CMPR map, where the preset orientation can be user-defined or system default.
  • multiple curved surfaces integrate a three-dimensional volume data according to a preset thickness, wherein the pixel values of some pixels or all pixel points on each surface can be calculated according to a preset method to obtain a target cut plane The pixel value of the corresponding pixel is determined according to the corresponding pixel value.
  • the three-dimensional volume data includes curved surface 1, curved surface 2, and curved surface 3, where curved surface 1, curved surface 2, and curved surface 3 may be completely coincident or partially coincident, and for the coincident area, select a number of pixels corresponding to each surface at several positions The pixel value of the point, and then calculate the pixel value of the corresponding pixel on each surface to obtain a target surface, in which the pixel value can be calculated by means of averaging or weighted summation. It can be in other ways, which is not specifically limited here.
  • the weighting system corresponding to each pixel in the weighted summation mode may be user-defined or the system default.
  • the calculation process of pixel values is exemplified by means of averaging: it is assumed that the corresponding pixel values of the corresponding pixels in the first vertical direction, second vertical direction, and third vertical direction of the curved surface 1 are a1, a2, and a3, respectively;
  • the corresponding pixel values of the curved surface 2 in the first vertical direction, the second vertical direction, and the third vertical direction are b1, b2, and b3, respectively, and the curved surface 3 is in the first vertical direction and the second vertical direction ,
  • the pixel values of the corresponding pixels in the third vertical direction are c1, c2, and c3, respectively, then the pixel values of the pixels corresponding to the target slice are (a1+b1+c1)/3, (a2+b2+ c2)/3, (a3+b3+c3)/3, that is, multiple surfaces are reconstructed to generate a surface through the calculation of pixel values.
  • there are many pixels corresponding to each curved surface which is only for illustration and is
  • Multi-surface reconstruction (Curved Multi-planar Reconstruction, CMPR) is to select a specific curved path in one dimension, and all voxels (three-dimensional volume data) on the path (Unit) is displayed on the same plane after reconstruction.
  • the curved path may be one or more, may be a curved path corresponding to a partial anatomical structure of a target spine such as a vertebral arch, a vertebral body, or a spinal cord, or may be a curved path corresponding to an entire target spine, which is not specifically limited herein.
  • the CMPR curve path can be selected with different thicknesses.
  • the thickness can be set manually (for example, the user selects different thicknesses when viewing different spinal structures), or it can be automatically set (for example, the ultrasound imaging system 10 can automatically calculate the thickness according to the preset anatomical structure, so that This thickness can enclose the predetermined anatomy).
  • doctors can use the CMPR diagram to display the entire anatomical structure of the target spine (such as the spine cross-section or coronal plane) on a two-dimensional plane to help the doctor observe the structure of the target spine from different angles and different thicknesses.
  • the realization of the CMPR map is to locate the plane corresponding to the curved path of the preset anatomical structure, and then use the plane corresponding to the curved path to automatically obtain the CMPR image of the preset anatomical structure:
  • multiple curved surfaces corresponding to the curved path are determined according to the curvilinear path of the target spine, and the curved surfaces are reconstructed to obtain the CMPR map of the target spine.
  • multiple curved surfaces corresponding to the curved path are determined according to the curved path of the vertebral arch and according to a preset thickness, and the multiple curved surfaces are reconstructed to obtain a CMPR graph of the vertebral arch.
  • multiple curved surfaces corresponding to the curved path are determined according to the curved path of the vertebral body and according to a preset thickness, and the multiple curved surfaces are reconstructed to obtain a CMPR graph of the vertebral body.
  • CMPR diagram for determining the predetermined anatomical structure according to multiple curved surfaces can refer to the above description, and will not be repeated here.
  • the CMPR map may also be pseudo-color marked to obtain a pseudo-color image of the preset anatomical structure, so that the preset anatomical structure displays It is clearer and easier for doctors to observe.
  • the display 104 in the ultrasound imaging system 10 may display at least one of the stereoscopic VR map, the two-dimensional slice map, and the multi-curved surface reconstruction CMPR map obtained through the above steps.
  • the process of displaying the target image can be triggered by the user's key press, or can be directly embedded in the ultrasound system, which is directly opened by default, or can be triggered by other conditions, which is not specifically limited here.
  • the embodiments shown in this application are applicable to both three-dimensional imaging and four-dimensional imaging. Among them, four-dimensional imaging is to dynamically display multiple three-dimensional spine volume data collected in real time through the application process dynamically.
  • the target image when displaying the target image, it may be displayed on the display 104 of the ultrasound imaging system 10 entirely, or the user may choose to display the VR image, two-dimensional slice view, or CMPR image of the target spine.
  • the method may further include: receiving a display instruction for the target image; and displaying the target image according to the display instruction.
  • the display 104 in the ultrasound imaging system 10 may display the target image according to the display instruction.
  • the method may further include: receiving a hiding instruction for the target image; and hiding the target image according to the hiding instruction.
  • the display 104 in the ultrasound imaging system 10 displays the target image, and the user does not need to observe the target image again, the hiding operation can be triggered.
  • the display 104 in the ultrasound imaging system 10 can according to the hiding instruction, Hide the target image.
  • acquiring three-dimensional volume data of a target spine determining a preset anatomical structure from the three-dimensional volume data; acquiring a target image of the preset anatomical structure and displaying the preset anatomical structure, wherein the target image It includes at least one of a three-dimensional VR map, a two-dimensional slice view and a multi-curved reconstruction CMPR map. That is, the ultrasound imaging system can automatically acquire the target image of the target spine, without the user's manual selection, you can intuitively see the stereogram, two-dimensional slice or multi-surface reconstruction of the target spine CMPR map, improve the efficiency and accuracy of ultrasound imaging, Effectively help doctors to carry out auxiliary diagnosis of diseases, improve work efficiency, and save doctors' time and energy.
  • an ultrasound imaging method is provided.
  • the processor can acquire the three-dimensional volume data of the fetus.
  • the three-dimensional volume data of the fetus may be obtained by real-time scanning of the aforementioned ultrasound imaging system, that is, the ultrasound probe transmits ultrasound waves to the fetus and receives ultrasound echoes to obtain ultrasound echo signals, and the processor processes the ultrasound echo signals to obtain fetal Three-dimensional volume data; the three-dimensional volume data of the fetus may also be pre-scanned and stored by the ultrasound imaging device and read in by the processor when the method of this embodiment is needed.
  • the processor may identify the spinal cone region from the fetal three-dimensional volume data based on the characteristics of the fetal spinal cone, determine the position of the spinal cone region according to the identified spinal cone region, and display the position of the spinal cone region.
  • the position of the conical region of the spinal cord can be displayed in various suitable ways, for example, by suitable symbols, colored regions, text, arrows, geometric shapes, and so on.
  • the position of the conical region of the spinal cord can be displayed on the three-dimensional image, it can also be displayed on the two-dimensional image, or it can be displayed on another location.
  • the determined position of the spinal cord cone region may be the position of the end of the spinal cord cone.
  • the position of the end of the spinal cord cone can be determined according to the identified spinal cord cone area, and the position of the end of the spinal cord cone can be displayed.
  • the position of the end of the spinal cord cone can be displayed in a variety of suitable ways, for example, suitable compound, color, dot, line, arrow, number, distance from suitable reference position, and so on.
  • the position of the end of the spinal cord cone can be displayed on a three-dimensional image, it can also be displayed on a two-dimensional image, or it can be displayed on another location.
  • the dotted frame A is the spinal cord cone area.
  • the virtual straight line where the arrow B is located and the virtual straight line where the arrow C is located indicate the end of the spinal cone, which is shown as a small dot in the figure.
  • the conical region of the spinal cord can be identified by the method of target matching.
  • at least two second candidate regions may be determined from the three-dimensional volume data of the fetus, the volume data features of the three-dimensional volume data of each second candidate region may be obtained, and each volume may be determined according to the volume data features of each second candidate region
  • the second degree of matching between the second candidate area and the spinal cord cone, and determining that the second candidate area with the highest second matching degree is the spinal cord cone area.
  • the fetal sagittal plane may be first identified from the three-dimensional volume data of the fetus, and then the spinal cone region may be identified from the sagittal plane image.
  • the sagittal image of the fetal spine can be determined from the three-dimensional volume data of the fetus based on the characteristics of the sagittal surface of the fetal spine, and then based on the characteristics of the spinal cone, in the sagittal image of the fetal spine Determine the spinal cone area.
  • the sagittal plane may be a median sagittal plane and/or a sagittal plane adjacent to the median sagittal plane.
  • the lumbar region can also be identified from the three-dimensional volume data of the fetus, and the position of the spinal cone region can be displayed relative to the lumbar region, so that the user can easily see the relative positional relationship between the spinal cone and the lumbar region .
  • the lumbar spine region can be identified from the three-dimensional volume data of the fetus, an ultrasound image of the lumbar spine region can be displayed, and the spinal cone region (e.g., spinal cone end, etc.) can be displayed relative to the ultrasound image of the lumbar spine region. s position.
  • displaying the position of the spinal cone region with respect to the ultrasound image of the lumbar vertebra region can include a variety of suitable ways, for example, the ultrasound image of the lumbar region and the position of the spinal cone region can be displayed at the same time, so that the user can directly see the relative
  • the positional relationship either by text or coincidence, shows the distance of the spinal cord cone area relative to the lumbar vertebral area, or by the symbols representing the lumbar vertebrae and spinal cord cone area, the relative positional relationship between the two, and so on.
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical, or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or software function unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium.
  • the technical solution of the present application essentially or part of the contribution to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to enable a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pathology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)

Abstract

本申请实施例公开了一种脊柱的成像方法以及超声成像***，用于自动生成并显示目标脊柱的目标图像，不需要手动选择，提高超声成像的效率以及准确度。可以包括：获取目标脊柱的三维体数据；从所述三维体数据中确定所述目标脊柱的预设解剖结构，其中，所述目标脊柱的预设解剖结构为所述目标脊柱的部分解剖结构或者全部解剖结构；获取所述预设解剖结构的目标图像，其中，所述目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种；显示所述目标图像。

Description

一种脊柱的成像方法以及超声成像*** 技术领域
本申请涉及医疗器械领域，尤其涉及一种脊柱的成像方法以及超声成像***。
背景技术
超声检查由于其安全、方便、无辐射、廉价等优势，在临床检查上具有广泛的应用，成为医生进行疾病诊断的主要辅助手段之一。医生通过超声成像技术可以观察人体内部组织结构，进行临床辅助诊断。
其中，脊柱是胎儿发育中非常重要的结构，也是产前检查的重要部位。近年来，三维超声在胎儿脊柱检查中得到广泛的应用，其优势在于通过一次扫描就能获取感兴趣区域的三维体数据，并可显示该三维体数据内的任何切面，图像直观。该三维体数据包括脊柱三维体数据，能够帮助医生更准确的定位脊柱异常节段。
但是，在获取脊柱三维体数据之后，医生需要频繁手动旋转、平移脊柱三维体数据，达到合适的观察角度。然后根据脊柱中椎弓、椎体等解剖学结构，手动调整感兴趣区域(volume of interest，VOI)的大小和位置，或利用曲面重建(Curve multi-planar reconstruction，CMPR)手动选择椎弓、椎体区域，得到标准的椎弓、椎体立体(Volume Rendering，VR)图像或脊柱矢状图像。这整个过程需要医生具有一定的经验，耗时费力。
发明内容
本申请实施例提供了一种脊柱的成像方法以及超声成像***，用于自动生成并显示目标脊柱的目标图像，不需要手动选择，提高超声成像的效率以及准确度。
本申请实施例的一个方面，提供了一种脊柱的成像方法，其特征在于，包括：获取胎儿的三维体数据；基于胎儿脊柱的特征，从所述三维体数据中识别出胎儿脊柱的图像区域；根据识别出的胎儿脊柱的图像区域获取所述胎儿脊柱 的目标图像，其中，所述目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种；显示所述目标图像。
本申请实施例的一方面，提供一种脊柱的成像方法，包括：获取扫描目标的三维体数据；从所述三维体数据中确定所述目标脊柱的预设解剖结构，其中，所述目标脊柱的预设解剖结构为所述目标脊柱的部分解剖结构或者全部解剖结构；获取所述预设解剖结构的目标图像，其中，所述目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种；显示所述目标图像。
本申请实施例的一方面，一种脊柱的成像方法，其特征在于，包括：获取胎儿的三维体数据；基于胎儿脊髓圆锥的特征，从所述胎儿的三维体数据中识别出脊髓圆锥区域；根据识别出的脊髓圆锥区域，确定脊髓圆锥区域的位置；显示所述脊髓圆锥区域的位置。
本申请实施例的一方面，提供一种超声成像***，包括：超声探头，所述超声探头向扫描目标发射超声波并接收超声回波，获得超声回波信号；处理器，所述处理器根据所述超声回波信号获得目标脊柱的三维体数据，并从所述三维体数据中确定所述目标脊柱的预设解剖结构，其中，所述目标脊柱的预设解剖结构为所述目标脊柱的部分解剖结构或者全部解剖结构；获取所述预设解剖结构的目标图像，其中，所述目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种；显示器，所述显示器显示所述目标图像。
本申请实施例的一方面，提供一种超声成像***，其特征在于，包括：超声探头，所述超声探头向胎儿发射超声波并接收超声回波，获得超声回波信号；处理器，所述处理器根据所述超声回波信号获得所述胎儿的三维体数据，基于胎儿脊髓圆锥的特征从所述胎儿的三维体数据中识别出脊髓圆锥区域，并根据识别出的脊髓圆锥区域，确定脊髓圆锥区域的位置；显示器，所述显示器显示所述脊髓圆锥区域的位置。
本申请实施例的一方面，提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行上述第一方面所提供的脊柱的成像方法。
本申请实施例提供的技术方案中，超声成像***获取目标脊柱的三维体数据，并从该三维体数据中确定该目标脊柱的预设解剖结构，其中，该预设解剖 结构可以是该目标脊柱的部分解剖结构，也可以是该目标脊柱的全部结构结构，例如，该预设解剖结构可以是该目标脊柱中的椎弓，椎体等。进一步，该超声成像***获取该预设解剖结构的目标图像并显示该目标图像，其中，该目标图像可以是VR图，也可以是二维标准切面图，也可以CMPR图等。可见，超声成像***可以自动获取目标脊柱的预设解剖结构的目标图像，不需要用户手动选择，就可以直观显示，提高超声成像的效率以及准确度，有效的帮助医生进行疾病辅助诊断，提升工作效率。
附图说明
图1为本申请实施例中一种可能的超声成像***的结构框图示意图；
图2为本申请实施例中脊柱的成像方法的一个实施例示意图；
图3为本申请实施例中目标脊柱的一个三维体数据的一个界面显示示意图；
图4为本申请实施例中在三维体数据中确定目标脊柱的一个界面显示示意图；
图5为本申请实施例中目标脊柱的一个VR图；
图6为本申请实施例中目标脊柱的一个二维切面图；
图7a为本申请实施例中目标脊柱的一个CMPR图；
图7b为本申请实施例中椎弓的一个CMPR图；
图7c为本申请实施例中椎体的一个CMPR图；
图8为本申请实施例中脊髓圆锥区域的一个示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域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，都属于本申请保护的范围。
图1为本申请实施例中的超声成像***10的结构框图示意图。该超声成像***10可以包括探头100，其中，该探头100可以是超声探头、发射/接收选择 开关101、发射/接收序列控制器102、处理器103、显示器104。发射/接收序列控制器102可以激励超声探头100向目标脊柱发射超声波，还可以控制超声探头100接收从目标脊柱返回的超声回波，从而获得超声回波信号/数据。处理器103对该超声回波信号/数据进行处理，以获得目标脊柱的组织相关参数和超声图像。处理器103获得的超声图像可以存储于存储器105中，这些超声图像可以在显示器104上显示。当然，该超声成像***10也可以不包括探头100、发射/接收选择开关101以及发射/接收序列控制器102，只需包括处理器103以及显示器104。即直接通过处理器103从其他设备获取该目标脊柱的超声图像或者相关参数，并通过显示器104进行显示，此处不做具体限定。
本申请实施例中，前述的超声成像***10的显示器104可为触摸显示屏、液晶显示屏等，也可以是独立于超声成像***10之外的液晶显示器、电视机等独立显示设备，也可为手机、平板电脑等电子设备上的显示屏。
本申请的一个可选实施例中，探头100可以是三维(3-dimension，3D)超声探头，也可以称为容积探头，可以接收从目标脊柱的不同角度返回的超声回波数据，得到目标脊柱的三维体数据。
本申请的一个可选实施例中，探头100也可以是二维超声探头，通过二维超声探头获取目标脊柱的三维体数据，具体实现方式是通过手动或者通过马达或者通过支撑臂控制二维超声探头移动来获取目标脊柱的三维数据。
本申请的一个可选实施例中，超声成像***10还可以包括机械扫描装置(图1中未示出)。该机械扫描装置可以带动探头100运动，使探头100可以接收从目标脊柱的不同角度返回的超声回波数据，以得到目标脊柱的三维体数据。
本申请的一个可选实施例中，探头100可以是独立存在的，也可以是设置在机械扫描装置上，由机械扫描装置带动探头100运动。
本申请的一个可选实施例中，探头100的声头部分可以是多个阵元组成的阵列，该多个为两个或两个以上。阵元可以用于将电信号转换为超声波，并发送超声波，以及接收返回的超声回波，将超声回波转换为电信号，以得到超声回波数据/信号。其中，该阵列的形状可以是直线排列，也可以是扇形排列等，具体可以根据实际应用场景调整。每个阵元通过接收发射电路的发射信号与接收电路发送的接收信号，进行超声波的发射或超声回波的接收。
本申请的一个可选实施例中，前述的超声成像***10的存储器105可为闪存卡、固态存储器、硬盘等。
本申请的一个可选实施例中，还提供一种计算机可读存储介质，该计算机可读存储介质存储有多条程序指令，该多条程序指令被处理器103调用执行后，可执行本申请各个实施例中脊柱的超声成像方法中的部分步骤或全部步骤或其中步骤的任意组合。
本申请的一个可选实施例中，该计算机可读存储介质可为存储器105，其可以是闪存卡、固态存储器、硬盘等非易失性存储介质。
本申请的一个可选实施例中，前述的超声成像***10的处理器103可以通过软件、硬件、固件或者其组合实现，可以使用电路、单个或多个专用集成电路(application specific integrated circuits，ASIC)、单个或多个通用集成电路、单个或多个微处理器、单个或多个可编程逻辑器件、或者前述电路或器件的组合、或者其他适合的电路或器件，从而使得该处理器103可以执行本申请的各个实施例中脊柱的超声成像方法的相应步骤。
在三维成像***中，三维可视化信息通常包括切面(或称为剖面，Multiple Planner Rendering，MPR)图像的显示及立体(Volume Rendering，VR)图像的显示，立体图像是指通过光线跟踪等方法对三维体数据渲染得到的图像，剖面图像是在三维体数据中将当前方位所在的一个平面显示出来。通常，临床上胎儿脊柱检查需要观察的是脊柱标准VR图或脊柱标准二维切面图。要想获得脊柱的标准二维切面图，医生需要通过调节X、Y、Z的平移和旋转来调整胎儿脊柱的三维体数据的方位，使得在该方位下能较好显示目标脊柱；同样，因为VR图是对感兴趣区域(Volume of Interest，VOI)内进行渲染，要想获得标准目标脊柱的VR图，医生除了需要调节目标脊柱的三维体数据的方位，还需要调节VOI的大小和位置。所以利用三维超声检查脊柱的过程中，往往需要医生对目标脊柱和三维超声调节有深入的理解，这极大的依赖于医生的经验，消耗临床检查时间，降低医生工作效率。
本申请提供了一种脊柱的超声自成像方法，可有效的帮助医生对胎儿脊柱进行诊断，提升工作效率。下面对本申请中脊柱的超声成像方法进行详细描述，请参阅图2，本申请实施例提供的一种脊柱的超声成像方法，该方法应用于上 述图1所示的超声成像***10，可以适用于包含触摸显示屏的超声成像***10，即利用接触触摸显示屏来执行输入触屏操作，也可以是其他包含显示屏的超声成像***10，即可以利用鼠标，轨迹球等进行输入操作，此处不做具体限定。该超声成像***10可利用超声回波数据生成三维体数据。本申请中脊柱的超声成像方法实施例包括：
201、获取目标脊柱的三维体数据。
本申请实施例中，超声成像***可以实时获取该目标脊柱的三维体数据，也可以从本地存储器或者云存储器中获取该目标脊柱的三维体数据，其中，该目标脊柱可以是任何胎儿，新生儿等待检测人体的脊柱，此处不做具体限定。
如图3所示，为超声成像***获取的目标脊柱的三维体数据，并控制该三维体数据以二维的形式显示在显示器上，当然也可以以三维的形式显示该三维体数据。其中，获取目标脊柱的三维体数据，可以包括：向目标脊柱发送超声波，接收目标脊柱返回的超声回波，根据超声回波确定该目标脊柱的三维体数据。
示例性的，发射/接收序列控制器102将一组经过延迟聚焦的脉冲发送到探头100，探头100向受测机体组织(包括目标脊柱)发射超声波，经过一定延时后接收从受测机体组织(包括目标脊柱)反射回来的带有组织信息(包括目标脊柱信息)的超声回波，并将此超声回波重新转换为电信号。发射/接收序列控制器102接收这些电信号，并将这些超声回波信号送入处理器103。超声回波信号在完成聚焦延时、加权和通道求和，再经过信号处理，再经过三维成像的处理，即可获取目标脊柱的三维体数据。
在本申请中的一个实施例中，目标脊柱的三维体数据可以是由前述图1中的超声成像***中，由探头向目标脊柱发送超声波，并接收从目标脊柱返回的超声回波得到的。具体地，因超声成像***中的探头可以是包括三维排列的阵元的三维探头，可以由该三维探头直接获取从目标脊柱返回的超声回波信号，以得到目标脊柱的三维体数据。或者，超声成像***还可以包括机械扫描装置，由该机械扫描装置带动探头运动，从不同角度接收从目标脊柱返回的超声回波信号，得到目标脊柱的三维体数据。
在本申请中的一个实施例中，目标脊柱的三维体数据也可以是从存储器中 获取到的。该三维体数据可以是在预置时间段内，通过超声成像***或其他超声成像设备中的三维超声探头，向目标脊柱发送超声波，并接收从目标脊柱返回的超声回波，得到目标脊柱的三维体数据后，将目标脊柱的三维体数据存储在存储器中。因此，本申请实施例中目标脊柱的三维体数据可以从存储器中读取得到。
可以理解的是，该存储器可以是本地存储器或者云存储器，或者其他方式的存储器，本申请实施例中不做具体限定。
在本申请中的一个实施例中，目标脊柱的三维体数据也可以从其他超声成像***拷贝获取，例如，A超声成像***从B超声成像***中的存储器中获取该目标脊柱的三维体数据，该三维体数据可以由该B超声成像***实时检测获取或者通过其他方式获取并存储，此处不做具体限定。
202、从三维体数据中确定预设解剖结构。
超声成像***获取目标脊柱的三维体数据后，进一步从该三维体数据中确定预设解剖结构，其中，该预设解剖结构为该目标脊柱的部分解剖结构或者全部解剖结构。例如，该预设解剖结构可以是椎弓或者椎体或者脊髓等该目标脊柱的部分解剖结构，该预设解剖结构也可以是该目标脊柱对应的全部解剖结构，即包括椎弓，椎体以及脊髓等目标脊柱的全部解剖结构。如图4所示，为三维体数据中确定目标脊柱的一个示意图，其中，该预设解剖结构为整个目标脊柱，即包括目标脊柱中的椎弓，椎体以及脊髓等全部解剖结构。在一种可能的实现方式中，可以通过框型来标注该目标脊柱，当然，也可以通过分界线，点，颜色等其他一种或者多种组合方式来标注，此处不做具体限定。
超声成像***从该三维体数据中确定预设解剖结构的方法可以是手动的，也可以是自动的，包括但不限于如下所示的方式：
(1)控制三维体数据显示于显示器，响应对三维体数据的输入操作以确定感兴趣区域，其中,感兴趣区域包括该预设解剖结构。
可选的，响应对三维体数据的输入操作以确定感兴趣区域，可以包括：接收对三维体数据绘制的目标框的输入操作,根据绘制的目标框确定感兴趣区域；或者,接收对三维体数据绘制的点或者线的输入操作,根据绘制的点或者线确定感兴趣区域。
例如：手动确定预设解剖结构可以指用户通过键盘、鼠标，轨迹球等工具，通过一定的工作流在三维体数据上点点、画线等操作，超声成像***响应用户的输入操作，确定预设解剖结构在三维体数据中的方向和位置。
(2)确定该预设解剖结构的特征信息，根据该预设解剖结构的特征信息从该三维体数据中确定该预设解剖结构。
可见，超声成像***可以提前确定该预设解剖结构的特征信息并进行本地存储，也可以从其他云存储器或者其他超声成像***获取该预设解剖结构的特征信息，当然，也可以实时获取该预设解剖结构的特征信息，其中，该特征信息表明该预设解剖结构的关键区别特征，例如，以预设解剖结构为椎弓为例，椎弓的形状为弓形，且回声为强回声，超声成像***可以通过该椎弓的特征信息从该三维体数据中识别出该椎弓。
其中，超声成像***根据该预设解剖结构的特征信息从该三维体数据中确定该预设解剖结构的方式有很多种，下面举例说明几种可能的实现方式：
A、利用预设学习算法确定预设解剖结构的检测模型；将三维体数据输入检测模型以根据该预设解剖结构的特征信息输出感兴趣区域，其中，感兴趣区域包括该预设解剖结构。
示例性的，可采用预设学习算法在目标脊柱的三维体数据中检测出目标脊柱中的椎弓或者椎体或者脊髓等部分解剖结构，以检测椎弓或者椎体为例，可以事先收集一定数量的椎弓或椎体图像(称为正样本)，以及收集一定数量的非椎弓或椎体图像(称为负样本)，然后基于预设学习算法，设计人工神经网络，确定椎弓或者椎体的检测模型。即利用椎弓或者椎体的检测模型自动学习出能够区分正样本和负样本的特征，利用这些特征在检测时遍历目标脊柱的三维体数据中所有可能的区域，计算该区域被判断为正样本的概率，选择概率最大的区域为感兴趣区域，该感兴趣区域包括椎弓或者椎体。当然，也可以以相同或者相似的方式从该三维体数据中检测出整个目标脊柱对应的解剖结构，例如，将该目标脊柱中的椎弓，椎体，脊髓等关键解剖结构检测出来，再根据该椎弓，椎体以及脊髓对应的全部区域确定该目标脊柱。
可以理解的是，常用的预设学习算法有Adaboost算法、支持向量机(Support Vector Machine，SVM)、神经网络算法、卷积神经网络算法 (Convolutional Neural Networks，CNN)、递归神经网络算法(Recurrent Neural Network，RNN)、FastRCNN，目标检测方法(Signal Shot Multiple Detector，SSD)等等，这些算法可用于从目标脊柱的三维体数据中确定目标脊柱。
B、利用图像分割方法对三维体数据进行图像分割，得到若干个候选区域；根据该预设解剖结构的特征信息以及若干个候选区域的图像特征，确定若干个候选区域是感兴趣区域的概率，其中，感兴趣区域包括预设解剖结构；从概率大于预设阈值的候选区域中确定预设解剖结构。
示例性的，可以通过图像分割方法，根据目标脊柱的三维体数据确定目标脊柱中的锥弓或者椎体或者脊髓等。这里以椎弓为例进行说明，目标脊柱中的椎弓，通常表现为强回声的弓形结构，可通过图像分割的方法将目标脊柱的三维体数据中的椎弓分割出来。首先对三维体数据进行二值化分割，即进行一些必要的形态学操作后可以得到很多候选区域，然后确定每个候选区域的图像特征，再根据每个候选区域的特征，判断该候选区域为椎弓的概率，从概率较高的区域中确定椎弓。例如：对每个候选区域根据形状、灰度等图像特征(例如椎弓的形状为弓形、强回声)判断该区域是椎弓的概率，选择概率较高的区域作为椎弓区域。可以理解的是，同样可以通过图像分割方法从该三维体数据中确定椎体或者其他目标脊柱的关键结构或者整个目标脊柱，具体确定过程与确定椎弓的过程相似，具体可参阅上述描述，此处不再赘述。
需要说明的是，也可以采用其它图像分割方法，例如水平集(Level Set)、图割(Graph Cut)、Snake、随机游走(Random walker)以及深度学习中的一些其他图像分割方法，如全卷积网络(Fully Convolutional Networks，FCN)、统一网络(Unity Networking，UNet)等，此处不再一一赘述。
C、利用模板匹配方法将三维体数据和预设解剖结构模板进行相似度匹配；根据该预设解剖结构的特征信息从三维体数据中确定相似度最高的区域；从相似度最高的区域中确定预设解剖结构。
示例性的，也可采用模板匹配的方法从目标脊柱的三维体数据中确定预设解剖结构。以椎体为例进行说明，目标脊柱中的椎体，通常呈短圆柱状，可以事先收集一些目标脊柱中的椎体数据建立椎体模板，在目标脊柱的三维体数据 中确定椎体时，可以遍历目标脊柱的三维体数据中所有可能的区域，和预设椎体模板进行相似度匹配，选择相似度最高的区域为感兴趣区域，该感兴趣区域包括该椎体。可以理解的是，同样可以通过模板匹配方法从该三维体数据中确定椎弓或者其他目标脊柱的关键结构或者整个目标脊柱，具体确定过程与确定椎体的过程相似，具体可参阅上述描述，此处不再赘述。
可以理解的是，自动根据目标脊柱的三维体数据确定预设解剖结构的方法有很多，通过以上一种或多种方法，也可以通过其他的方式来确定目标脊柱，此处不做具体限定，例如，在确定椎弓位置的基础上，也可以通过预设椎弓和椎体之间的空间距离，确定椎体的位置。又如，在确定椎***置的基础上，也可以通过预设椎体与椎弓之间的空间距离，确定椎弓的位置。又如，在确定椎弓和椎体的基础上，通过椎弓和椎体的位置，在预设范围内确定整个脊柱的位置，具体实现方式有很多种，此处不再一一赘述。
需要说明的是，这里根据目标脊柱的三维体数据确定的目标脊柱可以包括目标脊柱在三维体数据中的方向和位置，也可以是在某个二维切面中的方向和位置，此处不做具体限定。
可以理解的是，在目标脊柱的三维体数据确定预设解剖结构时，可以在超声成像***的显示屏上显示手动确定或者自动确定的选项，用户可根据实际需求来选择，例如，通过手动选项来支持手动确定预设解剖结构，通过自动选项来支持超声成像***自动确定该预设解剖结构。
203、获取预设解剖结构的目标图像。
其中，目标图像包括立体VR图、二维切面图和多曲面重建(curved Multi-planar reconstruction，CMPR)图中的至少一种，下面分别以该目标图像为其中一种进行示例性说明：
(1)目标图像为VR图时，获取预设解剖结构的目标图像，可以包括：在该三维体数据中将预设解剖结构旋转至目标方位；确定预设解剖结构对应的感兴趣区域的大小和位置；调节感兴趣区域的大小和位置以使得感兴趣区域包围预设解剖结构；对感兴趣区域进行渲染得到预设解剖结构的VR图。
示例性的，确定预设解剖结构VR图的方法：VR图是对VOI框内的区域进行渲染，即对预设解剖结构所在的区域进行渲染。要想获得预设解剖结构VR 图，除了知道椎弓或椎体或脊髓等关键结构的方向和位置，还需要设置VOI框的大小和位置，并调节该VOI框的大小和位置以使得该VOI框包围该预设解剖结构。例如，可以根据目标脊柱的长轴，或椎弓的位置，或椎体的位置将其旋转目标方位，然后调整VOI框的大小和位置，使VOI框把目标脊柱(如椎弓或椎体或脊髓或包含椎弓和椎体以及脊髓的目标脊柱)包围，即可得到椎弓或椎体或脊髓或包含椎弓和椎体以及脊髓的目标脊柱的VR图。如图5所示，为目标脊柱的一个VR图，即以预设解剖结构为整个目标脊柱为例，通过高亮的颜色等方式对该整个目标脊柱进行渲染得到该目标脊柱的VR图，使得该目标脊柱所在区域明显区别于该目标脊柱以外的其他区域。
(2)目标图像为二维切面图时，获取该预设解剖结构的目标图像，可以包括：从预设解剖结构中选取至少三个目标像素点；根据至少三个目标像素点的位置生成二维平面；从三维体数据中确定二维平面对应的二维切面图，将该二维平面对应的二维切面图确定为该预设解剖结构的二维切面图。
需要说明的是，这里获取预设解剖结构的二维切面图，通常指的是预设解剖结构的标准二维切面图，简称标准切面，该标准切面可以是正中矢状面图，冠状面图，横截面图等标准程度比较高的切面图。示例性的，通过上一步在三维体数据中检测到的预设解剖结构，即可生成一个平面，平面可以通过解方程或者拟合得到。其中，该平面包含椎弓或者椎体或者脊髓等目标脊柱的部分解剖结构，该平面也可以是包含椎弓，椎体以及脊髓等目标脊柱的全部解剖结构。例如，如果知道了标准切面中三个不共线的预设解剖结构的位置，即可根据空间中不共线的三点确定一个平面的数学定理，求解出平面方程；又如，如果知道了标准切面中大于三个预设解剖结构的位置，即可采用拟合的方法拟合出一个平面方程，拟合方法有很多，如最小二乘估计、Hough变换等。得到平面方程后，即可从三维体数据中取出该平面所对应的灰阶图像，从而得到预设解剖结构的标准切面。在该实施例中，二维切面采用了平面方程的表达方式，也可以用其它等价的表达方式，如空间中的一个点加一个法向量也可以表示二维切面，本申请实施例中不做具体限定。如图6所示，为目标脊柱的一个二维切面图，即从该三维体数据中确定该目标脊柱的整个解剖结构，并针对该整个解剖切面从三维体数据中确定对应的二维切面图。
(3)目标图像为CMPR图时，获取预设解剖结构的目标图像，可以包括：从该三维体数据中确定预设解剖结构的曲线路径，按照预设厚度从该三维体数据中确定该曲线路径对应的多个曲面，并将该多个曲面进行重建得到该预设解剖结构的CMPR图。
其中，超声成像***将该多个曲面进行重建得到该预设解剖结构的CMPR图可以包括：获取该多个曲面上目标像素点的像素值，其中，该目标像素点为该多个曲面分别在目标位置上对应的全部像素点或者部分像素点，根据该多个曲面上目标像素点的像素值确定预设解剖结构的目标曲面，将该目标曲面按照预设方向拉直得到该预设解剖结构的CMPR图，其中，该预设方位可以是用户自定义或者***默认。
可以理解的是，多个曲面按照预设厚度集成一个三维体数据，其中，可以将每个曲面上的部分像素点的像素值或者全部像素点的像素值按照预设方式进行计算得到一个目标切面对应的像素点的像素值，根据该对应的像素值确定该目标切面。例如，该三维体数据包括曲面1，曲面2，曲面3，其中，曲面1，曲面2，曲面3可以是完全重合或者部分重合，针对重合的区域，选取若干个位置上每个曲面对应的像素点的像素值，然后将每个曲面上对应的像素点的像素值进行计算，得到一个目标曲面，其中，像素值的计算方式可以是作平均的方式，也可以是加权求和的方式，也可以是其他方式，此处不做具体限定。其中，加权求和方式中每个像素点对应的权重***可以是用户自定义的或者是***默认的。
以作平均的方式举例说明像素值的计算过程：假设曲面1在第一竖直方向，第二竖直方向，第三竖直方向上对应的像素点的像素值分别为a1，a2，a3；曲面2在第一竖直方向，第二竖直方向，第三竖直方向上对应的像素点的像素值分别为b1，b2，b3，曲面3在第一竖直方向，第二竖直方向，第三竖直方向上对应的像素点的像素值分别为c1，c2，c3，则得到目标切面对应的像素点的像素值分别为(a1+b1+c1)/3，(a2+b2+c2)/3，(a3+b3+c3)/3，即通过像素值的计算将多个曲面重建生成一个曲面。需要说明的是，在实际应用中，每个曲面对应的像素点有很多个，此处只是为了举例说明，并没有限定。
示例性的，确定预设解剖结构的CMPR图像的方法：多曲面重建 (Curved Multi-planar reconstruction，CMPR)是在一个维度上选择特定的曲线路径，将该路径上所有体素(三维体数据的单位)重建后在同一平面上显示。该曲线路径可以是一个或者多个，可以是椎弓或者椎体或者脊髓等目标脊柱的部分解剖结构对应的曲线路径，也可以是整个目标脊柱对应的曲线路径，此处不做具体限定。CMPR的曲线路径可以选择不同的厚度，厚度可以手动设置(例如用户观察不同的脊柱结构时选择不同的厚度)，也可以自动设置(例如超声成像***10可根据预设解剖结构自动计算厚度，使得该厚度能够把预设解剖结构包住)。在临床上，医生可利用CMPR图将目标脊柱的全部解剖结构(如脊柱横截面或冠状面)在一个二维平面上显示，帮助医生从不同角度、不同厚度观察目标脊柱的结构。
CMPR图的实现是定位预设解剖结构的曲线路径对应的平面，然后利用曲线路径对应的平面自动获得预设解剖结构的CMPR图像：
如图7a所示，根据目标脊柱的曲线路径，并按照预设厚度确定曲线路径对应的多个曲面，将该多个曲面进行重建后得到该目标脊柱的CMPR图。
如图7b所示，根据椎弓的曲线路径，并按照预设厚度确定曲线路径对应的多个曲面，将该多个曲面进行重建后得到该椎弓的CMPR图。
如图7c所示，根据椎体的曲线路径，并按照预设厚度确定曲线路径对应的多个曲面，将该多个曲面进行重建后得到该椎体的CMPR图。
需要说明的是，根据多个曲面确定预设解剖结构的CMPR图可以参阅上述描述，此处不再赘述。
在一些可能的实现方式中，当确定预设解剖结构的CMPR图后，还可以将该CMPR图进行伪彩标记，得到该预设解剖结构的伪彩图，从而使得该预设解剖结构显示的更加清晰，便于医生观察。
204、显示目标图像。
超声成像***10中的显示器104可以显示经过上述步骤得到的立体VR图、二维切面图和多曲面重建CMPR图中的至少一种。
可以理解的是，显示目标图像的流程既可以通过用户按键方式触发，也可以直接嵌入到超声***中，默认直接打开，也可以是其他条件触发的显示，此处不做具体限定。本申请所示的实施例，既适用于三维成像，也适用于四维成 像。其中，四维成像是将实时采集到的多个三维脊柱体数据依次通过本申请流程动态的显示出来。
可选的，在显示目标图像的时候，可以在超声成像***10的显示器104上全部显示，也可以由用户选择显示目标脊柱的VR图、二维切面图、或者CMPR图。
可选的，VR图、二维切面图或者CMPR图分别都可以是一个或者多个。
可选的，显示目标图像之前，方法还可以包括：接收对目标图像的显示指令；根据显示指令显示目标图像。超声成像***10中的显示器104接收到显示指令后，可以根据显示指令，显示目标图像。
可选的，显示目标图像之后，方法还可以包括：接收对目标图像的隐藏指令；根据隐藏指令隐藏目标图像。超声成像***10中的显示器104在显示目标图像后，用户不需要再观察该目标图像的情况下，可以触发隐藏操作，超声成像***10中的显示器104接收到隐藏指令后，可以根据隐藏指令，隐藏目标图像。
在本申请实施例中，获取目标脊柱的三维体数据；从所述三维体数据中确定预设解剖结构；获取该预设解剖结构的目标图像并显示该预设解剖结构，其中，该目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种。即超声成像***可以自动获取目标脊柱的目标图像，不需要用户手动选择，就可以直观的看到目标脊柱的立体图、二维切面图或者多曲面重建CMPR图，提高超声成像的效率以及准确度，有效的帮助医生进行疾病辅助诊断，提升工作效率，节约医生的时间和精力。
一个实施例中，提供了一种超声成像方法。在该方法中，处理器可以获取胎儿的三维体数据。胎儿的三维体数据可以是通过前述超声成像***实时扫查获得的，即，超声探头向胎儿发射超声波并接收超声回波，获得超声回波信号，处理器处理该超声回波信号，获得胎儿的三维体数据；该胎儿的三维体数据也可以是由超声成像设备预先扫描获得并存储的，在需要的进行本实施例的方法时由处理器读入的。
然后，处理器可以基于胎儿脊髓圆锥的特征，从胎儿的三维体数据中识别出脊髓圆锥区域，根据识别出的脊髓圆锥区域，确定脊髓圆锥区域的位置，并 显示该脊髓圆锥区域的位置。这里，可以使用各种适合的方式显示脊髓圆锥区域的位置，例如通过适合的符号、彩色区域、文字、箭头、几何形状等等。脊髓圆锥区域的位置可以显示在三维图像上，也可以显示在二维图像上，或者也可以显示在其他位置处。
该实施例中，确定的脊髓圆锥区域的位置可以是脊髓圆锥末端的位置。例如，可以根据识别出的脊髓圆锥区域确定脊髓圆锥末端的位置，并将该脊髓圆锥末端的位置显示出来。脊髓圆锥末端的位置可以通过多种适合的方式显示，例如，适合的复合、颜色、圆点、线、箭头、数字、与适合的参考位置的距离，等等。脊髓圆锥末端的位置可以显示在三维图像上，也可以显示在二维图像上，或者也可以显示在其他位置处。
例如，如图8所示，虚线框A中即为脊髓圆锥区域，箭头B所在的虚拟直线和箭头C所在的虚拟直线指示出了脊髓圆锥末端，图中显示为一个小圆点。
该实施例中，可以通过目标匹配的方法识别脊髓圆锥区域。例如，可以从胎儿的三维体数据中确定至少两个第二候选区域，获取每一第二候选区域的三维体数据的体数据特征，根据每一第二候选区域的体数据特征，确定每一第二候选区域与脊髓圆锥的第二匹配度，并确定第二匹配度最高的第二候选区域为脊髓圆锥区域。
该实施例中，也可以先从胎儿的三维体数据中识别胎儿的矢状面，然后在从矢状面图像中识别脊髓圆锥区域。例如，可以根据经过胎儿脊柱的矢状面的特征，从胎儿的三维体数据中确定经过胎儿的脊柱的矢状面图像，然后基于脊髓圆锥的特征，在经过胎儿的脊柱的矢状面图像中确定脊髓圆锥区域。这里，矢状面可以为正中矢状面和/或邻近正中矢状面的矢状面。
该实施例中，还可以从胎儿的三维体数据中识别出腰椎区域，并相对于腰椎区域显示脊髓圆锥区域的位置，从而使得用户可以很方便地看到脊髓圆锥与腰椎之间的相对位置关系。例如，可以基于胎儿腰椎的特征，从胎儿的三维体数据中识别出腰椎区域，显示腰椎区域的超声图像，并相对于腰椎区域的超声图像显示脊髓圆锥区域(例如，脊髓圆锥末端，等等)的位置。这里，相对于腰椎区域的超声图像显示脊髓圆锥区域的位置可以包括多种适合的方式，例如，可以同时显示腰椎区域的超声图像和脊髓圆锥区域的位置，使得用户可以 直接看到二者的相对位置关系，或者通过文字或者符合等等显示脊髓圆锥区域相对于腰椎区域的距离，或者通过代表腰椎和脊髓圆锥区域的符号显示二者之间的相对位置关系，等等。
该实施例中，涉及的各个步骤(例如，识别脊髓圆锥区域、识别腰椎区域、识别胎儿的矢状面，等等)的具体方案可以参考前述各个实施例中的方法或者与其前述各个实施例中的类似步骤相同或者类似，这里不再一一详述。
在本申请所提供的几个实施例中，应该理解到，所揭露的***，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个***，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，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，仅为本申请的具体实施方式，但本申请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申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (37)

  1. 一种脊柱的成像方法，其特征在于，包括：
    获取胎儿的三维体数据；
    基于胎儿脊柱的特征，从所述三维体数据中识别出胎儿脊柱的图像区域；
    根据识别出的胎儿脊柱的图像区域获取所述胎儿脊柱的目标图像，其中，所述目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种；
    显示所述目标图像。
  2. 根据权利要求1所述的方法，其特征在于，基于胎儿脊柱的特征，从所述三维体数据中识别出胎儿脊柱的图像区域，包括：
    确定胎儿脊柱的特征信息；
    根据胎儿脊柱的特征信息从所述三维体数据中确定胎儿脊柱的图像区域。
  3. 根据权利要求2所述的方法，其特征在于，根据胎儿脊柱的特征信息从所述三维体数据中确定胎儿脊柱的图像区域包括：
    利用预设学习算法确定胎儿脊柱的检测模型；
    将所述三维体数据输入所述检测模型以根据胎儿脊柱的所述特征信息输出确定胎儿脊柱的图像区域。
  4. 根据权利要求2所述的方法，其特征在于，根据胎儿脊柱的特征信息从所述三维体数据中确定胎儿脊柱的图像区域包括：
    利用图像分割方法对所述三维体数据进行图像分割，得到若干个候选区域；
    根据胎儿脊柱的所述特征信息以及所述若干个候选区域的图像特征，确定所述若干个候选区域为胎儿脊柱区域的概率；
    从概率大于预设阈值的候选区域中确定胎儿脊柱的图像区域。
  5. 根据权利要求2所述的方法，其特征在于，根据胎儿脊柱的特征信息从所述三维体数据中确定胎儿脊柱的图像区域包括：
    利用模板匹配方法将所述三维体数据和胎儿脊柱模板进行相似度匹配，从所述三维体数据中确定与所述胎儿脊柱模板相似度高的区域；
    从所述相似度高的区域中确定所述胎儿脊柱的图像区域。
  6. 根据权利要求1-5中任一项所述的方法，其特征在于，所述目标图像 为VR图，所述根据识别出的胎儿脊柱的图像区域获取所述胎儿脊柱的目标图像包括：
    在所述三维体数据中将所述胎儿脊柱的图像区域旋转至目标方位；
    确定所述胎儿脊柱的图像区域对应的感兴趣区域的大小和位置；
    调节所述感兴趣区域的大小和位置以使得所述感兴趣区域包围所述胎儿脊柱的图像区域；
    对所述感兴趣区域进行渲染得到所述胎儿脊柱的VR图。
  7. 根据权利要求1-5中任一项所述的方法，其特征在于，所述目标图像为二维切面图，所述根据识别出的胎儿脊柱的图像区域获取所述胎儿脊柱的目标图像包括：
    从所述胎儿脊柱的图像区域选取至少三个目标像素点；
    根据所述至少三个目标像素点的位置生成二维平面；
    从所述三维体数据中确定所述二维平面对应的二维切面图；
    将所述二维平面对应的二维切面图像确定为所述胎儿脊柱的二维切面图。
  8. 根据权利要求1-5中任一项所述的方法，其特征在于，根据识别出的胎儿脊柱的图像区域获取所述胎儿脊柱的目标图像包括：
    从所述三维体数据中确定胎儿脊柱的图像区域的曲线路径；
    按照预设厚度从所述三维体数据中确定所述曲线路径对应的多个曲面；
    将所述多个曲面进行重建得到所述胎儿脊柱的图像区域的CMPR图。
  9. 根据权利要求8所述的方法，其特征在于，所述将所述多个曲面进行重建得到所述预设解剖结构的CMPR图，包括：
    获取所述多个曲面上目标像素点的像素值，其中，所述目标像素点为所述多个曲面分别在目标位置上对应的全部像素点或者部分像素点；
    根据所述多个曲面上目标像素点的像素值确定胎儿脊柱的图像区域的目标曲面；
    将所述目标曲面按照预设方向拉直得到所述胎儿脊柱的图像区域的CMPR图。
  10. 根据权利要求9所述的方法，其特征在于，根据所述多个曲面上目标像素点的像素值确定胎儿脊柱的图像区域的目标曲面，包括：
    根据所述多个曲面上目标像素点的像素值确定所述目标像素点的平均像素值；
    根据所述目标像素点的平均像素值确定所述目标曲面。
  11. 根据权利要求9所述的方法，其特征在于，所述根据所述多个曲面上目标像素点的像素值确定胎儿脊柱的图像区域的目标曲面，包括：
    将所述多个曲面上目标像素点的像素值进行加权求和得到所述目标曲面，其中，每个目标像素点对应的权重系数是用户自定义的或者***默认的。
  12. 根据权利要求8至11任一项所述的方法，所述方法还包括：
    将所述CMPR图进行伪彩标记，得到所述胎儿脊柱的图像区域的伪彩图。
  13. 一种脊柱的成像方法，其特征在于，包括：
    获取扫描目标的三维体数据；
    从所述三维体数据中确定预设解剖结构，其中，所述预设解剖结构为所述目标脊柱的部分解剖结构或者全部解剖结构；
    获取所述预设解剖结构的目标图像，其中，所述目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种；
    显示所述目标图像。
  14. 根据权利要求13所述的方法，其特征在于，所述从所述三维体数据中确定预设解剖结构，包括：
    控制所述三维体数据显示于显示器；
    响应对所述三维体数据的输入操作以确定感兴趣区域，其中,所述感兴趣区域包括所述预设解剖结构。
  15. 根据权利要求14所述的方法，其特征在于，所述响应对所述三维体数据的输入操作以确定感兴趣区域，包括：
    接收对所述三维体数据绘制的目标框的输入操作,根据绘制的所述目标框确定所述感兴趣区域；
    或者,接收对所述三维体数据绘制的点或者线的输入操作,根据绘制的所述点或者线确定所述感兴趣区域。
  16. 根据权利要求13所述的方法，其特征在于，所述从所述三维体数据中确定预设解剖结构，包括：
    确定所述预设解剖结构的特征信息；
    根据所述预设解剖结构的特征信息从所述三维体数据中确定所述预设解剖结构。
  17. 根据权利要求16所述的方法，其特征在于，所述根据所述预设解剖结构的特征信息从所述三维体数据中确定所述预设解剖结构，包括：
    利用预设学习算法确定所述预设解剖结构的检测模型；
    将所述三维体数据输入所述检测模型以根据所述预设解剖结构的特征信息输出感兴趣区域，其中，所述感兴趣区域包括所述预设解剖结构。
  18. 根据权利要求16所述的方法，其特征在于，所述根据所述预设解剖结构的特征信息从所述三维体数据中确定所述预设解剖结构，包括：
    利用图像分割方法对所述三维体数据进行图像分割，得到若干个候选区域；
    根据所述预设解剖结构的特征信息以及所述若干个候选区域的图像特征，确定所述若干个候选区域是感兴趣区域的概率，其中，所述感兴趣区域包括所述预设解剖结构；
    从概率大于预设阈值的候选区域中确定所述预设解剖结构。
  19. 根据权利要求16所述的方法，其特征在于，所述根据所述预设解剖结构的特征信息从所述三维体数据中确定所述预设解剖结构，包括：
    利用模板匹配方法将所述三维体数据和预设解剖结构模板进行相似度匹配；
    根据所述预设解剖结构的特征信息从所述三维体数据中确定相似度最高的区域；
    从所述相似度最高的区域中确定所述预设解剖结构。
  20. 根据权利要求13-19中任一项所述的方法，其特征在于，所述目标图像为VR图，所述获取所述预设解剖结构的目标图像，包括：
    在所述三维体数据中将所述预设解剖结构旋转至目标方位；
    确定所述预设解剖结构对应的感兴趣区域的大小和位置；
    调节所述感兴趣区域的大小和位置以使得所述感兴趣区域包围所述预设解剖结构；
    对所述感兴趣区域进行渲染得到所述预设解剖结构的VR图。
  21. 根据权利要求13-19中任一项所述的方法，其特征在于，所述目标图像为二维切面图，所述获取所述预设解剖结构的目标图像，包括：
    从所述预设解剖结构中选取至少三个目标像素点；
    根据所述至少三个目标像素点的位置生成二维平面；
    从所述三维体数据中确定所述二维平面对应的二维切面图；
    将所述二维平面对应的二维切面图确定为所述预设解剖结构的二维切面图。
  22. 根据权利要求13-19中任一项所述的方法，其特征在于，所述目标图像为CMPR图，所述获取所述预设解剖结构的目标图像，包括：
    从所述三维体数据中确定所述预设解剖结构的曲线路径；
    按照预设厚度从所述三维体数据中确定所述曲线路径对应的多个曲面；
    将所述多个曲面进行重建得到所述预设解剖结构的CMPR图。
  23. 根据权利要求22所述的方法，其特征在于，所述将所述多个曲面进行重建得到所述预设解剖结构的CMPR图，包括：
    获取所述多个曲面上目标像素点的像素值，其中，所述目标像素点为所述多个曲面分别在目标位置上对应的全部像素点或者部分像素点；
    根据所述多个曲面上目标像素点的像素值确定预设解剖结构的目标曲面；
    将所述目标曲面按照预设方向拉直得到所述预设解剖结构的CMPR图。
  24. 根据权利要求23所述的方法，其特征在于，根据所述多个曲面上目标像素点的像素值确定预设解剖结构的目标曲面，包括：
    根据所述多个曲面上目标像素点的像素值确定所述目标像素点的平均像素值；
    根据所述目标像素点的平均像素值确定所述目标曲面。
  25. 根据权利要求23所述的方法，其特征在于，所述根据所述多个曲面上目标像素点的像素值确定预设解剖结构的目标曲面，包括：
    将所述多个曲面上目标像素点的像素值进行加权求和得到所述目标曲面，其中，每个目标像素点对应的权重系数是用户自定义的或者***默认的。
  26. 根据权利要求22至25任一项所述的方法，所述方法还包括：
    将所述CMPR图进行伪彩标记，得到所述预设解剖结构的伪彩图。
  27. 根据权利要求13所述的方法，其特征在于，所述获取目标脊柱的三维体数据，包括：
    向所述目标脊柱发送超声波；
    接收所述目标脊柱返回的超声回波；
    根据所述超声回波确定所述三维体数据。
  28. 根据权利要求13所述的方法，其特征在于，所述显示所述目标图像之前，所述方法还包括：
    接收对所述目标图像的显示指令；
    根据所述显示指令显示所述目标图像。
  29. 根据权利要求13所述的方法，其特征在于，所述显示所述目标图像之后，所述方法还包括：
    接收对所述目标图像的隐藏指令；
    根据隐藏指令隐藏所述目标图像。
  30. 一种脊柱的成像方法，其特征在于，包括：
    获取胎儿的三维体数据；
    基于胎儿脊髓圆锥的特征，从所述胎儿的三维体数据中识别出脊髓圆锥区域；
    根据识别出的脊髓圆锥区域，确定脊髓圆锥区域的位置；
    显示所述脊髓圆锥区域的位置。
  31. 根据权利要求30所述的方法，其特征在于：
    根据识别出的脊髓圆锥区域确定脊髓圆锥区域的位置包括：根据识别出的脊髓圆锥区域确定脊髓圆锥末端的位置；
    显示所述脊髓圆锥区域的位置包括：显示所述脊髓圆锥末端的位置。
  32. 根据权利要求30或31所述的方法，其特征在于，基于胎儿脊髓圆锥的特征从所述胎儿的三维体数据中识别出脊髓圆锥区域包括：
    从所述胎儿的三维体数据中确定至少两个第二候选区域，获取每一第二候选区域的三维体数据的体数据特征；
    根据所述每一第二候选区域的体数据特征，确定每一第二候选区域与脊髓 圆锥的第二匹配度；
    确定第二匹配度最高的第二候选区域为所述脊髓圆锥区域。
  33. 根据权利要求30或31所述的方法，其特征在于，基于胎儿脊髓圆锥的特征从所述胎儿的三维体数据中识别出脊髓圆锥区域包括：
    根据经过胎儿脊柱的矢状面的特征，从所述胎儿的三维体数据中确定经过所述胎儿的脊柱的矢状面图像；
    基于脊髓圆锥的特征，在经过所述胎儿的脊柱的矢状面图像中确定脊髓圆锥区域。
  34. 根据权利要求33所述的方法，其特征在于：所述矢状面为正中矢状面和/或邻近正中矢状面的矢状面。
  35. 根据权利要求30至34中任意一项所述的方法，其特征在于，还包括：
    基于胎儿腰椎的特征，从所述胎儿的三维体数据中识别出腰椎区域；
    显示所述腰椎区域的超声图像；
    其中显示脊髓圆锥区域的所述位置包括：相对于所述腰椎区域的超声图像显示脊髓圆锥区域的所述位置。
  36. 一种超声成像***，其特征在于，包括：
    超声探头，所述超声探头向扫描目标发射超声波并接收超声回波，获得超声回波信号；
    处理器，所述处理器根据所述超声回波信号获得目标脊柱的三维体数据，并从所述三维体数据中确定所述目标脊柱的预设解剖结构，其中，所述目标脊柱的预设解剖结构为所述目标脊柱的部分解剖结构或者全部解剖结构；获取所述预设解剖结构的目标图像，其中，所述目标图像包括立体VR图、二维切面图和多曲面重建CMPR图中的至少一种；
    显示器，所述显示器显示所述目标图像。
  37. 一种超声成像***，其特征在于，包括：
    超声探头，所述超声探头向胎儿发射超声波并接收超声回波，获得超声回波信号；
    处理器，所述处理器根据所述超声回波信号获得所述胎儿的三维体数据，基于胎儿脊髓圆锥的特征从所述胎儿的三维体数据中识别出脊髓圆锥区域，并 根据识别出的脊髓圆锥区域，确定脊髓圆锥区域的位置；
    显示器，所述显示器显示所述脊髓圆锥区域的位置。
PCT/CN2018/124950 2018-12-28 2018-12-28 一种脊柱的成像方法以及超声成像*** WO2020133236A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2018/124950 WO2020133236A1 (zh) 2018-12-28 2018-12-28 一种脊柱的成像方法以及超声成像***
CN201880097198.6A CN112654301A (zh) 2018-12-28 2018-12-28 一种脊柱的成像方法以及超声成像***

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/124950 WO2020133236A1 (zh) 2018-12-28 2018-12-28 一种脊柱的成像方法以及超声成像***

Publications (1)

Publication Number Publication Date
WO2020133236A1 true WO2020133236A1 (zh) 2020-07-02

Family

ID=71128474

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/124950 WO2020133236A1 (zh) 2018-12-28 2018-12-28 一种脊柱的成像方法以及超声成像***

Country Status (2)

Country Link
CN (1) CN112654301A (zh)
WO (1) WO2020133236A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4183349A1 (en) * 2021-11-23 2023-05-24 Koninklijke Philips N.V. Fetal ultrasound image processing

Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1676104A (zh) * 2004-04-01 2005-10-05 株式会社美蒂森 用于形成3d超声图像的设备和方法
CN101496068A (zh) * 2006-07-25 2009-07-29 皇家飞利浦电子股份有限公司 用于弯曲多切片显示的方法和装置
CN101639858A (zh) * 2009-08-21 2010-02-03 深圳创维数字技术股份有限公司 基于目标区域匹配的图像检索方法
CN102113897A (zh) * 2009-12-31 2011-07-06 深圳迈瑞生物医疗电子股份有限公司 一种在图像中提取及测量感兴趣目标的方法及其装置
CN102842125A (zh) * 2011-05-09 2012-12-26 奥林巴斯株式会社 图像处理装置及图像处理方法
US8423124B2 (en) * 2007-05-18 2013-04-16 Siemens Aktiengesellschaft Method and system for spine visualization in 3D medical images
CN103263278A (zh) * 2013-01-23 2013-08-28 郑末晶 一种从超声图像上自动测量胎儿颈背透明物厚度的图像处理方法
CN103955698A (zh) * 2014-03-12 2014-07-30 深圳大学 从超声图像中自动定位标准切面的方法
CN105342641A (zh) * 2015-11-20 2016-02-24 深圳开立生物医疗科技股份有限公司 一种超声成像方法、装置及其超声设备
CN105433988A (zh) * 2015-12-28 2016-03-30 深圳开立生物医疗科技股份有限公司 一种目标图像识别方法、装置及其超声设备
CN106659480A (zh) * 2014-08-22 2017-05-10 株式会社日立制作所 超声波诊断图像生成装置以及方法
CN107316315A (zh) * 2017-05-04 2017-11-03 佛山市南海区广工大数控装备协同创新研究院 一种基于模板匹配的目标识别定位方法
CN108013892A (zh) * 2016-10-31 2018-05-11 北京东软医疗设备有限公司 多平面重建方法和***
CN108520519A (zh) * 2018-04-11 2018-09-11 上海联影医疗科技有限公司 一种图像处理方法、装置及计算机可读存储介质
CN108778142A (zh) * 2016-03-01 2018-11-09 皇家飞利浦有限公司 颈部褶皱半透明带的自动超声测量

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1676104A (zh) * 2004-04-01 2005-10-05 株式会社美蒂森 用于形成3d超声图像的设备和方法
CN101496068A (zh) * 2006-07-25 2009-07-29 皇家飞利浦电子股份有限公司 用于弯曲多切片显示的方法和装置
US8423124B2 (en) * 2007-05-18 2013-04-16 Siemens Aktiengesellschaft Method and system for spine visualization in 3D medical images
CN101639858A (zh) * 2009-08-21 2010-02-03 深圳创维数字技术股份有限公司 基于目标区域匹配的图像检索方法
CN102113897A (zh) * 2009-12-31 2011-07-06 深圳迈瑞生物医疗电子股份有限公司 一种在图像中提取及测量感兴趣目标的方法及其装置
CN102842125A (zh) * 2011-05-09 2012-12-26 奥林巴斯株式会社 图像处理装置及图像处理方法
CN103263278A (zh) * 2013-01-23 2013-08-28 郑末晶 一种从超声图像上自动测量胎儿颈背透明物厚度的图像处理方法
CN103955698A (zh) * 2014-03-12 2014-07-30 深圳大学 从超声图像中自动定位标准切面的方法
CN106659480A (zh) * 2014-08-22 2017-05-10 株式会社日立制作所 超声波诊断图像生成装置以及方法
CN105342641A (zh) * 2015-11-20 2016-02-24 深圳开立生物医疗科技股份有限公司 一种超声成像方法、装置及其超声设备
CN105433988A (zh) * 2015-12-28 2016-03-30 深圳开立生物医疗科技股份有限公司 一种目标图像识别方法、装置及其超声设备
CN108778142A (zh) * 2016-03-01 2018-11-09 皇家飞利浦有限公司 颈部褶皱半透明带的自动超声测量
CN108013892A (zh) * 2016-10-31 2018-05-11 北京东软医疗设备有限公司 多平面重建方法和***
CN107316315A (zh) * 2017-05-04 2017-11-03 佛山市南海区广工大数控装备协同创新研究院 一种基于模板匹配的目标识别定位方法
CN108520519A (zh) * 2018-04-11 2018-09-11 上海联影医疗科技有限公司 一种图像处理方法、装置及计算机可读存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4183349A1 (en) * 2021-11-23 2023-05-24 Koninklijke Philips N.V. Fetal ultrasound image processing
WO2023094102A1 (en) 2021-11-23 2023-06-01 Koninklijke Philips N.V. Fetal ultrasound image processing

Also Published As

Publication number Publication date
CN112654301A (zh) 2021-04-13

Similar Documents

Publication Publication Date Title
CN109791692B (zh) 使用来自感兴趣区域的不同视角的多个图像进行计算机辅助检测以提高检测准确度的***和方法
US11100665B2 (en) Anatomical measurements from ultrasound data
US11488298B2 (en) System and methods for ultrasound image quality determination
US20110255762A1 (en) Method and system for determining a region of interest in ultrasound data
US20110125016A1 (en) Fetal rendering in medical diagnostic ultrasound
KR102539901B1 (ko) 2차원 초음파 이미지를 음영화하는 방법 및 시스템
US9390546B2 (en) Methods and systems for removing occlusions in 3D ultrasound images
US11607200B2 (en) Methods and system for camera-aided ultrasound scan setup and control
CN111368586B (zh) 超声成像方法及***
KR102063374B1 (ko) 초음파 볼륨의 자동 정렬
US11278259B2 (en) Thrombus detection during scanning
US11250564B2 (en) Methods and systems for automatic measurement of strains and strain-ratio calculation for sonoelastography
EP1952359A1 (en) System and method for generating for display two-dimensional echocardiography views from a three-dimensional image
JP5982602B2 (ja) 解剖学的に知的な心エコー法における肺組織同定
EP3579757B1 (en) Ultrasound evaluation of anatomical features
WO2024093911A1 (zh) 一种超声成像方法及超声设备
KR102419310B1 (ko) 초음파 이미징 데이터로부터 태아 이미지들을 프로세싱 및 디스플레이하기 위한 방법들 및 시스템들
WO2020133236A1 (zh) 一种脊柱的成像方法以及超声成像***
WO2022133806A1 (zh) 胎儿颜面部容积图像修复方法和超声成像***
KR101024857B1 (ko) 3차원 초음파 영상에 컬러 모델링 처리를 수행하는 초음파 시스템 및 방법
WO2022134049A1 (zh) 胎儿颅骨的超声成像方法和超声成像***
JP2024522311A (ja) 不均一な肝脂肪を特定するためのシステム、方法、及び装置
CN114642451A (zh) 一种超声成像装置
CN116258736A (zh) 用于分割图像的***和方法
CN113939234A (zh) 超声波诊断装置、医用图像处理装置以及医用图像处理方法

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: 18945232

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 12/11/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 18945232

Country of ref document: EP

Kind code of ref document: A1