CN110610481A - Automatic brain stem segmentation method and system based on medical image - Google Patents

Automatic brain stem segmentation method and system based on medical image Download PDF

Info

Publication number
CN110610481A
CN110610481A CN201910721119.0A CN201910721119A CN110610481A CN 110610481 A CN110610481 A CN 110610481A CN 201910721119 A CN201910721119 A CN 201910721119A CN 110610481 A CN110610481 A CN 110610481A
Authority
CN
China
Prior art keywords
brainstem
image
region
segmentation
automatic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN201910721119.0A
Other languages
Chinese (zh)
Inventor
汪辉
伍亚军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SHENZHEN YORKTAL DMIT CO Ltd
Original Assignee
SHENZHEN YORKTAL DMIT CO Ltd
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 SHENZHEN YORKTAL DMIT CO Ltd filed Critical SHENZHEN YORKTAL DMIT CO Ltd
Priority to CN201910721119.0A priority Critical patent/CN110610481A/en
Publication of CN110610481A publication Critical patent/CN110610481A/en
Withdrawn legal-status Critical Current

Links

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
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20092—Interactive image processing based on input by user
    • G06T2207/20104—Interactive definition of region of interest [ROI]
    • 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
    • G06T2207/30016—Brain

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

The invention is suitable for the technical field of three-dimensional visualization processing of medical images and computer medical auxiliary diagnosis and treatment systems, and provides a brain stem automatic segmentation method based on medical images, which comprises the following steps: an image acquisition step of respectively acquiring a CT image and a digital human image; an image registration step, namely registering the CT image and the digital human image to obtain a registration result; a region of interest selection step, selecting a brainstem region of interest in the registration result; and an automatic segmentation step, namely automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result. The invention also correspondingly provides an automatic brainstem segmentation system based on the medical image. Therefore, the automatic brain stem segmentation method and the device can achieve the high-precision automatic brain stem segmentation effect, avoid manual drawing by doctors, waste time and labor and save precious time for the doctors.

Description

Automatic brain stem segmentation method and system based on medical image
Technical Field
The invention relates to the technical field of medical image three-dimensional visualization processing and computer medical auxiliary diagnosis and treatment systems, in particular to a method and a system for automatically segmenting a brainstem based on a medical image.
Background
Among various organ tissues of a human body, the brainstem is a very important sensitive organ, and the irradiation dose is required to be as small as possible in the radiation treatment process, otherwise, unpredictable complications such as brainstem damage and the like are easily caused. For brainstem segmentation, manual delineation is mainly performed on one layer of image at present, and although some semi-automatic or full-automatic methods appear, an effective method for automatically segmenting the brainstem for a CT image is lacked mainly aiming at an MR image. Therefore, the automatic brain stem delineation based on the CT image is further improved in clinical application.
In view of the above, the prior art is obviously inconvenient and disadvantageous in practical use, and needs to be improved.
Disclosure of Invention
In view of the above-mentioned drawbacks, the present invention aims to provide a method and a system for automatically segmenting a brainstem based on a medical image, which can achieve a high-precision automatic segmentation effect of the brainstem, and avoid time and labor waste caused by manual drawing by a doctor, thereby saving precious time for the doctor.
In order to achieve the above object, the present invention provides a method for automatically segmenting a brainstem based on a medical image, comprising:
an image acquisition step of respectively acquiring a CT image and a digital human image;
an image registration step, namely registering the CT image and the digital human image to obtain a registration result;
a region of interest selection step, selecting a brainstem region of interest in the registration result;
and an automatic segmentation step, namely automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result.
According to the automatic brain stem segmentation method based on the medical image, the image registration step further comprises the following steps:
a head region is selected in the CT image as a head region of interest.
According to the automatic brain stem segmentation method based on the medical image, the image registration step further comprises the following steps:
and performing fusion display on the registration result and the CT image.
According to the automatic brain stem segmentation method based on the medical image, the automatic segmentation algorithm is an edge-limited region growing algorithm.
According to the automatic brain stem segmentation method based on the medical image, the edge-limited region growing algorithm comprises the following steps:
the method comprises the following steps: taking the central point of each layer of image on the brain stem region of interest as a seed point, and putting the seed point into a seed point set SseedPerforming the following steps;
step two: from the set of seed points SseedTaking out one seed point, and judging whether the seed point is a non-edge point or not;
step three: if the seed point is the non-edge point (X, Y), all pixel points (X) in the neighborhood range of the non-edge point (X, Y)9 are takenn,Yn) (n is a neighborhood pixel index in the range of 1. ltoreq. n.ltoreq.9), where X-1 ≦ Xn＜＝X+1,Y-1＜＝YnAnd < Y +1, and sequentially judging each pixel point (X)n,Yn) Whether the image is a pixel point of the brainstem area or not; if the pixel point (X)n,Yn) Is the pixel point of the brainstem area, the pixel point (X) isn,Yn) Put into the SseedAnd a set S of pixels in the brainstem regionBrainPerforming the following steps;
step four: repeating the second step and the third step until no pixel point (X) existsn,Yn) Put into the collection S of the pixel points in the brainstem areaBrainObtaining the final set S of the pixel points in the brainstem areaBrainAnd finally, collecting the pixel points S in the brainstem areaBrainAnd obtaining the brain stem region segmentation result.
According to the automatic brain stem segmentation method based on the medical image, in the second step, the step of judging whether the seed points are non-edge points comprises the following steps: if the seed pointThe gradient G in the up, down, left and right directionsup＝abs(P-Pup)、Gdown＝abs(P-Pdown)、Gleft＝abs(P-Pleft) And Gright＝P-PrightAll are less than or equal to a preset edge gradient threshold value TedgeIf so, the seed point is considered as the non-edge point;
wherein P is the pixel value of the seed point, the Pup、Pdown、Pleft、PrightPixel values of pixel points adjacent to the seed point in four directions, namely, upper, lower, left and right directions are respectively set;
in the third step, each pixel point (X) is judged in turnn,Yn) The step of judging whether the pixel points are pixel points of the brainstem area comprises the following steps: if the pixel point (X)n,Yn) The pixel value of (A) is greater than or equal to a preset brainstem threshold value TBrainThen the pixel point (X) is consideredn,Yn) The pixel points of the brainstem area are obtained.
According to the automatic brain stem segmentation method based on the medical image, the preset edge gradient threshold value Tedge＝50；
The preset brainstem threshold TBrain＝150。
According to the automatic brain stem segmentation method based on the medical image, the automatic segmentation step is further followed by:
and a segmentation result fusion display step, namely performing fusion display on the segmentation result of the brainstem region and the CT image.
According to the automatic brain stem segmentation method based on the medical image, the step of fusing and displaying the segmentation result further comprises the following steps:
and performing three-dimensional reconstruction on the brain stem region segmentation result.
The invention also provides a brain stem automatic segmentation system based on medical images, which comprises:
the image acquisition module is used for respectively acquiring a CT image and a digital human image;
the image registration module is used for registering the CT image and the digital human image to obtain a registration result;
a region-of-interest selection module for selecting a brainstem region-of-interest in the registration result;
and the automatic segmentation module is used for automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result.
The automatic brain stem segmentation method based on the medical image comprises the following steps: respectively acquiring a CT image and a digital human image; registering the CT image and the digital human image to obtain a registration result; selecting a brainstem region of interest in the registration result; and automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result. Therefore, the automatic brain stem segmentation method and the device can achieve the high-precision automatic brain stem segmentation effect, avoid manual drawing by doctors, waste time and labor and save precious time for the doctors.
Drawings
FIG. 1 is a schematic structural diagram of an automatic brain stem segmentation system based on medical images according to the present invention;
FIG. 2 is a flow chart of the automatic brain stem segmentation method based on medical images according to the present invention;
FIG. 3 is a flow chart of a method for automatic segmentation of a brain stem based on medical images according to a preferred embodiment of the present invention;
FIG. 4 is a sagittal view of a CT image used in an embodiment of the medical image based automatic brainstem segmentation method of the present invention;
FIG. 5 is a coronal plane view of a CT image used in an embodiment of the medical image based automatic brainstem segmentation method of the present invention;
FIG. 6 is a cross-sectional view of a CT image used in an embodiment of the method for automatic segmentation of a medical image-based brainstem according to the present invention;
FIG. 7 is a sagittal view of a digital human image used by an embodiment of the medical image based automatic brainstem segmentation method of the present invention;
FIG. 8 is a coronal plane view of a digital human image used by an embodiment of the medical image based automatic brainstem segmentation method of the present invention;
FIG. 9 is a cross-sectional view of a digital human image used by an embodiment of the medical image-based brainstem automatic segmentation method of the present invention;
FIG. 10 is a schematic sagittal view of a selected region of interest of the head in a CT image for use in an embodiment of a method for automatic segmentation of the brainstem based on medical images according to the present invention;
FIG. 11 is a schematic coronal view of a selected region of interest of a head in a CT image used in an embodiment of a method for automatic segmentation of a brain stem based on medical images according to the present invention;
FIG. 12 is a cross-sectional view of a selected region of interest of the head in a CT image used in an embodiment of the method for automatic segmentation of the brainstem based on medical images according to the present invention;
FIG. 13 is a schematic sagittal plane view of the registration result of the automatic brain stem segmentation method based on medical images according to the embodiment of the present invention;
FIG. 14 is a coronal view of a registration result of an embodiment of the medical image based automatic brainstem segmentation method of the present invention;
FIG. 15 is a cross-sectional diagram illustrating the registration result of an embodiment of the automatic brain stem segmentation method based on medical images according to the present invention;
FIG. 16 is a schematic sagittal plane view of a region of interest selected from the registration results of the automatic brain stem segmentation method based on medical images according to the embodiment of the present invention;
FIG. 17 is a schematic coronal view of a region of interest selected from the registration results of an embodiment of the automatic brain stem segmentation method based on medical images;
FIG. 18 is a schematic cross-sectional diagram of the selection of the brainstem region of interest in the registration result of the automatic brain stem segmentation method based on medical images according to the embodiment of the present invention;
FIG. 19 is a schematic sagittal view of the segmentation result of the brainstem region of the embodiment of the automatic brain stem segmentation method based on medical images of the present invention;
FIG. 20 is a schematic coronal view of a segmentation result of an automatic brain stem segmentation method based on medical images according to an embodiment of the present invention;
FIG. 21 is a cross-sectional diagram of the segmentation result of the brain stem region according to the embodiment of the automatic brain stem segmentation method based on medical images;
FIG. 22 is a sagittal view schematically showing the brain stem region segmentation result fused with the CT image according to the embodiment of the automatic brain stem segmentation method based on medical images;
FIG. 23 is a schematic coronal plane view of a fusion display of a segmentation result of a brain stem region and a CT image according to an embodiment of the automatic brain stem segmentation method based on medical images;
FIG. 24 is a cross-sectional view schematically illustrating the brain stem region segmentation result and the CT image fusion display according to the embodiment of the automatic brain stem segmentation method based on medical images;
fig. 25 is a schematic diagram of a three-dimensional reconstruction of a brain stem region segmentation result according to an embodiment of the automatic brain stem segmentation method based on medical images.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, 4-24, in a first embodiment of the present invention, there is provided a medical image-based automatic brainstem segmentation system 100, comprising:
an image acquisition module 10, configured to acquire a CT image and a digital human image respectively; loading a CT image and a digital human image;
an image registration module 20, configured to register the CT image and the digital human image to obtain a registration result;
a region of interest selection module 30 for selecting a brainstem region of interest in the registration result;
and the automatic segmentation module 40 is used for automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result.
In this embodiment, the automatic brainstem segmentation system 100 based on medical images utilizes a high-resolution digital human image to automatically segment the brainstem on the basis of registering the digital human image and the CT image aiming at the characteristic that the boundary of the brainstem region in the CT image is very fuzzy, so that a high-precision automatic brainstem segmentation effect can be realized; by means of the method of registering the digital human image and the CT image, the image registering module 20 selects the digital human image as a moving image and the CT image as a fixed image for registering, and the CT image and the digital human image are three-dimensional images; preferably, the image registration module 20 performs registration and then performs fusion display on the registration result and the CT image. Preferably, before registering the CT image and the digital human image, the head region is selected as the head region of interest in the CT image, and when registering, the image registration module 20 only needs to register the head region of interest with the digital human image. The interesting region selection module 30 selects the interesting region of the brain stem, and the automatic segmentation module 40 performs automatic segmentation of the brain stem on the interesting region of the brain stem through an automatic segmentation algorithm to obtain a segmentation result of the brain stem region.
In the second embodiment of the present invention, the automatic segmentation algorithm used by the automatic segmentation module 40 is an edge-limited region growing algorithm, and the automatic segmentation module includes a region growing algorithm execution sub-module 41, and the edge-limited region growing algorithm is executed by the region growing algorithm execution sub-module 41. Excessive growth to areas other than the brainstem during regional growth is avoided.
Specifically, the edge-limited region growing algorithm executed by the region growing algorithm execution sub-module 41 includes the following steps:
the method comprises the following steps: taking the central point of each layer of image on the brain stem region of interest as a seed point, and putting the seed point into a seed point set SseedPerforming the following steps;
step two: from the set of seed points SseedTaking out one seed point, and judging whether the seed point is a non-edge point or not;
step three: if the seed point is the non-edge point (X, Y), all images in the neighborhood range of the non-edge point (X, Y)9 are takenPrime point (X)n,Yn) (n is a neighborhood pixel index in the range of 1. ltoreq. n.ltoreq.9), where X-1 ≦ Xn＜＝X+1,Y-1＜＝YnAnd < Y +1, and sequentially judging each pixel point (X)n,Yn) Whether the image is a pixel point of the brainstem area or not; if the pixel point (X)n,Yn) Is the pixel point of the brainstem area, the pixel point (X) isn,Yn) Put into the SseedAnd a set S of pixels in the brainstem regionBrainPerforming the following steps;
step four: repeating the second step and the third step until no pixel point (X) existsn,Yn) Put into the collection S of the pixel points in the brainstem areaBrainObtaining the final set S of the pixel points in the brainstem areaBrainAnd finally, collecting the pixel points S in the brainstem areaBrainAnd obtaining the brain stem region segmentation result. When all the pixel points (X) are judged to be finishedn,Yn) And putting the pixels belonging to the brainstem area into a brainstem area pixel point set SBrainIn this case, the set S of pixels in the brainstem areaBrainThe segmentation result of the brain stem region is obtained.
Preferably, in the second step, the step of judging whether the seed point is a non-edge point includes: if the gradient G of the seed point in the upper, lower, left and right directionsup＝abs(P-Pup)、Gdown＝abs(P-Pdown)、Gleft＝abs(P-Pleft) And Gright＝P-PrightAll are less than or equal to a preset edge gradient threshold value TedgeIf so, the seed point is considered as the non-edge point; the preset edge gradient threshold is preferably TedgeThe accuracy of the segmentation result of the brainstem region is improved by 50;
wherein P is the pixel value of the seed point, the Pup、Pdown、Pleft、PrightPixel values of adjacent pixel points in four directions, namely, upper, lower, left and right directions of the seed point are respectively;
in the third step, each pixel point (X) is judged in turnn,Yn) The step of judging whether the pixel points are pixel points of the brainstem area comprises the following steps: if the pixel point (X)n,Yn) The pixel value of (A) is greater than or equal to a preset brainstem threshold value TBrainThen the pixel point (X) is consideredn,Yn) Pixel points of the brainstem region; the preset brainstem threshold is preferably TBrain150, the accuracy of the segmentation result of the brainstem region is improved.
In the third embodiment of the present invention, the automatic brain stem segmentation system 100 based on medical images further includes a fusion display module 50 for performing fusion display of the segmentation result of the brain stem region and the CT image.
In a fourth embodiment of the present invention, the automatic brain stem segmentation system 100 based on medical images further comprises a three-dimensional reconstruction module 60 for performing three-dimensional reconstruction on the segmentation result of the brain stem region.
Referring to fig. 2 to 24, a fifth embodiment of the present invention further provides a method for automatically segmenting a brainstem based on a medical image, including:
step S201, respectively acquiring a CT image and a digital human image; loading a CT image and a digital human image;
step S202, registering the CT image and the digital human image to obtain a registration result;
step S203, selecting a brainstem region of interest in the registration result;
and S204, automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result.
In the embodiment, the automatic brain stem segmentation method based on the medical image mainly solves the problem that the existing brain stem segmentation on the CT image depends on time consumption of manual drawing of a doctor, the registration result of the digital human image and the CT image is utilized, preferably, the registration result and the CT image are fused and displayed, the region of interest of the brain stem is selected as the region of interest of the brain stem in the registration result, and the automatic segmentation of the brain stem is carried out on the region of interest of the brain stem by utilizing an automatic segmentation algorithm.
Preferably, the step S202 further includes:
and selecting a head region as a head region of interest in the CT image, and registering the head region of interest with the digital human image only when registering.
Preferably, the step S202 further includes:
and performing fusion display on the registration result and the CT image. And selecting an interested area of the brainstem as an interested area of the brainstem on the basis of fusion display of the registration result and the CT image, and automatically segmenting the brainstem on the interested area of the brainstem by utilizing an automatic segmentation algorithm.
In a sixth embodiment of the present invention, the automatic segmentation algorithm is an edge-limited region growing algorithm.
Specifically, the edge-limited region growing algorithm comprises the following steps:
the method comprises the following steps: taking the central point of each layer of image on the brain stem region of interest as a seed point, and putting the seed point into a seed point set SseedPerforming the following steps;
step two: from the set of seed points SseedTaking out one seed point, and judging whether the seed point is a non-edge point or not;
step three: if the seed point is the non-edge point (X, Y), all pixel points (X) in the neighborhood range of the non-edge point (X, Y)9 are takenn,Yn) (n is a neighborhood pixel index in the range of 1. ltoreq. n.ltoreq.9), where X-1 ≦ Xn＜＝X+1,Y-1＜＝YnAnd < Y +1, and sequentially judging each pixel point (X)n,Yn) Whether the image is a pixel point of the brainstem area or not; if the pixel point (X)n,Yn) Is the pixel point of the brainstem area, the pixel point (X) isn,Yn) Put into the SseedAnd a set S of pixels in the brainstem regionBrainPerforming the following steps;
step four: repeating the second step and the third step until no pixel point is put into the SBrainThen the final brain stem region segmentation result S is obtainedBrain。
Preferably, in the second step, the judgment is madeThe step of determining whether the seed point is a non-edge point comprises: if the gradient G of the seed point in the upper, lower, left and right directionsup＝abs(P-Pup)、
Gdown＝abs(P-Pdown)、Gleft＝abs(P-Pleft) And Gright＝P-PrightAll are less than or equal to a preset edge gradient threshold value TedgeIf so, the seed point is considered as the non-edge point; the preset edge gradient threshold is preferably Tedge＝50；
Wherein P is the pixel value of the seed point, the Pup、Pdown、Pleft、PrightPixel values of adjacent pixel points in four directions, namely, upper, lower, left and right directions of the seed point are respectively;
in the third step, each pixel point (X) is judged in turnn,Yn) The step of judging whether the pixel points are pixel points of the brainstem area comprises the following steps: if the pixel point (X)n,Yn) The pixel value of (A) is greater than or equal to a preset brainstem threshold value TBrainThen the pixel point (X) is consideredn,Yn) Pixel points of the brainstem region; the preset brainstem threshold is preferably TBrain＝150。
Referring to fig. 3, in the seventh embodiment of the present invention, after step S204, the method further includes:
and S301, performing fusion display on the brain stem region segmentation result and the CT image, namely displaying the brain stem segmentation result on the CT image by using fusion display.
In the eighth embodiment of the present invention, after the step S301, the method further includes: step S302 performs three-dimensional reconstruction on the segmentation result of the brainstem region, and refer to fig. 25, which is a three-dimensional schematic diagram of the brainstem obtained by performing three-dimensional reconstruction on the segmentation result of the brainstem region.
The implementation process of the preferred embodiment of the invention comprises the following steps:
in step S201, a CT brainstem three-dimensional image with an image size of 512 × 196 pixels is loaded, so that the three-dimensional image can be observed from three different angles, for example, three azimuth observation effects of the CT brainstem image are shown in fig. 1 to 3, which are a sagittal view, a coronal view and a cross-sectional view of the CT image; and loading digital human images, as shown in fig. 4-6, which are sagittal plane images, coronal plane images and cross sectional images of the digital human images, respectively; selecting a head region in the CT image as a head region of interest, as shown in fig. 10, which is a schematic diagram of selecting a head region of interest in a frame in a sagittal view of the CT image, and the region 10 in a white frame is the head region of interest; fig. 11 is a schematic diagram of a region of interest of the head framed in a coronal plane of a CT image, and a region 20 within a white frame is the region of interest of the head; fig. 12 is a schematic diagram of a region of interest of the head framed in a cross-sectional view of a CT image, and a region 30 within a white frame is the region of interest of the head; in step S202, the digital human image is registered as a moving image and the CT image is registered as a fixed image, and fig. 13 to 15 are a sagittal view, a coronal view, and a transverse view of the registration result, respectively; in step S203, a brainstem region of interest is selected from the registration result, fig. 16 is a schematic diagram of selecting a brainstem region of interest in a frame in a sagittal plane of the registration result, and a region 40 in a white frame is a brainstem region of interest; fig. 17 is a schematic diagram of a region of interest of the brainstem outlined in the coronal plane of the registration result, the region 50 within the white frame being the region of interest of the brainstem; fig. 18 is a schematic diagram of a brainstem region of interest framed in a cross-sectional view of the registration result, with the region 60 within the white frame being the brainstem region of interest; in step S204, the automatic segmentation algorithm automatically segments the brainstem on the brainstem region of interest to obtain a brainstem region segmentation result, and the regions 70, 80, and 90 within the white frame in fig. 19 to 21 are the brainstem region segmentation result. In step S301, the segmentation result of the brainstem region is displayed on the CT image by fusion display, as shown in fig. 22 to 24, wherein the regions 100, 110 and 120 in the white frame are the segmentation result of the brainstem region displayed on the CT image. In step S302, the brain stem region segmentation result is three-dimensionally reconstructed, and fig. 25 is an effect diagram obtained by three-dimensionally reconstructing the obtained brain stem segmentation result.
In summary, the invention utilizes the digital human image and the CT image for registration to solve the problem that the brain stem is difficult to be automatically segmented due to fuzzy brain stem region boundary in the CT image, and the method for automatically segmenting the brain stem based on the medical image comprises the following steps: respectively acquiring a CT image and a digital human image; registering the CT image and the digital human image to obtain a registration result; selecting a brainstem region of interest in the registration result; and automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result. Therefore, the automatic brain stem segmentation method and the device can achieve the high-precision automatic brain stem segmentation effect, avoid manual drawing by doctors, waste time and labor and save precious time for the doctors.
The present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof, and it should be understood that various changes and modifications can be effected therein by one skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (10)

1. A brain stem automatic segmentation method based on medical images is characterized by comprising the following steps:
an image acquisition step of respectively acquiring a CT image and a digital human image;
an image registration step, namely registering the CT image and the digital human image to obtain a registration result;
a region of interest selection step, selecting a brainstem region of interest in the registration result;
and an automatic segmentation step, namely automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result.
2. The medical image based automatic brainstem segmentation method according to claim 1, wherein the image registration step further comprises, before:
a head region is selected in the CT image as a head region of interest.
3. The medical image based automatic brainstem segmentation method according to claim 1, wherein the image registration step further comprises:
and performing fusion display on the registration result and the CT image.
4. The medical image-based automatic brainstem segmentation method according to claim 1, wherein the automatic segmentation algorithm is an edge-limited region growing algorithm.
5. The medical image-based automatic brainstem segmentation method according to claim 4, wherein the edge-limited region growing algorithm comprises the following steps:
the method comprises the following steps: taking the central point of each layer of image on the brain stem region of interest as a seed point, and putting the seed point into a seed point set SseedPerforming the following steps;
step two: from the set of seed points SseedTaking out one seed point, and judging whether the seed point is a non-edge point or not;
step three: if the seed point is the non-edge point (X, Y), all pixel points (X) in the neighborhood range of the non-edge point (X, Y)9 are takenn,Yn) (n is a neighborhood pixel index in the range of 1. ltoreq. n.ltoreq.9), where X-1 ≦ Xn＜＝X+1,Y-1＜＝YnAnd < Y +1, and sequentially judging each pixel point (X)n,Yn) Whether the image is a pixel point of the brainstem area or not; if the pixel point (X)n,Yn) Is the pixel point of the brainstem area, the pixel point (X) isn,Yn) Put into the SseedAnd a set S of pixels in the brainstem regionBrainPerforming the following steps;
step four: repeating the second step and the third step until no pixel point (X) existsn,Yn) Put into the collection S of the pixel points in the brainstem areaBrainObtaining the final set S of the pixel points in the brainstem areaBrainAnd finally, collecting the pixel points S in the brainstem areaBrainAnd obtaining the brain stem region segmentation result.
6. The method for automatically segmenting the brainstem based on the medical image according to claim 5, wherein in the second step, the step of judging whether the seed points are the non-edge points comprises the following steps: if the gradient G of the seed point in the upper, lower, left and right directionsup＝abs(P-Pup)、Gdown＝abs(P-Pdown)、Gleft＝abs(P-Pleft) And Gright＝P-PrightAll are less than or equal to a preset edge gradient threshold value TedgeIf so, the seed point is considered as the non-edge point;
wherein P is the pixel value of the seed point, the Pup、Pdown、Pleft、PrightPixel values of pixel points adjacent to the seed point in four directions, namely, upper, lower, left and right directions are respectively set;
in the third step, each pixel point (X) is judged in turnn,Yn) The step of judging whether the pixel points are pixel points of the brainstem area comprises the following steps: if the pixel point (X)n,Yn) The pixel value of (A) is greater than or equal to a preset brainstem threshold value TBrainThen the pixel point (X) is consideredn,Yn) The pixel points of the brainstem area are obtained.
7. The method according to claim 6, wherein the automatic brain stem segmentation method based on medical images,
the preset edge gradient threshold value Tedge＝50；
The preset brainstem threshold TBrain＝150。
8. The medical image-based automatic brainstem segmentation method according to claim 1, further comprising after the automatic segmentation step:
and a segmentation result fusion display step, namely performing fusion display on the segmentation result of the brainstem region and the CT image.
9. The automatic brain stem segmentation method based on medical images according to claim 8, wherein the segmentation result fusion display step further comprises:
and performing three-dimensional reconstruction on the brain stem region segmentation result.
10. An automatic segmentation system for implementing the automatic brain stem segmentation method based on medical images according to any one of claims 1 to 9, comprising:
the image acquisition module is used for respectively acquiring a CT image and a digital human image;
the image registration module is used for registering the CT image and the digital human image to obtain a registration result;
a region-of-interest selection module for selecting a brainstem region-of-interest in the registration result;
and the automatic segmentation module is used for automatically segmenting the brainstem on the brainstem region of interest through an automatic segmentation algorithm to obtain a brainstem region segmentation result.
CN201910721119.0A 2019-08-06 2019-08-06 Automatic brain stem segmentation method and system based on medical image Withdrawn CN110610481A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910721119.0A CN110610481A (en) 2019-08-06 2019-08-06 Automatic brain stem segmentation method and system based on medical image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910721119.0A CN110610481A (en) 2019-08-06 2019-08-06 Automatic brain stem segmentation method and system based on medical image

Publications (1)

Publication Number Publication Date
CN110610481A true CN110610481A (en) 2019-12-24

Family

ID=68890522

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910721119.0A Withdrawn CN110610481A (en) 2019-08-06 2019-08-06 Automatic brain stem segmentation method and system based on medical image

Country Status (1)

Country Link
CN (1) CN110610481A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112132854A (en) * 2020-09-22 2020-12-25 推想医疗科技股份有限公司 Image segmentation method and device and electronic equipment
CN112598669A (en) * 2021-03-04 2021-04-02 之江实验室 Lung lobe segmentation method based on digital human technology

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030194119A1 (en) * 2002-04-15 2003-10-16 General Electric Company Semi-automatic segmentation algorithm for pet oncology images
CN103473557A (en) * 2013-09-02 2013-12-25 杭州电子科技大学 Spatial-spectral integrated hyperspectral remote sensing image classification method
CN103489198A (en) * 2013-10-21 2014-01-01 钟映春 Method for partitioning brainstem areas automatically from MR (magnetic resonance) sequence images
CN103793905A (en) * 2012-11-02 2014-05-14 上海联影医疗科技有限公司 Method and device for determining direction of coronal view of brain image
CN104376551A (en) * 2014-08-25 2015-02-25 浙江工业大学 Color image segmentation method integrating region growth and edge detection
CN106611413A (en) * 2016-11-30 2017-05-03 上海联影医疗科技有限公司 Image segmentation method and system
CN107292889A (en) * 2017-06-14 2017-10-24 上海联影医疗科技有限公司 A kind of method of lesion segmentation, system and computer-readable recording medium
CN109509199A (en) * 2018-10-10 2019-03-22 华南理工大学 A kind of medical image Organizational Intelligence dividing method based on three-dimensional reconstruction
CN109903264A (en) * 2019-01-16 2019-06-18 深圳市旭东数字医学影像技术有限公司 The method for registering and system of Digital Human image and CT image
CN109949404A (en) * 2019-01-16 2019-06-28 深圳市旭东数字医学影像技术有限公司 Based on Digital Human and CT and/or the MRI image three-dimensional rebuilding method merged and system

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030194119A1 (en) * 2002-04-15 2003-10-16 General Electric Company Semi-automatic segmentation algorithm for pet oncology images
CN103793905A (en) * 2012-11-02 2014-05-14 上海联影医疗科技有限公司 Method and device for determining direction of coronal view of brain image
CN103473557A (en) * 2013-09-02 2013-12-25 杭州电子科技大学 Spatial-spectral integrated hyperspectral remote sensing image classification method
CN103489198A (en) * 2013-10-21 2014-01-01 钟映春 Method for partitioning brainstem areas automatically from MR (magnetic resonance) sequence images
CN104376551A (en) * 2014-08-25 2015-02-25 浙江工业大学 Color image segmentation method integrating region growth and edge detection
CN106611413A (en) * 2016-11-30 2017-05-03 上海联影医疗科技有限公司 Image segmentation method and system
CN107292889A (en) * 2017-06-14 2017-10-24 上海联影医疗科技有限公司 A kind of method of lesion segmentation, system and computer-readable recording medium
CN109509199A (en) * 2018-10-10 2019-03-22 华南理工大学 A kind of medical image Organizational Intelligence dividing method based on three-dimensional reconstruction
CN109903264A (en) * 2019-01-16 2019-06-18 深圳市旭东数字医学影像技术有限公司 The method for registering and system of Digital Human image and CT image
CN109949404A (en) * 2019-01-16 2019-06-28 深圳市旭东数字医学影像技术有限公司 Based on Digital Human and CT and/or the MRI image three-dimensional rebuilding method merged and system

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112132854A (en) * 2020-09-22 2020-12-25 推想医疗科技股份有限公司 Image segmentation method and device and electronic equipment
CN112132854B (en) * 2020-09-22 2021-11-09 推想医疗科技股份有限公司 Image segmentation method and device and electronic equipment
CN112598669A (en) * 2021-03-04 2021-04-02 之江实验室 Lung lobe segmentation method based on digital human technology
CN112598669B (en) * 2021-03-04 2021-06-01 之江实验室 Lung lobe segmentation method based on digital human technology

Similar Documents

Publication Publication Date Title
CN102068281B (en) Processing method for space-occupying lesion ultrasonic images
CN106056537B (en) A kind of medical image joining method and device
CN107203741B (en) Blood vessel extraction method, device and system
WO2018120644A1 (en) Blood vessel extraction method and system
US9384555B2 (en) Motion correction apparatus and method
CN107808156A (en) Area-of-interest exacting method
WO2019000455A1 (en) Method and system for segmenting image
CN109741344A (en) Blood vessel segmentation method, apparatus, medical imaging equipment and storage medium
CN103886621B (en) A kind of method for automatically extracting bed board
CN103279929B (en) A kind of prediction of the CT image metallic traces based on integral cosine and artifact minimizing technology
CN113223010B (en) Method and system for multi-tissue full-automatic segmentation of oral cavity image
CN106388843A (en) Medical imaging equipment and scanning method thereof
JP2004520923A (en) How to segment digital images
CN106530236B (en) Medical image processing method and system
CN107248162A (en) The method of preparation method and acute cerebral ischemia the image segmentation of acute cerebral ischemia Image Segmentation Model
CN110610481A (en) Automatic brain stem segmentation method and system based on medical image
CN115100067A (en) Image reconstruction method and device
CN106780491B (en) Initial contour generation method adopted in segmentation of CT pelvic image by GVF method
CN108961278B (en) Method and system for abdominal wall muscle segmentation based on image data
CN114419181A (en) CTA image reconstruction method and device, display method and device
CN114119950A (en) Artificial intelligence-based oral cavity curved surface fault layer dental image segmentation method
CN110400359B (en) Method, device and equipment for eliminating image artifacts and storage medium
CN110097610B (en) Speech synthesis system and method based on ultrasound and magnetic resonance imaging
CN111563876A (en) Medical image acquisition method and display method
JP2004234579A (en) Method and program for extracting tissue region of interest and image processor

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication

Application publication date: 20191224

WW01 Invention patent application withdrawn after publication