CN109410181B - Heart image segmentation method and device - Google Patents
Heart image segmentation method and device Download PDFInfo
- Publication number
- CN109410181B CN109410181B CN201811159285.8A CN201811159285A CN109410181B CN 109410181 B CN109410181 B CN 109410181B CN 201811159285 A CN201811159285 A CN 201811159285A CN 109410181 B CN109410181 B CN 109410181B
- Authority
- CN
- China
- Prior art keywords
- image
- heart
- blood vessel
- radius
- main body
- 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.)
- Active
Links
- 210000002216 heart Anatomy 0.000 title claims abstract description 401
- 238000000034 method Methods 0.000 title claims abstract description 80
- 238000003709 image segmentation Methods 0.000 title claims abstract description 43
- 210000004204 blood vessel Anatomy 0.000 claims abstract description 152
- 210000000115 thoracic cavity Anatomy 0.000 claims abstract description 118
- 210000004185 liver Anatomy 0.000 claims abstract description 76
- 230000009466 transformation Effects 0.000 claims abstract description 38
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000006243 chemical reaction Methods 0.000 claims abstract description 16
- 230000000747 cardiac effect Effects 0.000 claims description 45
- 238000003325 tomography Methods 0.000 claims description 40
- 238000000605 extraction Methods 0.000 claims description 14
- 210000003205 muscle Anatomy 0.000 claims description 13
- 230000000877 morphologic effect Effects 0.000 claims description 12
- 210000000038 chest Anatomy 0.000 claims description 11
- 230000001186 cumulative effect Effects 0.000 claims description 9
- 230000035040 seed growth Effects 0.000 claims description 8
- 238000009499 grossing Methods 0.000 claims description 5
- 230000001788 irregular Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000001464 adherent effect Effects 0.000 claims 1
- 230000011218 segmentation Effects 0.000 abstract description 30
- 239000000284 extract Substances 0.000 abstract description 4
- 210000001519 tissue Anatomy 0.000 description 10
- 230000005484 gravity Effects 0.000 description 6
- 210000005003 heart tissue Anatomy 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 210000000683 abdominal cavity Anatomy 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 210000005228 liver tissue Anatomy 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 210000004351 coronary vessel Anatomy 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000012010 growth Effects 0.000 description 1
- 210000002837 heart atrium Anatomy 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- 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
- G06T2207/20112—Image segmentation details
- G06T2207/20152—Watershed segmentation
-
- 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
- G06T2207/30048—Heart; Cardiac
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
Abstract
The application provides a heart image segmentation method and a heart image segmentation device, and the embodiment of the application firstly extracts a heart main body image from a thoracic cavity tomogram; then judging whether a liver image adhered to the heart is included in the heart main body image or not according to the shape of the heart main body image, under the condition that the heart main body image includes the liver image, performing distance transformation on the basis of the heart main body image to obtain a right local maximum point, and removing the liver image from the heart main body image by using a clustering algorithm to obtain the heart image; and finally, removing the blood vessel image from the heart image based on the local maximum point obtained by distance conversion of the heart image to obtain a heart target image. The heart target image obtained by the technical scheme has high accuracy and strong anti-interference performance, and the heart image is automatically segmented, so that the segmentation accuracy and stability are good.
Description
Technical Field
The present application relates to the field of data processing and image processing technologies, and in particular, to a cardiac image segmentation method and apparatus.
Background
The heart has a complex structure including structures such as a new atrium, a ventricle, a coronary artery, upper and lower vena cava, and a valve, and the gray scale of heart tissue is similar to that of surrounding tissues, so that when the heart image is segmented, the heart and other tissues are adhered together, which causes great difficulty in heart segmentation.
Existing cardiac image segmentation methods include image feature-based segmentation methods and prior model-based segmentation methods. The segmentation method based on the image features specifically performs image segmentation based on regions or edges. This method has a high segmentation speed, but is susceptible to noise, and is likely to excessively segment a cardiac image, and also generates many false edges during segmentation, resulting in low segmentation accuracy. The segmentation method based on the prior model is characterized in that the prior model is established by utilizing abundant medical knowledge and years of clinical experience of medical experts, and then the established model is utilized to segment the heart image. The method effectively improves the precision and speed of heart segmentation, but during heart segmentation, an initial model of a prior model needs to be selected manually, relevant setting parameters are set, the automation degree is low, the segmentation precision depends on the experience of a doctor to a great extent, and the segmentation precision is unstable.
Disclosure of Invention
In view of the above, an object of the present application is to provide a cardiac image segmentation method and apparatus, so as to improve the accuracy and the degree of automation of cardiac image segmentation and enhance the anti-interference performance of cardiac image segmentation.
In a first aspect, an embodiment of the present application provides a cardiac image segmentation method, including:
acquiring a thoracic cavity tomography image, and extracting a heart main body image from the thoracic cavity tomography image;
judging whether a liver image adhered to the heart is included in the heart main body image or not based on the shape of the heart main body image;
under the condition that the heart main body image comprises a liver image, performing distance transformation on the heart main body image to obtain a right local maximum point, and removing the liver image from the heart main body image by a clustering algorithm to obtain a heart image;
and removing the blood vessel image from the heart image based on the local maximum point obtained by distance conversion of the heart image to obtain a heart target image.
With reference to the first aspect, the present embodiments provide a first possible implementation manner of the first aspect, where the extracting a heart subject image from the thoracic tomography image includes:
carrying out mean centering on gray values of pixel points in the thoracic cavity tomogram, carrying out binarization processing according to data subjected to mean centering, and obtaining a binarization image with the largest area to obtain a mask image of the removed bed image;
and acquiring the minimum external rectangle of the thoracic cavity tomography image without the bed image to obtain the heart main body image.
With reference to the first possible implementation manner of the first aspect, the present application provides a second possible implementation manner of the first aspect, where the extracting a heart subject image from the thoracic tomography image further includes:
adjusting the gray value of a pixel point of which the gray value is outside a preset gray range to be 0 in the heart main body image, and performing linear proportion adjustment on the gray value of the pixel point of which the gray value is within the preset gray range;
carrying out binarization processing on the heart main body image after gray level adjustment, and removing muscle tissues in the heart main body image by using the obtained binarization image; wherein the predetermined gray scale range is determined according to gray scale values of pixel points in the heart real image.
With reference to the second possible implementation manner of the first aspect, the present application provides an example of a third possible implementation manner of the first aspect, where the extracting a heart subject image from the chest tomography image further includes:
according to a preset seed point in the thoracic cavity tomography image, taking the heart main body image of the removed muscle tissue as a mask, growing seeds, and removing a thoracic cavity outer contour image in the heart main body image after the muscle tissue is removed;
and performing expansion processing on the heart main body image without the thoracic cavity outline image, and connecting the separated heart images to obtain a final heart main body image.
With reference to the third possible implementation manner of the first aspect, this example provides a fourth possible implementation manner of the first aspect, where the determining whether the liver image adhered to the heart is included in the heart subject image includes:
representing the boundary of the heart subject image in complex form;
performing fourier transform on a boundary of a heart subject image represented in complex form;
determining the amplitude of the Fourier transform when a preset frequency is taken, and judging whether the amplitude is greater than a preset value;
determining the shape of the heart subject image as an irregular heart shape in the case that the amplitude is larger than the predetermined value, wherein the heart subject image comprises a liver image.
With reference to the third possible implementation manner of the first aspect, the present application provides a fifth possible implementation manner of the first aspect, where the removing the liver image from the heart subject image to obtain a heart image includes:
acquiring a distance between the local maximum point on the right side and a pixel point which is smaller than a distance value corresponding to the local maximum on the right side to obtain a pixel point of the liver image;
removing pixel points of the liver image from the heart main body image to obtain the heart main body image with the liver image removed;
and performing clustering operation on the heart main body image without the liver image to obtain a heart image.
With reference to the fifth possible implementation manner of the first aspect, the present application provides a sixth possible implementation manner of the first aspect, where the removing a liver image from the heart subject image to obtain a heart image further includes:
removing burrs of a boundary of the cardiac image by using a morphological operation;
determining a new prefabricated seed point based on the intersection of the heart main body image without the extrathoracic cavity image and the heart image, and performing seed growth on the heart image according to the new prefabricated seed point to determine the boundary of the heart image;
filling the heart image according to the acquired boundary to obtain a complete heart image;
and removing burrs of the boundary of the complete heart image by using morphological operation to obtain a final heart image.
With reference to the first aspect, an embodiment of the present application provides a seventh possible implementation manner of the first aspect, where the method further includes:
sequencing the thoracic cavity tomographic images according to the extension direction from the head to the foot and the shooting position of the thoracic cavity tomographic images;
determining the overlapping area of the heart image corresponding to the current thoracic cavity tomography image and the heart image corresponding to the last thoracic cavity tomography image;
and calculating the ratio of the overlapping area to the area of the heart image corresponding to the previous thoracic cavity tomographic image, and under the condition that the ratio is smaller than a preset ratio, judging that the current thoracic cavity tomographic image is an image obtained by shooting the edge layer of the heart, and completing heart image segmentation.
With reference to the first aspect, the present application provides an eighth possible implementation manner of the first aspect, where the removing, from the cardiac image, the blood vessel image based on the local maximum point obtained by performing the distance transformation on the cardiac image includes:
determining the radius of the blood vessel and the coordinates of the center point of the blood vessel based on the local maximum value point obtained by distance transformation of the heart image;
and determining the radius of a target blood vessel based on the radius of the blood vessel and the radius of the equivalent blood vessel, and removing the blood vessel from the heart image according to the radius of the target blood vessel and the coordinates of the center point of the blood vessel to obtain a heart target image.
With reference to the eighth possible implementation manner of the first aspect, this embodiment provides a ninth possible implementation manner of the first aspect, where the determining a vessel radius and a vessel center point coordinate based on a local maximum point obtained by performing distance transformation on the cardiac image includes:
performing a distance transform on the cardiac image;
and acquiring a local maximum point positioned at the bottom of the heart, taking the coordinate of the acquired local maximum point as the coordinate of the center point of the blood vessel, and taking the distance value corresponding to the acquired local maximum point as the radius of the blood vessel.
With reference to the eighth possible implementation manner of the first aspect, this application example provides a sixth possible implementation manner of the tenth aspect, where the method further includes a step of determining the equivalent vessel radius:
generating a blood vessel radius histogram based on the corresponding blood vessel radii of all the thoracic cavity tomograms;
generating a vessel radius cumulative histogram based on the vessel radius histogram;
normalizing the blood vessel radius cumulative histogram, and generating a blood vessel radius equalized histogram based on the normalized blood vessel radius cumulative histogram;
and determining the maximum value of the forward difference of the blood vessel radius equalization histogram to obtain the equivalent blood vessel radius.
With reference to the eighth possible implementation manner, the ninth possible implementation manner, or the tenth possible implementation manner of the first aspect, an example of the present application provides an eleventh possible implementation manner of the first aspect, where the determining a target vessel radius based on the vessel radius and an equivalent vessel radius includes:
under the condition that the difference value between the blood vessel radius corresponding to the previous thoracic cavity tomographic image of the current thoracic cavity tomographic image and the equivalent blood vessel radius is larger than the preset radius difference value, obtaining the blood vessel radii corresponding to all thoracic cavity tomographic images before the current thoracic cavity tomographic image, and replacing the blood vessel radius corresponding to the previous thoracic cavity tomographic image by using the blood vessel radius with the minimum equivalent blood vessel radius difference value;
under the condition that the difference value between the blood vessel radius corresponding to the next thoracic cavity tomographic image of the current thoracic cavity tomographic image and the equivalent blood vessel radius is larger than the preset radius difference value, obtaining the blood vessel radii corresponding to all thoracic cavity tomographic images after the current thoracic cavity tomographic image, and replacing the blood vessel radius corresponding to the next thoracic cavity tomographic image by using the blood vessel radius with the minimum equivalent blood vessel radius difference value;
and determining the target blood vessel radius based on the blood vessel radius corresponding to the previous thoracic cavity tomography image and the blood vessel radius corresponding to the next thoracic cavity tomography image.
With reference to the eleventh possible implementation manner of the first aspect, this application example provides a twelfth possible implementation manner of the first aspect, where the method further includes:
carrying out distance transformation on the heart target image to obtain a distance transformation image;
and removing blood vessels in the heart target image based on the distance transformation image and the watershed algorithm.
With reference to the twelfth possible implementation manner of the first aspect, an embodiment of the present application provides a thirteenth possible implementation manner of the first aspect, where the method further includes:
removing burrs of the boundary in the heart target image by using morphological operation;
representing boundary coordinates of the heart target image in a complex form;
and based on the boundary coordinates expressed in the complex form, smoothing the boundary of the heart target image with the boundary burrs removed by utilizing Fourier change, central transformation and inverse Fourier transformation to obtain a final heart target image.
In a second aspect, an embodiment of the present application further provides a heart segmentation apparatus, including:
the heart main body image extraction module is used for acquiring a thoracic cavity tomography image and extracting a heart main body image from the thoracic cavity tomography image;
a heart image extraction module, configured to determine whether a liver image adhered to a heart is included in the heart subject image based on a shape of the heart subject image, and remove the liver image from the heart subject image based on a right local maximum point obtained by performing distance transformation on the heart subject image when the liver image is included in the heart subject image, so as to obtain the heart image;
and the heart target image extraction module is used for removing the blood vessel image from the heart image based on the local maximum value point obtained by distance conversion of the heart image to obtain the heart target image.
In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the electronic device is running, the machine-readable instructions when executed by the processor performing the steps of the first aspect described above, or of the method for segmenting a cardiac image in any of the possible implementations of the first aspect.
In a fourth aspect, this embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor to perform the steps of the method for segmenting a cardiac image according to the first aspect, or any one of the possible implementations of the first aspect.
According to the heart image segmentation method and device provided by the embodiment of the application, a heart main body image is extracted from a thoracic cavity sectional image; then judging whether a liver image adhered to the heart is included in the heart main body image or not according to the shape of the heart main body image, and removing the liver image from the heart main body image to obtain the heart image based on a right local maximum point obtained by distance conversion of the heart main body image under the condition that the liver image is included in the heart main body image; and finally, removing the blood vessel image from the heart image based on the local maximum point obtained by distance conversion of the heart image to obtain a heart target image. The technical scheme has the advantages of high accuracy of the obtained heart target image, strong anti-interference performance, realization of automatic segmentation of the heart image, and good accuracy and stability of the segmentation.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
FIG. 1 shows a flow chart of a cardiac image segmentation method provided by an embodiment of the present application;
fig. 2A shows a flowchart of extracting a heart subject image in another heart image segmentation method provided in the embodiment of the present application;
fig. 2B is a flowchart illustrating a method for segmenting a cardiac image according to another embodiment of the present application, wherein the method determines whether a liver image adhered to the heart is included in a cardiac subject image;
fig. 2C is a flowchart illustrating a method for removing a liver image from a heart subject image according to another cardiac image segmentation method provided by an embodiment of the present application;
FIG. 3 is a flow chart illustrating a method for removing a blood vessel image from a heart image in another heart image segmentation method provided by an embodiment of the present application;
FIG. 4A is a schematic diagram of a cardiac image obtained by another cardiac image segmentation method provided by an embodiment of the present application;
FIG. 4B is a schematic diagram of a cardiac target image obtained by another cardiac image segmentation method provided by an embodiment of the present application;
fig. 5 is a schematic structural diagram illustrating a name of a cardiac image segmentation apparatus provided in an embodiment of the present application;
fig. 6 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
In view of the defects of low precision, large noise influence and low automation degree of heart image segmentation in the prior art, the embodiment of the application provides a heart image segmentation method and device, which can acquire an accurate heart image from a thoracic cavity tomogram, and the segmentation precision is not easily influenced by noise.
For the understanding of the present embodiment, a detailed description will be given first of all of a cardiac image segmentation method disclosed in the embodiments of the present application.
Example one
The embodiment discloses a heart image segmentation method which can extract an accurate heart target image from a chest tomography image obtained by shooting, and particularly comprises the following steps of:
s110, obtaining a thoracic cavity tomography image, and extracting a heart main body image from the thoracic cavity tomography image.
Here, the thoracic tomographic image is a sequence of thoracic tomographic images obtained by sequentially imaging a region including the thoracic cavity in the positive direction or negative direction of the axis (taking the supine position, the direction of the head being the positive direction of the Z axis). It should be noted that the method of the present embodiment is to extract a heart target image for one chest tomographic image. The heart target image obtained by the method of the embodiment is not a complete and three-dimensional heart image, and the complete and three-dimensional heart image can be obtained only by three-dimensionally reconstructing all the heart target images corresponding to the thoracic tomography image sequence.
In a specific implementation, the heart main body image may be obtained by performing threshold segmentation based on the gray-scale values of the pixels in the thoracic cavity tomographic image, and then removing the anterior external contour by seed growth.
And S120, judging whether a liver image adhered to the heart is included in the heart main body image or not based on the shape of the heart main body image.
Here, in the captured image, the heart image may have a region that is adhered to the liver image, and since the heart image has a more regular shape, it may be determined whether the liver image that is adhered to the heart is included in the heart subject image according to the shape of the heart subject image.
S130, under the condition that the heart main body image comprises the liver image, based on the right local maximum point obtained by distance conversion of the heart main body image, removing the liver image from the heart main body image to obtain the heart image.
Here, when whether a liver image adhering to the heart is included in the heart subject image or not, the heart subject image may be subjected to distance conversion to obtain more than one local maximum point, and since the liver is located on the right side of the heart, the right local maximum point may be used as the center of gravity of the liver, and the liver image may be removed from the heart subject image according to the center of gravity of the liver.
And S140, removing the blood vessel image from the heart image based on the local maximum value point obtained by distance conversion of the heart image to obtain a heart target image.
Here, the center of gravity of the blood vessel can be specified from the local maximum point obtained by distance conversion of the heart image, and the blood vessel image can be removed from the heart image from the center of gravity of the blood vessel.
According to the method, the technologies such as threshold segmentation, morphology, distance transformation, clustering algorithm and watershed algorithm are fused together to extract the heart, so that the advantages of the original algorithm are inherited, the defects of each algorithm when used independently are overcome, the heart target image with higher accuracy can be obtained, the anti-interference performance is high, the automatic segmentation of the heart image is realized, and the segmentation accuracy and stability are good.
Example two
Based on the previous embodiment, the present embodiment specifically discloses a specific implementation manner of extracting a heart subject image, removing a liver image, and the like in the heart image segmentation method. Specifically, the cardiac image segmentation method of the present embodiment includes:
s210, obtaining a thoracic cavity tomography image, and extracting a heart main body image from the thoracic cavity tomography image.
Here, as shown in fig. 2A, the extraction of the heart subject image may be achieved by the following steps:
s2101, carrying out mean centering on the gray values of the pixel points in the thoracic cavity tomographic image, and carrying out binarization processing according to the mean centered data to obtain a binarization image with the largest area so as to obtain a mask image with a bed image removed.
In the step, mean centering processing is carried out on the gray values of the pixels in the thoracic cavity tomogram according to the mean value and the variance of the gray values of the pixels in the current thoracic cavity tomogram. The mean-centered gray value is 16 bits, and in order to reduce the storage consumption space, the mean-centered gray value is converted into 8 bits. Then, the 8-bit gray scale value is binarized by using the Otsu method (OSTU).
The values of the chest, the outer contour and the bed image in the image after the binarization processing are 1, the rest are 0, and the chest outer contour comprises the heart and is a main shooting object, so the area of the binarization image of the chest outer contour is certainly larger than that of the binarization image of the bed, and the mask image with the bed image removed can be obtained by acquiring the binarization image with the largest area.
In addition, before the binarized image with the largest area is obtained, the binarized image may be filled first, that is, pixel points with a value of 0 but surrounding 1 are filled with 1.
S2102, acquiring the minimum circumscribed rectangle of the thoracic cavity tomography image of the removed bed image, and obtaining the heart main body image.
Here, the heart subject image corresponding to the minimum bounding rectangle is a region of interest (ROI) for subsequent algorithm processing, so as to reduce unnecessary memory consumption during the processing and reduce the algorithm running time.
S2103, adjusting the gray value of a pixel point of which the gray value is outside a preset gray range to be 0 in the heart main body image, and performing linear proportion adjustment on the gray value of the pixel point of which the gray value is within the preset gray range; wherein the predetermined gray scale range is determined according to gray scale values of pixel points in the heart real image.
In this step, a predetermined gray scale range is determined based on the gray scale value of the image of the normal human tissue and the gray scale value of the pixel point in the heart real image, so that the tissues such as the muscle and the like are removed from the heart main body image. Specifically, the predetermined gray scale range may be [ -150,100], and this step is to adjust the gray scale value of the pixel point with the gray scale value greater than 100 to 0, and adjust the gray scale value of the pixel point with the gray scale value less than-150 to 0.
In the step, the gray value of the pixel point with the gray value between-150 and 100 is adjusted to be between 0 and 255 according to the linear proportion, and then the gray value can be represented by 8 bits, so that the storage space of the gray value is reduced.
However, this step does not remove the liver image because empirically the range of pixel gray scale values of the heart overlaps with the range of pixel gray scale values of the liver, and the range does not vary much.
S2104, performing binarization processing on the heart main body image after gray level adjustment, and removing muscle tissues in the heart main body image by using the obtained binarization image; wherein the predetermined gray scale range is determined according to gray scale values of pixel points in the heart real image.
In this step, the Otsu method is used to perform binarization processing to obtain a binarized image of various tissues with similar gray levels, such as the external contour of the thoracic cavity, the heart, the blood vessel, the liver and the like, and the image of the tissue with muscle tissue equal to the gray level of the heart image but not similar to the gray level of the heart image is removed.
S2105, according to the preset seed points in the thoracic cavity tomography image, taking the heart main body image without the muscle tissue as a mask, growing the seeds, and removing the thoracic cavity outer contour image in the heart main body image without the muscle tissue.
In this step, a binary image of the heart, blood vessel, liver is obtained.
S2106, performing expansion processing on the heart main body image without the external outline image of the thoracic cavity, and connecting the separated heart images to obtain a final heart main body image.
In this step, the binarized image obtained in the previous step may be subjected to a plurality of consecutive expansion operations, and the expanded structural element may have a circular structure with a radius of 1. Thereby ensuring that the hearts in the separated state are connected into a whole. For example, 10 successive expansion operations may be performed to ensure process quality.
S220, judging whether a liver image adhered to the heart is included in the heart main body image or not based on the shape of the heart main body image.
Here, as shown in fig. 2B, it may be specifically determined whether the liver image adhered to the heart is included in the heart subject image by using the following steps:
s2201, representing the boundary of the heart subject image in a complex form.
In this step, the boundaries may be extracted in a clockwise order, and the extracted boundaries may be represented in a complex form.
For example, the boundary of the heart subject image is represented as: s (k) ═ x (k) + jy (k).
S2202, fourier transform is performed on the boundary of the heart subject image expressed in complex form.
S2203, determining an amplitude of the fourier transform when a predetermined frequency is taken, and determining whether the amplitude is greater than a predetermined value.
The predetermined frequency in this step can be flexibly set according to the requirements of the actual application scenario, for example, the predetermined frequency is set to 1.
S2204, in a case that the amplitude is greater than the predetermined value, determining that the shape of the heart subject image includes an irregular heart shape, and the heart subject image includes a liver image.
In this step, the larger the amplitude is, the less the shape of the heart subject image is close to the real image shape of the heart. Therefore, a predetermined value is set, and in the case where the amplitude is larger than the predetermined value, it indicates that the shape of the heart subject image is very dissimilar to the shape of the real image of the heart, and at this time, the liver image must be included in the heart subject image. In the case where the amplitude is less than or equal to the predetermined value, it indicates that the heart subject image is an independent heart image.
The predetermined value in this step can be flexibly set according to the requirements of the actual application scenario, for example, the predetermined frequency is set to 15.
And S230, under the condition that the heart main body image comprises the liver image, removing the liver image from the heart main body image based on the right local maximum point obtained by distance conversion of the heart main body image to obtain the heart image.
Here, as shown in fig. 2C, the liver image may be removed from the heart subject image by the following steps:
s2301, obtaining the distance between the right local maximum point and a pixel point smaller than the distance value corresponding to the right local maximum point, and obtaining the pixel point of the liver image.
In this step, the heart subject image is subjected to distance, and then the right local maximum point is found in the horizontal axis direction.
In this step, the right local maximum point is the center of gravity of the liver, and the gray value of the right local maximum point is the distance value from the right local maximum point to the edge of the liver, so that the pixel point of the liver image can be obtained based on the right local maximum point and the corresponding distance value.
S2302, removing the pixel points of the liver image from the heart main body image to obtain the heart main body image with the liver image removed.
And S2303, performing clustering operation on the heart subject image without the liver image to obtain a heart image.
S2304, removing burrs of the boundary of the heart image by using morphological operation.
In this step, morphological erosion expansion operation may be specifically performed on the cardiac image obtained in the previous step to remove boundary burrs.
S2305, determining new prefabricated seed points based on the intersection of the heart main body image without the chest external contour image and the heart image, performing seed growth on the heart image according to the new prefabricated seed points, and determining the boundary of the heart image.
In this step, the new pre-fabricated seed points are preferentially selected from the points which are located in the middle of the image and have uniform texture.
In this step, the condition of seed growth is that the growth satisfies the pixel points whose gray value is less than 0 and greater than-150.
S2306, filling the heart image according to the acquired boundary to obtain a complete heart image.
In this step, the cardiac images are filled to make the cardiac images more complete and continuous.
S2307, removing burrs of the boundary of the complete heart image by using morphological operation to obtain a final heart image.
In this step, the boundary burr can be removed by etching, expansion, opening, closing, or the like in morphology.
S240, removing the blood vessel image from the heart image based on the local maximum value point obtained by distance transformation of the heart image to obtain a heart target image.
EXAMPLE III
Based on the above embodiments, this embodiment specifically discloses a specific implementation manner of removing a blood vessel image from a heart image in a heart image segmentation method. Specifically, the cardiac image segmentation method of the present embodiment includes:
s310, obtaining a thoracic cavity tomography image, and extracting a heart main body image from the thoracic cavity tomography image.
And S320, judging whether a liver image adhered to the heart is included in the heart main body image or not based on the shape of the heart main body image.
S330, under the condition that the heart main body image comprises the liver image, based on the right local maximum point obtained by distance conversion of the heart main body image, removing the liver image from the heart main body image to obtain the heart image.
S340, removing the blood vessel image from the heart image based on the local maximum value point obtained by distance transformation of the heart image to obtain a heart target image.
Here, as shown in fig. 3, the blood vessel image in the heart image can be removed by specifically using the following steps:
and S3401, determining the radius of the blood vessel and the coordinates of the center point of the blood vessel based on the local maximum value point obtained by distance transformation of the heart image.
In this step, the following steps may be specifically used:
and S34011, performing distance transformation on the heart image.
And obtaining N local maximum value points in the image after distance conversion.
And S34012, acquiring a local maximum point positioned at the bottommost layer, taking the acquired coordinate of the local maximum point as the coordinate of the center point of the blood vessel, and taking the distance value corresponding to the acquired local maximum point as the radius of the blood vessel.
Because the blood vessel is positioned at the bottom layer of the heart, the local maximum value point at the bottom layer is obtained, namely the gravity center of the blood vessel is obtained, and the coordinates of the local maximum value point are taken as the coordinates of the center point of the blood vessel. The distance value corresponding to the local maximum point of the bottommost layer is the distance between the local maximum point of the bottommost layer and the edge of the blood vessel, and therefore the distance is used as the radius of the blood vessel.
And S3402, determining the radius of a target blood vessel based on the radius of the blood vessel and the radius of the equivalent blood vessel, and removing the blood vessel from the heart image according to the radius of the target blood vessel and the coordinates of the center point of the blood vessel to obtain a heart target image.
In this step, the target blood vessel radius may be determined specifically by using the following steps:
s34021, under the condition that the difference value between the radius of the blood vessel corresponding to the previous thoracic cavity tomographic image of the current thoracic cavity tomographic image and the equivalent blood vessel radius is larger than the preset radius difference value, obtaining the blood vessel radii corresponding to all thoracic cavity tomographic images before the current thoracic cavity tomographic image, and replacing the blood vessel radius corresponding to the previous thoracic cavity tomographic image by the blood vessel radius with the minimum equivalent blood vessel radius difference value.
The predetermined radius difference in this step can be flexibly set according to the requirements of the actual application scenario, for example, to 2 pixels.
S34022, under the condition that the difference value between the radius of the blood vessel corresponding to the next thoracic cavity tomogram of the current thoracic cavity tomogram and the equivalent blood vessel radius is larger than the preset radius difference value, obtaining the blood vessel radius corresponding to all thoracic cavity tomograms after the current thoracic cavity tomogram, and replacing the blood vessel radius corresponding to the next thoracic cavity tomogram by using the blood vessel radius with the minimum equivalent blood vessel radius difference value.
And S34023, determining the target blood vessel radius based on the blood vessel radius corresponding to the previous thoracic cavity tomography image and the blood vessel radius corresponding to the next thoracic cavity tomography image.
In this step, the weight of the blood vessel radius corresponding to the previous thoracic cavity tomographic image may be set to be 0.5, the weight of the blood vessel radius corresponding to the next thoracic cavity tomographic image may be set to be 0.5, and then the target blood vessel radius may be obtained by weighted summation of the blood vessel radius corresponding to the previous thoracic cavity tomographic image and the blood vessel radius corresponding to the next thoracic cavity tomographic image.
After determining the target vessel radius, the vessel may be removed from the cardiac image based on the target vessel radius and the vessel center point coordinates, in particular using the following steps: and calculating a distance transformation graph by taking the coordinate of the center point of the blood vessel as the center and the radius of the target blood vessel as the radius, and setting the gray level of a pixel point within the radius in the heart image as 0, thereby removing the influence of the blood vessel.
Further, the cardiac image segmentation method of the present embodiment further includes the following steps after obtaining the cardiac target image:
and S350, performing distance transformation on the heart target image to obtain a distance transformation image.
And S360, removing blood vessels in the heart target image based on the distance transformation image and the watershed algorithm.
The above steps S350-S360 remove blood vessels that cannot be removed in the cardiac target image.
Further, after the above steps are completed, the cardiac image segmentation method of the present embodiment further includes the following steps:
and S370, removing burrs of the boundary in the heart target image by using morphological operation.
In particular, the boundary burrs may be removed by a morphological erosion-dilation operation.
And S380, representing the boundary coordinates of the heart target image into a complex form.
When the method is implemented again, the boundary coordinates of the heart target image can be extracted clockwise and expressed as a complex number form: s (k) ═ x (k) + jy (k).
And S390, based on the boundary coordinates expressed in the complex form, smoothing the boundary of the heart target image with the boundary burrs removed by utilizing Fourier change, central transformation and inverse Fourier transformation to obtain a final heart target image.
During the specific implementation, the fourier transform and the center transform can be performed on the boundary, and the frequency of each 5 coefficients at the two sides of the center is taken to perform the inverse fourier transform, so that the boundary smoothing is realized, and the heart boundary delineation is completed.
It should be noted that, the methods of the above embodiments are all to extract the heart target image in one thoracic cavity tomographic image, and as can be seen from the above statements, one thoracic cavity tomographic image is only an image of one thoracic cavity tomographic image, and therefore, to obtain a complete and stereoscopic heart image, it is necessary to extract the heart target images corresponding to all thoracic cavity tomographic images. When the method is implemented again, a thoracic cavity tomographic image sequence can be obtained according to an axial position (taking a supine position, the direction of the head is the positive direction of the Z axis), and then a heart target image in each thoracic cavity tomographic image is extracted. When the heart target image is extracted, the upper layer edge or the lower layer edge of the heart needs to be judged, and after the heart target image at the upper layer edge or the lower layer edge is extracted, the heart target image can be spliced, so that a complete and three-dimensional heart image is obtained.
The heart image segmentation method of the present embodiment may further include the step of determining whether the chest tomographic image is an image captured at an upper layer edge or a lower layer edge of the heart as follows:
further, the method of this embodiment further includes the step of determining the equivalent vessel radius:
the method comprises the following steps of firstly, sequentially determining a heart image corresponding to each thoracic cavity sectional image according to an axial position (Z axis) sequence;
determining the overlapping area of the heart image corresponding to the current thoracic cavity sectional image and the heart image corresponding to the last thoracic cavity sectional image;
and thirdly, calculating the ratio of the overlapped area to the area of the heart image corresponding to the previous thoracic cavity tomographic image, and under the condition that the ratio is smaller than a preset ratio, judging that the current thoracic cavity tomographic image is an image obtained by shooting the edge layer of the heart, and finishing heart image segmentation.
The predetermined ratio can be flexibly set according to the actual application scenario, for example, the predetermined ratio can be set to 0.2.
It should be noted that after the image extraction of one thoracic cavity tomographic image is performed, the image extraction of the next thoracic cavity tomographic image is performed based on the above-mentioned new pre-made seed points, and the extraction step is the same as the image extraction step of the previous thoracic cavity tomographic image.
Further, the heart image segmentation method of the present embodiment further includes the following step of calculating an equivalent vessel radius:
step one, generating a blood vessel radius histogram based on blood vessel radii corresponding to all thoracic cavity tomograms;
in the step, all the thoracic cavity tomographic images are traversed in a sequence forward direction, a distance transformation graph of each corresponding heart image is calculated, N local maximum value points of each distance transformation graph are found, the minimum extreme value point of the vertical coordinate in the N local maximum value points is taken as a blood vessel center coordinate point of the heart image, and the blood vessel center coordinates of all the heart images and corresponding distance values (blood vessel radius) are stored in an array.
Generating a vessel radius cumulative histogram based on the vessel radius histogram;
step three, normalizing the blood vessel radius cumulative histogram, and generating a blood vessel radius equalization histogram based on the normalized blood vessel radius cumulative histogram;
and step four, determining the maximum value of the forward difference of the blood vessel radius equalization histogram to obtain the equivalent blood vessel radius.
In summary, the cardiac image segmentation method in the above embodiment adopts threshold segmentation as the first step of cardiac segmentation, extracts the cardiac subject image, and omits some regions affected by noise. And secondly, taking the heart main body image as a seed point extracted from the next thoracic cavity tomogram through morphological processing, and obtaining a heart coarse segmentation object through a seed growth algorithm. Thirdly, removing the liver image by using a seed growth algorithm, a clustering algorithm and the like to obtain a heart image, wherein fig. 4A is the heart image after the liver is removed. And fourthly, removing the influence of noise such as blood vessels by using algorithms such as watershed and the like to obtain a heart target image, wherein fig. 4B is the image of the heart with the blood vessels removed, namely the heart target image.
In the heart image segmentation method in the above embodiment, the automatic threshold segmentation algorithm can segment the heart from most tissues of the abdominal cavity, but because the automatic threshold segmentation does not consider the spatial structure characteristics of the heart, a part of muscle tissues and some liver tissues at the edge of the abdominal cavity with similar gray levels are also extracted together, resulting in heart over-segmentation. In addition, a part of heart peripheral tissues are missed to be segmented due to the fact that the gray level of the heart peripheral tissues is lower than the gray level of the heart internal tissues, and in order to solve the problem of heart over-segmentation, clustering analysis is conducted on the over-segmented heart tissues, and the heart is segmented from muscle or liver tissues. In order to solve the problem of missing segmentation of the heart tissue, the segmented heart tissue is used as a seed point, and seed growth segmentation is carried out to miss the segmented heart tissue. When the missing segmentation and the over-segmentation are completed, the heart blood vessels need to be segmented, the blood vessels of all thoracic cavity sectional images are segmented one by adopting a method for positioning the centers of the blood vessels and a method for calculating the equivalent blood vessel diameter by determining the radius of the blood vessels. And removing blood vessels, removing boundary burrs and smoothing boundaries of the segmented heart by using algorithms such as watershed calculation and the like, so as to obtain a high-precision heart target image. The method and the device are not easily influenced by noise, and the problems of missing segmentation, over-segmentation and the like existing in the heart image segmentation in the prior art are solved. Meanwhile, as shown in fig. 4B, the image of the heart target obtained by the embodiment of the present application has smooth edges and no burrs, and is a high-precision and high-aesthetic image.
Based on the same technical concept, embodiments of the present application further provide a cardiac image segmentation apparatus, an electronic device, a computer storage medium, and the like, which can be seen in the following embodiments.
Example four
The present embodiment discloses a heart image segmentation apparatus, as shown in fig. 5, including:
a heart main body image extraction module 501, configured to acquire a thoracic cavity tomographic image and extract a heart main body image from the thoracic cavity tomographic image;
a heart image extraction module 502, configured to determine whether a liver image adhered to a heart is included in the heart subject image based on the shape of the heart subject image, and remove the liver image from the heart subject image based on a right local maximum point obtained by performing distance transformation on the heart subject image when the liver image is included in the heart subject image, so as to obtain a heart image;
a heart target image extracting module 503, configured to remove the blood vessel image from the heart image based on the local maximum point obtained by performing distance transformation on the heart image, so as to obtain a heart target image.
EXAMPLE five
The present embodiment discloses an electronic device, as shown in fig. 6, including: a processor 601, a memory 602, and a bus 603, wherein the memory 602 stores machine-readable instructions executable by the processor 601, and when the electronic device is operated, the processor 601 and the memory 602 communicate via the bus 603.
The machine readable instructions, when executed by the processor 601, perform the steps of the push method of the following taxi order:
acquiring a thoracic cavity tomography image, and extracting a heart main body image from the thoracic cavity tomography image;
judging whether a liver image adhered to the heart is included in the heart main body image or not based on the shape of the heart main body image;
removing the liver image from the heart main body image to obtain a heart image based on a right local maximum point obtained by distance conversion of the heart main body image under the condition that the heart main body image comprises the liver image;
and removing the blood vessel image from the heart image based on the local maximum point obtained by distance conversion of the heart image to obtain a heart target image.
EXAMPLE six
The present embodiment discloses a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, performs the steps of the cardiac image segmentation method of the above embodiments.
Embodiments of the present application further provide a computer program product for cardiac image segmentation, which includes a computer-readable storage medium storing non-volatile program code executable by a processor, where the program code includes instructions for performing the method described in the foregoing method embodiments, and specific implementation thereof can be found in the method embodiments, and will not be described herein again.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the exemplary embodiments of the present application, and are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (14)
1. A method of cardiac image segmentation, comprising:
acquiring a thoracic cavity tomography image, and extracting a heart main body image from the thoracic cavity tomography image;
judging whether a liver image adhered to the heart is included in the heart main body image or not based on the shape of the heart main body image;
removing the liver image from the heart main body image to obtain a heart image based on a right local maximum point obtained by distance conversion of the heart main body image under the condition that the heart main body image comprises the liver image;
removing the blood vessel image from the heart image based on the local maximum point obtained by distance transformation of the heart image to obtain a heart target image;
the removing the blood vessel image from the heart image based on the local maximum point obtained by distance transformation of the heart image comprises:
determining the radius of the blood vessel and the coordinates of the center point of the blood vessel based on the local maximum value point obtained by distance transformation of the heart image;
and determining the radius of a target blood vessel based on the radius of the blood vessel and the radius of the equivalent blood vessel, and removing the blood vessel from the heart image according to the radius of the target blood vessel and the coordinates of the center point of the blood vessel to obtain a heart target image.
2. The method of claim 1, wherein said extracting a heart subject image from said chest tomography image comprises:
carrying out mean centering on gray values of pixel points in the thoracic cavity tomogram, carrying out binarization processing according to mean centered data, and obtaining a binarization image with the largest area to obtain a mask image with a bed removed;
and acquiring the minimum external rectangle of the thoracic cavity tomography image without the bed image to obtain the heart main body image.
3. The method of claim 2, wherein the extracting a heart subject image from the chest tomography image further comprises:
adjusting the gray value of a pixel point of which the gray value is outside a preset gray range to be 0 in the heart main body image, and performing linear proportion adjustment on the gray value of the pixel point of which the gray value is within the preset gray range;
carrying out binarization processing on the heart main body image after gray level adjustment, and removing muscle tissues in the heart main body image by using the obtained binarization image; wherein the predetermined gray scale range is determined according to gray scale values of pixel points in the heart real image.
4. The method of claim 3, wherein the extracting a heart subject image from the chest tomography image further comprises:
according to a preset seed point in the thoracic cavity tomography image, taking the heart main body image without the muscle tissue as a mask, growing seeds, and removing a thoracic cavity outer contour image in the heart main body image without the muscle tissue;
and performing expansion processing on the heart main body image without the thoracic cavity outline image, and connecting the separated heart images to obtain a final heart main body image.
5. The method of claim 4, wherein the determining whether the image of the heart subject includes an image of a liver that is adherent to the heart comprises:
representing the boundary of the heart subject image in complex form;
performing fourier transform on a boundary of a heart subject image represented in complex form;
determining the amplitude of the Fourier transform when a preset frequency is taken, and judging whether the amplitude is greater than a preset value;
determining the shape of the heart subject image as an irregular heart shape in the case that the amplitude is larger than the predetermined value, wherein the heart subject image comprises a liver image.
6. The method of claim 4, wherein removing the liver image from the heart subject image to obtain a heart image comprises:
acquiring pixel points which are less than the distance value of the corresponding point of the right local maximum and have the distance to the right local maximum point to obtain pixel points of the liver image;
removing pixel points of the liver image from the heart main body image to obtain the heart main body image with the liver image removed;
and performing clustering operation on the heart main body image without the liver image to obtain a heart image.
7. The method of claim 6, wherein removing the liver image from the heart subject image results in a heart image, further comprising:
removing burrs of a boundary of the cardiac image by using a morphological operation;
determining a new prefabricated seed point based on the intersection of the heart main body image without the extrathoracic cavity image and the heart image, and performing seed growth on the heart image according to the new prefabricated seed point to determine the boundary of the heart image;
filling the heart image according to the acquired boundary to obtain a complete heart image;
and removing burrs of the boundary of the complete heart image by using morphological operation to obtain a final heart image.
8. The method of claim 1, further comprising:
sequencing the thoracic cavity tomographic images according to the extension direction from the head to the foot and the shooting position of the thoracic cavity tomographic images;
determining the overlapping area of the heart image corresponding to the current thoracic cavity tomography image and the heart image corresponding to the last thoracic cavity tomography image;
and calculating the ratio of the overlapping area to the area of the heart image corresponding to the previous thoracic cavity tomographic image, and under the condition that the ratio is smaller than a preset ratio, judging that the current thoracic cavity tomographic image is an image obtained by shooting the edge layer of the heart, and completing heart image segmentation.
9. The method of claim 1, wherein determining the vessel radius and the vessel center point coordinates based on the local maximum point obtained by distance transformation of the cardiac image comprises:
performing a distance transform on the cardiac image;
and acquiring a local maximum point positioned at the bottom of the heart, taking the coordinate of the acquired local maximum point as the coordinate of the center point of the blood vessel, and taking the distance value corresponding to the acquired local maximum point as the radius of the blood vessel.
10. The method according to claim 1, further comprising the step of determining the equivalent vessel radius:
generating a blood vessel radius histogram based on the corresponding blood vessel radii of all the thoracic cavity tomograms;
generating a vessel radius cumulative histogram based on the vessel radius histogram;
normalizing the blood vessel radius cumulative histogram, and generating a blood vessel radius equalized histogram based on the normalized blood vessel radius cumulative histogram;
and determining the maximum value of the forward difference of the blood vessel radius equalization histogram to obtain the equivalent blood vessel radius.
11. The method of any one of claims 1, 9 or 10, wherein determining a target vessel radius based on the vessel radius and an equivalent vessel radius comprises:
under the condition that the difference value between the blood vessel radius corresponding to the previous thoracic cavity tomographic image of the current thoracic cavity tomographic image and the equivalent blood vessel radius is larger than the preset radius difference value, obtaining the blood vessel radii corresponding to all thoracic cavity tomographic images before the current thoracic cavity tomographic image, and replacing the blood vessel radius corresponding to the previous thoracic cavity tomographic image by using the blood vessel radius with the minimum equivalent blood vessel radius difference value;
under the condition that the difference value between the blood vessel radius corresponding to the next thoracic cavity tomographic image of the current thoracic cavity tomographic image and the equivalent blood vessel radius is larger than the preset radius difference value, obtaining the blood vessel radii corresponding to all thoracic cavity tomographic images after the current thoracic cavity tomographic image, and replacing the blood vessel radius corresponding to the next thoracic cavity tomographic image by using the blood vessel radius with the minimum equivalent blood vessel radius difference value;
and determining the target blood vessel radius based on the blood vessel radius corresponding to the previous thoracic cavity tomography image and the blood vessel radius corresponding to the next thoracic cavity tomography image.
12. The method of claim 11, further comprising:
carrying out distance transformation on the heart target image to obtain a distance transformation image;
and removing blood vessels in the heart target image based on the distance transformation image and the watershed algorithm.
13. The method of claim 12, further comprising:
removing burrs of the boundary in the heart target image by using morphological operation;
representing boundary coordinates of the heart target image in a complex form;
and based on the boundary coordinates expressed in the complex form, smoothing the boundary of the heart target image by utilizing Fourier transformation, center transformation and inverse Fourier transformation to obtain the final heart target image.
14. A cardiac image segmentation apparatus, comprising:
the heart main body image extraction module is used for acquiring a thoracic cavity tomography image and extracting a heart main body image from the thoracic cavity tomography image;
the heart image extraction module is used for judging whether a liver image adhered to the heart is included in the heart main body image or not based on the shape of the heart main body image, and under the condition that the liver image is included in the heart main body image, performing distance transformation based on the heart main body image to obtain a right local maximum point, and removing the liver image from the heart main body image by a clustering algorithm to obtain a heart image;
the heart target image extraction module is used for removing the blood vessel image from the heart image based on the local maximum point obtained by distance conversion of the heart image to obtain a heart target image;
the heart target image extraction module comprises:
the first determining module is used for determining the radius of the blood vessel and the coordinates of the center point of the blood vessel based on the local maximum value point obtained by distance transformation of the heart image;
and the second determination module is used for determining the radius of a target blood vessel based on the radius of the blood vessel and the equivalent radius of the blood vessel, and removing the blood vessel from the heart image according to the radius of the target blood vessel and the coordinates of the center point of the blood vessel to obtain a heart target image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811159285.8A CN109410181B (en) | 2018-09-30 | 2018-09-30 | Heart image segmentation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811159285.8A CN109410181B (en) | 2018-09-30 | 2018-09-30 | Heart image segmentation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109410181A CN109410181A (en) | 2019-03-01 |
| CN109410181B true CN109410181B (en) | 2020-08-28 |
Family
ID=65465800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811159285.8A Active CN109410181B (en) | 2018-09-30 | 2018-09-30 | Heart image segmentation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109410181B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111445449B (en) * | 2020-03-19 | 2024-03-01 | 上海联影智能医疗科技有限公司 | Method, device, computer equipment and storage medium for classifying region of interest |
| CN111899244B (en) * | 2020-07-30 | 2021-02-12 | 推想医疗科技股份有限公司 | Image segmentation method, network model training method, device and electronic equipment |
| CN112652011B (en) * | 2020-12-25 | 2023-04-14 | 北京阅影科技有限公司 | Extraction method and extraction device of target blood vessel and computer-readable storage medium |
| CN113989303A (en) * | 2021-11-04 | 2022-01-28 | 中广核贝谷科技有限公司 | A method and device for rapidly segmenting objects in three-dimensional CT images |
| CN118154670B (en) * | 2024-05-10 | 2024-08-20 | 慧诺云谱(海南)科技有限公司 | Ear shape processing method and device for ear image, electronic equipment and medium |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101013503A (en) * | 2007-01-26 | 2007-08-08 | 清华大学 | Method for segmenting abdominal organ in medical image |
| EP1949336A1 (en) * | 2005-11-09 | 2008-07-30 | Koninklijke Philips Electronics N.V. | Automated stool removal method for medical imaging |
| WO2009001378A1 (en) * | 2007-06-26 | 2008-12-31 | Im3D S.P.A. | Morphological correction method of juxta- vascular pulmonary nodules segmentation in cat images |
| CN101533518A (en) * | 2009-04-21 | 2009-09-16 | 山东大学 | Method for re-establishing surface of three dimensional target object by unparallel dislocation image sequence |
| CN101727666A (en) * | 2008-11-03 | 2010-06-09 | 深圳迈瑞生物医疗电子股份有限公司 | Image segmentation method and device, and method for judging image inversion and distinguishing front side and back side of sternum |
| CN101826209A (en) * | 2010-04-29 | 2010-09-08 | 电子科技大学 | Canny model-based method for segmenting three-dimensional medical image |
| WO2010079519A9 (en) * | 2009-01-09 | 2010-10-14 | Im3D S.P.A | Method and system for the automatic recognition of lesions in a set of breast magnetic resonance images |
| CN103745470A (en) * | 2014-01-08 | 2014-04-23 | 兰州交通大学 | Wavelet-based interactive segmentation method for polygonal outline evolution medical CT (computed tomography) image |
| CN103839269A (en) * | 2014-03-21 | 2014-06-04 | 南京大学 | Image segmentation method based on quaternion and fuzzy C-means clustering |
| CN104036484A (en) * | 2013-03-06 | 2014-09-10 | 株式会社东芝 | Image segmentation device, image segmentation method and medical image equipment |
| CN104484877A (en) * | 2014-12-12 | 2015-04-01 | 山东大学 | AML cell segmentation method based on Meanshift cluster and morphological operations |
| CN104680127A (en) * | 2014-12-18 | 2015-06-03 | 闻泰通讯股份有限公司 | Gesture identification method and gesture identification system |
| CN104751178A (en) * | 2015-03-31 | 2015-07-01 | 上海理工大学 | Pulmonary nodule detection device and method based on shape template matching and combining classifier |
| CN104992445A (en) * | 2015-07-20 | 2015-10-21 | 河北大学 | Automatic division method for pulmonary parenchyma of CT image |
| CN106355599A (en) * | 2016-08-30 | 2017-01-25 | 上海交通大学 | Non-fluorescent eye fundus image based automatic segmentation method for retinal blood vessels |
| CN106846349A (en) * | 2017-02-26 | 2017-06-13 | 哈尔滨理工大学 | A kind of prostate Magnetic Resonance Image Segmentation method based on level set |
-
2018
- 2018-09-30 CN CN201811159285.8A patent/CN109410181B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1949336A1 (en) * | 2005-11-09 | 2008-07-30 | Koninklijke Philips Electronics N.V. | Automated stool removal method for medical imaging |
| CN101013503A (en) * | 2007-01-26 | 2007-08-08 | 清华大学 | Method for segmenting abdominal organ in medical image |
| WO2009001378A1 (en) * | 2007-06-26 | 2008-12-31 | Im3D S.P.A. | Morphological correction method of juxta- vascular pulmonary nodules segmentation in cat images |
| CN101727666A (en) * | 2008-11-03 | 2010-06-09 | 深圳迈瑞生物医疗电子股份有限公司 | Image segmentation method and device, and method for judging image inversion and distinguishing front side and back side of sternum |
| WO2010079519A9 (en) * | 2009-01-09 | 2010-10-14 | Im3D S.P.A | Method and system for the automatic recognition of lesions in a set of breast magnetic resonance images |
| CN101533518A (en) * | 2009-04-21 | 2009-09-16 | 山东大学 | Method for re-establishing surface of three dimensional target object by unparallel dislocation image sequence |
| CN101826209A (en) * | 2010-04-29 | 2010-09-08 | 电子科技大学 | Canny model-based method for segmenting three-dimensional medical image |
| CN104036484A (en) * | 2013-03-06 | 2014-09-10 | 株式会社东芝 | Image segmentation device, image segmentation method and medical image equipment |
| CN103745470A (en) * | 2014-01-08 | 2014-04-23 | 兰州交通大学 | Wavelet-based interactive segmentation method for polygonal outline evolution medical CT (computed tomography) image |
| CN103839269A (en) * | 2014-03-21 | 2014-06-04 | 南京大学 | Image segmentation method based on quaternion and fuzzy C-means clustering |
| CN104484877A (en) * | 2014-12-12 | 2015-04-01 | 山东大学 | AML cell segmentation method based on Meanshift cluster and morphological operations |
| CN104680127A (en) * | 2014-12-18 | 2015-06-03 | 闻泰通讯股份有限公司 | Gesture identification method and gesture identification system |
| CN104751178A (en) * | 2015-03-31 | 2015-07-01 | 上海理工大学 | Pulmonary nodule detection device and method based on shape template matching and combining classifier |
| CN104992445A (en) * | 2015-07-20 | 2015-10-21 | 河北大学 | Automatic division method for pulmonary parenchyma of CT image |
| CN106355599A (en) * | 2016-08-30 | 2017-01-25 | 上海交通大学 | Non-fluorescent eye fundus image based automatic segmentation method for retinal blood vessels |
| CN106846349A (en) * | 2017-02-26 | 2017-06-13 | 哈尔滨理工大学 | A kind of prostate Magnetic Resonance Image Segmentation method based on level set |
Non-Patent Citations (2)
| Title |
|---|
| Chen L C等.Predictive Watershed:A Fast Watershed Algorithm for Videc Segmentation.《IEEE Trans on Circuit and Systems for Video Technology》.2003,第13卷(第5期),第453-461页. * |
| 宁纪锋等.基于偏微分方程和分水岭算法的图像分割.《模式识别与人工智能》.2008,第21卷(第5期),第664-669页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109410181A (en) | 2019-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109410181B (en) | Heart image segmentation method and device | |
| CN107563983B (en) | Image processing method and medical imaging device | |
| CN106447645B (en) | Device and method for detecting and quantifying coronary artery calcification in enhanced CT images | |
| CN105913432B (en) | Aorta extracting method and device based on CT sequence images | |
| CN111951277B (en) | Coronary artery segmentation method based on CTA image | |
| CN112288786B (en) | Electronic device and non-transitory computer-readable storage medium | |
| US9875581B2 (en) | Automated 3D reconstruction of the cardiac chambers from MRI or ultrasound | |
| CN111612743B (en) | A Method of Extracting Coronary Artery Centerline Based on CT Image | |
| CN101964118B (en) | A 3D Segmentation Method of Intravascular Ultrasound Image Sequence | |
| CN110223271B (en) | Automatic level set segmentation method and device for blood vessel image | |
| CN107292312B (en) | Tumor CT image processing method | |
| US20070031019A1 (en) | System and method for coronary artery segmentation of cardiac CT volumes | |
| CN111047607B (en) | Method for automatically segmenting coronary artery | |
| CN112308846B (en) | Blood vessel segmentation method and device and electronic equipment | |
| CN110648338B (en) | Image segmentation method, readable storage medium, and image processing apparatus | |
| JP2008521473A (en) | Multi-element vascular segmentation | |
| CN102068281A (en) | Processing method for space-occupying lesion ultrasonic images | |
| CN111815663A (en) | Hepatic vessel segmentation system based on Hessian matrix and gray scale method | |
| CN112136157A (en) | Method, system and computer program for segmenting tooth pulp regions from images | |
| CA3104562A1 (en) | Method and computer program for segmentation of optical coherence tomography images of the retina | |
| CN110111354B (en) | Liver medical image artery region segmentation method and system | |
| CN112529918B (en) | Method, device and equipment for segmenting brain room area in brain CT image | |
| CN112308845B (en) | Left ventricle segmentation method and device and electronic equipment | |
| CN112862850A (en) | Digital silhouette angiography image processing method and device | |
| CN111093506A (en) | Motion compensated heart valve reconstruction |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |