CN112184632A - Image processing method, device and computer storage medium - Google Patents
Image processing method, device and computer storage medium Download PDFInfo
- Publication number
- CN112184632A CN112184632A CN202010935129.7A CN202010935129A CN112184632A CN 112184632 A CN112184632 A CN 112184632A CN 202010935129 A CN202010935129 A CN 202010935129A CN 112184632 A CN112184632 A CN 112184632A
- Authority
- CN
- China
- Prior art keywords
- image
- region
- pericardium
- segmentation
- target image
- 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.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 30
- 230000011218 segmentation Effects 0.000 claims abstract description 124
- 210000003516 pericardium Anatomy 0.000 claims abstract description 112
- 238000012545 processing Methods 0.000 claims abstract description 62
- 230000002107 myocardial effect Effects 0.000 claims abstract description 58
- 230000000877 morphologic effect Effects 0.000 claims abstract description 31
- 238000000034 method Methods 0.000 claims abstract description 30
- 210000004165 myocardium Anatomy 0.000 claims abstract description 27
- 230000009467 reduction Effects 0.000 claims description 28
- 230000003628 erosive effect Effects 0.000 claims description 20
- 238000004458 analytical method Methods 0.000 claims description 13
- 230000007797 corrosion Effects 0.000 claims description 11
- 238000005260 corrosion Methods 0.000 claims description 11
- 238000003708 edge detection Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 8
- 238000011282 treatment Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 230000004660 morphological change Effects 0.000 claims description 4
- 238000002591 computed tomography Methods 0.000 description 51
- 230000000747 cardiac effect Effects 0.000 description 20
- 238000010586 diagram Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 238000002600 positron emission tomography Methods 0.000 description 8
- 210000002615 epidermis Anatomy 0.000 description 5
- 238000003759 clinical diagnosis Methods 0.000 description 3
- 230000010339 dilation Effects 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 238000002604 ultrasonography Methods 0.000 description 3
- 208000025584 Pericardial disease Diseases 0.000 description 2
- 210000004204 blood vessel Anatomy 0.000 description 2
- 238000002059 diagnostic imaging Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000011946 reduction process Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000000916 dilatatory effect Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012549 training 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
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T7/12—Edge-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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10104—Positron emission tomography [PET]
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (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)
- Image Processing (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The invention discloses an image processing method, which comprises the following steps: obtaining a designated image, wherein the designated image is used for representing a myocardial segmentation region; performing morphological processing on the designated image to obtain a region image, wherein the region image is used for representing a pericardium buffer region corresponding to the myocardial segmentation region; and segmenting the region image based on a two-channel segmentation network to obtain a target image, wherein the target image is used for representing a pericardium segmentation image corresponding to the myocardial segmentation region. By applying the method, the pericardium image can be obtained according to the myocardium segmentation image.
Description
Technical Field
The present invention relates to the field of medical imaging technologies, and in particular, to an image processing method, an image processing apparatus, and a computer storage medium.
Background
The pericardium is the pericardium. The pericardium is an approximately conical fibrous serosal sac, is wrapped outside the heart and the root of a great vessel entering and exiting the heart, has the functions of preventing the heart cavity from being excessively enlarged and stabilizing blood volume, and can be said to be an extremely important protective film of the heart. However, the damage to human body is also very large when the pericardium is diseased, and the pericardial disease is a series of diseases causing physiological or pathological changes of the pericardium to cause clinical symptoms. In clinical diagnosis, because the CT value of the pericardium is not obvious compared with the background, it is relatively difficult to acquire a pericardium image in a CT examination, so that the pericardial disease is often easily missed in clinical diagnosis.
Disclosure of Invention
Embodiments of the present invention provide an image processing method, an apparatus, and a computer storage medium, which have an effect of acquiring a pericardium image from a myocardium segmentation image.
An embodiment of the present invention provides an image processing method, which includes: obtaining a designated image, wherein the designated image is used for representing a myocardial segmentation region; performing morphological processing on the designated image to obtain a region image, wherein the region image is used for representing a pericardium buffer region corresponding to the myocardial segmentation region; and segmenting the region image based on a two-channel segmentation network to obtain a target image, wherein the target image is used for representing a pericardium segmentation image corresponding to the myocardial segmentation region.
In an embodiment, after the segmenting the region image based on the two-channel segmentation network to obtain the target image, the method further includes: and carrying out noise reduction processing on the target image to obtain a pericardium noise reduction image corresponding to the target image.
In an implementation manner, the denoising processing is performed on the target image to obtain a pericardium denoising image corresponding to the target image, and the denoising processing includes: performing connectivity analysis on the target image to obtain a connected domain; screening the connected domains according to the areas, determining the connected domains meeting a set area threshold as growth regions, and deleting the connected domains not meeting the set area threshold; and fitting according to the growing region to obtain a pericardium noise reduction image corresponding to the target image.
In an embodiment, the performing morphological processing on the designated image to obtain a region image includes: performing expansion processing on the designated image to obtain an expanded image; carrying out corrosion treatment on the specified image to obtain a corrosion image; and integrating the expansion image and the erosion image to determine the region image.
In one embodiment, the two-channel segmentation network comprises a first channel based on edge detection features and a second channel based on an image sample, the image sample being used for characterizing the image sample containing pericardium; correspondingly, the segmenting the region image based on the two-channel segmentation network to obtain the target image includes: predicting the area image according to the first channel to obtain first prediction information; predicting the area image according to the second channel to obtain second prediction information; the target image is determined based on the first prediction information and the second prediction information.
In an embodiment, the method further comprises: obtaining an original image; and segmenting the original image through a segmentation network to obtain a specified image.
Another aspect of an embodiment of the present invention provides an image processing apparatus, including: an acquisition module, configured to acquire a specified image, where the specified image is used to represent a myocardial segmentation region; a morphological change module, configured to perform morphological processing on the designated image to obtain a region image, where the region image is used to represent a pericardium buffer region corresponding to the myocardium segmentation region; and the segmentation module is used for segmenting the region image based on a dual-channel segmentation network to obtain a target image, and the target image is used for representing a pericardium segmentation image corresponding to the myocardial segmentation region.
In an embodiment, the segmentation module, the apparatus further includes: and the noise reduction module is used for carrying out noise reduction processing on the target image to obtain a pericardium noise reduction image corresponding to the target image.
In an embodiment, the noise reduction module includes: the analysis submodule is used for carrying out communication analysis on the target image to obtain a communication domain; the screening submodule is used for screening the connected domains according to the areas, determining the connected domains meeting the set area threshold value as growth regions, and deleting the connected domains not meeting the set area threshold value; and the fitting submodule is used for fitting according to the growing region to obtain a pericardium noise reduction image corresponding to the target image.
In an embodiment, the form change module includes: the expansion submodule is used for performing expansion processing on the specified image to obtain an expanded image; the corrosion submodule is used for carrying out corrosion treatment on the specified image to obtain a corrosion image; and the integration submodule is used for integrating the expansion image and the erosion image and determining the area image.
In one embodiment, the two-channel segmentation network comprises a first channel based on edge detection features and a second channel based on an image sample, the image sample being used for characterizing the image sample containing pericardium; correspondingly, the segmentation module comprises: the first channel submodule is used for predicting the area image according to the first channel to obtain first prediction information; the second channel submodule is used for predicting the area image according to the second channel to obtain second prediction information; a determination sub-module for determining a target image based on the first prediction information and the second prediction information.
In an embodiment, the apparatus further comprises: the acquisition module is also used for acquiring an original image; the segmentation module is further used for segmenting the original image through a segmentation network to obtain a specified image.
Another aspect of the invention provides a computer-readable storage medium comprising a set of computer-executable instructions which, when executed, perform any of the image processing methods described above.
In the image processing method, the image processing device and the computer storage medium provided by the embodiment of the invention, the original image is segmented to obtain the designated image, the designated image is morphologically processed to obtain the region image, and the region image is segmented based on the dual-channel segmentation network to obtain the target image. And performing noise reduction processing on the target image to determine a pericardium noise reduction image. By applying the image processing method, the image processing equipment and the computer storage medium provided by the embodiment of the invention, the medical image of the heart can be processed, so that the pericardium image can be quickly obtained, and the image acquisition efficiency is improved.
Drawings
The above and other objects, features and advantages of exemplary embodiments of the present invention will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the invention are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:
in the drawings, the same or corresponding reference numerals indicate the same or corresponding parts.
FIG. 1 is a schematic diagram of an implementation flow of an image processing method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a process of obtaining a region image according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a target image obtaining process of an image processing method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating a designated image obtaining process of an image processing method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a noise reduction process of an image processing method according to an embodiment of the present invention;
fig. 6 is a block diagram of an image processing apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart illustrating an implementation of an image processing method according to an embodiment of the present invention.
Referring to fig. 1, in one aspect, an embodiment of the present invention provides an image processing method, where the method includes: an operation 101 of obtaining a designated image, wherein the designated image is used for representing a myocardial segmentation region; operation 102, performing morphological processing on the designated image to obtain a region image, where the region image is used to represent a pericardium buffer region corresponding to the myocardium segmentation region; in operation 103, the region image is segmented based on the two-channel segmentation network to obtain a target image, and the target image is used for representing a pericardium segmentation image corresponding to the myocardium segmentation region.
The embodiment of the invention aims to provide an image processing method for a user, which is mainly applied to the field of medical images and is used for quickly generating a corresponding target image according to a specified image, wherein the target image can be used for representing a pericardium segmentation image corresponding to a myocardial segmentation region. Specifically, the image processing method provided by the embodiment of the present invention may perform morphological processing on the acquired designated image to obtain the region image, where the morphological processing is used to process the designated image, that is, to distinguish the connected regions in the designated image to obtain the region image. And then, the regional image is segmented by using a two-channel segmentation network to obtain a target image.
Specifically explaining the image processing method in conjunction with the operation process, when the image processing method provided by the embodiment of the present invention is applied to perform image processing, in operation 101, the designated image may be any one of a Computed Tomography (CT), a Positron Emission Tomography (PET), an ultrasound image, or other medical images. Further, the designated images may be obtained by a corresponding medical imager, such as by at least one of an electronic computed tomography scanner, an ultrasound machine, or the like, directly capturing, scanning, and/or image pre-processing the target object, and in one implementation, the designated images may be obtained by segmenting a positron emission tomography image (PET image), wherein the PET image may be used to characterize the heart, including a heart region image and a non-heart background region image. The heart region image includes a myocardium region image and a non-myocardium region image, and the non-myocardium region image may include an valve region image. Further, the non-cardiac background region may include a blood vessel image. The PET image used for representing the heart is segmented, namely images except the myocardial area image in the PET image used for representing the heart are deleted to obtain a designated image, the designated image is the PET image used for representing the heart and is obtained by segmenting the non-heart background area image and the non-myocardial area image, and the designated image can be used for representing the myocardial segmented area.
In operation 102 of the method, after the designated image is acquired, the designated image is processed based on morphology to obtain an area image. Wherein the morphological processing may be applied to simplify the image data, removing the designated area while preserving the basic shape features of the image data, to obtain the desired image. Common morphological treatments are: dilation, erosion, opening and closing, morphological processing can be applied in the environment of noise suppression, feature extraction, edge detection, image segmentation, shape recognition, texture analysis, image restoration and reconstruction, image compression and the like in image processing. In an embodiment of the present invention, morphological processing is used to acquire a region image in a specified image. Further, the type of the region image may be any one of an electronic computed tomography image, a positron emission tomography image, an ultrasound image, or other medical images, but the embodiment of the present invention does not limit the type of the region image, but it should be clear that the region image may be segmented, and since the region image is obtained by morphological processing of the designated image, the image type of the region image may be the same as the designated image.
Fig. 2 is a schematic diagram of a regional image acquisition process of an image processing method according to an embodiment of the present invention.
Referring to fig. 2, in an embodiment, the morphological processing of the designated image to obtain the region image includes: operation 201, performing expansion processing on the designated image to obtain an expanded image; operation 202, performing erosion processing on the designated image to obtain an eroded image; in operation 203, the dilated image and the erosion image are integrated to determine a region image.
In operation 201, a dilation process is performed on a specified image to obtain a dilated image. Namely, the content in the designated image is expanded to obtain an expanded image. The expansion process may be used to eliminate some tiny dark spots and lines. In one implementation, when the designated image is a CT image of the myocardial segmented region, the dilated image is obtained by dilating the contour of the image of the myocardial segmented region to eliminate unnecessary fine dark spots and dark lines in the image.
Correspondingly, in operation 202, the designated image is subjected to erosion processing, and the obtaining of the erosion image specifically includes: the content in the designated image is corroded, and the corrosion treatment can eliminate some tiny bright spots and bright lines. In one implementation, when the designated image is a CT image of the myocardial segmented area, the myocardial segmented area image is eroded, that is, the edge of the myocardial segmented area image is eroded to eliminate some fine bright spots and bright lines in the middle of the image, so as to obtain an eroded image of the myocardial segmented area image.
Finally, the dilated image and the erosion image are integrated according to operation 203 to determine a regional image. Specifically, after obtaining the dilated image obtained by performing dilation processing on the designated image and the erosion image obtained by performing erosion processing on the designated image, it is necessary to perform integration processing on the dilated image and the erosion image in order to obtain the region image, and specifically, when the designated image is a myocardium segmentation region image and the region image is a pericardium buffer region corresponding to the myocardium segmentation region image, the integration processing may be a morphological gradient map which is a difference between the dilated image and the erosion image, wherein the morphological gradient is used to retain an edge contour of the object after the morphological processing, and when the designated image is a myocardium segmentation region image, the myocardial segmentation region image is morphologically processed to obtain the dilated image of the myocardium segmentation region and the erosion image of the myocardial segmentation region, and the dilated image of the myocardial segmentation region and the erosion image of the myocardial segmentation region are differentiated, a morphological gradient map of the myocardium segmented region image, that is, a pericardium buffer region image, is obtained, which retains an edge contour as the myocardium segmented region, and thus, the pericardium buffer region image may be a myocardium epidermis region image including a pericardium.
In an implementation case, the designated image may be a CT image obtained indirectly by an electronic computed tomography scanner, and a CT image obtained by directly photographing and scanning the target object and then performing image preprocessing to characterize the myocardial segmentation region by the electronic computed tomography scanner. And carrying out morphological processing on the CT image to obtain a region image corresponding to the CT image. In the method, the region image may be a CT image used for characterizing a pericardium buffer region corresponding to the myocardial segmentation region, it should be noted that the pericardium buffer region corresponding to the myocardial segmentation region and characterized by the region image is a blurred region including a pericardium, further, the pericardium buffer region may be a myocardial epidermis region including a pericardium, specifically, the pericardium buffer region includes a pericardium region and a non-pericardium region connected to the pericardium region. The non-pericardial region may be a portion of the myocardial region or a fat region of the pericardium that is adhered to the pericardium, among others. That is, the region image may include a pericardium region corresponding to the myocardium segmentation region and a non-pericardium region connected to the pericardium region.
To obtain a target image, a region image is segmented based on a two-channel segmentation network to obtain a target image, the target image being used to characterize a pericardium segmentation image corresponding to a myocardium segmentation region in operation 103. The method comprises the steps that the two-channel segmentation network is one of image segmentation networks, the two-channel segmentation network is formed by training based on edge detection features corresponding to image samples and the image samples and is used for segmenting regional images to obtain target images, in the method, the image samples are images containing pericardium buffer regions, and the two-channel segmentation network is used for segmenting the images containing the pericardium buffer regions to obtain images containing the pericardium. In one implementation case, the region image is a CT image for characterizing a pericardium buffer region corresponding to the myocardium segmentation region, and the CT image for characterizing the pericardium buffer region corresponding to the myocardium segmentation region is segmented based on a two-channel segmentation network, so as to obtain a target image corresponding to the CT image for characterizing the pericardium buffer region. The type of the target image is the same as that of the region image, and the target image is used for representing a pericardium segmentation image corresponding to the myocardium segmentation region.
Fig. 3 is a schematic diagram of a target image obtaining process of an image processing method according to an embodiment of the present invention.
Referring to fig. 3, in one possible embodiment, a two-channel segmentation network includes a first channel based on edge detection features and a second channel based on an image sample used to characterize the pericardium-containing image sample; accordingly, operation 103 is performed to segment the region image based on the two-channel segmentation network to obtain the target image, and includes: operation 301, predicting the region image according to the first channel to obtain first prediction information; operation 302, predicting the region image according to the second channel to obtain second prediction information; in operation 303, a target image is determined based on the first prediction information and the second prediction information.
Specifically, in order to segment the pericardium buffer area image to obtain a segmented pericardium image, i.e., a target image, in operation 103, the segmenting the area image based on the two-channel segmentation network, and the obtaining the target image includes: operation 301, predicting the region image according to the first channel to obtain first prediction information; specifically, the first channel is a channel that performs prediction based on edge detection features corresponding to the image samples, that is, the first channel is used to predict the area image according to the edge detection features to obtain first prediction information. Accordingly, in operation 302, the area image is predicted according to the second channel, and second prediction information is obtained. The second channel is a channel for performing prediction based on an image sample, and it is to be understood that the image sample is used to characterize an image sample containing a pericardium.
Accordingly, in an implementable case, the second prediction information may also be a scheme for segmenting the region image. In operation 303, a target image is determined based on the first prediction information and the second prediction information. Specifically, when the first prediction information is one scheme for dividing the region image and the second prediction information is the same as the other scheme for dividing the region image, the region image is divided based on the two schemes for dividing the region image, and the target image is finally obtained.
Fig. 4 is a schematic diagram of a designated image obtaining process of an image processing method according to an embodiment of the present invention.
Referring to fig. 4 in one possible implementation, the method further includes: operation 401, obtaining an original image; in operation 402, the original image is segmented by the segmentation network to obtain a designated image.
The image processing method according to the embodiment of the present invention further includes an operation 401 of obtaining an original image. The raw images may be cardiac CT images obtained by electronic computed tomography, including CT images of cardiac regions as well as CT images of non-cardiac background regions. It should be noted that the kind of the original image is not limited in the embodiment of the present invention. By segmenting the cardiac CT image, a given image can be obtained. Further, in operation 402, the original image is segmented by a segmentation network to obtain a specified image. In one implementation, the original image may be a cardiac CT image obtained by an electronic computed tomography scanner, and the original image includes a cardiac region CT image and a CT image of a non-cardiac background region, wherein the cardiac region CT image may include a CT image for characterizing a myocardial region and a CT image of a non-cardiac background region, and the CT image of a non-cardiac background region may include a CT image of a blood vessel. Furthermore, the designated image may be a CT image for characterizing the myocardial segmentation region, and in operation 402, the segmentation network is used to segment the cardiac CT image, and the images other than the CT image of the myocardial region in the cardiac CT image are deleted to obtain the CT image for characterizing the myocardial segmentation region, i.e., the designated image.
To facilitate further understanding of the above implementation method, a specific implementation scenario is provided below, in which an image processing method provided by an embodiment of the present invention may obtain a pericardium segmentation image according to an original image. Specifically, in an implementation scenario of the embodiment of the present invention, the original image may be a cardiac CT image directly obtained by an electronic computed tomography scanner, and the designated image may be a myocardial segmentation region image, in an implementation case, in order to obtain a pericardium segmentation image, the cardiac CT image is segmented according to a segmentation network to obtain the myocardial segmentation region image, where the cardiac CT image includes CT images used for representing the myocardial region image and a non-myocardial region image, and the segmentation network is used for deleting the non-myocardial region in the cardiac CT image to obtain the myocardial segmentation region image, that is, the designated image. And performing morphological processing on the image of the myocardial segmentation region according to a morphological processing principle to obtain a region image, wherein in an implementable case, the region image can be a pericardium buffer region image corresponding to the image of the myocardial segmentation region, the image of the central envelope buffer region is a myocardial epidermis region including a pericardium, the myocardial epidermis region includes the pericardium region and a non-pericardium region connected with the pericardium region, the image of the myocardial envelope buffer region is segmented by a dual-channel segmentation network trained by using edge detection features corresponding to the image sample and the image sample, and finally the image of the pericardium segmentation corresponding to the myocardial segmentation region is obtained.
In an embodiment, after the region image is segmented based on the two-channel segmentation network to obtain the target image, the method further includes: and carrying out noise reduction processing on the target image to obtain a pericardium noise reduction image corresponding to the target image.
Because the acquisition of the segmented pericardium image is obtained by processing the image obtained by the medical imaging instrument, the obtained segmented pericardium image has image noise. Since image noise is unnecessary or redundant interference information present in image data and the presence of image noise seriously affects the imaging quality of an image, it is necessary to perform noise reduction processing on a pericardial segmented image to obtain a pericardial noise-reduced image corresponding to the pericardial segmented image in order to perform imaging optimization on the obtained pericardial segmented image. The same pericardium segmented image is the same as the pericardium noise-reduced image in image type, and when the pericardium segmented image is a CT image, the pericardium noise-reduced image is also a CT image.
Fig. 5 is a schematic view of a noise reduction process of an image processing method according to an embodiment of the present invention.
Referring to fig. 5, in an embodiment, performing noise reduction processing on a target image to obtain a pericardium noise-reduced image corresponding to the target image includes: operation 501, performing connectivity analysis on the target image to obtain a connected domain; operation 502, screening connected domains according to area, determining the connected domains meeting a set area threshold as growth regions, and deleting the connected domains not meeting the set area threshold; and operation 503, performing fitting according to the growth area to obtain a pericardium noise reduction image corresponding to the target image.
Specifically, in order to perform noise reduction on the target image and obtain a pericardium noise-reduced image corresponding to the target image, in operation 501, connectivity analysis is performed on the target image to obtain a connected domain. In an implementation case, the target image may be a segmented pericardium image, and after obtaining the segmented pericardium image, the segmented pericardium image is first subjected to connected domain analysis to determine a plurality of connected domains distributed on the segmented pericardium image.
In operation 502, connected domains are screened according to area, connected domains satisfying a set area threshold are determined as growth regions, and connected domains not satisfying the set area threshold are deleted. The method comprises the steps that a plurality of acquired connected regions distributed on a target image, namely a pericardium segmentation image, are classified according to the area size, the connected regions are determined to be growing regions under the condition that the area of the connected regions distributed on the pericardium segmentation image meets a preset area threshold, and the connected regions are deleted on the pericardium segmentation image under the condition that the area of the connected regions distributed on the pericardium segmentation image does not exceed the preset area threshold. And finally, performing operation 503, fitting according to the growing region, and obtaining a pericardium noise reduction image corresponding to the target image. Namely, the growing region is used as a fitting seed, and other regions on the target image are fitted according to the growing region, so that the pericardium noise reduction image is finally obtained. In an implementation case, the other region may be a region corresponding to the deleted connected component, and the growing region is used as a fitting seed, and the other region on the target image is fitted according to the growing region, that is, the growing region is extracted to restore the region from which the connected component is deleted, where the restoring includes filling the region from which the connected component is deleted with the growing region.
In a specific implementation, an embodiment of the present invention provides an image processing method, which has a function of quickly obtaining a pericardium segmentation image from a cardiac CT image. Specifically, the original image may be a cardiac CT image obtained by an electronic computer tomography scanner, the designated image may be a myocardium segmentation region image, and the region image may be a pericardium buffer region image corresponding to the myocardium segmentation region image. In an embodiment of the present invention, an image processing method is first required to obtain a cardiac CT image including a CT image representing a myocardial region image and a non-myocardial region image. The heart CT image is segmented by a segmentation network, and non-myocardial regions are deleted to obtain a myocardial segmented region image. Further, performing morphological processing on the image of the myocardial segmentation region, namely performing expansion processing on the image of the myocardial segmentation region to obtain an expanded image, correspondingly performing erosion processing on the image of the myocardial segmentation region to obtain an erosion image, and determining a morphological gradient map corresponding to the image of the myocardial segmentation region according to the expanded image and the erosion image. The morphological gradient map is an image obtained by subtracting the dilated image and the erosion image, and is used for preserving the edge contour of the object after morphological processing, so that the morphological gradient map obtained from the myocardium segmentation region image is a myocardium epidermis region image including a pericardium, that is, a pericardium buffer region image. And then, according to the two-channel segmentation network, deleting the non-pericardium part in the pericardium buffer area image to obtain a pericardium segmentation image. Finally, in order to better apply the segmented images of the pericardium to clinical diagnosis, the segmented images of the pericardium need to be denoised, and specifically, denoising the segmented images of the pericardium to obtain a denoised image of the pericardium includes: performing connectivity analysis on the pericardial segmented images to obtain connected domains distributed on the pericardial segmented images, classifying the connected domains according to the areas of the connected domains, determining the connected domains as growth areas when the areas of the connected domains meet a preset threshold, deleting the connected domains on the pericardial segmented images when the areas of the connected domains do not meet the preset threshold, and finally fitting the target images according to the growth areas by using the growth areas as fitting seeds to finally obtain the pericardial noise reduction images.
Fig. 6 is a block diagram of an image processing apparatus according to an embodiment of the present invention.
Another aspect of the embodiments of the present invention provides an image processing apparatus with reference to fig. 6, the apparatus including: an obtaining module 601, configured to obtain a designated image, where the designated image is used to represent a myocardial segmentation region; a morphological change module 604, configured to perform morphological processing on the designated image to obtain a region image, where the region image is used to represent a pericardium buffer region corresponding to the myocardium segmentation region; a segmentation module 602, configured to segment the region image based on a two-channel segmentation network to obtain a target image, where the target image is used to represent a pericardium segmentation image corresponding to the myocardium segmentation region.
In an implementation manner, after the segmentation module 602 segments the region image based on the two-channel segmentation network to obtain the target image, the apparatus further includes: and the denoising module 603 is configured to perform denoising processing on the target image to obtain a pericardium denoising image corresponding to the target image.
In an implementation manner, the denoising module 603 performs denoising processing on the target image to obtain a pericardium denoised image corresponding to the target image, and includes: an analysis submodule 6031 configured to perform connectivity analysis on the target image to obtain a connected domain; a screening submodule 6032, configured to screen connected domains according to areas, determine a connected domain that meets a set area threshold as a growth region, and delete a connected domain that does not meet the set area threshold; and the fitting submodule 6033 is configured to perform fitting according to the growth area to obtain a pericardium noise reduction image corresponding to the target image.
In one embodiment, the morphological change module 604 performs morphological processing on the designated image to obtain a region image, and the morphological processing includes: an expansion sub-module 6041 configured to perform expansion processing on the specified image to obtain an expanded image; a corrosion submodule 6042, configured to perform corrosion processing on the specified image to obtain a corrosion image; and an integrating sub-module 6043 for integrating the expansion image and the erosion image to determine a region image.
In one embodiment, the two-channel segmentation network comprises a first channel based on edge detection features and a second channel based on an image sample, wherein the image sample is used for characterizing the image sample containing the pericardium;
correspondingly, the segmentation module 602 segments the region image based on the two-channel segmentation network to obtain the target image, including: the first channel sub-module 6021 is configured to predict the region image according to the first channel to obtain first prediction information; a second channel sub-module 6022, configured to predict the area image according to a second channel to obtain second prediction information; a determination sub-module 6023 for determining the target image based on the first prediction information and the second prediction information.
In one embodiment, the apparatus further comprises: the obtaining module 601 is further configured to obtain an original image; the segmentation module 602 is further configured to segment the original image through a segmentation network to obtain a specified image. Another aspect of the invention provides a computer-readable storage medium comprising a set of computer-executable instructions which, when executed, perform any of the image processing methods described above.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. An image processing method, characterized in that the method comprises:
obtaining a designated image, wherein the designated image is used for representing a myocardial segmentation region;
performing morphological processing on the designated image to obtain a region image, wherein the region image is used for representing a pericardium buffer region corresponding to the myocardial segmentation region;
and segmenting the region image based on a two-channel segmentation network to obtain a target image, wherein the target image is used for representing a pericardium segmentation image corresponding to the myocardial segmentation region.
2. The method of claim 1, wherein after the segmentation of the region image based on the two-pass segmentation network to obtain a target image, the method further comprises:
and carrying out noise reduction processing on the target image to obtain a pericardium noise reduction image corresponding to the target image.
3. The method according to claim 2, wherein the denoising processing is performed on the target image to obtain a pericardium denoising image corresponding to the target image, and the method comprises:
performing connectivity analysis on the target image to obtain a connected domain;
screening the connected domains according to the areas, determining the connected domains meeting a set area threshold as growth regions, and deleting the connected domains not meeting the set area threshold;
and fitting according to the growing region to obtain a pericardium noise reduction image corresponding to the target image.
4. The method according to claim 1, wherein the performing morphological processing on the designated image to obtain a region image comprises:
performing expansion processing on the designated image to obtain an expanded image;
carrying out corrosion treatment on the specified image to obtain a corrosion image;
and integrating the expansion image and the erosion image to determine the region image.
5. The method of claim 1, wherein the two-channel segmentation network comprises a first channel based on edge detection features and a second channel based on an image sample used to characterize the pericardium-containing image sample;
correspondingly, the segmenting the region image based on the two-channel segmentation network to obtain the target image includes:
predicting the area image according to the first channel to obtain first prediction information;
predicting the area image according to the second channel to obtain second prediction information;
determining a target image based on the first prediction information and the second prediction information.
6. The method of claim 1, further comprising:
obtaining an original image;
and segmenting the original image through a segmentation network to obtain a specified image.
7. An image processing apparatus, characterized in that the apparatus comprises:
an acquisition module, configured to acquire a specified image, where the specified image is used to represent a myocardial segmentation region;
a morphological change module, configured to perform morphological processing on the designated image to obtain a region image, where the region image is used to represent a pericardium buffer region corresponding to the myocardium segmentation region;
and the segmentation module is used for segmenting the region image based on a dual-channel segmentation network to obtain a target image, and the target image is used for representing a pericardium segmentation image corresponding to the myocardial segmentation region.
8. The apparatus of claim 7, further comprising:
and the noise reduction module is used for carrying out noise reduction processing on the target image to obtain a pericardium noise reduction image corresponding to the target image.
9. The apparatus of claim 7, wherein the noise reduction module comprises:
the analysis submodule is used for carrying out communication analysis on the target image to obtain a communication domain;
the screening submodule is used for screening the connected domains according to the areas, determining the connected domains meeting the set area threshold value as growth regions, and deleting the connected domains not meeting the set area threshold value;
and the fitting submodule is used for fitting according to the growing region to obtain a pericardium noise reduction image corresponding to the target image.
10. A computer-readable storage medium comprising a set of computer-executable instructions that, when executed, perform the image processing method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010935129.7A CN112184632B (en) | 2020-09-08 | 2020-09-08 | Image processing method, device and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010935129.7A CN112184632B (en) | 2020-09-08 | 2020-09-08 | Image processing method, device and computer storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112184632A true CN112184632A (en) | 2021-01-05 |
| CN112184632B CN112184632B (en) | 2024-04-16 |
Family
ID=73925471
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010935129.7A Active CN112184632B (en) | 2020-09-08 | 2020-09-08 | Image processing method, device and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112184632B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140029835A1 (en) * | 2012-07-27 | 2014-01-30 | Samsung Electronics Co., Ltd. | Method and apparatus for determining blood flow required, method and apparatus for producing blood flow image, and method and apparatus for processing myocardial perfusion image |
| CN109993797A (en) * | 2019-04-04 | 2019-07-09 | 广东三维家信息科技有限公司 | Door and window method for detecting position and device |
| CN110378868A (en) * | 2019-05-29 | 2019-10-25 | 心医国际数字医疗***(大连)有限公司 | Pericardium dividing method based on CT sequence and the method from pericardium formation center layer slice coarse localization pericardium region |
| CN111105424A (en) * | 2019-12-19 | 2020-05-05 | 广州柏视医疗科技有限公司 | Lymph node automatic delineation method and device |
-
2020
- 2020-09-08 CN CN202010935129.7A patent/CN112184632B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140029835A1 (en) * | 2012-07-27 | 2014-01-30 | Samsung Electronics Co., Ltd. | Method and apparatus for determining blood flow required, method and apparatus for producing blood flow image, and method and apparatus for processing myocardial perfusion image |
| CN109993797A (en) * | 2019-04-04 | 2019-07-09 | 广东三维家信息科技有限公司 | Door and window method for detecting position and device |
| CN110378868A (en) * | 2019-05-29 | 2019-10-25 | 心医国际数字医疗***(大连)有限公司 | Pericardium dividing method based on CT sequence and the method from pericardium formation center layer slice coarse localization pericardium region |
| CN111105424A (en) * | 2019-12-19 | 2020-05-05 | 广州柏视医疗科技有限公司 | Lymph node automatic delineation method and device |
Non-Patent Citations (1)
| Title |
|---|
| 茅正冲 等: "一种基于混合高斯模型的运动目标检测新算法", 《科学技术与工程》, vol. 13, no. 14, pages 4070 - 4078 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112184632B (en) | 2024-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110706246B (en) | Blood vessel image segmentation method and device, electronic equipment and storage medium | |
| CN107527341B (en) | Method and system for processing angiography image | |
| John et al. | Multilevel thresholding based segmentation and feature extraction for pulmonary nodule detection | |
| CN110610498A (en) | Mammary gland molybdenum target image processing method, system, storage medium and equipment | |
| Sreedevi et al. | A novel approach for removal of pectoral muscles in digital mammogram | |
| US20110002523A1 (en) | Method and System of Segmenting CT Scan Data | |
| JPH07504056A (en) | Methods for identifying objects using data processing techniques | |
| JP2011517986A (en) | Automatic detection and accurate segmentation of abdominal aortic aneurysms | |
| CN116309676A (en) | Coronary vessel image segmentation method and device, electronic equipment and storage medium | |
| CN112862850A (en) | Digital silhouette angiography image processing method and device | |
| CN116309647B (en) | Method for constructing craniocerebral lesion image segmentation model, image segmentation method and device | |
| CN113034522A (en) | CT image segmentation method based on artificial neural network | |
| Makandar et al. | A review on preprocessing techniques for digital mammography images | |
| Ihsan et al. | A median filter with evaluating of temporal ultrasound image for impulse noise removal for kidney diagnosis | |
| CN112184632B (en) | Image processing method, device and computer storage medium | |
| Bania et al. | Adaptive trimmed median filter for impulse noise detection and removal with an application to mammogram images | |
| KR20000056228A (en) | An image processing method for a liver and a spleen from tomographical image | |
| CN115775219A (en) | Medical image segmentation method, system, electronic device, and medium | |
| Hamed et al. | A proposed model for denoising breast mammogram images | |
| Saoji et al. | Speckle and rician noise removal from medical images and Ultrasound images | |
| Hou et al. | Research on tumor image segmentation in medical imaging based on extremum adaptive median filtering | |
| Ganesan et al. | Hybrid contrast enhancement approach for medical image | |
| CN113658172B (en) | Image processing method and device, computer readable storage medium and electronic device | |
| Prakash | Medical image processing methodology for liver tumour diagnosis | |
| US12016696B2 (en) | Device for the qualitative evaluation of human organs |
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 |