CN112766258B - Image segmentation method, system, electronic device and computer readable storage medium - Google Patents
Image segmentation method, system, electronic device and computer readable storage medium Download PDFInfo
- Publication number
- CN112766258B CN112766258B CN202011643646.3A CN202011643646A CN112766258B CN 112766258 B CN112766258 B CN 112766258B CN 202011643646 A CN202011643646 A CN 202011643646A CN 112766258 B CN112766258 B CN 112766258B
- Authority
- CN
- China
- Prior art keywords
- region
- interest
- dimensional
- medical image
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000003709 image segmentation Methods 0.000 title claims abstract description 27
- 230000011218 segmentation Effects 0.000 claims abstract description 62
- 238000004590 computer program Methods 0.000 claims description 3
- 238000012986 modification Methods 0.000 abstract description 5
- 230000004048 modification Effects 0.000 abstract description 5
- 238000012937 correction Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 18
- 238000002595 magnetic resonance imaging Methods 0.000 description 5
- 210000001519 tissue Anatomy 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 5
- 238000002591 computed tomography Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000002597 diffusion-weighted imaging Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000001788 irregular Effects 0.000 description 3
- 210000000988 bone and bone Anatomy 0.000 description 2
- 238000002059 diagnostic imaging Methods 0.000 description 2
- 238000009792 diffusion process Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 206010028980 Neoplasm Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 210000004351 coronary vessel Anatomy 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention provides an image segmentation method, an image segmentation system, electronic equipment and a computer readable storage medium. Because the data of the medical image in the three-dimensional region is input into the segmentation model, the target region segmented by the segmentation model cannot exceed the three-dimensional region, so that segmentation errors can be avoided, segmentation precision is improved, workload of modification and correction of a later physical operator is reduced, and misdiagnosis of a patient caused by inaccurate segmentation is avoided; meanwhile, the images input into the segmentation model are smaller, so that the segmentation time and the operation amount can be reduced, and the segmentation efficiency is improved.
Description
Technical Field
The present invention relates to the field of medical image technology, and in particular, to an image segmentation method, system, electronic device, and computer readable storage medium.
Background
Medical images refer to techniques and procedures for non-invasively acquiring images of internal tissue of a human body or a portion of a human body for medical or medical research. It comprises the following two relatively independent directions: a medical imaging system (MEDICAL IMAGING SYSTEM) and medical image processing (MEDICAL IMAGE processing). The former refers to the image forming process, including problems of imaging mechanism, imaging equipment, imaging system analysis and the like; the latter means that the obtained medical image is further processed, for example, the medical image is subjected to denoising, segmentation and the like, and the medical image is used as an auxiliary means for subsequent clinical diagnosis, treatment and the like.
The segmentation of medical images is currently divided into two types: one is to manually delineate and segment the region of interest; the other is to automatically draw out the region of interest and divide the region of interest by using a neural network model. The manual sketching mode requires a physical engineer to sketch the medical image, is time-consuming and labor-consuming, and has the problem that sketching standards of different physical engineers are not uniform. The automatic sketching mode needs to learn the medical images of a large number of treated cases, and because the input medical images are often much larger than the actually required sketched region of interest, the situation that other regions are sketched by mistake during automatic sketching often occurs, so that the operation amount of correcting after automatic segmentation is relatively large.
Disclosure of Invention
The invention aims to provide an image segmentation method, an image segmentation system, electronic equipment and a computer readable storage medium, which are used for solving the problems of time consumption, labor consumption or inaccuracy in the process of segmenting medical images.
In order to achieve the above object, the present invention provides a segmentation method of an image, comprising:
Providing a medical image;
Setting a three-dimensional region characterizing the region of interest in the medical image; and
And intercepting the data of the medical image in the three-dimensional region and inputting the data into a preset segmentation model to segment out a target region in the region of interest.
Optionally, the medical image is a three-dimensional image, and the step of setting a three-dimensional region characterizing the region of interest in the medical image includes:
Selecting a medical image with the region of interest on a view, and outlining a two-dimensional region;
selecting a medical image with the region of interest on another view, and outlining another two-dimensional region; and
And fitting the three-dimensional area by using the two outlined two-dimensional areas.
Alternatively, both two-dimensional regions are rectangular, and the three-dimensional region is a cuboid.
Optionally, the three-dimensional region is a smallest cuboid that accommodates two-dimensional regions.
Optionally, the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
And respectively selecting two medical images of the boundary of the region of interest from the three views, and enclosing the medical images selected from the three views into the three-dimensional region.
Optionally, the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
selecting a medical image with the region of interest on a view, and outlining a two-dimensional region; and
And selecting two medical images of the boundary of the region of interest on the view, wherein the boundary of the two-dimensional region and the two selected medical images define the three-dimensional region.
Optionally, the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
selecting a closed region characterizing the region of interest in the medical image by using a tolerance method or an energy field method; and
And taking the smallest cuboid containing the closed area as the three-dimensional area.
The invention also provides an image segmentation system, which comprises:
An image providing module for providing a medical image;
The region selection module is used for setting a three-dimensional region representing the region of interest in the medical image; and
The input module is used for intercepting the medical image in the three-dimensional area and inputting the medical image into a preset segmentation model so as to segment a target area in the region of interest.
The present invention also provides an electronic device,
A memory for storing one or more programs; and
The one or more programs, when executed by the one or more processors, cause the one or more processors to implement a segmentation method for the image.
The present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of segmentation of an image.
In the image segmentation method, system, electronic equipment and computer readable storage medium provided by the invention, a three-dimensional area representing an interested area is firstly set in a medical image, and data of the medical image in the three-dimensional area are intercepted and input into a preset segmentation model so as to segment a target area in the interested area. Because the data of the medical image in the three-dimensional region is input into the segmentation model, the target region segmented by the segmentation model cannot exceed the three-dimensional region, so that segmentation errors can be avoided, segmentation precision is improved, workload of modification and correction of a later physical operator is reduced, and misdiagnosis of a patient caused by inaccurate segmentation is avoided; meanwhile, the images input into the segmentation model are smaller, so that the segmentation time and the operation amount can be reduced, and the segmentation efficiency is improved.
Drawings
FIG. 1 is a flowchart of an image segmentation method according to a first embodiment of the present invention;
FIGS. 2 a-2 c are flow charts illustrating selecting a medical image having a region of interest over a first field of view and delineating a first two-dimensional region according to a first embodiment of the present invention;
FIGS. 3 a-3 c are flow charts illustrating selecting a medical image with a region of interest over a second field of view and delineating a second two-dimensional region according to a first embodiment of the present invention;
FIG. 4 is a schematic diagram of fitting a first two-dimensional region and a second two-dimensional region to a three-dimensional region according to an embodiment of the present invention;
FIG. 5a is a schematic diagram of fitting a first two-dimensional region and a third two-dimensional region to a three-dimensional region according to an embodiment of the present invention;
FIG. 5b is a schematic diagram of fitting a second two-dimensional region and a third two-dimensional region to a three-dimensional region according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a target region in a region of interest segmented using a segmentation model according to an embodiment of the present invention;
Fig. 7a is a schematic diagram of two medical images where a boundary of a region of interest is selected on a first field of view according to a second embodiment of the present invention;
fig. 7b is a schematic diagram of two medical images where a boundary of a region of interest is selected on a second field of view according to a second embodiment of the present invention;
fig. 7c is a schematic diagram of two medical images where a boundary of a region of interest is selected on a third field of view according to a second embodiment of the present invention;
FIG. 8 is a schematic diagram of defining the three-dimensional region by using the boundary of the two-dimensional region and two selected medical images according to a fourth embodiment of the present invention;
Fig. 9 is a block diagram of an image segmentation system according to a fifth embodiment of the present invention;
wherein, the reference numerals are as follows:
10-an image providing module; 20-an area selection module; 30-an input module;
d1—a first two-dimensional region; d2—a second two-dimensional region; d3—a third two-dimensional region; h-three-dimensional region;
Q1, Q2-center point.
Detailed Description
Specific embodiments of the present invention will be described in more detail below with reference to the drawings. The advantages and features of the present invention will become more apparent from the following description. It should be noted that the drawings are in a very simplified form and are all to a non-precise scale, merely for convenience and clarity in aiding in the description of embodiments of the invention.
To solve the problems in the prior art, the present embodiment provides an image segmentation method, system, electronic device, and computer-readable storage medium.
It should be noted that, the image segmentation method according to the embodiment of the present invention may be applied to the image segmentation system of the similar problem of the present embodiment, and the image segmentation system may be configured on an electronic device. The electronic device may be a personal computer, a mobile terminal, etc., and the mobile terminal may be a hardware device with various operating systems, such as a mobile phone, a tablet computer, etc.
Example 1
Fig. 1 is a flowchart of an image segmentation method according to an embodiment of the present invention. As shown in fig. 1, the present embodiment provides an image segmentation method, which includes step S100, step S200, and step S300.
Step S100: a medical image is provided.
In particular, the medical image in this embodiment is, for example, a three-dimensional image, which has several medical images in at least three fields of view, which may be a field of view along the front face of the patient, a field of view along the side face of the patient, or a field of view perpendicular to the front face and the side face of the patient. It will be appreciated that the medical image may be obtained by directly capturing medical images over three fields of view of the patient; or only medical images on one visual field of the patient can be shot, and medical images on the other two visual fields are acquired in a three-dimensional reconstruction mode, so that the medical images are obtained.
The medical image is an image of a specific mode, can be an image of a certain resolution, and can be an image of a plurality of modes or a plurality of resolutions. The medical image may be a magnetic resonance imaging (Magnetic Resonance Imaging, MRI) image, an electronic computed tomography (Computed Tomography, CT) image, a positron emission computed tomography (Positron Emission Computed Tomography, PET) image, an ultrasound image, or the like. Taking the medical image as an MRI image as an example, the MRI image may be a T1 weighted image, a T2 weighted image, an apparent Diffusion coefficient (Apparent Diffusion Coefficient, ADC), or a Diffusion weighted imaging (Diffusion WEIGHTED IMAGING, DWI) image.
In particular, the computer device may obtain the medical image by performing a three-dimensional reconstruction of the data of the part to be examined of the patient acquired by the scanning device. Of course, the medical image may be reconstructed in advance and stored in the computer device, and the medical image is read directly from the memory of the computer device when the medical image is required to be used. Of course, the computer device may also acquire the medical image from an external device. For example, the medical image is stored in a cloud, and when the medical image is needed to be used, the computer device acquires the medical image of the patient from the cloud. The present embodiment is not limited to the manner in which the medical image is acquired.
In this embodiment, the medical image is a tomographic image sequence, but the present invention is not limited thereto.
Step S200: a three-dimensional region characterizing the region of interest is set in the medical image.
Specifically, selecting a medical image with the region of interest on a visual field, and outlining a first two-dimensional region; selecting a medical image with the region of interest on the other view, and outlining a second two-dimensional region; the three-dimensional region is fitted with the first two-dimensional region and the second two-dimensional region. In this embodiment, the view along the front and side surfaces of the patient is taken as a first view, the view along the front surface of the patient is taken as a second view, the view along the side surface of the patient is taken as a third view, and the first, second and third views are mutually perpendicular to each other, but not limited thereto.
Fig. 2a to 2c are flowcharts for selecting a medical image with a region of interest on a first field of view and outlining a first two-dimensional region according to this embodiment. As shown in fig. 2a, the physicist views several medical images over the first field of view and selects one medical image with the region of interest from the several medical images of the first field of view. As shown in fig. 2b, the physicist finds the center point Q1 of the region of interest in the selected medical image, and it should be understood that, since the region of interest is generally irregular, the center point Q1 referred to in this embodiment is only the center of the region of interest observed by the human eye, and is not necessarily the absolute center position of the region of interest. As shown in fig. 2c, the first two-dimensional region D1 is delineated in the medical image centered on the center point Q1, the first two-dimensional region D1 being capable of characterizing the region of interest over the first field of view.
Fig. 3a to 3c are flowcharts for selecting a medical image with a region of interest on a second field of view and outlining a second two-dimensional region according to this embodiment. As shown in fig. 3a, the physicist views several medical images in the second field of view and selects one medical image with the region of interest from among the several medical images in the second field of view, where the selected region of interest in the medical image should coincide with the selected region of interest in the medical image in the first field of view. As shown in fig. 3b, the physicist finds the center point Q2 of the region of interest in the selected medical image, it should be understood that, since the region of interest is generally irregular, the center point Q2 referred to in this embodiment is only the center of the region of interest as observed by the human eye, and is not necessarily the absolute center position of the region of interest. As shown in fig. 3c, the second two-dimensional region D2 is delineated in the medical image centered on the center point Q2, the second two-dimensional region D2 being capable of characterizing the region of interest over the second field of view.
In this embodiment, the first two-dimensional area D1 and the second two-dimensional area D2 have preset shapes, and the preset shapes may be rectangular or elliptical, or may be specific shapes preset according to different tissue characteristics, so as to better adapt to the segmentation requirements of different tissues. For example, for tumors, the preset shape may be circular or elliptical; for a bone, the preset shape may be the shape of the model to which the bone corresponds. In this embodiment, the first two-dimensional area D1 and the second two-dimensional area D2 are both rectangular, preferably, the first two-dimensional area D1 is a smallest rectangle capable of characterizing the region of interest on the first field of view, and the second two-dimensional area D2 is a smallest rectangle capable of characterizing the region of interest on the second field of view, so as to facilitate subsequent processing, which will be described later.
It should be understood that, in this embodiment, the first two-dimensional area D1 and the second two-dimensional area D2 are delineated by searching for the center point of the region of interest, so that the region of interest can be approximately defined by the first two-dimensional area D1 and the second two-dimensional area D2 in the corresponding field of view, however, the present invention is not limited thereto, and as long as the manner of delineating the first two-dimensional area D1 and the second two-dimensional area D2 that characterize the region of interest is within the scope of the present invention, for example, the center point of the region of interest is not searched, but the first two-dimensional area D1 and the second two-dimensional area D2 are delineated directly.
In this embodiment, the three-dimensional region is a cuboid capable of accommodating the first two-dimensional region D1 and the second two-dimensional region D2. Further, the first two-dimensional region D1 is a smallest rectangular box on the first field of view that can characterize the region of interest, and the second two-dimensional region D2 is a smallest rectangular box on the second field of view that can characterize the region of interest, the three-dimensional region being a smallest cuboid that can characterize the region of interest.
Fig. 4 is a schematic diagram of fitting a first two-dimensional region and a second two-dimensional region into a three-dimensional region according to the present embodiment. As shown in fig. 4, since the first view and the second view are perpendicular to each other, the first two-dimensional area D1 and the second two-dimensional area D2 should also be perpendicular to each other. The first two-dimensional area D1 and the second two-dimensional area D2 may define a surface of a cuboid on the first view field and the second view field, and then the opposite sides of the second two-dimensional area D2 are taken as boundaries on the first view field and the first two-dimensional area D1 is taken as boundaries on the second view field, so that a cuboid can be fit as the three-dimensional area H, and the three-dimensional area H is a smallest cuboid capable of accommodating the first two-dimensional area D1 and the second two-dimensional area D2.
Fig. 5a is a schematic diagram of fitting a first two-dimensional region and a third two-dimensional region into a three-dimensional region according to the present embodiment. As an alternative embodiment, as shown in fig. 5a, a medical image with a region of interest may also be selected on a third field of view and a third two-dimensional region D3 is delineated, and since the first field of view and the third field of view are perpendicular to each other, the first two-dimensional region D1 and the third two-dimensional region D3 should also be perpendicular to each other. The first two-dimensional area D1 and the third two-dimensional area D3 may define surfaces of a cuboid on the first view and the third view, and then the opposite sides of the third two-dimensional area D3 are used as boundaries on the first view, and the first two-dimensional area D1 is used as boundaries on the third view, and a cuboid may also be fitted as the three-dimensional area H, where the three-dimensional area H is a smallest cuboid capable of accommodating the first two-dimensional area D1 and the third two-dimensional area D3.
Fig. 5b is a schematic diagram of fitting a three-dimensional region by using the second two-dimensional region and the third two-dimensional region according to the present embodiment. As an alternative embodiment, as shown in fig. 5b, a medical image with a region of interest may also be selected on a third field of view and a third two-dimensional region D3 is delineated, and since the second field of view and the third field of view are perpendicular to each other, the second two-dimensional region D2 and the third two-dimensional region D3 should also be perpendicular to each other. The second two-dimensional area D2 and the third two-dimensional area D3 may define surfaces of a cuboid on the second view field and the third view field, and then the opposite sides of the third two-dimensional area D3 are taken as boundaries on the second view field, and a cuboid may be fitted on the third view field with the second two-dimensional area D2 as boundaries, as the three-dimensional area H, where the three-dimensional area H is a smallest cuboid capable of accommodating the second two-dimensional area D2 and the third two-dimensional area D3.
In another embodiment, the first two-dimensional area D1, the second two-dimensional area D2 and the third two-dimensional area D3 may be preset angles, and the preset angles may be determined by the tissue to be segmented according to the need. For example, for an image of the heart, the division of the heart may be determined based on the bottom surface (rear view) and the diaphragm surface (rear view); the pericardial cavity is divided according to left side view and front view; the division of the coronary artery may be determined by the sternal and diaphragmatic surfaces, or the left and right anterior oblique surfaces.
Step S300: and intercepting the data of the medical image in the three-dimensional region and inputting the data into a preset segmentation model to segment out a target region in the region of interest.
Specifically, the three-dimensional region is taken as a boundary to intercept data of a medical image in the three-dimensional region, and the intercepted data of the medical image is input into the segmentation model, so that a target region in the region of interest is segmented. Fig. 6 is a schematic diagram of a target area in a region of interest segmented by using a segmentation model according to the present embodiment. As shown in fig. 6, the target area in the region of interest divided by the division model does not exceed the three-dimensional area, so that a division error can be avoided, the division precision is improved, the workload of modification and correction of a later physical operator is reduced, and misdiagnosis of a patient caused by inaccurate division is avoided. In addition, since the data of the medical image is intercepted, the image input into the segmentation model is smaller, the segmentation time and the operation amount can be reduced, and the segmentation efficiency is improved.
In this embodiment, the segmentation model may be, for example, a pre-trained neural network model, and after the data of the medical image is input into the segmentation model, the segmentation model may automatically segment the target region in the region of interest. The segmentation model may be any of existing segmentation models, and may be, for example, a U-Net model, DEEPMEDIC model, a V-Net model, PSPNet model, deepLab model, or the like.
In this embodiment, since the training sample of the segmentation model is typically a natural tomographic image sequence, its shape is typically rectangular; therefore, a rectangular region of interest is outlined and the segmentation model is input, so that a segmentation result can be obtained more quickly and accurately, and the probability of segmentation failure and error is smaller.
Further, after the region of interest is segmented, a physicist can evaluate the segmentation result of the image, and feedback information generated after evaluation is fed back to the segmentation model for online learning. It should be appreciated that since the deep-learning model is easily over-fitted to the training set, and medical images of different hospitals are often different, the accuracy and universality of the segmentation model can be enhanced by applying an online learning manner.
After the region of interest is segmented, the physical operator may perform post-processing on the segmented image, for example, manually adding, deleting or modifying the delineated region, to obtain a final segmentation result, which is not described herein in detail.
Example two
The difference from the first embodiment is that, when the three-dimensional region characterizing the region of interest is set in the medical image, two medical images where the boundary of the region of interest is located are selected on three fields of view, and the medical images selected on the three fields of view enclose the three-dimensional region.
Fig. 7a is a schematic diagram of two medical images where a boundary of a region of interest is selected on a first field of view according to the present embodiment. As shown in fig. 7a, the physicist views several medical images on the first field of view and selects two medical images of the first field of view where the boundary of the region of interest is located from among the several medical images. That is, of the several medical images on the first field of view, these two medical images are the medical images where the upper and lower boundaries (relative to the human body) of the region of interest are located. And setting the two medical images selected on the first visual field as a top layer and a bottom layer respectively.
Fig. 7b is a schematic diagram of two medical images where the boundary of the region of interest is selected on the second field of view according to the present embodiment. As shown in fig. 7b, the physicist views several medical images on the second field of view and selects two medical images of the second field of view where the border of the region of interest is located from among the several medical images. That is, of the several medical images on the second field of view, these two medical images are the medical images where the front-back boundary (with respect to the human body) of the region of interest is located. The two medical images selected on the second field of view are set as a front boundary and a rear boundary, respectively.
Fig. 7c is a schematic diagram of two medical images where the boundary of the region of interest is selected on the third field of view according to the present embodiment. As shown in fig. 7c, the physicist views several medical images in the third field of view and selects two medical images in which the boundary of the region of interest is located from among the several medical images in the third field of view. That is, of the several medical images on the third field of view, these two medical images are the medical images where the left and right boundaries (with respect to the human body) of the region of interest are located. And setting the two medical images selected on the third visual field as an inner boundary and an outer boundary respectively.
It should be understood that the six selected medical images may be enclosed to form a cuboid as the three-dimensional region, which is equivalent to selecting six surfaces of the three-dimensional region, each surface of the three-dimensional region being a part of one selected medical image.
In this embodiment, the three-dimensional area is enclosed by selecting two medical images on three fields of view, so that compared with the embodiment, a physicist does not need to manually outline a rectangular frame, the workload is smaller, and the efficiency is higher.
Example III
The difference between the first embodiment and the second embodiment is that when the three-dimensional region characterizing the region of interest is set in the medical image, the present embodiment selects a closed region characterizing the region of interest in the medical image by using a tolerance method or an energy field method, and uses a minimum cuboid capable of accommodating the closed region as the three-dimensional region.
Specifically, the region of interest generally has different features compared to other regions, while the features inside the region of interest are consistent, for example, gray levels of pixels or contrast of pixels, and so on, based on this, when a closed region characterizing the region of interest is selected in the medical image by using a tolerance method in this embodiment, the features inside the region of interest are regarded as a whole (different from other regions), for example, a physicist selects one pixel, and all pixels around the selected pixel consistent with the selected pixel features can be automatically drawn as the closed region.
Further, when the structure and energy of the tissue inside the region of interest are different from those of other regions, and a closed region representing the region of interest is selected from the medical image by using an energy field method, the region of interest is used as an energy field, so that the closed region representing the region of interest is selected.
The closed region selected in the medical image by means of tolerance method or energy field method is usually an irregular pattern, and in order to avoid the irregularity of the medical image to be subsequently intercepted, in this embodiment, the smallest square frame capable of accommodating the selected region of interest is taken as the three-dimensional region.
It should be understood that the above-mentioned selection of a pixel by a physical operator is only one way to provide the feature, and the present invention is not limited to providing the feature by selecting a pixel, and the physical operator may also manually input the feature, which is not explained here.
Compared with the first embodiment and the second embodiment, the method is more convenient, the workload of a physicist is smaller, and the efficiency is higher.
Example IV
The difference from the first, second and third embodiments is that, when the three-dimensional region characterizing the region of interest is set in the medical image, the present embodiment firstly selects the medical image having the region of interest on a view, outlines a two-dimensional region, then selects two medical images where the boundary of the region of interest is located on the view, and the boundary of the two-dimensional region and the two selected medical images define the three-dimensional region.
Specifically, as shown in fig. 2a to 2c, the physicist views several medical images in the first field of view, selects one medical image having the region of interest from the several medical images in the first field of view, and draws the first two-dimensional region D1 from the medical images. As shown in fig. 7a, the physicist views the several medical images in the first view, and selects two medical images in which the boundary of the region of interest is located from the several medical images in the first view, and sets the two medical images selected in the first view as a top layer and a bottom layer respectively.
Fig. 8 is a schematic diagram of defining the three-dimensional region by using the boundary of the two-dimensional region and the two selected medical images according to the present embodiment. As shown in fig. 8, the first two-dimensional area D1 is taken as a circumferential boundary, and the two medical images (top layer and bottom layer in fig. 8) selected on the first field of view are taken as upper and lower boundaries, so that the three-dimensional area can be defined.
It should be appreciated that, as an alternative embodiment, the three-dimensional area may be defined by using two medical images of the boundary between the second two-dimensional area D2 and the region of interest selected on the second field of view, or the three-dimensional area may be defined by using two medical images of the boundary between the third two-dimensional area D3 and the region of interest selected on the third field of view, which are not illustrated here.
Example five
The embodiment provides an image segmentation system. Fig. 9 is a block diagram of the image segmentation system according to the present embodiment, and as shown in fig. 9, the image segmentation system includes:
An image providing module 10 for providing a medical image;
a region selection module 20 for setting a three-dimensional region characterizing the region of interest in the medical image; and
The input module 30 is configured to intercept data of the medical image in the three-dimensional region and input the data into a preset segmentation model to segment a target region in the region of interest.
Further, the embodiment also provides an electronic device, which can be used for dividing images. The electronic device includes:
one or more processors;
A memory for storing one or more programs;
The program or programs, when executed by the processor or processors, cause the processor or processors to implement the image segmentation method as proposed in the above embodiment.
In this embodiment, the processor and the memory are both one, and the processor and the memory may be connected by a bus or other means.
The memory is used as a computer readable storage medium for storing a software program, a computer executable program and modules, such as program instructions/modules corresponding to the image segmentation method in the embodiment of the present invention. The processor executes various functional applications and data processing of the electronic device by running the software programs, instructions and modules stored in the memory, that is, implements the above-described image segmentation method.
The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for a function; the storage data area may store data created according to the use of the electronic device, etc. Further, the memory of the image segmentation method may include a high-speed random access memory, and may further include a nonvolatile memory, such as at least one magnetic disk storage device, a flash memory device, or other nonvolatile solid-state storage device. In some examples, the memory may further include memory remotely located with respect to the processor, the remote memory being connectable to the electronic device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The electronic device according to the present embodiment and the image segmentation method according to the foregoing embodiment belong to the same inventive concept, and technical details not described in detail in the present embodiment may be referred to the foregoing embodiment, and the present embodiment has the same beneficial effects as the foregoing embodiment.
The present embodiment also provides a computer-readable storage medium having stored thereon a computer program which, when executed by the processor, implements the image segmentation method as set forth in the above embodiments.
From the above description of embodiments, it will be clear to a person skilled in the art that the present invention may be implemented by means of software and necessary general purpose hardware, but of course also by means of hardware, although in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a FLASH Memory (FLASH), a hard disk, or an optical disk of a computer, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method of the embodiments of the present invention.
In summary, in the segmentation method, system, electronic device and computer readable storage medium provided in the present embodiment, a three-dimensional region representing a region of interest is first set in a medical image, and data of the medical image in the three-dimensional region is intercepted and input into a preset segmentation model to segment a target region in the region of interest. Because the data of the medical image in the three-dimensional region is input into the segmentation model, the target region segmented by the segmentation model cannot exceed the three-dimensional region, so that segmentation errors can be avoided, segmentation precision is improved, workload of modification and correction of a later physical operator is reduced, and misdiagnosis of a patient caused by inaccurate segmentation is avoided; meanwhile, the images input into the segmentation model are smaller, so that the segmentation time and the operation amount can be reduced, and the segmentation efficiency is improved.
It should be noted that, in the present specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment is mainly described in a different point from other embodiments. In particular, for apparatus, electronic devices, computer readable storage medium embodiments, since they are substantially similar to method embodiments, the description is relatively simple, and relevant references are made to the partial description of method embodiments.
The foregoing is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any person skilled in the art will make any equivalent substitution or modification to the technical solution and technical content disclosed in the invention without departing from the scope of the technical solution of the invention, and the technical solution of the invention is not departing from the scope of the invention.
Claims (6)
1. A method of segmenting an image, comprising:
Providing a medical image;
setting a three-dimensional region characterizing a region of interest in the medical image; and
Intercepting data of the medical image in the three-dimensional region and inputting the data into a preset segmentation model to segment a target region in the region of interest;
the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
Selecting a medical image with the region of interest on a view, and outlining a two-dimensional region;
selecting a medical image with the region of interest on another view, and outlining another two-dimensional region; and
Fitting the three-dimensional region by using the two drawn two-dimensional regions;
Or the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
respectively selecting two medical images of the boundary of the region of interest from the three views, and enclosing the medical images selected from the three views into the three-dimensional region;
Or the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
selecting a medical image with the region of interest on a view, and outlining a two-dimensional region; and
And selecting two medical images of the boundary of the region of interest on the view, wherein the boundary of the two-dimensional region and the two selected medical images define the three-dimensional region.
2. The method of segmenting an image according to claim 1, wherein both two-dimensional regions are rectangular, and the three-dimensional region is a cuboid.
3. The image segmentation method according to claim 1 or 2, characterized in that the three-dimensional region is a smallest cuboid accommodating two-dimensional regions.
4. A segmentation system for an image, comprising:
An image providing module for providing a medical image;
The region selection module is used for setting a three-dimensional region representing the region of interest in the medical image; and
The input module is used for intercepting the data of the medical image in the three-dimensional area and inputting the data into a preset segmentation model so as to segment a target area in the region of interest;
the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
Selecting a medical image with the region of interest on a view, and outlining a two-dimensional region;
selecting a medical image with the region of interest on another view, and outlining another two-dimensional region; and
Fitting the three-dimensional region by using the two drawn two-dimensional regions;
Or the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
respectively selecting two medical images of the boundary of the region of interest from the three views, and enclosing the medical images selected from the three views into the three-dimensional region;
Or the step of setting a three-dimensional region characterizing the region of interest in the medical image comprises:
selecting a medical image with the region of interest on a view, and outlining a two-dimensional region; and
And selecting two medical images of the boundary of the region of interest on the view, wherein the boundary of the two-dimensional region and the two selected medical images define the three-dimensional region.
5. An electronic device, the electronic device comprising:
One or more processors; and
A memory for storing one or more programs; and
The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of segmentation of an image as claimed in any one of claims 1-3.
6. A computer-readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements a method of segmentation of an image as claimed in any one of claims 1-3.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011643646.3A CN112766258B (en) | 2020-12-31 | 2020-12-31 | Image segmentation method, system, electronic device and computer readable storage medium |
US17/559,473 US20220207742A1 (en) | 2020-12-30 | 2021-12-22 | Image segmentation method, device, equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011643646.3A CN112766258B (en) | 2020-12-31 | 2020-12-31 | Image segmentation method, system, electronic device and computer readable storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112766258A CN112766258A (en) | 2021-05-07 |
CN112766258B true CN112766258B (en) | 2024-07-02 |
Family
ID=75698721
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011643646.3A Active CN112766258B (en) | 2020-12-30 | 2020-12-31 | Image segmentation method, system, electronic device and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112766258B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107530881A (en) * | 2015-01-29 | 2018-01-02 | Abb瑞士股份有限公司 | 3D for robot application is split |
CN111260642A (en) * | 2020-02-12 | 2020-06-09 | 上海联影医疗科技有限公司 | Image positioning method, device, equipment and storage medium |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100268225A1 (en) * | 2009-04-15 | 2010-10-21 | Tyco Healthcare Group Lp | Methods for Image Analysis and Visualization of Medical Image Data Suitable for Use in Assessing Tissue Ablation and Systems and Methods for Controlling Tissue Ablation Using Same |
KR101727670B1 (en) * | 2015-06-22 | 2017-04-18 | 한국광기술원 | Device and method for medical image segmentation based on user interactive input |
KR101760287B1 (en) * | 2016-02-05 | 2017-07-25 | 한국광기술원 | Device and method for medical image segmentation |
CN110415792B (en) * | 2019-05-31 | 2022-03-25 | 上海联影智能医疗科技有限公司 | Image detection method, image detection device, computer equipment and storage medium |
CN111192356B (en) * | 2019-12-30 | 2023-04-25 | 上海联影智能医疗科技有限公司 | Method, device, equipment and storage medium for displaying region of interest |
-
2020
- 2020-12-31 CN CN202011643646.3A patent/CN112766258B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107530881A (en) * | 2015-01-29 | 2018-01-02 | Abb瑞士股份有限公司 | 3D for robot application is split |
CN111260642A (en) * | 2020-02-12 | 2020-06-09 | 上海联影医疗科技有限公司 | Image positioning method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN112766258A (en) | 2021-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107133996B (en) | Method for generating an attenuation map for PET data reconstruction and PET/CT system | |
CN107909622B (en) | Model generation method, medical imaging scanning planning method and medical imaging system | |
CN109567843B (en) | Imaging scanning automatic positioning method, device, equipment and medium | |
CN110960241A (en) | Method and device for determining scanning parameters of medical image scanning and computer equipment | |
US8655040B2 (en) | Integrated image registration and motion estimation for medical imaging applications | |
US10796498B2 (en) | Image processing apparatus, image processing method, and non-transitory computer-readable medium | |
US10311585B2 (en) | Mechanism for advanced structure generation and editing | |
CN113647967A (en) | Control method, device and system of medical scanning equipment | |
CN111563496A (en) | Continuous learning for automatic view planning for image acquisition | |
CN111568451A (en) | Exposure dose adjusting method and system | |
WO2011163414A2 (en) | Mechanism for advanced structure generation and editing | |
US20240075631A1 (en) | Methods and systems for positioning robots and adjusting postures | |
CN114943714A (en) | Medical image processing system, medical image processing apparatus, electronic device, and storage medium | |
CN111243052A (en) | Image reconstruction method and device, computer equipment and storage medium | |
CN113989231A (en) | Method and device for determining kinetic parameters, computer equipment and storage medium | |
WO2023232068A1 (en) | Systems and methods for image processing | |
CN112365479A (en) | PET parameter image processing method, device, computer equipment and storage medium | |
CN112766258B (en) | Image segmentation method, system, electronic device and computer readable storage medium | |
CN105787922B (en) | A kind of method and apparatus for realizing automatic MPR batch processing | |
CN111968112A (en) | CT three-dimensional positioning image acquisition method and device and computer equipment | |
US9401051B2 (en) | Mechanism for dynamically propagating real-time alterations of medical images | |
CN110473241A (en) | Method for registering images, storage medium and computer equipment | |
CN113538419B (en) | Image processing method and system | |
CN116168097A (en) | Method, device, equipment and medium for constructing CBCT sketching model and sketching CBCT image | |
CN109671131B (en) | Image correction method, device, medical image equipment and storage medium |
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 |