Disclosure of Invention
The invention provides an automatic labeling method and device for abnormal cells, electronic equipment and a computer readable storage medium, and aims to improve the accuracy of abnormal cell labeling and reduce the calculation and storage pressure.
In order to achieve the above object, the present invention provides an automatic labeling method for abnormal cells, comprising:
acquiring cytopathology pictures, carrying out Gaussian convolution smoothing treatment on the cytopathology pictures for preset times, generating a plurality of cytopathology pictures, and obtaining an image pyramid according to the cytopathology pictures;
Detecting and fitting the image pyramid to obtain a low-power fitting region of interest;
Mapping the low-power fitting interested region to the obtained cytopathology picture, performing image amplification operation, generating a fitting high-power image, and obtaining all image coordinates of the interested region in the fitting high-power image;
Dividing the region of interest in the fitting high-power image by using the image coordinates to generate a divided high-power image;
Labeling and cutting the segmented high-power image through a self-adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set;
mapping the abnormal cell labeling set to the cytopathology picture to obtain an abnormal cell labeling picture.
Optionally, the performing gaussian convolution smoothing on the cytopathic pictures for a preset number of times to generate a plurality of cytopathic pictures, and obtaining an image pyramid according to the cytopathic pictures includes:
step A: performing expansion operation on the acquired cytopathology picture to obtain an a-group b-th layer image, wherein initial values of a and b are 1, and performing Gaussian convolution smoothing operation on the a-group b-th layer image by using a Gaussian convolution function to obtain an a-group b+1-th layer image, wherein the Gaussian convolution function is as follows:
Wherein sigma represents a smoothing factor, G(x,y,σ) represents the convolution of x and y, and x and y represent image coordinates;
And (B) step (B): multiplying the smoothing factor sigma by a scaling factor k to obtain a new smoothing factor k sigma, performing Gaussian convolution smoothing operation on the a-group b+1th layer image by using the Gaussian convolution function through the new smoothing factor k sigma to obtain the a-group b+2th layer image, and repeating the steps until the a-group L layer image is obtained, wherein L is a predefined value;
Step C: performing downsampling operation on the (a+1) th group of layer L images to obtain (a+1) th group of layer B images, performing Gaussian convolution smoothing operation on the (a+1) th group of layer B images by using the Gaussian convolution function to obtain (a+1) th group of layer b+1 images, and repeating the step B until the (a+1) th group of layer L images are obtained;
Step D: performing the cyclic operation of the step C and the step B on the result obtained in the step C until an O group L-th layer image is obtained, wherein O is a predefined value;
Step E: and combining the b-th layer images from the a-th group to the L-th layer images from the O-th group to generate the image pyramid.
Optionally, detecting and fitting the image pyramid by a hough transform circle detection method to obtain the low-power fitting region of interest, including:
Performing gray conversion on the image pyramid to generate a gray image;
performing binarization processing on the gray level image to obtain a contour image;
a random selection method is adopted from the contour image, so that a candidate region is obtained;
And identifying the region of interest in the candidate region by using a Hough circle transformation detection method, and fitting the region of interest to obtain the low-power fitting region of interest.
Optionally, the identifying the region of interest in the candidate region by using a hough circle transformation detection method, fitting the region of interest to obtain the low-power fitting region of interest, including:
Detecting the edges of the candidate region in the image space of the candidate region by an edge detection algorithm to obtain n edge pixel point sets;
mapping the n sets of edge pixel points to a parameter space with a predefined value r as a radius:
wherein r is a predefined value, θ∈ [0,2π), x and y represent coordinates (x, y) corresponding to the n sets of edge pixel points, and (a, b) represent coordinates of reference points in the parameter space;
Counting all coordinate points in the parameter space, traversing theta, and mapping the edge pixel points on the image space to a circle when the parameter space is the circle, wherein the circle is used as a Hough circle, and the region of interest is formed by the Hough circle;
fitting the interested region to generate a low-power fitting interested region.
Optionally, labeling and cutting the segmented high-power image by an adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set, which includes:
passing each pixel i of the segmented high-power image through the formula:
Converting into a real number between 0 and 1 to obtain a normalized value h;
Pre-compensating the normalized value h by using a predefined gamma correction compensation value to obtain a pre-compensation constant value f;
performing inverse normalization calculation on the pre-compensation constant value f through a formula f of 256-0.5 to obtain a corrected image of the segmented high-power image;
Forming a binary group by the image gray level of the correction image and the gray level of a pixel point of a coordinate in the correction image, calculating the mean value and variance of all the binary groups, establishing a two-dimensional maximum inter-class variance model through the mean value and the variance, calculating the two-dimensional maximum inter-class variance model through a self-adaptive particle clustering algorithm, and generating an optimal threshold value of the correction image;
dividing the corrected image by utilizing the optimal threshold value to generate a background and foreground divided image;
and performing open operation and close operation processing on the background and foreground segmented image to generate an abnormal cell labeling set.
In order to solve the above problems, the present invention also provides an automatic labeling device for abnormal cells, the device comprising:
The image pyramid generation module is used for carrying out Gaussian convolution smoothing processing on the cytopathology pictures for preset times to generate a plurality of cytopathology pictures, and obtaining an image pyramid according to the cytopathology pictures;
the detection fitting module is used for detecting and fitting the image pyramid by a Hough transformation circle detection method to obtain a low-power fitting region of interest;
The mapping and segmentation module is used for mapping the low-power fitting region of interest to the acquired cytopathology picture and performing image amplification operation to generate a fitting high-power image, acquiring all image coordinates of the region of interest in the fitting high-power image, and segmenting the region of interest in the fitting high-power image by utilizing the image coordinates to generate a segmented high-power image;
the labeling cutting module is used for labeling and cutting the segmented high-power image through a self-adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set;
and the abnormal generation module is used for mapping the abnormal cell labeling set to the cytopathology picture to obtain an abnormal cell labeling picture.
Preferably, the performing gaussian convolution smoothing on the cytopathic pictures for a preset number of times to generate a plurality of cytopathic pictures, and obtaining an image pyramid according to the cytopathic pictures includes:
step A: performing expansion operation on the acquired cytopathology picture to obtain an a-group b-th layer image, wherein initial values of a and b are 1, and performing Gaussian convolution smoothing operation on the a-group b-th layer image by using a Gaussian convolution function to obtain an a-group b+1-th layer image, wherein the Gaussian convolution function is as follows:
Wherein sigma represents a smoothing factor, G(x,y,σ) represents the convolution of x and y, and x and y represent image coordinates;
And (B) step (B): multiplying the smoothing factor sigma by a scaling factor k to obtain a new smoothing factor k sigma, performing Gaussian convolution smoothing operation on the a-group b+1th layer image by using the Gaussian convolution function through the new smoothing factor k sigma to obtain the a-group b+2th layer image, and repeating the steps until the a-group L layer image is obtained, wherein L is a predefined value;
Step C: performing downsampling operation on the (a+1) th group of layer L images to obtain (a+1) th group of layer B images, performing Gaussian convolution smoothing operation on the (a+1) th group of layer B images by using the Gaussian convolution function to obtain (a+1) th group of layer b+1 images, and repeating the step B until the (a+1) th group of layer L images are obtained;
Step D: performing the cyclic operation of the step C and the step B on the result obtained in the step C until an O group L-th layer image is obtained, wherein O is a predefined value;
Step E: and combining the b-th layer images from the a-th group to the L-th layer images from the O-th group to generate the image pyramid.
Preferably, labeling and cutting the segmented high-power image by an adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set, which comprises the following steps:
passing each pixel i of the segmented high-power image through the formula:
Converting into a real number between 0 and 1 to obtain a normalized value h;
Pre-compensating the normalized value h by using a predefined gamma correction compensation value to obtain a pre-compensation constant value f;
performing inverse normalization calculation on the pre-compensation constant value f through a formula f of 256-0.5 to obtain a corrected image of the segmented high-power image;
Forming a binary group by the image gray level of the correction image and the gray level of a pixel point of a coordinate in the correction image, calculating the mean value and variance of all the binary groups, establishing a two-dimensional maximum inter-class variance model through the mean value and the variance, calculating the two-dimensional maximum inter-class variance model through a self-adaptive particle clustering algorithm, and generating an optimal threshold value of the correction image;
dividing the corrected image by utilizing the optimal threshold value to generate a background and foreground divided image;
and performing open operation and close operation processing on the background and foreground segmented image to generate an abnormal cell labeling set.
In order to solve the above-mentioned problems, the present invention also provides an electronic apparatus including:
a memory storing at least one instruction; and
And the processor executes the instructions stored in the memory to realize the automatic labeling method of the abnormal cells.
In order to solve the above-mentioned problems, the present invention also provides a computer-readable storage medium having stored therein at least one instruction that is executed by a processor in an electronic device to implement the above-mentioned abnormal cell automatic labeling method.
According to the embodiment of the invention, the original cell pathology picture is subjected to Gaussian convolution smoothing treatment to obtain the image pyramid composed of a plurality of cell case pictures with higher resolution from top to bottom, the pathology data features in the cell case pictures are more obvious, and the accuracy of abnormal cell labeling is improved; further, as the resolution of the image pixels in the image pyramid is large, the general image processing method cannot directly use a computer to perform image analysis processing, so that the embodiment of the invention fits the pixels to a low-power area, reduces the pixels and the resolution of the image, and further releases the storage pressure of the computer while not affecting the detection accuracy of subsequent abnormal cells; in addition, after the low-power fitting interested region is obtained, the embodiment of the invention is further changed into fitting the high-power image through the amplifying operation, the resolution of the high-power image is high, and the accuracy of marking and cutting by using an algorithm can be effectively improved. Therefore, the automatic labeling method, the automatic labeling device, the electronic equipment and the computer readable storage medium for the abnormal cells can solve the problems of low accuracy of abnormal cell labeling and high calculation and storage pressure.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the following detailed description of the embodiments of the present application will be given with reference to the accompanying drawings. However, those of ordinary skill in the art will understand that in various embodiments of the present application, numerous technical details have been set forth in order to provide a better understanding of the present application. The claimed application may be practiced without these specific details and with various changes and modifications based on the following embodiments.
The method and the device can be applied to the field of intelligent medical treatment so as to promote construction of intelligent cities, and the method and the device aim at marking abnormal cells through self-adaptive threshold segmentation, are used for improving accuracy of abnormal cell marking and reducing calculation and storage pressure, for example, in a cancer cell screening process, data characteristics after layering are more obvious through layering of cytopathology pictures, marking the abnormal cells in the layered cytopathology pictures, mapping the abnormal cells in low-pixel images in cervical cell pathology pictures into high-pixel images through a coordinate mapping method, and are used for subsequent abnormal cell marking segmentation.
The implementation details of the automatic labeling method for abnormal cells according to the present embodiment are specifically described below, and the following description is provided only for convenience of understanding, and is not necessary to implement the present embodiment.
Referring to fig. 1, a flow chart of an automatic labeling method for abnormal cells according to an embodiment of the invention is shown. The method may be performed by an apparatus, which may be implemented in software and/or hardware.
In this embodiment, the method for automatically labeling abnormal cells includes:
s1, obtaining cytopathology pictures, carrying out Gaussian convolution smoothing treatment on the cytopathology pictures for preset times, generating a plurality of cytopathology pictures, and obtaining an image pyramid according to the cytopathology pictures.
For example, in the field of cervical cancer screening, the cytopathology image according to embodiments of the present invention includes a cervical cytopathology image.
In detail, the performing gaussian convolution smoothing on the cytopathic pictures for a preset number of times to generate a plurality of cytopathic pictures, and obtaining an image pyramid according to the cytopathic pictures includes:
step A: performing expansion operation on the acquired cytopathology picture to obtain an a-group b-th layer image, wherein initial values of a and b are 1, and performing Gaussian convolution smoothing operation on the a-group b-th layer image by using a Gaussian convolution function to obtain an a-group b+1-th layer image, wherein the Gaussian convolution function is as follows:
Wherein sigma represents a smoothing factor, G(x,y,σ) represents the convolution of x and y, and x and y represent image coordinates;
And (B) step (B): multiplying the smoothing factor sigma by a scaling factor k to obtain a new smoothing factor k sigma, performing Gaussian convolution smoothing operation on the a-group b+1th layer image by using the Gaussian convolution function through the new smoothing factor k sigma to obtain the a-group b+2th layer image, and repeating the steps until the a-group L layer image is obtained, wherein L is a predefined value;
Step C: performing downsampling operation on the (a+1) th group of layer L images to obtain (a+1) th group of layer B images, performing Gaussian convolution smoothing operation on the (a+1) th group of layer B images by using the Gaussian convolution function to obtain (a+1) th group of layer b+1 images, and repeating the step B until the (a+1) th group of layer L images are obtained;
Step D: performing the cyclic operation of the step C and the step B on the result obtained in the step C until an O group L-th layer image is obtained, wherein O is a predefined value;
Step E: and combining the b-th layer images from the a-th group to the L-th layer images from the O-th group to generate the image pyramid. Wherein each of the a-th to O-th layer images is a picture layer of the image pyramid.
After Gaussian convolution smoothing treatment, the cytopathology picture can be split into a plurality of cytopathology pictures with higher and higher resolution from top to bottom, so that the pathological data features are more obvious, and the accuracy of abnormal cell labeling is improved.
S2, detecting and fitting the image pyramid through a Hough transformation circle detection method to obtain a low-power fitting region of interest.
In detail, the S2 includes: performing gray conversion on the image pyramid to generate a gray image; performing binarization processing on the gray level image to obtain a contour image; a random selection method is adopted from the contour image, so that a candidate region is obtained; and identifying the region of interest in the candidate region by using a Hough circle transformation detection method, and fitting the region of interest to obtain the low-power fitting region of interest.
Further, the identifying the region of interest in the candidate region by using the hough circle transformation detection method, fitting the region of interest to obtain the low-power fitting region of interest, including:
step a, detecting the edges of the candidate areas in the image space of the candidate areas through an edge detection algorithm to obtain n edge pixel point sets;
step b, mapping the n edge pixel point sets to a parameter space by taking a predefined value r as a radius:
wherein r is a predefined value, θ∈ [0,2π), x and y represent coordinates (x, y) corresponding to the n sets of edge pixel points, and (a, b) represent coordinates of reference points in the parameter space;
c, counting all coordinate points in the parameter space, traversing theta, and when the edge pixel points on the image space are mapped to a circle in the parameter space, taking the circle as a Hough circle, and forming the region of interest through the Hough circle;
And d, fitting the region of interest to generate a low-power fitting region of interest.
According to the embodiment of the invention, the region of interest of the cytopathology picture can be identified and fitted through a Hough circle transformation detection method by marking the region at will by a doctor.
And S3, mapping the low-power fitting interested region to the acquired cytopathology picture, performing image amplification operation, generating a fitting high-power image, and acquiring all image coordinates of the interested region in the fitting high-power image.
In the embodiment of the invention, the high-power image size of the cytopathology picture is large, and the general image processing method is difficult to load for direct processing and can not directly use a computer for image analysis processing, so that the embodiment of the invention generates a fitting high-power image by mapping the low-power fitting region of interest onto the acquired high-power image, and acquires the coordinate information of the region of interest image in the fitting high-power image, wherein the coordinate information of the region of interest image is used for dividing the high-power image subsequently.
In the embodiment of the invention, the mapping is to divide a low-resolution image into a plurality of subareas, multiply the subarea coordinates with a multiple, and add each subarea to a high-resolution image, wherein the multiple is the magnification of the high-resolution image relative to the low-resolution image.
S4, segmenting out the region of interest in the fitting high-power image by utilizing the image coordinates, and generating a segmented high-power image.
The invention can segment the cell region of interest by fitting all the image coordinates of the region of interest, and generate a segmented high-power image with a size of 3000x3000, for example.
And S5, labeling and cutting the segmented high-power image through a self-adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set.
In detail, the S5 includes:
step I, passing each pixel i of the segmented high-power image through the formula:
Converting into a real number between 0 and 1 to obtain a normalized value h;
Step II, pre-compensating the normalized value h by utilizing a predefined gamma correction compensation value to obtain a pre-compensating constant value f;
step III, performing inverse normalization calculation on the pre-compensation constant value f through a formula f of 256-0.5 to obtain a corrected image of the segmented high-power image;
Step IV, forming a binary group by the image gray level of the corrected image and the gray level of the pixel point of the coordinates in the corrected image, calculating the mean value and variance of all the binary groups, establishing a two-dimensional maximum inter-class variance model through the mean value and the variance, calculating the two-dimensional maximum inter-class variance model through a self-adaptive particle clustering algorithm, and generating an optimal threshold value of the corrected image;
v, dividing the corrected image by utilizing the optimal threshold value to generate a background and foreground divided image;
And step VI, carrying out open operation and close operation treatment on the background and foreground segmented image to generate an abnormal cell labeling set.
In the embodiment of the invention, open operation calculation is sequentially performed on the background and foreground segmented image through a morphological algorithm, isolated small points in the segmented image are removed, a first image set is generated, the first image set is sequentially performed on the closed operation calculation through the morphological algorithm, small cracks among image cells in the first image set are filled, cavities in the first image set are removed, and therefore foreground cells in the first image set are separated from background areas in the first image set, and an abnormal cell labeling set is generated.
Conventional cell image segmentation methods are broadly divided into two categories: the basic principle of the region-based segmentation method is to divide adjacent regions with similar characteristics into one class, the region-based segmentation method generally divides the region by taking the place with abrupt change of gray level or structure as an edge, and the image is divided by adopting the edge-based segmentation method through an adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set.
And S6, mapping the abnormal cell labeling set onto the cytopathology picture to obtain an abnormal cell labeling picture.
In the embodiment of the invention, the abnormal cell labeling set is used for carrying out positioning segmentation identification on target cells by a deep learning method, generating an abnormal cell labeling graph, and carrying out edge adjustment by adopting an edge smoothing method if the edges of the abnormal cell labeling graph generate saw teeth.
It should be emphasized that, to further ensure the privacy and security of the abnormal cell annotation set, the abnormal cell annotation set may also be stored in a node of a blockchain.
FIG. 2 is a functional block diagram showing an automatic labeling device for abnormal cells according to the present invention.
The automatic labeling device 100 for abnormal cells according to the embodiment of the invention can be installed in electronic equipment. The automatic labeling device for abnormal cells can comprise a detection fitting module 101, a mapping and dividing module 102, a labeling and cutting module 103 and an abnormality generating module 104 according to the implemented functions. The module of the present invention may also be referred to as a unit, meaning a series of computer program segments capable of being executed by the processor of the electronic device and of performing fixed functions, stored in the memory of the electronic device.
In the present embodiment, the functions concerning the respective modules/units are as follows:
The image pyramid generation module 101 is configured to obtain a cytopathic picture, perform gaussian convolution smoothing on the cytopathic picture for a preset number of times, generate a plurality of cytopathic pictures, and obtain an image pyramid according to the cytopathic picture;
The detection fitting module 102 is configured to detect and fit the image pyramid by using a hough transform circle detection method, so as to obtain a low-power fitting region of interest;
The mapping and segmentation module 103 is configured to map the low-power fitting region of interest onto the obtained cytopathology image and perform image amplification operation, generate a fitting high-power image, obtain all image coordinates of the region of interest in the fitting high-power image, and segment the region of interest in the fitting high-power image by using the image coordinates to generate a segmented high-power image;
The labeling cutting module 104 is configured to label and cut the segmented high-power image through an adaptive threshold segmentation algorithm, so as to obtain an abnormal cell labeling set;
The anomaly generation module 105 is configured to map the anomaly cell label set onto the cytopathology image, so as to obtain an anomaly cell label image.
In detail, the specific implementation steps of each module of the automatic abnormal cell labeling device are as follows:
The image pyramid generation module 101 acquires cytopathology pictures, performs Gaussian convolution smoothing processing on the cytopathology pictures for preset times, generates a plurality of cytopathology pictures, and obtains an image pyramid according to the cytopathology pictures.
For example, in the field of cervical cancer screening, the cytopathology image according to embodiments of the present invention includes a cervical cytopathology image.
In detail, the performing gaussian convolution smoothing on the cytopathic pictures for a preset number of times to generate a plurality of cytopathic pictures, and obtaining an image pyramid according to the cytopathic pictures includes:
step A: performing expansion operation on the acquired cytopathology picture to obtain an a-group b-th layer image, wherein initial values of a and b are 1, and performing Gaussian convolution smoothing operation on the a-group b-th layer image by using a Gaussian convolution function to obtain an a-group b+1-th layer image, wherein the Gaussian convolution function is as follows:
Wherein sigma represents a smoothing factor, G(x,y,σ) represents the convolution of x and y, and x and y represent image coordinates;
And (B) step (B): multiplying the smoothing factor sigma by a scaling factor k to obtain a new smoothing factor k sigma, performing Gaussian convolution smoothing operation on the a-group b+1th layer image by using the Gaussian convolution function through the new smoothing factor k sigma to obtain the a-group b+2th layer image, and repeating the steps until the a-group L layer image is obtained, wherein L is a predefined value;
Step C: performing downsampling operation on the (a+1) th group of layer L images to obtain (a+1) th group of layer B images, performing Gaussian convolution smoothing operation on the (a+1) th group of layer B images by using the Gaussian convolution function to obtain (a+1) th group of layer b+1 images, and repeating the step B until the (a+1) th group of layer L images are obtained;
Step D: performing the cyclic operation of the step C and the step B on the result obtained in the step C until an O group L-th layer image is obtained, wherein O is a predefined value;
Step E: and combining the b-th layer images from the a-th group to the L-th layer images from the O-th group to generate the image pyramid. Wherein each of the a-th to O-th layer images is a picture layer of the image pyramid.
After Gaussian convolution smoothing treatment, the cytopathology picture can be split into a plurality of cytopathology pictures with higher and higher resolution from top to bottom, so that the pathological data features are more obvious, and the accuracy of abnormal cell labeling is improved.
The detection fitting module 102 detects and fits the image pyramid by a hough transform circle detection method to obtain a low-power fitting region of interest.
In detail, the detection fitting module 102: performing gray conversion on the image pyramid to generate a gray image; performing binarization processing on the gray level image to obtain a contour image; a random selection method is adopted from the contour image, so that a candidate region is obtained; and identifying the region of interest in the candidate region by using a Hough circle transformation detection method, and fitting the region of interest to obtain the low-power fitting region of interest.
Further, the identifying the region of interest in the candidate region by using the hough circle transformation detection method, fitting the region of interest to obtain the low-power fitting region of interest, including:
step a, detecting the edges of the candidate areas in the image space of the candidate areas through an edge detection algorithm to obtain n edge pixel point sets;
step b, mapping the n edge pixel point sets to a parameter space by taking a predefined value r as a radius:
wherein r is a predefined value, θ∈ [0,2π), x and y represent coordinates (x, y) corresponding to the n sets of edge pixel points, and (a, b) represent coordinates of reference points in the parameter space;
c, counting all coordinate points in the parameter space, traversing theta, and when the edge pixel points on the image space are mapped to a circle in the parameter space, taking the circle as a Hough circle, and forming the region of interest through the Hough circle;
And d, fitting the region of interest to generate a low-power fitting region of interest.
According to the embodiment of the invention, the region of interest of the cytopathology picture can be identified and fitted through a Hough circle transformation detection method by marking the region at will by a doctor.
The mapping and segmentation module 103 maps the low-power fitting interested region to the obtained cytopathology picture and performs image amplification operation, generates a fitting high-power image, and obtains all image coordinates of the interested region in the fitting high-power image.
In the embodiment of the invention, the high-power image size of the cytopathology picture is large, and the general image processing method is difficult to load for direct processing and can not directly use a computer for image analysis processing, so that the embodiment of the invention generates a fitting high-power image by mapping the low-power fitting region of interest onto the acquired high-power image, and acquires the coordinate information of the region of interest image in the fitting high-power image, wherein the coordinate information of the region of interest image is used for dividing the high-power image subsequently.
In the embodiment of the invention, the mapping is to divide a low-resolution image into a plurality of subareas, multiply the subarea coordinates with a multiple, and add each subarea to a high-resolution image, wherein the multiple is the magnification of the high-resolution image relative to the low-resolution image. Further, the mapping segmentation module 103 segments out the region of interest in the fitted high-power image by using the image coordinates, and generates a segmented high-power image.
The invention can segment the cell region of interest by fitting all the image coordinates of the region of interest, and generate a segmented high-power image with a size of 3000x3000, for example.
The labeling and cutting module 104 performs labeling and cutting on the segmented high-power image through a self-adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set.
In detail, the labeling and cutting the segmented high-power image to obtain an abnormal cell labeling set includes:
step I, passing each pixel i of the segmented high-power image through the formula:
Converting into a real number between 0 and 1 to obtain a normalized value h;
Step II, pre-compensating the normalized value h by utilizing a predefined gamma correction compensation value to obtain a pre-compensating constant value f;
step III, performing inverse normalization calculation on the pre-compensation constant value f through a formula f of 256-0.5 to obtain a corrected image of the segmented high-power image;
Step IV, forming a binary group by the image gray level of the corrected image and the gray level of the pixel point of the coordinates in the corrected image, calculating the mean value and variance of all the binary groups, establishing a two-dimensional maximum inter-class variance model through the mean value and the variance, calculating the two-dimensional maximum inter-class variance model through a self-adaptive particle clustering algorithm, and generating an optimal threshold value of the corrected image;
v, dividing the corrected image by utilizing the optimal threshold value to generate a background and foreground divided image;
And step VI, carrying out open operation and close operation treatment on the background and foreground segmented image to generate an abnormal cell labeling set.
In the embodiment of the invention, open operation calculation is sequentially performed on the background and foreground segmented image through a morphological algorithm, isolated small points in the segmented image are removed, a first image set is generated, the first image set is sequentially performed on the closed operation calculation through the morphological algorithm, small cracks among image cells in the first image set are filled, cavities in the first image set are removed, and therefore foreground cells in the first image set are separated from background areas in the first image set, and an abnormal cell labeling set is generated.
Conventional cell image segmentation methods are broadly divided into two categories: the basic principle of the region-based segmentation method is to divide adjacent regions with similar characteristics into one class, the region-based segmentation method generally divides the region by taking the place with abrupt change of gray level or structure as an edge, and the image is divided by adopting the edge-based segmentation method through an adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set.
The anomaly generation module 105 maps the anomaly cell label set onto the cytopathology image to obtain an anomaly cell label image.
In the embodiment of the invention, the abnormal cell labeling set is used for carrying out positioning segmentation identification on target cells by a deep learning method, generating an abnormal cell labeling graph, and carrying out edge adjustment by adopting an edge smoothing method if the edges of the abnormal cell labeling graph generate saw teeth.
It should be emphasized that, to further ensure the privacy and security of the abnormal cell annotation set, the abnormal cell annotation set may also be stored in a node of a blockchain.
FIG. 3 is a schematic diagram of an electronic device for implementing the automatic labeling method of abnormal cells according to the present invention.
The electronic device 1 may comprise a processor 10, a memory 11 and a bus, and may further comprise a computer program, such as an abnormal cell automatic labeling program 12, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may in other embodiments also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD) or the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used not only for storing application software installed in the electronic device 1 and various data such as a code of an abnormal cell automatic labeling program or the like, but also for temporarily storing data that has been output or is to be output.
The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various parts of the entire electronic device using various interfaces and lines, executes or executes programs or modules (for example, executes an abnormal cell automatic labeling program or the like) stored in the memory 11, and invokes data stored in the memory 11 to perform various functions of the electronic device 1 and process data.
The bus may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
Fig. 3 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 1 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.
Further, the electronic device 1 may also comprise a network interface, optionally the network interface may comprise a wired interface and/or a wireless interface (e.g. WI-FI interface, bluetooth interface, etc.), typically used for establishing a communication connection between the electronic device 1 and other electronic devices.
The electronic device 1 may optionally further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 1 and for displaying a visual user interface.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
The automatic labeling program 12 for abnormal cells stored in the memory 11 of the electronic device 1 is a combination of a plurality of instructions, and when running in the processor 10, it can be implemented:
acquiring cytopathology pictures, carrying out Gaussian convolution smoothing treatment on the cytopathology pictures for preset times, generating a plurality of cytopathology pictures, and obtaining an image pyramid according to the cytopathology pictures;
detecting and fitting the image pyramid by a Hough transform circle detection method to obtain a low-power fitting region of interest;
Mapping the low-power fitting interested region to the obtained cytopathology picture, performing image amplification operation, generating a fitting high-power image, and obtaining all image coordinates of the interested region in the fitting high-power image;
Dividing the region of interest in the fitting high-power image by using the image coordinates to generate a divided high-power image;
Labeling and cutting the segmented high-power image through a self-adaptive threshold segmentation algorithm to obtain an abnormal cell labeling set;
mapping the abnormal cell labeling set to the cytopathology picture to obtain an abnormal cell labeling picture.
Specifically, the specific implementation method of the above instructions by the processor 10 may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein.
Further, the integrated modules/units of the electronic device 1 may be stored in a non-volatile computer readable storage medium if implemented in the form of software functional units and sold or used as a stand alone product. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
Further, the computer-usable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created from the use of blockchain nodes, and the like.
The blockchain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, encryption algorithm and the like. The blockchain (Blockchain), essentially a de-centralized database, is a string of data blocks that are generated in association using cryptographic methods, each of which contains information from a batch of network transactions for verifying the validity (anti-counterfeit) of its information and generating the next block. The blockchain may include a blockchain underlying platform, a platform product services layer, an application services layer, and the like.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. A plurality of units or means recited in the system claims can also be implemented by means of software or hardware by means of one unit or means. The terms second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.