CN111751266A - Mark-free cell nucleus scattering inversion method based on multi-scale gray level co-occurrence matrix - Google Patents
Mark-free cell nucleus scattering inversion method based on multi-scale gray level co-occurrence matrix Download PDFInfo
- Publication number
- CN111751266A CN111751266A CN202010663410.XA CN202010663410A CN111751266A CN 111751266 A CN111751266 A CN 111751266A CN 202010663410 A CN202010663410 A CN 202010663410A CN 111751266 A CN111751266 A CN 111751266A
- Authority
- CN
- China
- Prior art keywords
- cell nucleus
- cell
- image
- scattering
- volume
- 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.)
- Pending
Links
- 210000003855 cell nucleus Anatomy 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000011159 matrix material Substances 0.000 title claims abstract description 33
- 210000004027 cell Anatomy 0.000 claims abstract description 70
- 238000001914 filtration Methods 0.000 claims abstract description 17
- 238000013178 mathematical model Methods 0.000 claims abstract description 7
- 238000010586 diagram Methods 0.000 claims description 8
- 210000004940 nucleus Anatomy 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 7
- 206010028980 Neoplasm Diseases 0.000 claims description 6
- 201000011510 cancer Diseases 0.000 claims description 6
- 210000004881 tumor cell Anatomy 0.000 claims description 6
- 238000012886 linear function Methods 0.000 claims description 5
- 230000003287 optical effect Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 230000003321 amplification Effects 0.000 claims 1
- 238000003199 nucleic acid amplification method Methods 0.000 claims 1
- 230000000717 retained effect Effects 0.000 claims 1
- 238000000354 decomposition reaction Methods 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 11
- 210000000265 leukocyte Anatomy 0.000 description 10
- 239000013307 optical fiber Substances 0.000 description 7
- 239000011521 glass Substances 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 238000000149 argon plasma sintering Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000000523 sample Substances 0.000 description 4
- WSFSSNUMVMOOMR-UHFFFAOYSA-N Formaldehyde Chemical compound O=C WSFSSNUMVMOOMR-UHFFFAOYSA-N 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000001413 cellular effect Effects 0.000 description 3
- 201000010099 disease Diseases 0.000 description 3
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 238000006073 displacement reaction Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000007935 neutral effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 239000006285 cell suspension Substances 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 210000003743 erythrocyte Anatomy 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 210000003463 organelle Anatomy 0.000 description 2
- 231100000915 pathological change Toxicity 0.000 description 2
- 230000036285 pathological change Effects 0.000 description 2
- 210000005259 peripheral blood Anatomy 0.000 description 2
- 239000011886 peripheral blood Substances 0.000 description 2
- 230000000704 physical effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 208000005443 Circulating Neoplastic Cells Diseases 0.000 description 1
- 206010035226 Plasma cell myeloma Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 210000001772 blood platelet Anatomy 0.000 description 1
- 230000024245 cell differentiation Effects 0.000 description 1
- 230000010261 cell growth Effects 0.000 description 1
- 230000004663 cell proliferation Effects 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 210000000805 cytoplasm Anatomy 0.000 description 1
- 230000001086 cytosolic effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000007850 fluorescent dye Substances 0.000 description 1
- 238000001215 fluorescent labelling Methods 0.000 description 1
- 230000012010 growth Effects 0.000 description 1
- 238000009169 immunotherapy Methods 0.000 description 1
- 230000002779 inactivation Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 230000005499 meniscus Effects 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 201000000050 myeloid neoplasm Diseases 0.000 description 1
- 230000008689 nuclear function Effects 0.000 description 1
- 210000004976 peripheral blood cell Anatomy 0.000 description 1
- 230000035755 proliferation Effects 0.000 description 1
- 238000000763 radioactive isotope labeling Methods 0.000 description 1
- 238000000790 scattering method Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 210000002966 serum Anatomy 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Optical investigation techniques, e.g. flow cytometry
- G01N15/1429—Signal processing
- G01N15/1433—Signal processing using image recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/148—Wavelet transforms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N2015/1006—Investigating individual particles for cytology
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N15/14—Optical investigation techniques, e.g. flow cytometry
- G01N2015/1493—Particle size
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Dispersion Chemistry (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a label-free cell nucleus scattering inversion method based on a multi-scale gray level co-occurrence matrix, which comprises the steps of collecting a side scattering image of a cell, carrying out Gabor filtering decomposition on the image to obtain a multi-scale multi-direction filtering image, calculating the gray level co-occurrence matrix of the Gabor filtering image to obtain an entropy value, constructing mathematical models between the size and the entropy value of the cell nucleus in different scales and different directions, selecting the mathematical model in the optimal scale and direction, and inverting the cell nucleus volume through the entropy values of the mathematical model and the side scattering image. The invention is based on the lateral two-dimensional scattering image of the cell, inverts the quantitative rule between the size of the cell nucleus volume and the lateral scattering light thereof, more simply and accurately searches the quantitative rule between the size of the cell nucleus volume and the lateral scattering light thereof, and realizes more accurate and faster inversion of the cell nucleus volume.
Description
Technical Field
The invention belongs to the field of unmarked cell scattering detection research, and relates to an unmarked cell nucleus scattering inversion method based on a multi-scale gray level co-occurrence matrix.
Background
The nucleus is the control pivot of cell life activity, and the structural nuclear function of the nucleus is damaged, which causes serious consequences, often causes the abnormality of cell growth, proliferation, differentiation and the like, and then causes the generation of diseases. Compared with normal cells, tumor cells and cancer cells have the advantages of vigorous proliferation and growth, active metabolic activity and a plurality of abnormal morphological structures of cell nuclei. In cancer cells and tumor cells, the nucleus is usually larger and the nuclear to cytoplasmic ratio is increased. The shape of the nucleus is represented by: elongation, jagged edges, dimpling, long bud, lobulation, meniscus, and the like. In myeloma cells, even binuclear cells are present which divide only the nucleus but not the cytoplasm. Therefore, the size of the cell nucleus is in important connection with the pathological changes of the cell to the cancer cell and the tumor cell, and if the trend that the size of the cell nucleus in the cell is increased and the nuclear-to-cytoplasmic ratio is increased can be detected in time, whether the cell nucleus is changed into the tumor cell or the cancer cell can be diagnosed in time, so that the patient can be diagnosed and treated in time.
Cellular light scattering is an important non-invasive means of detection. Light scattering results from the interaction of electromagnetic waves with a medium, and the scattered waves carry a great deal of information about the properties of the medium. The optical scattering information of the cell not only contains the common characteristics of the cells of the same kind, but also contains the individual characteristics of the cell, and can accurately reflect the physical characteristics of the biological cell in a non-intervention state, so that the cell is known as 'cell fingerprint' information.
The method of detecting human peripheral blood cells (circulating tumor cells, white blood cells, red blood cells, etc.) by using fluorescence labeling, radioactive isotope labeling and other methods is a widely used detection method in clinical diagnosis and biological research. The cellular immune marker detection technology is an intervention indirect detection technology based on biochemical means, and has the problems of biotoxicity and inactivation hazard, complex operation process, expensive target probe, serious environmental pollution and the like. The method for realizing non-contact, label-free and non-intervention measurement of cells by using the optical scattering method is undoubtedly a more direct, more effective and more green medical diagnosis technology, and has important research significance and potential clinical application value in the fields of cell canceration state tracking, cellular immunotherapy, clinical accurate medical treatment and the like.
The study of cell light scattering shows that the volume size of cells and cell nuclei has a close relationship with the corresponding two-dimensional scattering, and in order to explore the close relationship, researchers propose a series of indexes to analyze the acquired two-dimensional scattering image, such as spot size and spot area distribution, average light intensity, gray level co-occurrence matrix and other methods in the two-dimensional scattering image. The above series of methods can be used for analyzing the relation of the cell volume change when the two-dimensional scattering image has large change due to the cell volume change. However, in the case of pathological changes of cancer cells and tumor cells, the size of the cells is not changed much, but the volume of the cell nucleus is changed drastically. According to the experimental conclusion, when the cell volume is not changed, the influence of the change of the cell nucleus volume in the cell on the two-dimensional scattering image is very small, so that the traditional classification method is difficult to achieve high resolution precision.
In the aspect of inversion research of two-dimensional scattering images and biological cell features, FengYuan Ming et al propose a method for calculating a gray level co-occurrence matrix of the two-dimensional scattering images so as to extract a series of texture feature values of the two-dimensional scattering images, and train a Support Vector Machine (SVM) by using the obtained feature values so as to realize three types of cell classification (compatibility of constellation and gray level co-occurrence matrix for analysis of cell scatter patterns). The method has more processing steps and large calculation amount, and a machine learning model needs to be established, so that the rule information corresponding to the cell volume cannot be found. More importantly, the gray level co-occurrence matrix can only reflect global information and cannot perform enlarged representation on detail information caused by the change of cell nuclei, so that only three types of different cells can be classified, and cells with variation caused by the change of cell nuclei of the same type are difficult to identify. However, many diseases directly cause the change of cell nuclei inside cells, so whether the cell nuclei inside the cells generate abnormal volume increase or the like relative to the cell nuclei of normal cells is inferred through a cell scattering inversion rule, and the method has a vital significance for providing a powerful basis for clinical disease diagnosis. Therefore, the research provides a multi-scale gray level co-occurrence matrix-based unmarked cell nucleus and a scattered light inversion method thereof, which can definitely acquire two-dimensional scattering image information and an inversion rule of biological intrinsic physical characteristics, aiming at the problems of the current cell nucleus detection method.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a marker-free cell nucleus based on a multi-scale gray level co-occurrence matrix and a scattered light inversion method thereof, which can perform multi-scale multi-directional decomposition on a two-dimensional scattering image and have statistical characteristics, and inverts the volume characteristics of the cell nucleus under a non-intervention and marker-free state by combining cell intrinsic optical information carried in the marker-free cell two-dimensional light scattering image with a Gabor wavelet transformation and gray level co-occurrence matrix method.
The invention is realized by adopting the following technical scheme:
a beam of laser irradiates on a cell, the laser is scattered after passing through the cell, and physical information such as the shape and the size of organelles such as the cell, a cell nucleus and the like is carried in the scattered light. According to the invention, the lateral scattered light of the cells is measured and is collected in a two-dimensional scattering spectrum intensity mode, so that a lateral two-dimensional scattering image is obtained. And filtering and denoising the acquired side scattering image by a digital image processing method, then carrying out Gabor filtering on the two-dimensional side scattering image to obtain a multi-scale and multi-direction characteristic diagram, calculating a gray level co-occurrence matrix of the characteristic diagram and solving an entropy value. Finally, the linear inversion relation with higher resolution is obtained between the entropy value of the cell side scattering image and the cell nucleus volume.
Under the condition of not carrying out any intervention, contact and damage on cells, the intrinsic scattering physical property of the cells is adopted, the two-dimensional intrinsic scattering image method with the intrinsic scattering physical property and the intrinsic scattering biological characteristic information superior to one-dimensional space scattering is adopted, the two-dimensional intrinsic scattering image is subjected to multi-scale and multi-directional decomposition, then the single characteristic value entropy technology with statistical properties is adopted, the effective inversion of the cell nucleus volume can be realized, and the accurate identification of the cells (fixed or living bodies) in the green environment is completed.
Drawings
FIG. 1 is a flow chart of inversion of unmarked nuclei and their scattered light based on multi-scale gray level co-occurrence matrix.
FIG. 2 is a diagram of a system for acquiring a cell side scatter plot. The system comprises a Laser 1, a neutral density filter ND 2, an objective lens 3 with the numerical aperture of 10 times of 0.25, an optical fiber coupler 4, a three-coordinate displacement table 5, a 6-stage depth-of-field microscope, an optical fiber 7, a glass slide 8 and a computer 9.
FIG. 3 is a cytographic, two-dimensional side scatter, Gabor filter, nuclear size, and entropy of five groups of white blood cells from the experiment.
FIG. 4 is a linear function image of entropy and nuclear volume of the two-dimensional forward scattering images of five groups of white blood cells in FIG. 3 after Gabor transformation.
Detailed Description
The implementation process of the invention is shown in fig. 1, firstly, white blood cell samples with the same cell volume and size and different nuclei are extracted, then a scattering acquisition system shown in fig. 2 is adopted to acquire a side scattering spectrogram after laser passes through the white blood cells, and the side scattering spectrogram carries morphological information of organelles (cell nuclei) in the cells. The method comprises the steps of denoising and gray processing the acquired side scattering images by a digital image processing method, decomposing the processed side scattering images into multi-scale and multi-direction Gabor filtering images through Gabor wavelet change, amplifying the side scattering information of white blood cells, calculating gray level co-occurrence matrixes for each group of filtering images respectively, and solving an entropy value. Finally, the entropy value of the side scattering diagram of the white blood cells and the cell nucleus volume obtained by a mathematical fitting method have a linear inversion rule as shown in figure 4.
The invention mainly aims at the side scattering images of cells, Gabor wavelet transformation is carried out on each group of side scattering images in 5 scales and 8 directions, and then the rule between the cell volume and the forward scattering images is inverted by a gray level co-occurrence matrix method. The Gabor wavelet transform takes a Gabor function as a wavelet basis function, can well give consideration to the resolution capability of signals in time domain and frequency domain, is similar to the visual characteristics of human, has excellent space locality and direction selectivity, can grasp the space frequency of a plurality of directions and local structural characteristics in the local area of an image, and enhances the local scattering information generated by the change of cell nucleus volume while keeping the total side scattering information.
The combination of Gabor wavelet transformation and gray level co-occurrence matrix is equivalent to adding a magnifying glass to the gray level co-occurrence matrix, amplifying detail signals generated by the volume change of cell nuclei, and then carrying out probability statistics on the information by using the gray level co-occurrence matrix, thereby digitizing the texture information. The implementation process of the invention mainly comprises three parts, namely acquisition of a cell side scattering image, Gabor wavelet change of the side scattering image, and calculation of a gray level co-occurrence matrix of a Gabor filter image so as to extract characteristic value entropy, and finding out the scale and the direction with the best linear relation between the characteristic value entropy and the cell nucleus volume so as to establish a mathematical model between the characteristic value and the cell nucleus volume.
Leukocytes in human peripheral blood were used as the measurement target. Removing platelets, serum and red blood cells by adopting a standard human body peripheral blood leukocyte separation method to obtain a plurality of spherical-like leukocytes, and then fixing by adopting formaldehyde to prepare a test sample.
In the acquisition of a forward two-dimensional scatter image of a cell. The forward scattering light measurement system comprises a laser light source with the wavelength of 632nm, an objective lens, a neutral density filter, a three-coordinate displacement table, an optical fiber coupler, a glass slide, a single-mode optical fiber, an ultra-field-depth microscope and the like. A cell suspension of a certain concentration is prepared, the cell suspension is sucked onto a glass slide by using a pipette gun, and the glass slide is placed in the center of a microscope lens. Turning on the white light illumination source of the super-depth-of-field microscope, finding the sample cell under the white light illumination, moving the sample cell to the center of the field of view through the electric stage of the microscope, and then rotating the lens of the super-depth-of-field microscope to keep the lens and the optical fiber in a vertical state as shown in fig. 2. And turning off a white light illumination light source, turning on a 632nm laser light source, enabling the laser to pass through a neutral density filter to enable stray light in the laser, enabling a light beam after filtering to pass through an objective lens, adjusting a three-coordinate displacement table behind the objective lens, focusing the light beam into an optical fiber coupler, and enabling the laser to excite cells to be detected on the glass slide through an optical fiber. And finally, adjusting the focal length of the microscope to enable the microscope to be in an out-of-focus state, wherein the microscope can observe a clear scattering image in the out-of-focus state. And finally, collecting the side scattering image of the cell by using a CCD (charge coupled device) of the super-field-depth microscope and storing the side scattering image into a computer.
The specific processes of extracting the angular second moment from the gray level co-occurrence matrix of the forward scattering image and establishing a mathematical model are as follows:
A. through the system for acquiring the side scattering images, a plurality of groups of cell side scattering images with the same cell volume size are acquired, the cell nucleus volumes are V1, V2, … and Vn, and the cell nucleus volumes are sequentially increased, namely F1, F2, … and Fn.
B. Respectively carrying out Gabor wavelet changes in 5 scales and 8 directions on the side scattering images, and filtering each side scattering image to obtain 40 groups of filtering images F1 with different scales and different directions1，F12，…F140； F21，F22，…F240；...；Fn1，Fn2，…Fn40。
C. Calculating the gray level co-occurrence matrix P1 of the Gabor filter diagram1，P12，…P140；P21， P22，…P240；...；Pn1，Pn2，…Pn40And respectively calculating the entropy values K11， K12，…K140；K21，K22，…K240；...；Kn1，Kn2，…Kn40。
D. By matlab software, the cell nucleus volumes V1, V2, …, Vn and the entropy value K1i，K2i，…Kni(i ═ 1, 2.., 40). Performing data fitting to obtain a function model V of the cell nucleus volumes V and the entropy values K of 40 groupsi(K)(i＝1，2，...，40)。
E. Searching the function model V (K) with the best linear relation from the 40 groups of function models, namely, the best linear function model in the invention is V4(K)。
F. Measuring the side scattering image Fx with unknown cell nucleus volume by the same method in G, and obtaining the entropy Kx of the image by the same method1，Kx2，…Kx40To determine the entropy Kx in the direction of the optimal scale4Substituting the optimal linear function model V4(K) The volume Vx of the unknown cell is obtained.
The Gabor transformation and gray level co-occurrence matrix and eigenvalue K extraction method comprises the following steps:
(1) and carrying out gray level processing on the collected color picture to obtain a gray level image. The data form of the gray image F is a two-dimensional array, F (x, y) represents the gray value at (x, y) in the image F, and the gray value is an integer within the interval [0,255], wherein 0 corresponds to black and 255 corresponds to white.
(2) Gabor filtering is carried out on the processed picture, and a two-dimensional Gabor filter is selected for filtering, wherein the parameters of the two-dimensional filter are that the wavelength (filtering scale) lambda is 14, 28, 42, 56 and 70 in sequence, and the direction theta is 0,
N, other parameters, e.g. phase
Spatial aspect ratio
The bandwidth σ is kept constant at 2 pi. By selecting the filter parameters, the image can be decomposed into 40 groups of 5 filtered images with different scales and 8 different directions.
(3) And calculating the gray level co-occurrence matrix of the 40 groups of filtered images, wherein the number of the stages for calculating the gray level co-occurrence matrix is set to be 16, the sampling step length of the GLCM is d equal to 1, and the sampling directions are four directions of 0 degrees, 45 degrees, 90 degrees and 135 degrees on the image. And calculating a gray level co-occurrence matrix of the filtering image by adopting the parameters, and solving a characteristic entropy value K. The formula for calculating the entropy value is:
(4) and calculating the mean value entropy of the entropy values in the four directions as the characteristic value of the filtering image. The mean entropy calculation formula in the four directions is:
Claims (5)
1. the method is characterized in that an optical detection system is used for collecting a two-dimensional side scattering image with nuclear cells, after image processing methods such as denoising and gray processing are carried out, Gabor wavelet transformation is adopted to decompose the two-dimensional side scattering image into a multi-scale and multidirectional characteristic map, the gray level co-occurrence matrix of the characteristic map is calculated to obtain an entropy value of the characteristic map, a mathematical model of a nuclear volume V and an entropy value k is established through a data fitting method, the V (k) is 346.91k-842.19, the size of the nuclear volume can be inverted by the mathematical model, and whether the cells are diseased into tumor cells or are deteriorated into cancer cells due to severe increase of the nuclear volume is judged through comparison with the size of the nuclear volume of normal cells.
2. The label-free cell nucleus scattering inversion method based on the multi-scale gray level co-occurrence matrix as claimed in claim 1, wherein Gabor wavelet transform filtering is performed on the two-dimensional side scattering image, the scattering image is decomposed in multiple scales and directions, and global features of the two-dimensional side scattering image are enhanced in different directions with different resolutions, so that tiny local features of the two-dimensional side scattering image generated due to cell nucleus volume changes can be better distinguished.
3. The label-free cell nucleus scattering inversion method based on the multi-scale gray level co-occurrence matrix according to claim 1, characterized in that a feature extraction method combining Gabor wavelet transformation and a gray level co-occurrence matrix is provided, so that the amplification of details in different directions of a side scattering image by a Gabor filter is realized, and the probability statistics of the global pixel points of the scattering image by the gray level co-occurrence matrix is retained to extract features.
4. The method for label-free nucleus scattering inversion based on the multi-scale gray-scale co-occurrence matrix as claimed in claim 1, wherein the Gabor filter adopts a two-dimensional Gabor filter, and the parameters of the filter are selected, wherein the wavelength (filter scale) λ is 14, 28, 42, 56 and 70 in sequence, and the direction θ is 0, 0 and,
Pi, phase
Space crossbar
The bandwidth σ is 2 pi. When the filter scale λ is 14, direction
When other parameters are unchanged, the linear relation between the entropy value characteristic extracted from the filtered characteristic diagram and the cell nucleus volume is better.
5. The label-free cell nucleus scattering inversion method based on the multi-scale gray level co-occurrence matrix is characterized by comprising the following specific steps of:
A. acquiring a plurality of groups of cell side scattering images F1, F2, … and Fn with the same cell volume and the cell nucleus volume of V1, V2, … and Vn and the cell nucleus volume sequentially increased by a side scattering image acquisition system;
B. respectively carrying out Gabor wavelet changes in 5 scales and 8 directions on the side scattering images, and filtering each side scattering image to obtain 40 groups of side scattering images with different scales and different directionsFilter map F11，F12，…F140；F21，F22，…F240；...；Fn1，Fn2，…Fn40；
C. Calculating the gray level co-occurrence matrix P1 of the Gabor filter diagram1，P12，…P140；P21，P22，…P240；...；Pn1，Pn2，…Pn40And respectively calculating the entropy values K11，K12，…K140；K21，K22，…K240；...；Kn1，Kn2，…Kn40；
D. By matlab software, the cell nucleus volumes V1, V2, …, Vn and the entropy value K1i，K2i，…Kni(i ═ 1, 2.., 40). Performing data fitting to obtain a function model V of the cell nucleus volumes V and the entropy values K of 40 groupsi(K)(i＝1，2，...，40)；
E. Searching the function model V (K) with the best linear relation from the 40 groups of function models, namely, the best linear function model in the invention is V4(K)；
F. Measuring the unknown side scatter diagram Fx of the cell nucleus volume by the same method in A, and obtaining the entropy Kx of the image by the same method1，Kx2，…Kx40To determine the entropy Kx in the direction of the optimal scale4Substituting the optimal linear function model V4(K) The volume Vx of the unknown cell is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010663410.XA CN111751266A (en) | 2020-07-10 | 2020-07-10 | Mark-free cell nucleus scattering inversion method based on multi-scale gray level co-occurrence matrix |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010663410.XA CN111751266A (en) | 2020-07-10 | 2020-07-10 | Mark-free cell nucleus scattering inversion method based on multi-scale gray level co-occurrence matrix |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111751266A true CN111751266A (en) | 2020-10-09 |
Family
ID=72710919
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010663410.XA Pending CN111751266A (en) | 2020-07-10 | 2020-07-10 | Mark-free cell nucleus scattering inversion method based on multi-scale gray level co-occurrence matrix |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111751266A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101943663A (en) * | 2010-07-09 | 2011-01-12 | 董珂 | Measuring analytical system and measuring analytical method for distinguishing diffraction image of particles automatically |
| CN105181649A (en) * | 2015-10-09 | 2015-12-23 | 山东大学 | Novel label-free pattern recognition cytometer method |
| CN106023188A (en) * | 2016-05-17 | 2016-10-12 | 天津大学 | Breast tumor feature selection method based on Relief algorithm |
| US20180188173A1 (en) * | 2015-12-22 | 2018-07-05 | University Of Maryland, College Park | Cell classification based on mechanical signature of nucleus |
| CN108272533A (en) * | 2017-12-26 | 2018-07-13 | 中国科学院苏州生物医学工程技术研究所 | The skin modeling method in skin wound region |
| CN109859181A (en) * | 2019-01-29 | 2019-06-07 | 桂林电子科技大学 | A kind of PCB welding point defect detection method |
-
2020
- 2020-07-10 CN CN202010663410.XA patent/CN111751266A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101943663A (en) * | 2010-07-09 | 2011-01-12 | 董珂 | Measuring analytical system and measuring analytical method for distinguishing diffraction image of particles automatically |
| CN105181649A (en) * | 2015-10-09 | 2015-12-23 | 山东大学 | Novel label-free pattern recognition cytometer method |
| US20180188173A1 (en) * | 2015-12-22 | 2018-07-05 | University Of Maryland, College Park | Cell classification based on mechanical signature of nucleus |
| CN106023188A (en) * | 2016-05-17 | 2016-10-12 | 天津大学 | Breast tumor feature selection method based on Relief algorithm |
| CN108272533A (en) * | 2017-12-26 | 2018-07-13 | 中国科学院苏州生物医学工程技术研究所 | The skin modeling method in skin wound region |
| CN109859181A (en) * | 2019-01-29 | 2019-06-07 | 桂林电子科技大学 | A kind of PCB welding point defect detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10025271B2 (en) | Method and system for detecting and/or classifying cancerous cells in a cell sample | |
| US7756305B2 (en) | Fast 3D cytometry for information in tissue engineering | |
| US7587078B2 (en) | Automated image analysis | |
| US10083340B2 (en) | Automated cell segmentation quality control | |
| Netten et al. | Fluorescent dot counting in interphase cell nuclei | |
| Ginley et al. | Unsupervised labeling of glomerular boundaries using Gabor filters and statistical testing in renal histology | |
| CN108388853B (en) | Stepwise reconstruction and counting method for leucocyte and platelet coexistence hologram | |
| Lin et al. | Automatic detection and characterization of quantitative phase images of thalassemic red blood cells using a mask region-based convolutional neural network | |
| Meyer et al. | Automated cell analysis in 2D and 3D: A comparative study | |
| Moazzen et al. | Metaphase finding with deep convolutional neural networks | |
| CN116097086A (en) | Cellular diagnostic and analytical methods | |
| del Milagro Fernández-Carrobles et al. | Automatic handling of tissue microarray cores in high-dimensional microscopy images | |
| CN111751263A (en) | Mark-free cell two-dimensional scattering image inversion method based on gray level co-occurrence matrix | |
| US20220383629A1 (en) | Label-free cell classification and screening system based on hybrid transfer learning | |
| Durkee et al. | Pseudo-spectral angle mapping for automated pixel-level analysis of highly multiplexed tissue image data | |
| CN115689984A (en) | Single cell feature extraction and analysis method and system based on mitochondria | |
| CN111751266A (en) | Mark-free cell nucleus scattering inversion method based on multi-scale gray level co-occurrence matrix | |
| CN112945835A (en) | Cell accurate three-dimensional shape recovery method based on depth change point diffusion function | |
| Bapure | Automated image analysis for nuclear morphometry using h&e and feulgen stains in prostate biopsies | |
| Voros et al. | Correlative Fluorescence and Raman Microscopy to Define Mitotic Stages at the Single-Cell Level: Opportunities and Limitations in the AI Era | |
| Delikoyun et al. | 2 Deep learning-based cellular image analysis for intelligent medical diagnosis | |
| Qiu et al. | Automated detection of analyzable metaphase chromosome cells depicted on scanned digital microscopic images | |
| Tang et al. | Fast post-processing pipeline for optical projection tomography | |
| Mahalakshmi et al. | KNN Model for Cancer Prediction Using Stem Cells | |
| D'Almeida | Development of lens-free holographic microscopes using multiheight and multispectral phase recovery methods |
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 |