CN113935983A - Full-automatic quantitative analysis method for perinuclear lysosome distribution - Google Patents
Full-automatic quantitative analysis method for perinuclear lysosome distribution Download PDFInfo
- Publication number
- CN113935983A CN113935983A CN202111263263.8A CN202111263263A CN113935983A CN 113935983 A CN113935983 A CN 113935983A CN 202111263263 A CN202111263263 A CN 202111263263A CN 113935983 A CN113935983 A CN 113935983A
- Authority
- CN
- China
- Prior art keywords
- region
- cell
- lysosome
- single cell
- perinuclear
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 210000003712 lysosome Anatomy 0.000 title claims abstract description 128
- 230000001868 lysosomic effect Effects 0.000 title claims abstract description 128
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004445 quantitative analysis Methods 0.000 title claims abstract description 20
- 210000004027 cell Anatomy 0.000 claims abstract description 181
- 210000002472 endoplasmic reticulum Anatomy 0.000 claims abstract description 55
- 230000002132 lysosomal effect Effects 0.000 claims abstract description 20
- 238000013135 deep learning Methods 0.000 claims abstract description 13
- 210000003855 cell nucleus Anatomy 0.000 claims description 82
- 239000011159 matrix material Substances 0.000 claims description 22
- 230000011218 segmentation Effects 0.000 claims description 10
- 230000002093 peripheral effect Effects 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 5
- 238000003384 imaging method Methods 0.000 claims description 5
- 230000001413 cellular effect Effects 0.000 claims description 4
- 238000003709 image segmentation Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000007405 data analysis Methods 0.000 abstract description 4
- 230000008569 process Effects 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 abstract description 3
- 239000000284 extract Substances 0.000 abstract description 2
- 210000004940 nucleus Anatomy 0.000 description 10
- 230000033001 locomotion Effects 0.000 description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 210000000170 cell membrane Anatomy 0.000 description 2
- 210000000805 cytoplasm Anatomy 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 210000004688 microtubule Anatomy 0.000 description 2
- 230000035790 physiological processes and functions Effects 0.000 description 2
- 108090000623 proteins and genes Proteins 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 208000015439 Lysosomal storage disease Diseases 0.000 description 1
- 206010027476 Metastases Diseases 0.000 description 1
- 102000029749 Microtubule Human genes 0.000 description 1
- 108091022875 Microtubule Proteins 0.000 description 1
- 206010028980 Neoplasm Diseases 0.000 description 1
- 206010064390 Tumour invasion Diseases 0.000 description 1
- 239000000427 antigen Substances 0.000 description 1
- 108091007433 antigens Proteins 0.000 description 1
- 102000036639 antigens Human genes 0.000 description 1
- 230000006907 apoptotic process Effects 0.000 description 1
- 210000003050 axon Anatomy 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000009400 cancer invasion Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000021164 cell adhesion Effects 0.000 description 1
- 210000005056 cell body Anatomy 0.000 description 1
- 230000012292 cell migration Effects 0.000 description 1
- 230000033077 cellular process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000000942 confocal micrograph Methods 0.000 description 1
- 230000001086 cytosolic effect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004064 dysfunction Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002073 fluorescence micrograph Methods 0.000 description 1
- 235000003642 hunger Nutrition 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000003834 intracellular effect Effects 0.000 description 1
- 235000010335 lysozyme Nutrition 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 230000009401 metastasis Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000004770 neurodegeneration Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 210000003463 organelle Anatomy 0.000 description 1
- 230000020477 pH reduction Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 244000052769 pathogen Species 0.000 description 1
- 230000008506 pathogenesis Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 210000002955 secretory cell Anatomy 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 230000037351 starvation Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 208000011580 syndromic disease Diseases 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
本发明提出了一种核周溶酶体分布全自动定量分析方法，包括：在内质网图像中识别并提取内质网区域，用所述内质网区域近似代表单细胞区域；在细胞核图像中分割细胞核区域，得到所有细胞核区域；在溶酶体图像中提取所有溶酶体区域；针对每一个单细胞区域，确定该单细胞区域中的细胞核区域、溶酶体区域；根据该单细胞的实际形态设定相应的核周区域；根据该单细胞的核周区域和该单细胞区域中的溶酶体位置和半径，确定该单细胞的核周区域的溶酶体分布。本发明基于深度学习技术设计了一个自动化分析流程，实现了精准的核周溶酶体分布定量分析，解决了目前人工手动计算的难题，使高通量大数据分析成为可能。
The present invention provides a fully automatic quantitative analysis method for perinuclear lysosome distribution, comprising: identifying and extracting the endoplasmic reticulum region in the endoplasmic reticulum image, and using the endoplasmic reticulum region to approximately represent the single cell region; Segment the nuclear area in the middle to obtain all the nuclear areas; extract all the lysosomal areas in the lysosomal image; for each single-cell area, determine the nuclear area and lysosomal area in the single-cell area; according to the single-cell area The actual shape sets the corresponding perinuclear region; according to the perinuclear region of the single cell and the lysosome position and radius in the single cell region, the lysosome distribution in the perinuclear region of the single cell is determined. The invention designs an automatic analysis process based on the deep learning technology, realizes accurate quantitative analysis of perinuclear lysosome distribution, solves the current problem of manual calculation, and makes high-throughput big data analysis possible.
Description
Technical Field
The invention relates to the technical field of cell analysis, in particular to a full-automatic quantitative analysis method for perinuclear lysosome distribution.
Background
The distribution of lysosomes in cells is in a certain state, e.g., lysosomes in senescent cells generally accumulate in the perinuclear region. In the related field, an effective means for realizing accurate quantitative analysis of the perinuclear lysosome distribution is still lacked, and the technical difficulty is to accurately define the cell center and the cell boundary and accurately depict the perinuclear region, such as the quantification of the far nuclear region of the near nuclear region. Most of the existing work is to manually designate and divide the areas manually, and high-flux big data analysis is difficult to realize. How to carry out automated analysis on the distribution condition of the perinuclear lysosome by means of a computer is a technical problem to be solved urgently by the technical personnel in the field.
Disclosure of Invention
The invention aims to solve the technical problem of providing a full-automatic quantitative analysis method for perikaryosome distribution, which is used for automatically analyzing the perikaryosome distribution condition based on a deep learning technology in the field of artificial intelligence.
The technical scheme adopted by the invention is that the full-automatic quantitative analysis method for the perinuclear lysosome distribution comprises the following steps:
step 1, respectively carrying out different fluorescent marks on cell nucleus, endoplasmic reticulum and lysosome, and then carrying out imaging by using a confocal microscope to obtain a cell nucleus image, an endoplasmic reticulum image and a lysosome image which comprise a plurality of cells;
step 2, identifying and extracting an endoplasmic reticulum region in the endoplasmic reticulum image, and using the endoplasmic reticulum region to approximately represent a single cell region; dividing cell nucleus areas in the cell nucleus image to obtain all the cell nucleus areas; extracting all lysosome regions and the positions and the radii thereof in the lysosome image;
step 3, aiming at each single cell region, determining the cell nucleus region in the single cell region based on the single cell region and all the cell nucleus regions, and determining the lysosome region in the single cell region and the position and the radius of the lysosome region based on the single cell region and all the lysosome regions;
step 4, respectively extracting the boundary of the single cell region, the boundary of the cell nucleus region and the center of the cell nucleus aiming at each single cell region; setting a corresponding perinuclear region between the boundary of the cell nucleus region and the boundary of the cell nucleus region according to the actual form of the single cell;
and 5, aiming at each single cell region, determining the lysosome distribution of the perinuclear region of the single cell according to the position and the radius of the perinuclear region of the single cell and the lysosome region in the single cell region.
Optionally, in the step 2, an endoplasmic reticulum region is identified and extracted in the endoplasmic reticulum image by using a deep learning-based example segmentation model MaskRCNN, and the endoplasmic reticulum region is used to approximately represent a single cell region.
Optionally, in the step 2, the image segmentation model UNet based on deep learning is used to segment the cell nucleus regions in the cell nucleus image, so as to obtain all the cell nucleus regions; and extracting all lysosome regions, positions and radii thereof in the lysosome image by using a point detection method of difference of Gaussian functions.
Optionally, in step 3, for each single-cell region, determining a nucleus region in the single-cell region based on the single-cell region and all nucleus regions includes:
expressing the endoplasmic reticulum area in the form of an endoplasmic reticulum area element matrix, wherein an element 1 represents a cell area, and an element 0 represents an area outside the cell;
the cell nucleus area is represented by a cell nucleus area element matrix, wherein the cell nucleus area is represented by an element 1, and the area outside the cell nucleus is represented by an element 0;
and for each single cell, multiplying the endoplasmic reticulum area element matrix of the single cell with corresponding elements in all the cell nucleus area element matrices to obtain the cell nucleus area in the single cell area.
Optionally, in step 3, for each single-cell region, determining a lysosomal region in the single-cell region based on the single-cell region and all lysosomal regions includes:
expressing the endoplasmic reticulum area in the form of an endoplasmic reticulum area element matrix, wherein an element 1 represents a cell area, and an element 0 represents an area outside the cell;
expressing the lysosome region in the form of a lysosome region element matrix, wherein the lysosome region is represented by an element 1, and the region outside the lysosome is represented by an element 0;
and aiming at each single cell, multiplying the endoplasmic reticulum region element matrix of the single cell with corresponding elements in all lysosome region element matrices to obtain the lysosome region in the single cell region.
Optionally, in the step 4, for each single cell region, the single cell region boundary, the cell nucleus region boundary, and the cell nucleus center are respectively extracted by using image morphology operation.
Optionally, in step 4, setting a corresponding perinuclear region between the boundary of the nuclear region and the boundary of the cellular region according to the actual morphology of the single cell, including:
for each single cell region, taking the center of a cell nucleus as the center of the cell, and radiating rays to the periphery of the cell, wherein any ray is intersected with the boundary of the cell nucleus region and the boundary of the cell region;
and setting segmentation points for line segments from the cell nucleus region boundary to the cell region boundary on each ray according to a set proportion, connecting the segmentation points of each ray together to form the outer boundary of the cell nucleus peripheral region, wherein the inner boundary of the cell nucleus peripheral region is the cell nucleus region boundary.
Optionally, the size of the nuclear peripheral region is adjusted by adjusting the value of the set ratio.
Optionally, in step 5, determining the lysosome distribution of the perinuclear region of the single cell according to the perinuclear region of the single cell and the lysosome position and radius in the single cell region, including:
determining the lysosome number of the perinuclear region of the single cell and the lysosome number of the single cell region according to the perinuclear region of the single cell and the lysosome position and radius of the single cell region;
a parameter indicative of lysosomal distribution in the perinuclear region of the single cell, comprising at least one of:
the ratio of the number of lysosomes in the perinuclear region of a single cell to the number of lysosomes in the single cell region;
the proportion of the area of lysosomes within the perinuclear region of a single cell to the area of lysosomes within that single cell region.
The present invention also provides a computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described fully-automatic quantitative analysis method for perinuclear lysosomal distribution.
By adopting the technical scheme, the invention at least has the following advantages:
1) the full-automatic quantitative analysis method for the perinuclear lysosome distribution is an automatic analysis process based on a human deep learning technology, realizes accurate quantitative analysis of the perinuclear lysosome distribution, solves the problem of manual calculation at present, and makes high-throughput big data analysis possible.
2) The invention obtains the cell nucleus and endoplasmic reticulum area in the cell by the multi-channel imaging technology, further detects and extracts the cell nucleus and endoplasmic reticulum area by the deep learning technology, and uses the cell nucleus and endoplasmic reticulum area to approximately depict the cell center and the cell boundary, thereby solving the problem that the space structure of the cell is difficult to be accurately depicted by the imaging and image analysis technology in the fluorescence image.
3) The invention flexibly defines the range of the perinuclear region by setting a parameter, so that the calculation of the perinuclear particle distribution is more robust, and the perinuclear region can be accurately defined under different conditions (cell size and cell shape).
Drawings
FIG. 1 is a flow chart of a fully automatic quantitative analysis method for perinuclear lysosomal distribution according to a first embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating the principle of a fully automatic quantitative analysis method for perinuclear lysosomal distribution according to a second embodiment of the present invention;
fig. 3 is a schematic flow chart of a full-automatic quantitative analysis method for perinuclear lysosomal distribution according to a third embodiment of the present invention.
Detailed Description
To further explain the technical means and effects of the present invention adopted to achieve the intended purpose, the present invention will be described in detail with reference to the accompanying drawings and preferred embodiments.
Lysozymes are considered to be typical degradation organelles. In recent years, however, they have been found to be involved in many other cellular processes, including killing intracellular pathogens, antigen display, plasma membrane repair, cell adhesion and migration, tumor invasion and metastasis, apoptotic cell death, metabolic signaling, and gene regulation. In addition, lysosomal dysfunction has been shown to be not limited to rare lysosomal storage disorder diseases, but also to be associated with common diseases such as cancer and neurodegeneration. In modern biological concepts, lysosomes have been thought to perform substance and information transfer functions, closely related to the physiological state of cells. Lysosomes do not function with their spatial distribution and motor characteristics. In non-polarized cells, they are mainly concentrated in the central region of the center of the microtubule tissue around the nucleus, with a small number of peripheral lysosomes, reaching the plasma membrane and the cell processes. Polarized cells such as neurons, secretory cells, etc., lysosomes are distributed in all cytoplasm including soma, axons and dendritic domains. When cells function abnormally, the spatial distribution of lysosomes will change. Lysosomes in senescent cells, for example, generally accumulate in the perinuclear region. Lysosomes in cells undergo bidirectional movement between the center and periphery of the cell along microtubules. The motion of the lysosome is tightly regulated and its motion is also altered with changes in physiological state. For example, cytoplasmic acidification causes lysosomes to alkalize and return them to their central position, starvation, expression of key genes of huntington's syndrome leads to lysosomal motor arrest, and the like. Therefore, the distribution and movement ability of lysosomes in cytoplasm are crucial to the development of cells, and the interference of the distribution and movement of lysosomes is directly related to the pathogenesis of various diseases. However, in the related art, there is still a lack of effective means for realizing accurate quantitative analysis of the perinuclear lysosome distribution, and the technical difficulty is to precisely define the cell center and the cell boundary, and precisely delineate the quantification of the perinuclear region, such as the far nuclear region of the near nuclear region. Most of the existing work is to manually designate and divide the areas manually, and high-flux big data analysis is difficult to realize. In order to solve the technical problem, the invention designs an automatic analysis process based on the deep learning technology in the field of artificial intelligence.
In a first embodiment of the present invention, a method for fully automatically and quantitatively analyzing the distribution of lysosomes around the nucleus, as shown in fig. 1, comprises the following specific steps:
step 1, respectively carrying out different fluorescent marks on cell nucleus, endoplasmic reticulum and lysosome, and then carrying out imaging by using a confocal microscope to obtain a cell nucleus image, an endoplasmic reticulum image and a lysosome image which comprise a plurality of cells;
step 2, identifying and extracting an endoplasmic reticulum region in the endoplasmic reticulum image, and using the endoplasmic reticulum region to approximately represent a single cell region; dividing cell nucleus areas in the cell nucleus image to obtain all the cell nucleus areas; extracting all lysosome regions and the positions and the radii thereof in the lysosome image;
optionally, in the step 2, identifying and extracting an endoplasmic reticulum region in the endoplasmic reticulum image by using an example segmentation model MaskRCNN based on deep learning, and approximately representing a single-cell region by using the endoplasmic reticulum region;
in the step 2, dividing the cell nucleus regions in the cell nucleus image by using an image segmentation model UNet based on deep learning to obtain all the cell nucleus regions; and extracting all lysosome regions, positions and radii thereof in the lysosome image by using a point detection method of difference of Gaussian functions.
Step 3, aiming at each single cell region, determining the cell nucleus region in the single cell region based on the single cell region and all the cell nucleus regions, and determining the lysosome region in the single cell region and the position and the radius of the lysosome region based on the single cell region and all the lysosome regions;
optionally, in step 3, for each single-cell region, determining a nucleus region in the single-cell region based on the single-cell region and all nucleus regions includes:
expressing the endoplasmic reticulum area in the form of an endoplasmic reticulum area element matrix, wherein an element 1 represents a cell area, and an element 0 represents an area outside the cell;
the cell nucleus area is represented by a cell nucleus area element matrix, wherein the cell nucleus area is represented by an element 1, and the area outside the cell nucleus is represented by an element 0;
and for each single cell, multiplying the endoplasmic reticulum area element matrix of the single cell with corresponding elements in all the cell nucleus area element matrices to obtain the cell nucleus area in the single cell area.
In step 3, for each single-cell region, determining a lysosomal region in the single-cell region based on the single-cell region and all lysosomal regions, including:
expressing the endoplasmic reticulum area in the form of an endoplasmic reticulum area element matrix, wherein an element 1 represents a cell area, and an element 0 represents an area outside the cell;
expressing the lysosome region in the form of a lysosome region element matrix, wherein the lysosome region is represented by an element 1, and the region outside the lysosome is represented by an element 0;
and aiming at each single cell, multiplying the endoplasmic reticulum region element matrix of the single cell with corresponding elements in all lysosome region element matrices to obtain the lysosome region in the single cell region.
Step 4, respectively extracting the boundary of the single cell region, the boundary of the cell nucleus region and the center of the cell nucleus aiming at each single cell region; setting a corresponding perinuclear region between the boundary of the cell nucleus region and the boundary of the cell nucleus region according to the actual form of the single cell;
optionally, in the step 4, for each single cell region, the single cell region boundary, the cell nucleus region boundary, and the cell nucleus center are respectively extracted by using image morphology operation.
In step 4, setting a corresponding perinuclear region between the boundary of the nuclear region and the boundary of the cellular region according to the actual morphology of the single cell, including:
for each single cell region, taking the center of a cell nucleus as the center of the cell, and radiating rays to the periphery of the cell, wherein any ray is intersected with the boundary of the cell nucleus region and the boundary of the cell region;
and setting segmentation points for line segments from the cell nucleus region boundary to the cell region boundary on each ray according to a set proportion, connecting the segmentation points of each ray together to form the outer boundary of the cell nucleus peripheral region, wherein the inner boundary of the cell nucleus peripheral region is the cell nucleus region boundary.
Optionally, the size of the nuclear peripheral region is adjusted by adjusting the value of the set ratio.
And 5, aiming at each single cell region, determining the lysosome distribution of the perinuclear region of the single cell according to the perinuclear region of the single cell and the position and the radius of the lysosome in the single cell region.
Optionally, in step 5, determining the lysosome distribution of the perinuclear region of the single cell according to the perinuclear region of the single cell and the lysosome position and radius in the single cell region, including:
determining the lysosome number of the perinuclear region of the single cell and the lysosome number of the single cell region according to the perinuclear region of the single cell and the lysosome position and radius of the single cell region;
a parameter indicative of lysosomal distribution in the perinuclear region of the single cell, comprising at least one of:
the ratio of the number of lysosomes in the perinuclear region of a single cell to the number of lysosomes in the single cell region;
the proportion of the area of lysosomes within the perinuclear region of a single cell to the area of lysosomes within that single cell region.
A second embodiment of the present invention is based on the above embodiments, and an application example of the present invention is described with reference to fig. 2 to 3.
As shown in fig. 3, the fully automatic quantitative analysis method for perinuclear lysosomal distribution according to the embodiment of the present invention includes the following specific steps:
step S1: the nucleus, endoplasmic reticulum and lysosome are respectively marked with different fluorescence.
Step S2: by high resolution multi-channel confocal microscopy images of the nucleus, endoplasmic reticulum and lysosome were obtained separately, each image containing a number of cells, typically more than 10, as shown in figure 2.
Step S3: identifying and extracting an endoplasmic reticulum region by using an example segmentation model MaskRCNN based on deep learning by means of the characteristic that the endoplasmic reticulum occupies the whole cell region and taking an endoplasmic reticulum image as an object, approximately depicting a single cell region by using the endoplasmic reticulum region of a single cell, and approximately depicting the single cell region by using the endoplasmic reticulum region of the single cell
Is shown as
The mask element matrix of an individual cell, i.e., the endoplasmic reticulum region element matrix,
element 1 in (1) represents a cell region, and 0 represents an extracellular background region.
Step S4: segmenting the nuclear image by using an image segmentation model UNet based on deep learning, segmenting and extracting all nuclear regions, and using
And (4) showing.
Step S5: dividing and extracting corresponding second cells among all the nucleus regions obtained in step S4 using the single-cell region obtained in step S3
Nuclear area element matrix of individual cells:
wherein, in the step (A),
representing corresponding elementsMultiplication.
Step S6: dividing lysosome image by using single cell region obtained in step S3, and extracting corresponding step
Lysosomal region element matrix within individual cells:
wherein, in the step (A),
all the lysosome regions and the positions and the radii thereof are extracted from the lysosome image by using a point detection method of difference of gaussian function (difference of gaussian), and once all the lysosome regions are determined, the position and the radius of each lysosome region are determined.
Step S7: extracting the single cell region obtained in step S3 by image morphology
Boundary of single cell region
Extracting the following
Nuclear boundary of individual single cell
And the center of the nucleus
Detecting using a point detection method based on difference of Gaussian functions
In the single cell region
Location and radius of individual lysosomes
。
Step S8: by nuclear centre
As the center of the cell, the perinuclear region is defined by the radiation emitted toward the periphery of the cell. In particular,
a) using nuclear centres
As a point on a straight line, an arbitrary point on a cell boundary
As another point on the straight line, a geometric equation of the straight line is obtained using a two-point method:
wherein
，
，
。
b) The intersection of this line with the nuclear boundary was calculated:
。
c) the perinuclear region is defined based on the nuclear boundary and the cell boundary, because the perinuclear region is between the nuclear region boundary and the cell boundary and takes the nuclear boundary as the inner boundary, using one parameter
Defining the extent of this area, i.e. indicated by the dashed line in fig. 2Outer boundary of perinuclear region:
wherein
。
Close to 0, meaning that the closer the region is to the nucleus, the smaller the region,
approaching 1 means that the closer the region is to the cell boundary, the larger the region.
d) Repeating a) -c), traversing all points on the cell boundary
，
Represents the total number of all points on the cell boundary, to obtain
Closed borders of the perinuclear region of individual cells
。
Step S9: and (4) calculating the lysosome distribution of each cell nucleus periphery according to the lysosome number in the cell nucleus periphery region obtained in the step (S8) and the corresponding single cell region obtained in the step (S7). Two indices are defined here, the number of lysosomes in the perinuclear region in the whole cell range in the lysosome is proportional:
wherein
Is the nuclear cycle
The number of lysosomes within the region is,
is the number of lysosomes throughout the cell; the area fraction of lysosomes in the perinuclear region in lysosomes over the entire cellular range:
wherein
Is the nuclear cycle
The total area of the lysosome within the region,
is the total area of all lysosomes within the cell. By defining different parameters
And the lysosome distribution rule of the corresponding perinuclear region can be calculated and quantitatively depicted.
In the third embodiment of the present invention, the procedure of the fully automatic quantitative analysis method for perinuclear lysosome distribution in this embodiment is the same as that in the first and second embodiments, but the difference is that in terms of engineering implementation, this embodiment can be implemented by software plus a necessary general hardware platform, and certainly, the present embodiment can also be implemented by hardware, but the former is a better implementation mode in many cases. With this understanding in mind, the methods of the present invention may be embodied in the form of a computer software product stored on a storage medium (e.g., ROM/RAM, magnetic or optical disk) and including instructions that cause an apparatus to perform the methods of the embodiments of the invention.
While the invention has been described in connection with specific embodiments thereof, it is to be understood that it is intended by the appended drawings and description that the invention may be embodied in other specific forms without departing from the spirit or scope of the invention.
Claims (10)
1. A full-automatic quantitative analysis method for perinuclear lysosome distribution is characterized by comprising the following steps:
step 1, respectively carrying out different fluorescent marks on cell nucleus, endoplasmic reticulum and lysosome, and then carrying out imaging by using a confocal microscope to obtain a cell nucleus image, an endoplasmic reticulum image and a lysosome image which comprise a plurality of cells;
step 2, identifying and extracting an endoplasmic reticulum region in the endoplasmic reticulum image, and using the endoplasmic reticulum region to approximately represent a single cell region; dividing cell nucleus areas in the cell nucleus image to obtain all the cell nucleus areas; extracting all lysosome regions and the positions and the radii thereof in the lysosome image;
step 3, aiming at each single cell region, determining the cell nucleus region in the single cell region based on the single cell region and all the cell nucleus regions, and determining the lysosome region in the single cell region and the position and the radius of the lysosome region based on the single cell region and all the lysosome regions;
step 4, respectively extracting the boundary of the single cell region, the boundary of the cell nucleus region and the center of the cell nucleus aiming at each single cell region; setting a corresponding perinuclear region between the boundary of the cell nucleus region and the boundary of the cell nucleus region according to the actual form of the single cell;
and 5, aiming at each single cell region, determining the lysosome distribution of the perinuclear region of the single cell according to the perinuclear region of the single cell and the position and the radius of the lysosome in the single cell region.
2. The fully automated quantitative analysis method of perinuclear lysosomal distribution according to claim 1, characterized in that in step 2, endoplasmic reticulum regions are identified and extracted in the endoplasmic reticulum image using an example segmentation model based on deep learning, with which single cell regions are approximately represented.
3. The method for fully automatically and quantitatively analyzing the perinuclear lysosomal distribution according to claim 1, wherein in the step 2, the nuclear regions are segmented in the nuclear image by using an image segmentation model based on deep learning to obtain all the nuclear regions; and extracting all lysosome regions, positions and radii thereof in the lysosome image by using a point detection method of difference of Gaussian functions.
4. The method according to claim 1, wherein the step 3 of determining the nuclear region of each single-cell region based on the single-cell region and all nuclear regions comprises:
expressing the endoplasmic reticulum area in the form of an endoplasmic reticulum area element matrix, wherein an element 1 represents a cell area, and an element 0 represents an area outside the cell;
the cell nucleus area is represented by a cell nucleus area element matrix, wherein the cell nucleus area is represented by an element 1, and the area outside the cell nucleus is represented by an element 0;
and for each single cell, multiplying the endoplasmic reticulum area element matrix of the single cell with corresponding elements in all the cell nucleus area element matrices to obtain the cell nucleus area in the single cell area.
5. The method according to claim 1, wherein the step 3 of determining the lysosome region in each single-cell region based on the single-cell region and all the lysosome regions comprises:
expressing the endoplasmic reticulum area in the form of an endoplasmic reticulum area element matrix, wherein an element 1 represents a cell area, and an element 0 represents an area outside the cell;
expressing the lysosome region in the form of a lysosome region element matrix, wherein the lysosome region is represented by an element 1, and the region outside the lysosome is represented by an element 0;
and aiming at each single cell, multiplying the endoplasmic reticulum region element matrix of the single cell with corresponding elements in all lysosome region element matrices to obtain the lysosome region in the single cell region.
6. The method according to claim 1, wherein in step 4, the boundary of the single-cell region, the boundary of the nuclear region, and the center of the nuclear region are extracted by image morphology for each single-cell region.
7. The method according to claim 1, wherein the step 4 of setting the corresponding perinuclear regions between the boundaries of the nuclear region and the boundaries of the cellular region according to the actual morphology of the single cell comprises:
for each single cell region, taking the center of a cell nucleus as the center of the cell, and radiating rays to the periphery of the cell, wherein any ray is intersected with the boundary of the cell nucleus region and the boundary of the cell region;
and setting segmentation points for line segments from the cell nucleus region boundary to the cell region boundary on each ray according to a set proportion, connecting the segmentation points of each ray together to form the outer boundary of the cell nucleus peripheral region, wherein the inner boundary of the cell nucleus peripheral region is the cell nucleus region boundary.
8. The method according to claim 7, wherein the size of the perinuclear region is adjusted by adjusting the value of the set ratio.
9. The method according to claim 1, wherein the step 5 of determining the lysosome distribution in the perinuclear region of the single cell based on the location and radius of the lysosome in the perinuclear region of the single cell and the single-cell region comprises:
determining the lysosome number of the perinuclear region of the single cell and the lysosome number of the single cell region according to the perinuclear region of the single cell and the lysosome position and radius of the single cell region;
a parameter indicative of lysosomal distribution in the perinuclear region of the single cell, comprising at least one of:
the ratio of the number of lysosomes in the perinuclear region of a single cell to the number of lysosomes in the single cell region;
the proportion of the area of lysosomes within the perinuclear region of a single cell to the area of lysosomes within that single cell region.
10. A computer storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the fully automated method of quantitative analysis of perinuclear lysosomal distribution according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111263263.8A CN113935983B (en) | 2021-10-28 | 2021-10-28 | Full-automatic quantitative analysis method for distribution of perinuclear lysosomes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111263263.8A CN113935983B (en) | 2021-10-28 | 2021-10-28 | Full-automatic quantitative analysis method for distribution of perinuclear lysosomes |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113935983A true CN113935983A (en) | 2022-01-14 |
| CN113935983B CN113935983B (en) | 2024-08-02 |
Family
ID=79284663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111263263.8A Active CN113935983B (en) | 2021-10-28 | 2021-10-28 | Full-automatic quantitative analysis method for distribution of perinuclear lysosomes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113935983B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101484806A (en) * | 2006-05-17 | 2009-07-15 | 协乐民公司 | Method for automated tissue analysis |
| US20100172569A1 (en) * | 2008-12-22 | 2010-07-08 | Olympus Corporation | Cell image analysis apparatus, cell image analysis method, and program |
| US20100279341A1 (en) * | 2000-10-24 | 2010-11-04 | Tissuegnostics Gmbh | Methods and system for analyzing cells |
| CN104813366A (en) * | 2012-12-04 | 2015-07-29 | 通用电气公司 | Systems and methods for using immunostaining mask to selectively refine ISH analysis results |
| CN104992435A (en) * | 2015-06-24 | 2015-10-21 | 广西师范大学 | Cervix uteri single cell image segmentation algorithm |
| CN109507406A (en) * | 2018-11-30 | 2019-03-22 | 苏州深析智能科技有限公司 | A kind of cellular morphology intelligent extract method, device and analysis method |
-
2021
- 2021-10-28 CN CN202111263263.8A patent/CN113935983B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100279341A1 (en) * | 2000-10-24 | 2010-11-04 | Tissuegnostics Gmbh | Methods and system for analyzing cells |
| CN101484806A (en) * | 2006-05-17 | 2009-07-15 | 协乐民公司 | Method for automated tissue analysis |
| US20100172569A1 (en) * | 2008-12-22 | 2010-07-08 | Olympus Corporation | Cell image analysis apparatus, cell image analysis method, and program |
| CN104813366A (en) * | 2012-12-04 | 2015-07-29 | 通用电气公司 | Systems and methods for using immunostaining mask to selectively refine ISH analysis results |
| CN104992435A (en) * | 2015-06-24 | 2015-10-21 | 广西师范大学 | Cervix uteri single cell image segmentation algorithm |
| CN109507406A (en) * | 2018-11-30 | 2019-03-22 | 苏州深析智能科技有限公司 | A kind of cellular morphology intelligent extract method, device and analysis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113935983B (en) | 2024-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11756198B1 (en) | Systems and methods for training a statistical model to predict tissue characteristics for a pathology image | |
| US11756318B2 (en) | Convolutional neural networks for locating objects of interest in images of biological samples | |
| US12229959B2 (en) | Systems and methods for determining cell number count in automated stereology z-stack images | |
| Godinez et al. | A multi-scale convolutional neural network for phenotyping high-content cellular images | |
| JP6192747B2 (en) | Machine learning system based on tissue objects for automatic scoring of digital hall slides | |
| WO2023284341A1 (en) | Deep learning-based context-sensitive detection method for urine formed element | |
| US20230186659A1 (en) | Machine learning models for cell localization and classification learned using repel coding | |
| CN113393443B (en) | HE pathological image cell nucleus segmentation method and system | |
| US20120207360A1 (en) | Systems and Methods for Object Identification | |
| Alegro et al. | Automating cell detection and classification in human brain fluorescent microscopy images using dictionary learning and sparse coding | |
| Dzyubachyk et al. | Automated analysis of time-lapse fluorescence microscopy images: from live cell images to intracellular foci | |
| CN118443413B (en) | Intelligent evaluation method and terminal for esophageal cancer tissue PD-L1 immunohistochemistry | |
| US20230420133A1 (en) | Methods and systems for training a model to diagnose abnormalittes in tissue samples | |
| US8693743B1 (en) | Analysis and display of multiple biomarker co-expression in cells and tissues | |
| CN113935983A (en) | Full-automatic quantitative analysis method for perinuclear lysosome distribution | |
| US20240412541A1 (en) | Automated identification of necrotic regions in digital images of multiplex immunofluorescence stained tissue | |
| US20240221360A1 (en) | Machine-learning techniques for predicting phenotypes in duplex digital pathology images | |
| Pati et al. | Mitosis detection under limited annotation: A joint learning approach | |
| CN113052806B (en) | Canceration degree grading system | |
| Alegro et al. | Deep learning for Alzheimer’s disease: Mapping large-scale histological tau protein for neuroimaging biomarker validation | |
| CN116704503B (en) | Machine learning-based three-dimensional lung slice cell classification and protein quantification method | |
| EP4246468A1 (en) | Method for training an artificial intelligence system, method for recognizing and classifying cells for pathological cell and tissue examination, classifying system, computer-readable medium | |
| US20250095852A1 (en) | Method for training an artificial intelligence system, method for recognizing and classifying cells for pathological cell and tissue examination, classifying system, computer-readable medium | |
| CN115049714A (en) | Method and system for registering in-situ hybridization fluorescence microscopic panoramic image and constructing microenvironment | |
| CN118334054A (en) | A method for evaluating resistance to soybean gray leaf spot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |