CN110490971B - Method for reconstructing cell dynamic characteristic three-dimensional image under biological microscope - Google Patents
Method for reconstructing cell dynamic characteristic three-dimensional image under biological microscope Download PDFInfo
- Publication number
- CN110490971B CN110490971B CN201910758074.4A CN201910758074A CN110490971B CN 110490971 B CN110490971 B CN 110490971B CN 201910758074 A CN201910758074 A CN 201910758074A CN 110490971 B CN110490971 B CN 110490971B
- Authority
- CN
- China
- Prior art keywords
- cell
- image
- dimensional
- dynamic
- reconstructing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06T2207/10061—Microscopic image from scanning electron microscope
-
- 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)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Graphics (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method for reconstructing a three-dimensional image of dynamic characteristics of cells under a biological microscope, which comprises the following steps: acquiring a reference image of an optical microscope; debugging the optical axis of the optical microscope to the central line, preparing a cell sample, and regularly shooting a cell microscopic image; and step three, performing three-dimensional reconstruction on the two qualified two-dimensional cell microscopic images. The implementation method of the third step specifically comprises the following steps: (1) dynamic object segmentation and position calculation; (2) constructing a disparity map; and (3) constructing a dynamic cell image by using the three-dimensional characteristics. The invention uses a stereo dynamic object segmentation method, determines the optical axis and correction of the deviation, and estimates the 3D characteristics of the cell by combining the disparity map and the 2D characteristics of the initial image so as to complete the reconstruction of the dynamic characteristic three-dimensional image of the cell. The invention can quickly determine the actual value of the distance and the outline of the cell, and retain the value of all source information, thereby effectively solving the problem analysis of the power.
Description
Technical Field
The invention belongs to the technical field of medical microscope image processing, and particularly relates to a method for reconstructing a three-dimensional image of dynamic cell characteristics under a biological microscope.
Background
Cells are the basic unit of life formation, and life research needs to be started from cells. The general biological microscopy laboratory in colleges and universities can be used for cell research, but the workload of manually observing the number, the shapes and the like of cells is too large, and since the modern computer appears, the existing optical microscopy laboratory can be refitted by using a computer image processing technology to realize cell research automation, so that a large amount of experimental funds are saved, and the existing laboratory resources are fully utilized.
Blood cell morphology examination is an effective means for diagnosing diseases, and is one of the important contents of blood routine examination. In recent years, with the improvement of clinical diagnosis technology and the popularization and use of instruments such as blood analyzers, the clinical laboratory works more conveniently and efficiently, however, the detection results of the analyzers have certain differences, and the accuracy of the diagnosis results can be effectively improved through the traditional blood smear staining examination. At present, the test results of a fully automatic blood analyzer can provide more than ten or even more than twenty test indexes, but the morphology of peripheral blood cells is still regarded clinically as an indispensable test item.
With the continuous development of medical and health technologies in China, the application of blood cell analyzer equipment is more and more extensive, and the efficiency and the accuracy of diagnosis can be further improved by analyzing a blood sample of a patient and combining related analysis results. Currently, the blood cell analyzer used by medical institutions in China is still to be perfected in function, cells have diversified morphological characteristics, and the specific structure and morphology of the cells cannot be deeply analyzed in an instrument detection mode. Therefore, the blood cell sample of the patient cannot meet the high-quality diagnosis work target simply by analyzing the blood cell sample through an instrument detection mode, the experiment result and the clinical performance of the patient are deeply researched and analyzed in combination with an artificial microscope detection mode, the accuracy of diagnosis work can be further improved in cooperation with the result of the artificial microscope detection, and the microscope reconstructed image at the moment can play a role.
With the widespread use of image processing and object recognition techniques in the field of biomedical engineering, microscopy has also gradually transitioned from manual to automated analysis. In recent years, an automatic detection system has become a new research hotspot in the biomedical field, and the image processing speed is greatly improved by utilizing the high performance of a computer, so that the clinical pathological judgment time is greatly shortened.
The automatic microscopy system can be generally divided into image acquisition, image processing, identification and other processes, and in consideration of the characteristics of complex background, more interference, mutual adhesion of cells, different non-homogeneous cell forms and the like of a hemocyte microscopy image, the problem of 3D image reconstruction of a moving object of biological cells has several problems and needs to be solved by a specific method. This is mainly diffusion, low contrast boundary features, and changes in the complex cellular object itself. Furthermore, this type of scenario is dynamic. Therefore, after analyzing the system characteristics, the stereoscopic monitoring of medical objects requires the development of new methods and concepts for 3D reconstruction. The invention develops research deeply into a blood cell microscopic image segmentation method, and reconstructs a 3D image by using a parallax image method.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a method for reconstructing a three-dimensional image of dynamic characteristics of cells under a biological microscope, which can quickly determine the actual value of the distance and the outline of the cells, and retain the values of all source information, thereby effectively solving the dynamic problem analysis.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: a method for reconstructing a three-dimensional image of dynamic characteristics of cells under a biological microscope comprises the following steps:
the method comprises the following steps: acquiring a reference image of an optical microscope, calculating anisotropic characteristics of brightness gradient for each local area of the acquired microscopic image, scanning the local area, acquiring an anisotropic distribution map of the area, and then creating an anisotropic map for the whole image, namely the reference image of the optical microscope, wherein any optical system has an error, namely a system error, and the microscopic image needs to be corrected in order to obtain an optimal optical image of the system, so that the reference image of the microscopic image needs to be acquired, and the microscopic image of the cell can be acquired more accurately through correction;
step two: debugging the optical axis of the optical microscope to a central line, preparing a cell sample, and regularly shooting a cell microscopic image; before observing the sample, the optical axis of the microscope should be adjusted to the central line, otherwise, the observation effect of the cell sample is affected due to uneven brightness of the visual field, and the phenomenon that the image is dark at one side and light at the other side is generated during shooting. Because the central lines of the optical paths of the objective lenses with different multiples are not on the same line, the optical axis of the objective lens is required to be readjusted once when the objective lens is replaced;
step three: and performing three-dimensional reconstruction on the two qualified two-dimensional cell microscopic images, and regarding the same target cell image, regarding the two images shot by the left camera and the right camera as qualified two-dimensional cell images.
The implementation method of the third step specifically comprises the following steps:
(1) Dynamic target segmentation and position calculation;
(2) Constructing a disparity map;
(3) And constructing a dynamic cell image by using the three-dimensional characteristics.
Preferably, step two is specifically obtained by: placing a chemokine in the lower chamber; observing the concentration gradient change of the upper chamber and the lower chamber; the migration images of the cells were taken at intervals of several minutes to obtain microscopic images of the cells.
Further, the dynamic object segmentation and position calculation are obtained by: creating a background image, firstly representing the sequence image as a cube, then constructing a contour for the brightness of each pixel, penetrating all image sequences along a straight line, then constructing a histogram according to the values, determining a median, and distributing the pixels corresponding to the median to obtain the background image; determining the position of the dynamic target; the background image and the current image are subjected to difference operation, and when the background image is known, the position of the dynamic target is easy to identify, so that the difference operation of the images can be calculated.
Furthermore, the determination of the position of the dynamic object uses the robust threshold segmentation method in the binarization method, which aims to select a threshold value that minimizes the proportion of the combined chromatic dispersion between classes that define a part of the histogram above the threshold value, to determine the geometric properties according to their classification, to remove features outside the geometric properties at certain intervals, and to correct the defects of the object shape, mainly by morphological operations and filling.
Preferably, the algorithms used for constructing the disparity map include shadow removal, watershed algorithms and mathematical morphology.
Preferably, the three-dimensional features used to obtain the dynamic cell image using the three-dimensional features include the volume, outer surface and contour of the cell.
Further, the volume of the cell is equal to the sum of all pixels at the horizontal and vertical scale factor increments, the determination of the volume is done by first determining the coordinates of the centroid, the horizontal coordinate (X) of the centroid is defined as the sum of each pixel at the X coordinate divided by the number of pixels of the target, and the vertical coordinate (Y) of the centroid is also determined in the same manner:
n is the total number of pixels of the target cell.
The Z coordinate typically corresponds to the depth of the image space. The centroid coordinates are the plane between the upper and lower parts, since the cell model of this design is symmetric with respect to the boundary, and the symmetry properties are used to estimate the centroid, which is determined by the upper and lower parts, and the volume of the entire cell. According to the symmetry principle, the centroid of the lower cell is equal to the centroid of the upper cell. Thus, in the disparity map, the Z-coordinate of the centroid corresponds to the average of the cell boundaries:
where n is border Is the number of pixels at the cell boundary, I wh Is the intensity of each pixel on the disparity map.
The volume of the cell is equal to the sum of all pixels over the horizontal and vertical scale factor increments:
sw, sh are suitable scaling factors.
The volume is made up of two parts, the volume of the upper cell is equal to the weighted area of the statistical intensity:
I border is the average pixel of the object boundary.
The volume of the lower cells can be calculated according to the ratio:
I max is the largest pixel of the disparity map.
Thus, the total volume can be estimated:
Volume total ＝Volume 1 +Volume 2
further, the outer surface of the cell corresponds to Zhou Changzhi and within the depth coordinate cut-out region of each image space; the two types of perimeters are 4 connected boundary points (N4) and 8 connected boundary points (N8). In practice, 8-connectivity is defined as the basis for the measurement. 8 connectivity, the number of diagonal connections N4-N8, and the remaining N8- (N4-N8) pixels define the general perimeter as follows:
a rough estimate of a surface may use the inter-layer distance, and a more accurate estimate is to construct an equidistant surface in order to find the sum of all perimeters. First, the outer surface is calculated, starting from the boundary of the cell, the surface of this part is determined as the sum of the perimeters of all layers:
second part surface the same principle of symmetry is used to define the second part of the surface:
thus, the overall surface is estimated as follows:
Surface total ＝Surface 1 +Surface 2
further, the contour calculation of the cell adopts a newly generated shape convex hull algorithm; to normalize the profile representation, it is necessary to use convex features. These layers are characterized by a relationship defined by a convex boundary with a perimeter of:
the peripheral surface can be calculated using the same alternative principle, and therefore, can be described as convex cells in 3D mode:
in particular, this property can also describe a smooth profile.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention can quickly determine the actual value of the distance and the outline of the cell and retain the value of all source information.
2. The invention can effectively solve the problem of power analysis.
3. In the present invention, the biological cell objects are not static, they are moved in the field of view, and projections can be obtained from different angles.
Drawings
Fig. 1 is a flow chart diagram of the present invention.
Fig. 2 (a) and (b) are schematic diagrams of the detection steps of the background pixel value of the present invention.
FIGS. 3 (a), (b), and (c) are schematic diagrams showing the morphological processing of the cell image of the present invention.
Fig. 4 is a rough disparity diagram illustration of the present invention.
Fig. 5 is a schematic view of the improved parallax after the morphological processing of the present invention.
FIG. 6 is a schematic representation of the effect of 3D reconstitution of cells of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples.
As shown in FIG. 1, the invention provides a method for reconstructing a three-dimensional image of dynamic characteristics of cells under a biological microscope, which comprises the following steps:
the method comprises the following steps: acquiring a reference image of an optical microscope, calculating anisotropic characteristics of brightness gradient for each local area of the acquired microscopic image, scanning the local area, acquiring an anisotropic distribution map of the area, and then creating an anisotropic map for the whole image, namely the reference image of the optical microscope, wherein any optical system has an error, namely a system error, and the microscopic image needs to be corrected in order to obtain an optimal optical image of the system, so that the reference image of the microscopic image needs to be acquired, and the microscopic image of the cell can be acquired more accurately through correction;
step two: debugging the optical axis of the optical microscope to a central line, preparing a cell sample, and regularly shooting a cell microscopic image; before observing the sample, the optical axis of the microscope should be adjusted to the central line, otherwise, the observation effect of the cell sample is affected due to uneven brightness of the visual field, and the phenomenon that the image is dark at one side and light at the other side is generated during shooting. The specific debugging method is that the field diaphragm at the light source outlet is reduced first, and the height of the condenser is raised and lowered until the diaphragm edge is clear. Then gradually enlarging the diaphragm to the edge of the visual field, observing whether the distance between the edge of the diaphragm and the periphery of the visual field is equal, and rotating two obliquely-arranged screw rods on the condenser for adjustment if the optical axis is found to have deviation. Because the central lines of the optical paths of the objective lenses with different multiples are not on the same line, the optical axis of the objective lens is required to be readjusted once when the objective lens is replaced;
step three: and performing three-dimensional reconstruction on the two qualified two-dimensional cell microscopic images, wherein the two images shot by the left camera and the right camera are qualified two-dimensional cell images for the same target cell.
The implementation method of the third step specifically comprises the following steps:
(1) The method comprises the steps of dynamic target segmentation and position calculation, wherein the dynamic target segmentation and position calculation comprises the steps of creating a background image, as shown in fig. 2 (a), firstly representing a sequence image as a cube, then constructing a contour for the brightness of each pixel, passing through all image sequences along a straight line, as shown in fig. 2 (b), then constructing a histogram according to the pixel values, determining a median, and distributing pixels corresponding to the median to obtain the background image, wherein when noise is applied to the image, a camera can change in a manner of facing the image, and when the noise is small, the background image can be removed through median filtering; determining the position of the dynamic target; the background image and the current image are subjected to difference operation, when the background image is known, the position of the dynamic target is easy to identify, and therefore, only the difference operation of the image needs to be calculated;
determining the position of a dynamic object by using the robust threshold segmentation method in the binarization method, such as the binary image after threshold shown in fig. 3 (a), the method aims to select a threshold, minimize the ratio of the combined dispersion between classes, which define a part of the histogram on the threshold, and determine the geometric attributes according to their classification, such as the image after deleting small objects shown in fig. 3 (b), and the features outside the geometric characteristics at certain intervals are removed, such as the image shown in fig. 3 (c), and the defects of the object shape are corrected mainly by means of morphological operations and filling, so as to obtain a morphologically modified cell image;
(2) Constructing a rough disparity map and an improved disparity map after morphological processing, wherein the disparity map is constructed by an algorithm comprising a shadow removal method to obtain the rough disparity map, and obtaining the improved disparity map through a watershed algorithm and a mathematical morphology operation, and a cell 3D map can be obtained by reconstructing the disparity map for multiple times, and for specific cells, because of different motions and increment speeds, if the cell 3D reconstructed image has an unsatisfactory three-dimensional effect after the set number of times of reconstructing the disparity map is reached, the number of times of reconstructing the disparity map can be properly increased or decreased to generate a 3D cell image meeting the requirements of a client;
(3) A dynamic cell image is constructed using three-dimensional features, the used features including the volume, outer surface and contour of the cell, to obtain a 3D effect map after cell reconstruction, as shown in fig. 6.
The second step is specifically obtained in the following way: placing a chemokine in the lower chamber; observing the concentration gradient change of the upper chamber and the lower chamber; shooting the migration image of the cell at intervals of 15 minutes to obtain a cell microscopic image; although the membrane between the upper and lower chambers has small pores, the pore size is smaller, generally smaller than 8um, so that the lower chamber contains chemokines, and in the absence of the upper chamber, a concentration gradient initially exists between the upper and lower chambers, and when the cells move downward upon receiving a signal, the concentration gradient between the upper and lower chambers gradually decreases with time, and finally reaches the same level with the upper and lower chambers, and then the migration of the cells is stopped. Therefore, the test time is generally 6-10 hours, which is not suitable for being too long, and the cell movement speed is different under the action of the chemotactic factors, so that if the shooting time is too long, some cells die, and the die cells do not meet the original cell requirement of 3D reconstruction.
The dynamic cell image is constructed by utilizing the three-dimensional characteristics, and the specific calculation method comprises the following steps:
the volume of the cell is equal to the sum of all pixels over the horizontal and vertical scale factor increments:
sw, sh are suitable scaling factors. Specifically, the calculation is performed in the following manner:
first, the coordinates of the centroid need to be determined, with the horizontal coordinate (X) of the centroid defined as the sum of each pixel of the X coordinate divided by the number of pixels of the target, and the vertical coordinate (Y) of the centroid also determined in the same manner:
n is the total number of pixels of the target cell.
The Z coordinate typically corresponds to the depth of the image space. The centroid coordinates are the plane between the upper and lower parts, since the cell model of this design is symmetric with respect to the boundary, and the symmetry properties are used to estimate the centroid, which is determined by the upper and lower parts, and the volume of the entire cell. According to the symmetry principle, the centroid of the lower cell is equal to the centroid of the upper cell. Thus, in the disparity map, the Z-coordinate of the centroid corresponds to the average of the cell boundaries:
where n is border Is the number of pixels at the cell boundary, I wh Is the intensity of each pixel on the disparity map.
The volume is made up of two parts, the volume of the upper cell is equal to the weighted area of the statistical intensity:
I border is the average pixel of the object boundary.
The volume of the lower cells can be calculated according to the ratio:
I max is the largest pixel of the disparity map.
Thus, the total volume can be estimated:
Volume total ＝Volume 1 +Volume 2
the outer surface of the cell corresponds to Zhou Changzhi within the depth coordinate cut-out region of each image space and is defined as the distance between boundary pixels and the target contour feature length. The two types of perimeters are 4 connected boundary points (N4) and 8 connected boundary points (N8). In practice, 8-connectivity is defined as the basis for the measurement. Under 8 communication, the number of diagonal connections N4-N8 and the remaining number of N8- (N4-N8) pixels define a general perimeter as follows:
a rough estimate of a surface may use the inter-layer distance, and a more accurate estimate is to construct an equidistant surface in order to find the sum of all perimeters. First, the outer surface is calculated, starting from the boundary of the cell, and the surface of this part is determined as the sum of the perimeters of all layers:
second part surface the same principle of symmetry is used to define the second part of the surface:
thus, the overall surface is estimated as follows:
Surface total ＝Surface 1 +Surface 2
secondly, calculating the outline of the cell by adopting a newly generated shape convex hull algorithm; to normalize the profile representation, it is necessary to use convex features. These layers are characterized by a relationship of convex boundaries with a perimeter of:
the peripheral surface can be calculated using the same alternative principle, and therefore can be described as convex cells in 3D mode:
in particular, this property can also describe a smooth profile.
In this embodiment, the interval time for taking the cell microscopic image is only for taking the qualified cell microscopic image, and may be modified or adjusted according to the practice.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the present invention in any way. Any simple modification, change and equivalent changes of the above embodiments according to the technical essence of the invention are still within the protection scope of the technical solution of the invention.
Claims (8)
1. A method for reconstructing dynamic cell characteristic three-dimensional images under a biological microscope is characterized by comprising the following steps:
the method comprises the following steps: acquiring a reference image of the optical microscope for correcting the microscopic image, thereby acquiring an accurate microscopic image of the cell;
step two: debugging the optical axis of the optical microscope to a central line, preparing a cell sample, and regularly shooting a cell microscopic image;
and the cell microscopic image is obtained by: placing a chemokine in the lower chamber; observing the concentration gradient change of the upper chamber and the lower chamber; shooting the migration image of the cell at intervals of a plurality of minutes to obtain a cell microscopic image;
step three: the three-dimensional reconstruction method is used for performing three-dimensional reconstruction on two qualified two-dimensional cell microscopic images, and comprises the following specific steps:
dynamic target segmentation and position calculation;
constructing a disparity map;
and constructing a dynamic cell image by using the three-dimensional characteristics.
2. The method for reconstructing the dynamic cell characteristic three-dimensional image under the biological microscope as claimed in claim 1, wherein the dynamic object segmentation and position calculation are obtained by: creating a background image; determining the position of the dynamic target; and performing difference operation on the background image and the current image.
3. The method for reconstructing the three-dimensional image of the dynamic cell features under the biomicroscope according to claim 2, wherein the determining the position of the dynamic target uses the Otsu threshold segmentation method in the binarization method.
4. The method for reconstructing the three-dimensional image of the dynamic features of the cells under the biological microscope as claimed in claim 1, wherein the algorithm for constructing the disparity map comprises a shadow removal method, a watershed algorithm and mathematical morphology.
5. The method of claim 1, wherein the three-dimensional features used for obtaining the dynamic cell image by using the three-dimensional features include volume, outer surface and contour of the cell.
6. The method of claim 5, wherein the volume of the cell is equal to the sum of all pixels in horizontal and vertical scale factor increments.
7. The method of claim 5, wherein the outer surface of the cell corresponds to Zhou Changzhi within the depth coordinate cut-out region of each image space.
8. The method of claim 5, wherein the contour calculation of the cell adopts a newly generated convex hull shape algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910758074.4A CN110490971B (en) | 2019-08-16 | 2019-08-16 | Method for reconstructing cell dynamic characteristic three-dimensional image under biological microscope |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910758074.4A CN110490971B (en) | 2019-08-16 | 2019-08-16 | Method for reconstructing cell dynamic characteristic three-dimensional image under biological microscope |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110490971A CN110490971A (en) | 2019-11-22 |
| CN110490971B true CN110490971B (en) | 2023-01-31 |
Family
ID=68549813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910758074.4A Active CN110490971B (en) | 2019-08-16 | 2019-08-16 | Method for reconstructing cell dynamic characteristic three-dimensional image under biological microscope |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110490971B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111312373B (en) * | 2020-01-19 | 2023-08-18 | 浙江树人学院（浙江树人大学） | Automatic labeling method for PET/CT image fusion |
| CN112837261B (en) * | 2020-07-22 | 2022-11-11 | 杭州思柏信息技术有限公司 | Cell data labeling method and system integrating scanning data and optical image display |
| US11416990B1 (en) * | 2020-10-18 | 2022-08-16 | Aixmed, Inc. | Method and system to obtain cytology image in cytopathology |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5106966B2 (en) * | 2007-09-28 | 2012-12-26 | パナソニックヘルスケア株式会社 | Culture observation system |
| WO2009121137A1 (en) * | 2008-04-01 | 2009-10-08 | Lilian Soon | Method for imaging directional cell migration |
| US9418421B1 (en) * | 2015-09-26 | 2016-08-16 | Nastaran Neishaboori | Automation of biopsy specimen handling |
| WO2017151978A1 (en) * | 2016-03-02 | 2017-09-08 | Arizona Boad Of Regents On Behalf Of Arizona State University | Live-cell computed tomography |
| US10423819B2 (en) * | 2017-10-31 | 2019-09-24 | Chung Yuan Christian University | Method and apparatus for image processing and visualization for analyzing cell kinematics in cell culture |
-
2019
- 2019-08-16 CN CN201910758074.4A patent/CN110490971B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110490971A (en) | 2019-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110490971B (en) | Method for reconstructing cell dynamic characteristic three-dimensional image under biological microscope | |
| CN109272492B (en) | Method and system for processing cytopathology smear | |
| Zanella et al. | Cells segmentation from 3-D confocal images of early zebrafish embryogenesis | |
| CN105719295B (en) | A kind of intracranial hemorrhage region segmentation method and system based on three-dimensional super voxel | |
| Agarwal et al. | Automatic glaucoma detection using adaptive threshold based technique in fundus image | |
| CN105261015B (en) | Eye fundus image blood vessel automatic division method based on Gabor filter | |
| CN101238987A (en) | Processing method of CT cerebral hemorrhage image | |
| KR102140385B1 (en) | Cell-zone labeling apparatus and cell-zone detecting system including the same apparatus | |
| Yan et al. | Automated and accurate detection of soma location and surface morphology in large-scale 3D neuron images | |
| CN106157279A (en) | Eye fundus image lesion detection method based on morphological segment | |
| CN113781455B (en) | Cervical cell image anomaly detection method, device, equipment and medium | |
| CN110517273B (en) | Cytology image segmentation method based on dynamic gradient threshold | |
| CN116645384A (en) | Stem cell area rapid segmentation method based on artificial intelligence | |
| CN110264472B (en) | SD-OCT-based fruit hypodermal cell nondestructive imaging method | |
| CN106991660B (en) | The three dimensional ultrasonic image data methods of sampling decomposed based on modified Octree | |
| CN105869169A (en) | Automatic dividing method of tumor issue micro array image | |
| Su et al. | Learning based automatic detection of myonuclei in isolated single skeletal muscle fibers using multi-focus image fusion | |
| CN111311628A (en) | Full-automatic high-performance leukocyte segmentation method | |
| Ferro et al. | Multinuclear cell analysis using laplacian of gaussian and delaunay graphs | |
| CN104263804A (en) | Myocardial cell mechanical beating detection method based on optical image analysis | |
| Bhuiyan et al. | Drusen quantification for early identification of age related macular degeneration (AMD) using color fundus imaging | |
| CN112945835A (en) | Cell accurate three-dimensional shape recovery method based on depth change point diffusion function | |
| Ning et al. | Research on cell counting method based on flood fill algorithm | |
| On et al. | 3D Reconstruction of Phase Contrast Images Using Focus Measures | |
| CN110246125A (en) | Teat placement automatic testing method based on ABUS coronal image |
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 |