CN111767809A - Intelligent cell identification method based on laser confocal microscopy - Google Patents
Intelligent cell identification method based on laser confocal microscopy Download PDFInfo
- Publication number
- CN111767809A CN111767809A CN202010558516.3A CN202010558516A CN111767809A CN 111767809 A CN111767809 A CN 111767809A CN 202010558516 A CN202010558516 A CN 202010558516A CN 111767809 A CN111767809 A CN 111767809A
- Authority
- CN
- China
- Prior art keywords
- cells
- cell
- dimensional
- interpolation
- laser confocal
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- 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
- G06T2207/20—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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
-
- 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
- G06T2207/30096—Tumor; Lesion
Abstract
The invention discloses a cell intelligent identification method based on laser confocal microscopy, which comprises the following steps: (1) culturing and carrying out fluorescent staining treatment on the cells; (2) acquiring a slice image of a cell multi-fluorescence channel by using a laser confocal microscope; (3) carrying out graying, denoising and threshold segmentation on the slice image in sequence, picking out a cell region, and storing grayscale data of the same fluorescence channel into a three-dimensional matrix; (4) determining and marking each cell area by adopting an area random growth model; (5) carrying out interpolation three-dimensional reconstruction on the marked cells in sequence, and calculating three-dimensional morphological parameters of each cell; (6) repeating the processes from (1) to (5) by adopting different cells, and learning the three-dimensional morphological characteristics of the cells by using a machine learning method to finally realize the identification of various cells. The identification method provided by the invention can quickly and effectively identify various cells based on the three-dimensional morphological characteristics of the cells acquired by the laser confocal microscopy technology.
Description
Technical Field
The invention relates to the field of image processing and artificial intelligence recognition, in particular to a cell intelligent recognition method based on a laser confocal microscopy technology.
Background
The principle of confocal microscopy was proposed by the american scientist Marvin Minsky in 1957, and the american company Meridian realized commercialization of Laser Scanning Confocal Microscopy (LSCM) after 1987. The method has the advantages of high resolution, high contrast, nondestructive tomography and the like, and is widely applied to the research of biological tissues, cells and other various biological particles.
Different from the traditional optical microscope, the laser confocal microscope can acquire a two-dimensional plane image with extremely small depth of field when shooting a cell image. While continuously moving the sample in a direction perpendicular to the two-dimensional planar image, optical slices of the sample at different depths, i.e., slice images, can be obtained. Therefore, the imaging position and the shooting stepping distance can be selected according to the sample condition to obtain a series of two-dimensional slice images covering the whole cell, and then the three-dimensional information of the cell and the ultrastructure thereof is obtained through the three-dimensional reconstruction technology, so that people can more intuitively know the subcellular structure. Theoretically, three-dimensional morphological parameters of the cells are extracted, and a corresponding feature library is established, so that different cells can be distinguished.
However, in most of the current cell identification studies, two-dimensional optical images are used. For example: in 2003, Chengdong and the like firstly extract regional characteristics such as the area, the central moment, the circularity, the slenderness, the nuclear pulp ratio and the like of cells, then a single prototype mode characterization method is adopted, a minimum distance classifier in a cluster analysis technology is utilized, a discrimination function is constructed, and the discrimination function is used as a basis for identifying types, so that three types of squamous cells are effectively identified. Reference documents: cheng Yong, Fu De Sheng, cell identification method based on image region characteristics and realization [ J ]. atmospheric science report, 2003, 026(005): 712-. In 2006, the He Miao et al used a digital camera to photograph the cervical cell smear under an optical microscope, then extracted 15 morphological characteristic parameters and 12 colorimetric characteristic parameters of cervical cells and cell nuclei, and performed radial basis artificial neural network recognition on 700 cervical cells, and obtained a higher recognition rate. Reference documents: the application of radial basis artificial neural network in cervical cell image recognition [ J ]. the university of Chinese medicine, 2006, 035(001):79-81.
When a laser confocal microscope is adopted for three-dimensional reconstruction, the resolution of the obtained cells and the ultrastructure is higher. However, the collected multi-cell slice image generally needs to be manually selected and then three-dimensionally reconstructed. This process consumes a lot of time and effort of researchers, so intelligent segmentation and labeling of cell images is important, and an ideal cell labeling and counting method needs to be found. In addition, morphological parameter characteristics after three-dimensional reconstruction of the cells are more accurate to record the ultrastructure of the cells than two-dimensional morphological characteristics under an optical microscope. Therefore, when a machine learning method is used for identifying cells of different types or states, extraction of three-dimensional morphological parameters is expected to achieve a higher identification rate.
Disclosure of Invention
The invention aims to provide an intelligent cell identification method, which is used for intelligently segmenting a multi-cell image acquired by using a laser confocal microscopy technology and realizing identification of cells of different types or states.
In order to achieve the purpose, the technical scheme of the invention is as follows: an intelligent cell identification method based on laser confocal microscopy is characterized by comprising the following steps:
(1) culturing and carrying out fluorescent staining treatment on the cells;
(2) acquiring a slice image of a cell multi-fluorescence channel by using a laser confocal microscope;
(3) carrying out graying, denoising and threshold segmentation on the slice image in sequence, picking out a cell region, and storing grayscale data of the same fluorescence channel into a three-dimensional matrix;
(4) determining and marking each cell area by adopting an area random growth model;
(5) carrying out interpolation three-dimensional reconstruction on the marked cells in sequence, and calculating three-dimensional morphological parameters of each cell;
(6) repeating the processes from (1) to (5) by adopting different cells, and learning the three-dimensional morphological characteristics of the cells by using a machine learning method to finally realize the identification of various cells.
In the implementation of the region random growth model in the step (4), firstly, a cell region is picked out in the step (3) and a voxel is randomly selected, then the cell region grows randomly in each direction of nearest neighbor, whether the grown voxel is in the cell or not is judged until the grown voxel covers the whole cell, and then the cell is labeled. Finally, this process is repeated outside the area of labeled cells until all cells have been labeled.
And (5) performing interpolation three-dimensional reconstruction, namely inserting gray level images with the number being the ratio relationship into adjacent layer cutting images after the integration according to the ratio relationship between the stepping distance when the laser confocal microscope shoots the cell layer cutting images and the pixel size in the layer cutting images, and selecting one of nearest neighbor interpolation, linear interpolation, bilinear interpolation and bicubic interpolation by using a gray level interpolation method.
And (5) the three-dimensional morphological parameters of the cells comprise the volume, the membrane surface area, the membrane perimeter, the specific surface area, the isovolumetric sphere radius, the sphericity and the heterotypic index of the whole cells and the fluorescently-labeled ultrastructure thereof, and the volume ratio of the ultrastructure to the cells.
And (6) selecting the machine learning method from artificial neural network, support vector machine, linear and logistic regression, naive Bayes, decision tree and random forest.
Compared with the prior art, the invention mainly progresses as follows: (1) the laser confocal microscope is adopted to obtain the slice images of the cell multi-fluorescence channels, intelligent segmentation and labeling of the multi-cell laser confocal microscope images are achieved, a region random growth model is adopted in the segmentation process, each cell region and label are determined, voxels of different cells in a three-dimensional space can be accurately labeled, and a large amount of time consumed by manual selection of cells for three-dimensional reconstruction is saved. (2) The introduction of various three-dimensional morphological parameters and the combination of a machine learning method can realize the identification of cells of different types or states, and compared with the cell identification by using two-dimensional morphological parameters, the three-dimensional morphological parameters carry larger information amount and have higher identification accuracy.
Detailed Description
The present invention is further illustrated by the following specific embodiments.
The invention relates to a cell intelligent identification method based on laser confocal microscopy, which can realize the identification of cells of different types or states. The process of intelligently identifying human breast cancer cells and normal breast cells comprises the following steps:
(1) human breast cancer cells (MCF-7) were stored at 37 ℃ in 5% CO2In a humidified incubator, culture was performed using DMEM medium containing 10% Fetal Bovine Serum (FBS). Then, cells were stained with Syto-61, which demarcates the nuclear region and shows red color after fluorescence excitation, and M-7510, which were labeledColor, the latter marks mitochondria, and the color is green after fluorescence excitation;
(2) acquiring laminar-cut images of red and green fluorescence channels of the cells by using a laser confocal microscope, wherein the pixel size in a two-dimensional plane is 0.1 micrometer, the shooting stepping interval is 0.5 micrometer, and the slice images cover the whole cell area;
(3) and reading the slice image into a computer, and sequentially carrying out graying, denoising and threshold segmentation. The denoising method adopts a wiener filtering mode and judges whether the pixel is an impurity pixel according to the gray level overlapping condition of the red channel and the green channel. The threshold segmentation adopts an OTSU algorithm, combines morphological operation, segments all cell areas from the background, and sets the gray value of the background area to zero;
(4) each cell region was identified and labeled using a region random growth model. Firstly, selecting a voxel at random in the cell region selected in the step (3), then randomly growing in each direction of nearest neighbor, judging whether the grown voxel is in the cell until the whole cell is covered, and then labeling the cell. Finally, this process is repeated outside the area of labeled cells until all cells have been labeled;
(5) and (3) sequentially carrying out interpolation three-dimensional reconstruction on the marked cells, inserting 5 gray level images into adjacent tangent images, and selecting linear interpolation by a gray level interpolation method. And then calculating three-dimensional morphological parameters of each cell, including the volume of the cell, the nucleus and the mitochondria, the membrane surface area, the membrane perimeter, the specific surface area, the isovolumetric sphere radius, the sphericity and the heterotypic index, and the volume ratio of the nucleus, the mitochondria and the cell.
(6) And (3) repeating the processes from (1) to (5) by adopting normal human mammary cells, taking part of the obtained three-dimensional morphological parameter data as a training set, training by adopting a BP (back propagation) neural network model, and using other data as model tests to continuously improve the neural network model and improve the recognition rate, thereby finally realizing the accurate recognition of the normal human mammary cells and MCF-7 cells.
At present, the conventional method for cell identification adopts a two-dimensional morphological parameter provided by a traditional optical microscope and a machine learning method, and the identification accuracy is easily influenced by an observation angle. The method of the invention is adopted to carry out cell recognition on the breast cancer cells and the normal breast cells of the human body, the cell recognition accuracy can be improved to more than 98 percent, and the method can be applied to cell recognition of other two or more different types or states by improving a machine learning method.
The above detailed description is provided for the cell intelligent identification method based on the laser confocal microscopy, and a person skilled in the art may change the specific implementation and application scope according to the idea of the embodiment of the present invention. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (5)
1. An intelligent cell identification method based on laser confocal microscopy is characterized by comprising the following steps:
(1) culturing and carrying out fluorescent staining treatment on the cells;
(2) acquiring a slice image of a cell multi-fluorescence channel by using a laser confocal microscope;
(3) carrying out graying, denoising and threshold segmentation on the slice image in sequence, picking out a cell region, and storing grayscale data of the same fluorescence channel into a three-dimensional matrix;
(4) determining and marking each cell area by adopting an area random growth model;
(5) carrying out interpolation three-dimensional reconstruction on the marked cells in sequence, and calculating three-dimensional morphological parameters of each cell;
(6) repeating the processes from (1) to (5) by adopting different cells, and learning the three-dimensional morphological characteristics of the cells by using a machine learning method to finally realize the identification of various cells.
2. The method for intelligently identifying cells based on confocal laser microscopy as claimed in claim 1, wherein: in the implementation of the region random growth model in the step (4), firstly, a cell region is picked out in the step (3) and a voxel is randomly selected, then the cell region grows in each direction of nearest neighbor randomly, whether the grown voxel is in the cell or not is judged until the whole cell is covered, then the cell is labeled, and finally, the process is repeated outside the labeled cell region until all the cells are labeled.
3. The method for intelligently identifying cells based on confocal laser microscopy as claimed in claim 1, wherein: and (5) performing interpolation three-dimensional reconstruction, namely inserting gray level images with the number being the ratio relationship into adjacent layer cutting images after the integration according to the ratio relationship between the stepping distance when the laser confocal microscope shoots the cell layer cutting images and the pixel size in the layer cutting images, and selecting one of nearest neighbor interpolation, linear interpolation, bilinear interpolation and bicubic interpolation by using a gray level interpolation method.
4. The method for intelligently identifying cells based on confocal laser microscopy as claimed in claim 1, wherein: and (5) the three-dimensional morphological parameters of the cells comprise the volume, the membrane surface area, the membrane perimeter, the specific surface area, the isovolumetric sphere radius, the sphericity and the heterotypic index of the whole cells and the fluorescently-labeled ultrastructure thereof, and the volume ratio of the ultrastructure to the cells.
5. The method for intelligently identifying cells based on confocal laser microscopy as claimed in claim 1, wherein: and (6) selecting the machine learning method from artificial neural network, support vector machine, linear and logistic regression, naive Bayes, decision tree and random forest.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010558516.3A CN111767809A (en) | 2020-06-18 | 2020-06-18 | Intelligent cell identification method based on laser confocal microscopy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010558516.3A CN111767809A (en) | 2020-06-18 | 2020-06-18 | Intelligent cell identification method based on laser confocal microscopy |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111767809A true CN111767809A (en) | 2020-10-13 |
Family
ID=72721087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010558516.3A Pending CN111767809A (en) | 2020-06-18 | 2020-06-18 | Intelligent cell identification method based on laser confocal microscopy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111767809A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112945835A (en) * | 2021-01-22 | 2021-06-11 | 西安交通大学 | Cell accurate three-dimensional shape recovery method based on depth change point diffusion function |
| CN113160392A (en) * | 2021-05-14 | 2021-07-23 | 电子科技大学成都学院 | Optical building target three-dimensional reconstruction method based on deep neural network |
| CN116778483A (en) * | 2023-08-25 | 2023-09-19 | 泰州骆华生物科技有限公司 | Cell death type identification method based on reflection confocal microscope technology |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103994964A (en) * | 2014-05-23 | 2014-08-20 | 天津大学 | Quantitative analysis method aiming at apoptotic cell morphology of fluorescence microscopic image |
| CN106056118A (en) * | 2016-06-12 | 2016-10-26 | 合肥工业大学 | Recognition and counting method for cells |
| US20190130161A1 (en) * | 2017-10-31 | 2019-05-02 | Chung Yuan Christian University | Method and apparatus for image processing and visualization for analyzing cell kinematics in cell culture |
| CN110211108A (en) * | 2019-05-29 | 2019-09-06 | 武汉兰丁医学高科技有限公司 | A kind of novel abnormal cervical cells automatic identifying method based on Feulgen colouring method |
| CN110264472A (en) * | 2019-05-29 | 2019-09-20 | 浙江科技学院 | A kind of nearly subcutaneous cell noninvasive imaging method of fruit based on SD-OCT |
| CN110516584A (en) * | 2019-08-22 | 2019-11-29 | 杭州图谱光电科技有限公司 | A kind of Auto-counting of Cells method based on dynamic learning of microscope |
-
2020
- 2020-06-18 CN CN202010558516.3A patent/CN111767809A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103994964A (en) * | 2014-05-23 | 2014-08-20 | 天津大学 | Quantitative analysis method aiming at apoptotic cell morphology of fluorescence microscopic image |
| CN106056118A (en) * | 2016-06-12 | 2016-10-26 | 合肥工业大学 | Recognition and counting method for cells |
| US20190130161A1 (en) * | 2017-10-31 | 2019-05-02 | Chung Yuan Christian University | Method and apparatus for image processing and visualization for analyzing cell kinematics in cell culture |
| CN110211108A (en) * | 2019-05-29 | 2019-09-06 | 武汉兰丁医学高科技有限公司 | A kind of novel abnormal cervical cells automatic identifying method based on Feulgen colouring method |
| CN110264472A (en) * | 2019-05-29 | 2019-09-20 | 浙江科技学院 | A kind of nearly subcutaneous cell noninvasive imaging method of fruit based on SD-OCT |
| CN110516584A (en) * | 2019-08-22 | 2019-11-29 | 杭州图谱光电科技有限公司 | A kind of Auto-counting of Cells method based on dynamic learning of microscope |
Non-Patent Citations (3)
| Title |
|---|
| YUHUA WEN 等: "Quantitative analysis and comparison of 3D morphology between viable and apoptotic MCF-7 breast cancer cells and characterization of nuclear fragmentation", 《PLOS ONE》 * |
| 曹玉珍 等: "BM-MSCs的CNN 特征映射与活性评价模型研究", 《天津大学学报(自然科学与工程技术版)》 * |
| 汪舒亭: "人体细胞光散射特性及衍射成像研究", 《中国优秀博硕士学位论文全文数据库(硕士)基础科学辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112945835A (en) * | 2021-01-22 | 2021-06-11 | 西安交通大学 | Cell accurate three-dimensional shape recovery method based on depth change point diffusion function |
| CN113160392A (en) * | 2021-05-14 | 2021-07-23 | 电子科技大学成都学院 | Optical building target three-dimensional reconstruction method based on deep neural network |
| CN113160392B (en) * | 2021-05-14 | 2022-03-01 | 电子科技大学成都学院 | Optical building target three-dimensional reconstruction method based on deep neural network |
| CN116778483A (en) * | 2023-08-25 | 2023-09-19 | 泰州骆华生物科技有限公司 | Cell death type identification method based on reflection confocal microscope technology |
| CN116778483B (en) * | 2023-08-25 | 2023-10-31 | 泰州骆华生物科技有限公司 | Cell death type identification method based on reflection confocal microscope technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111767809A (en) | Intelligent cell identification method based on laser confocal microscopy | |
| CN109272492B (en) | Method and system for processing cytopathology smear | |
| Song et al. | A deep learning based framework for accurate segmentation of cervical cytoplasm and nuclei | |
| CN111986150B (en) | The method comprises the following steps of: digital number pathological image Interactive annotation refining method | |
| US7587078B2 (en) | Automated image analysis | |
| CN110136149A (en) | Leucocyte positioning and dividing method based on deep neural network | |
| CN102297873A (en) | Method for identifying cancer cell images by soft X-ray microscopic imaging | |
| WO2023078041A1 (en) | Tumor tissue pathological classification system and method based on adaptive proportional learning | |
| Signolle et al. | Wavelet-based multiscale texture segmentation: Application to stromal compartment characterization on virtual slides | |
| CN115100474B (en) | Thyroid gland puncture image classification method based on topological feature analysis | |
| He et al. | Local and global Gaussian mixture models for hematoxylin and eosin stained histology image segmentation | |
| Pourakpour et al. | Automated mitosis detection based on combination of effective textural and morphological features from breast cancer histology slide images | |
| Zhang et al. | Research on application of classification model based on stack generalization in staging of cervical tissue pathological images | |
| CN115206495A (en) | Renal cancer pathological image analysis method and system based on CoAtNet deep learning and intelligent microscopic device | |
| CN111210447A (en) | Method and terminal for hierarchical segmentation of hematoxylin-eosin staining pathological image | |
| Nazeran et al. | Biomedical image processing in pathology: a review | |
| Viswanath et al. | Grading of mammalian cumulus oocyte complexes using machine learning for in vitro embryo culture | |
| Chaudhury et al. | Diagnosis of invasive ductal carcinoma using image processing techniques | |
| Tosun et al. | Unsupervised tissue image segmentation through object-oriented texture | |
| Liimatainen et al. | Supervised method for cell counting from bright field focus stacks | |
| CN115274093A (en) | Method and system for generating reference pathology data set containing automatic labeling file | |
| RU2785607C1 (en) | Method for recognising the structure of blood and bone marrow blast nuclei | |
| Niemisto et al. | A K-means segmentation method for finding 2-D object areas based on 3-D image stacks obtained by confocal microscopy | |
| Pławiak-Mowna et al. | On effectiveness of human cell nuclei detection depending on digital image color representation | |
| Rojas et al. | Improved segmentation of meristem cells by an automated optimisation algorithm |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201013 |