CN115165710B - Rapid scanning method and device for cells - Google Patents
Rapid scanning method and device for cells Download PDFInfo
- Publication number
- CN115165710B CN115165710B CN202211092303.1A CN202211092303A CN115165710B CN 115165710 B CN115165710 B CN 115165710B CN 202211092303 A CN202211092303 A CN 202211092303A CN 115165710 B CN115165710 B CN 115165710B
- Authority
- CN
- China
- Prior art keywords
- sub
- scanning
- region
- subregion
- determining
- 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
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/10—Investigating individual particles
- G01N2015/1006—Investigating individual particles for cytology
-
- G—PHYSICS
- 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
Abstract
The application provides a rapid scanning method and a rapid scanning device for cells, wherein the method comprises the following steps: the method comprises the steps of pre-scanning a target scanning area to obtain a first cell image, determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size, determining a secondary scanning area based on cell association characteristics in each target subregion, clustering the secondary scanning area based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subregions, determining the scanning priority of each secondary scanning subregion based on the number of fields of view in each secondary scanning subregion and the mean variance of the corresponding Gaussian mixture model, and sequentially scanning the secondary scanning subregions based on the scanning priority to obtain a second cell image, so that the acquisition efficiency of the cell image meeting the detection requirement is improved.
Description
Technical Field
The present application relates to the field of image processing technologies, and in particular, to a method and an apparatus for rapidly scanning cells.
Background
The correlation between the abnormal cell (CAC) and the lung nodule is analyzed based on the characteristics of the CAC in the peripheral blood, and is an important subject of clinical research at present. To perform this research, it is first necessary to accurately find a very small number of CACs from a large number of background blood cells, and in the prior art, a microscopic device is usually used to obtain a cell image corresponding to a blood sample, and then an artificial intelligence algorithm is used to analyze the cell image to detect CACs.
However, CAC assays have corresponding requirements for the number of cells in the cell image and the cell type. The prior art generally uses a microscope to scan blood samples field by field according to a preset moving route to obtain cell images for CAC detection. However, since the cells in the blood sample are not uniformly distributed, some regions even have no cells, and abnormal cells clumping that interfere with CAC detection also exist in the sample, the efficiency of obtaining a cell image is very low by scanning according to a preset moving route, and the efficiency and accuracy of subsequent CAC detection are reduced because some cells collected by the scanning method are abnormal cells clumping.
Disclosure of Invention
The application provides a rapid cell scanning method and a rapid cell scanning device, which are used for realizing efficient acquisition of cell images meeting CAC detection requirements.
The application provides a rapid scanning method of cells, which comprises the following steps:
pre-scanning a target scanning area to obtain a first cell image;
determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of target sub-regions is the largest, and the target sub-regions refer to sub-regions with the number of clustered cells lower than a preset threshold;
determining secondary scanning areas based on cell association characteristics in each target subarea, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas;
determining the scanning priority of each secondary scanning subarea based on the number of the visual fields in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model, and scanning the secondary scanning subareas in sequence based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a preset value.
According to the rapid scanning method for the cells provided by the application, the scanning priority of each secondary scanning sub-region is determined based on the number of the fields of view in each secondary scanning sub-region and the variance mean value of the corresponding gaussian mixture model, and the method specifically comprises the following steps:
determining the sorting index of each secondary scanning subarea based on the number of the visual fields in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model;
determining the scanning priority of each secondary scanning subarea based on the sorting index of each secondary scanning subarea;
and the scanning priority of each secondary scanning subarea is positively correlated with the corresponding sorting index.
According to the rapid scanning method for the cells provided by the application, the scanning of the secondary scanning sub-regions in sequence based on the scanning priority comprises the following steps:
determining the scanning sequence of each secondary scanning sub-area based on the scanning priority, and determining a target secondary scanning sub-area to be scanned currently based on the scanning sequence;
scanning the target secondary scanning sub-area based on a preset route;
and the preset route is a route which takes the central position of the target secondary scanning subarea as a starting point and moves correspondingly in a spiral or itinerant manner.
According to the rapid scanning method for the cells, the determining of the optimal subregion size corresponding to the first cell image based on the preset threshold segmentation algorithm specifically includes:
determining a subregion size set corresponding to the first cell image based on the corresponding field sizes of the pre-scanning and the secondary scanning respectively;
dividing the first cell image based on different subregion sizes in the subregion size set to obtain a plurality of subregion sets corresponding to different subregion sizes, and determining a segmentation threshold value set corresponding to each subregion in each subregion set based on a preset threshold value segmentation algorithm;
and determining the proportion of the target sub-regions in different sub-region sets based on the segmentation threshold value set corresponding to each sub-region in each sub-region set, and taking the sub-region size corresponding to the sub-region set with the maximum proportion of the target sub-regions as the optimal sub-region size.
According to the rapid scanning method for cells provided by the application, the determining the proportion of the target sub-regions in different sub-region sets based on the segmentation threshold value set corresponding to each sub-region in each sub-region set specifically comprises:
for any subregion set, determining a first subregion containing cells in the subregion set based on the segmentation threshold value set corresponding to each subregion in the subregion set; the segmentation threshold value set corresponding to the first sub-region comprises first to fourth segmentation threshold values which are sequentially arranged from small to large;
determining the indicator number of the clustered cells of each first subregion based on the second to fourth segmentation threshold values corresponding to each first subregion in the subregion set;
and determining target sub-regions in the sub-region set based on the indicator number of the clustered cells of each first sub-region, and determining the proportion of the target sub-regions in the sub-region set based on the number of the target sub-regions.
According to the rapid cell scanning method provided by the application, the molecules with clumped cell indicators in the first sub-area are: the first subregion corresponds to a third segmentation threshold squared, and the denominator of the indicator number of the clumped cells in the first subregion is: and correspondingly, the target sub-region in the sub-region set is the first sub-region of which the clumped cell indicator number is in a preset interval.
According to the rapid scanning method for cells provided by the application, the secondary scanning area is determined based on the cell association characteristics in each target sub-area, and the method specifically comprises the following steps:
determining a first target subarea with the cell pixel ratio larger than a first threshold value based on the cell pixel ratio in each target subarea;
a second target sub-region that does not contain abnormal clumping cells and has a cell number greater than a second threshold is determined based on the cell contour in the first target sub-region, and a rescanning region is determined based on the second target sub-region.
The present application further provides a rapid scanning device for cell images, comprising:
the first cell image acquisition module is used for pre-scanning the target scanning area to obtain a first cell image;
the region dividing module is used for determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of a target sub-region is the largest, and the target sub-region refers to the sub-region with the number of the clustered cells lower than a preset threshold value;
the secondary scanning area determining module is used for determining a secondary scanning area based on the cell association characteristics in each target subarea and clustering the secondary scanning area based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas;
the second cell image acquisition module is used for determining the scanning priority of each secondary scanning sub-region based on the number of the visual fields in each secondary scanning sub-region and the variance mean value of the corresponding Gaussian mixture model, and scanning the secondary scanning sub-regions in sequence based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a predetermined value.
The present application also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for rapid scanning of cells as described in any of the above.
The present application also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method for rapid scanning of cells as described in any of the above.
The present application also provides a computer program product comprising a computer program which, when executed by a processor, performs the steps of the method for rapid scanning of cells as described in any one of the above.
According to the rapid cell scanning method and device, a target scanning area is pre-scanned to obtain a first cell image; determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of a target sub-region is the largest, and the target sub-region refers to the sub-region with the number of the clustered cells lower than a preset threshold value; determining secondary scanning areas based on cell association characteristics in each target subarea, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas; determining the scanning priority of each secondary scanning subarea based on the number of fields of view in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model, and sequentially scanning the secondary scanning subareas based on the scanning priority to obtain a second cell image; the cell number in the second cell image is a preset value, so that the area with dense cells and no abnormal conglomerated cells can be quickly determined and preferentially scanned, and the cell image acquisition efficiency meeting the CAC detection requirement is improved.
Drawings
In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow diagram of a method for rapid scanning of cells provided herein;
FIG. 2 is a schematic representation of a first cell image provided herein;
FIG. 3 is a schematic flow chart of a method for determining an optimal subregion size provided in the present application;
fig. 4 is a schematic flowchart of a method for determining a secondary scanning area provided in the present application;
FIG. 5 is a schematic view of a secondary scanning area provided herein;
FIG. 6 is a flowchart illustrating a method for determining scanning priority of a sub-area scanned twice according to the present application;
FIG. 7 is a schematic flow chart of a threshold segmentation algorithm provided herein;
FIG. 8 is a schematic diagram of a fast scanning device for cell images provided in the present application;
fig. 9 is a schematic structural diagram of an electronic device provided in the present application.
Detailed Description
To make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below with reference to the drawings in the present application, and it is obvious that the described embodiments are some, but not all embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Fig. 1 is a schematic flow chart of a rapid cell scanning method provided in the present application, as shown in fig. 1, the method includes:
In particular, it is understood that the scanning device of the embodiment of the present application may be any microscopic device, and the corresponding sample may be any sample in which CAC may exist. The following will describe the scheme of the present application in detail by taking the scanning device as a pathological section scanner as an example. The pathological section scanner comprises a high-speed objective table and a high-speed scanning camera, and can realize the rapid collection of cell images with different resolutions by adjusting the magnification of the objective lens. The embodiment of the application comprises two cell image scanning processes, namely pre-scanning and secondary scanning. The purpose of the pre-scan is to determine the distribution of cells in the sample, so as to determine the area of the secondary scan, which is to obtain an image of the cells that meets the CAC detection requirements. Based on the background art, CAC detection has corresponding requirements for both the number of cells and the type of cells in a cell image. However, during the process of preparing the sample slide, the cells may be aggregated (i.e., clumped cells) due to the tension of the liquid. When the number of cells in the clumped cells exceeds a certain number threshold (namely the abnormal clumped cells), the calculation amount of a subsequent CAC detection model is increased sharply, the CAC detection efficiency is reduced, and meanwhile, the CAC detection accuracy in the abnormal clumped cells is low. Meanwhile, cell aggregation and other factors can also cause uneven cell distribution in the sample, and partial areas even have no cells. Therefore, the existing scanning mode according to the preset moving route leads to low efficiency of obtaining cell images, and partial abnormal clustered cells exist in the corresponding cell images, which leads to reduced efficiency and precision of subsequent CAC detection. Based on this, in the embodiment of the present application, a pre-scan is first performed on a target scan area to obtain an image of all cells corresponding to a sample (i.e., a first cell image), and fig. 2 is a schematic diagram of the first cell image provided by the present application, where a bright color area is a cell, and a cell distribution condition can be determined based on the first cell image, so as to exclude an area where there are few cells and abnormal clustered cells, thereby ensuring that a cell image meeting CAC detection requirements can be quickly obtained during a secondary scan.
It can be understood that, because the collection field of view of the scanning camera is limited, the pathological section scanner needs to scan the sample field by field to obtain the scanned image corresponding to each field, and then based on the coordinates corresponding to each field, the scanned images corresponding to each field are spliced to obtain the complete cell distribution area image. It should be noted that, because the pre-scan is only for acquiring the cell distribution, and the requirement on the resolution of the first cell image is not high, the objective lens magnification used in the pre-scan in the embodiment of the present application is low, and based on this, it can be ensured that the field of view corresponding to a single scan of the pathological section scanner is as large as possible, so as to improve the acquisition efficiency of the first cell image.
Specifically, after acquiring the first cell image, the embodiment of the present application further performs area division on the first cell image, so as to quickly determine the secondary scanning area. The conventional area division method generally determines the size of the sub-area corresponding to the division based on the size of the field of view corresponding to the scanning camera, and since the field of view of the scanning camera is rectangular, the size of the corresponding sub-area is generally obtained by equally dividing the camera field of view by the length M and equally dividing the camera field of view by the width N. However, the present inventors have found that the distribution of cells in the subregions obtained by different region division methods (corresponding to different sizes of subregions) is greatly different. In the worst case, each sub-region obtained by region division includes a large number of clumped cells, and based on the foregoing, the secondary scanning needs to eliminate the interference of abnormal clumped cells, and if each sub-region includes a large number of clumped cells, the efficiency and accuracy of the subsequent CAC detection will be seriously affected. Therefore, it is necessary to determine the appropriate size of the sub-region to ensure that most of the sub-regions obtained by dividing the region include fewer clumped cells, so as to avoid the interference of abnormal clumped cells.
Based on this, in the embodiment of the present application, first, an optimal sub-region size corresponding to the first cell image is determined based on a preset threshold segmentation algorithm, and the first cell image is divided into a plurality of sub-regions with the same size based on the optimal sub-region size. The proportion of the target subarea in the subareas is the largest, and the target subarea is the subarea with the number of the clustered cells lower than a preset threshold value. The inventor of the application finds that the segmentation threshold corresponding to the cell image sub-region can indirectly reflect the number of the clustered cells in the sub-region, so that whether the number of the clustered cells is lower than a preset threshold can be judged by judging whether the segmentation threshold corresponding to the cell image sub-region meets a preset requirement, and then a target sub-region with the number of the clustered cells lower than the preset threshold and the proportion of the target sub-region are determined. Dividing the first cell image based on the optimal subregion size can obtain as many target subregions as possible for secondary scanning, and ensure the efficiency of secondary scanning while avoiding collecting abnormal clustered cells by secondary scanning. The preset threshold segmentation algorithm may adopt a histogram method, an inter-maximum variance (OTSU) method, a maximum entropy threshold method, an iterative threshold segmentation method, and the like, which is not specifically limited in this embodiment of the present application.
Fig. 3 is a schematic flow chart of the method for determining an optimal subregion size provided in the present application, and as shown in fig. 3, the specific steps of determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm in the embodiment of the present application include:
specifically, based on the foregoing, since the pre-scan is only for obtaining the cell distribution, and the requirement on the resolution of the first cell image is not high, the objective lens magnification used in the pre-scan in the embodiment of the present application is low. In order to ensure the accuracy of CAC detection, the resolution requirement of the cell image obtained by the secondary scanning is higher, so that the magnification of the objective lens adopted by the secondary scanning is higher than that of the pre-scanning, and it can be understood that the magnification of the objective lens corresponding to the pre-scanning and the secondary scanning can be set according to actual needs, which is not specifically limited in the embodiment of the present application. Correspondingly, assuming that the magnification of the objective lens corresponding to the secondary scanning is a times of the magnification of the objective lens corresponding to the pre-scanning, based on the corresponding relationship between the magnification and the size of the visual field, the length and width of the visual field corresponding to the secondary scanning are both 1/a of the visual field corresponding to the pre-scanning. Based on the principle, the embodiment of the application may determine the set of sizes of the sub-regions corresponding to the first cell image based on the sizes of the fields of view corresponding to the pre-scan and the secondary scan, respectively. It is noted that the size of the sub-region should be an integer multiple of the size of the field of view corresponding to the secondary scan, so that the secondary scan can scan the complete sub-region. For example, if the objective lens magnification of the pre-scan is 10 times and the objective lens magnification of the secondary scan is 40 times, the length and width of the field of view corresponding to the pre-scan are 4 times the length and width of the field of view corresponding to the secondary scan, respectively (accordingly, the area of the field of view corresponding to the pre-scan is 16 times the area of the field of view corresponding to the secondary scan). Based on this, the size of the sub-region corresponding to the first cell image may be 1, 2, 4, 8 and 16 times the size of the field of view corresponding to the pre-scan. It can be understood that the size of the sub-region corresponding to the first cell image is obtained by performing a proportional segmentation on the length and width of the field of view corresponding to the pre-scan based on the proportional relationship between the length and width of the field of view corresponding to the pre-scan and the length and width of the field of view corresponding to the secondary scan.
specifically, after the sub-region size set corresponding to the first cell image is determined, the first cell image may be divided based on different sub-region sizes in the sub-region size set to obtain a plurality of sub-region sets corresponding to different sub-region sizes, and then the division threshold value set corresponding to each sub-region in each sub-region set is determined based on a preset threshold value division algorithm.
Specifically, based on the partition threshold value set corresponding to each sub-region in each sub-region set, it may be determined whether the sub-region is the target sub-region. Based on this, the number of target sub-regions in each sub-region set can be obtained through statistics, and the total number of the sub-regions corresponding to each sub-region set is known, so that the proportion of the target sub-regions in different sub-region sets can be determined based on the number of the target sub-regions and the total number of the sub-regions in each sub-region set, the sub-region set with the maximum proportion of the target sub-regions is further determined, and the size of the sub-region corresponding to the sub-region set is used as the optimal sub-region size. Based on steps 1201-1203, the embodiment of the application can quickly and accurately determine the optimal size of the subregion corresponding to the first cell image.
The preset threshold segmentation algorithm described in the embodiment of the present application is specifically a four-threshold segmentation algorithm, that is, four segmentation thresholds are output after being processed by the algorithm. Based on this, further, the determining the proportion of the target sub-regions in different sub-region sets based on the partition threshold value set corresponding to each sub-region in each sub-region set specifically includes:
s1, for any subregion set, determining a first subregion containing cells in the subregion set based on a segmentation threshold value set corresponding to each subregion in the subregion set; the segmentation threshold value set corresponding to the first sub-region comprises first to fourth segmentation threshold values which are sequentially arranged from small to large;
s2, determining the indicator number of the clustered cells of each first subregion based on second to fourth segmentation threshold values corresponding to each first subregion in the subregion set;
s3, determining target sub-regions in the sub-region set based on the indication number of the clustered cells of each first sub-region, and determining the proportion of the target sub-regions in the sub-region set based on the number of the target sub-regions.
Specifically, it can be understood that, in the embodiment of the present application, the proportion of the target sub-regions in each sub-region set needs to be determined respectively, and the determination procedures of the proportion of the target sub-regions corresponding to each sub-region set are the same. The inventor finds out through research that the segmentation threshold value set corresponding to each subregion not only can reflect the number of clustered cells in the subregion, but also can reflect whether the subregion contains cells. If the sub-region contains cells, the corresponding segmentation threshold set comprises first to fourth segmentation thresholds which are arranged from small to large, and if the sub-region does not contain cells, the corresponding segmentation threshold set only comprises the first segmentation threshold. Based on the foregoing embodiment, it can be seen that the efficiency of obtaining cell images by the secondary scanning is reduced for the region not containing cells, and based on this, in this embodiment, for any subregion set, the first subregion containing cells in the subregion set is determined based on the segmentation threshold value set corresponding to each subregion in the subregion set, and based on this step, the influence of the subregion not containing cells on the secondary scanning can be excluded.
And after the first sub-region in the sub-region set is determined, determining the indicator number of the clustered cells of each first sub-region based on the second to fourth segmentation threshold values corresponding to each first sub-region in the sub-region set. The clumped cell indicator number molecules of the first subregion are: the first subregion corresponds to a third segmentation threshold squared, and the denominator of the indicator number of the clumped cells in the first subregion is: the inventor of the present application finds, through research, that when the indicator number of the clumped cells in the sub-region is in the interval of 1 to 3, the number of the clumped cells in the sub-region is lower than the preset threshold value. Based on the above, whether the first sub-region is the target sub-region can be determined according to the indicator number of the clumped cells of each first sub-region, and then the number and the proportion of the target sub-regions in the sub-region set are obtained through statistics. Based on the steps S1 to S3, the proportion of the target sub-regions in each sub-region set can be determined quickly and accurately, and the determination efficiency of the optimal sub-region size corresponding to the first cell image is further improved.
And step 130, determining secondary scanning areas based on the cell association characteristics in each target sub-area, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning sub-areas.
Specifically, as can be seen from the foregoing, the proportion of the target sub-region obtained by dividing the first cell image based on the optimal sub-region size is the largest, so that the number of regions available for the secondary scanning is the largest, and the efficiency of the secondary scanning is ensured while avoiding the secondary scanning from acquiring abnormal clumped cells. However, the inventors of the present application have found that, although the number of the cells in the target sub-region is lower than the preset threshold, the probability of the abnormal cells in the target sub-region is greatly reduced, but complete elimination of the abnormal cells in the target sub-region cannot be guaranteed, and the total number of the cells in a part of the target sub-region is small. Therefore, if the target sub-region is directly used as the secondary scanning region, it is still unavoidable that abnormal clumped cells are collected by the secondary scanning, and the efficiency of the secondary scanning cannot be improved to the maximum extent. Based on this, the embodiment of the present application further performs screening on the target sub-regions based on the cell association features in each target sub-region to determine the secondary scanning region that meets the requirement. The cell association characteristics comprise cell pixel ratio and cell outline, the cell pixel ratio refers to the ratio of the number of pixel points corresponding to cells in the sub-region to the total number of pixel points, the cell number in the sub-region can be indirectly reflected, and the higher the cell pixel ratio is, the more cells in the sub-region are indicated. After the preset threshold segmentation algorithm is used for processing, the pixel value interval of the pixel points corresponding to the cells in each target subregion can be obtained, and then the pixel points corresponding to the cells in the target subregion and the pixel points corresponding to the non-cells are determined based on the corresponding image detection algorithm, so that the cell pixel ratio corresponding to the target subregion can be obtained. The cell contour is a cell edge contour, which can be obtained by a circle detection algorithm, and if an agglomerated cell exists, the corresponding contours of each single cell in the agglomerated cell are merged, and only the external contour is reserved. Based on the cell contour, the location, number, and area of the cells in the target region can be determined. For the conglomerated cells, the cell area is obviously larger than that of a single cell, so that the number of single cells in the conglomerated cells can be roughly determined based on the cell area, and the abnormal conglomerated cells are further excluded. Correspondingly, fig. 4 is a schematic flow chart of the method for determining a secondary scanning area provided in the present application, and as shown in fig. 4, the step of determining the secondary scanning area based on the cell-associated features in each target sub-area includes the following steps:
Specifically, in the embodiment of the present application, first, the first target sub-region with the cell pixel ratio greater than the first threshold is determined based on the cell pixel ratio in each target sub-region, and based on this, the target sub-region with a large number of cells can be obtained, so as to ensure the efficiency of the secondary scanning. However, the cell-pixel ratio of the first target sub-region may be larger than the first threshold due to the presence of abnormal clumping cells, so the embodiment of the present application further determines, based on the cell contour in the first target sub-region, a second target sub-region that does not include abnormal clumping cells and has a cell number larger than the second threshold, and determines, based on the second target sub-region, a secondary scanning region, that is, a region set formed by the second target sub-region. Based on this, it can be ensured that abnormal clustered cells are not contained in the secondary scanning area, and the number of cells in the secondary scanning area is large, so that on the basis of avoiding the secondary scanning to acquire the abnormal clustered cells, the secondary scanning can be ensured to quickly obtain a cell image meeting the requirement of the number of cells, and the efficiency of the secondary scanning is further improved to the maximum extent. It can be understood that the first threshold and the second threshold may be set to appropriate values according to actual situations, and the specific values thereof are not limited in the embodiment of the present application.
Fig. 5 is a schematic view of a secondary scanning area provided by the present application, and it can be known from fig. 5 that, in the secondary scanning area, the field distributions of different areas are different, and some areas have concentrated fields (corresponding cells are denser, and cell images are acquired faster), and some areas have relatively dispersed fields (corresponding cells are relatively sparse, and cell images are acquired relatively slower), so that after the secondary scanning area is determined, the secondary scanning area is further clustered based on a gaussian mixture model in the embodiment of the present application to obtain a plurality of corresponding secondary scanning sub-areas, and then, based on the field distributions of different secondary scanning sub-areas, a scanning priority can be determined to preferentially scan the secondary scanning sub-areas with relatively concentrated fields (i.e., cells are more dense). Based on this, the embodiment of the present application can further ensure that the secondary scanning can rapidly acquire the second cell image satisfying the cell number requirement. Specifically, firstly, the visual field coordinates corresponding to the secondary scanning area are standardized, then, B Gaussian mixture models are determined based on the visual field distribution condition of the secondary scanning area, and clustering analysis is performed based on the B Gaussian mixture models, so that the corresponding B secondary scanning sub-areas can be obtained. It can be understood that B is a positive integer, and the value of B can be freely set according to the view distribution, and the value is not specifically limited in the embodiment of the present application.
Specifically, after a plurality of secondary scanning sub-regions are determined, the scanning priority of each secondary scanning sub-region is determined based on the number of fields of view in each secondary scanning sub-region and the variance mean of the corresponding gaussian mixture model, and the secondary scanning sub-regions are scanned in sequence based on the scanning priority, so as to quickly acquire a second cell image. It can be understood that the second cell image is an image for subsequent CAC detection, and the number of cells in the second cell image is a preset value, where the preset value is the number of cells that need to be met by CAC detection. Fig. 6 is a schematic flowchart of a method for determining scanning priority of secondary scanning sub-regions provided in the present application, and as shown in fig. 6, the determining the scanning priority of each secondary scanning sub-region based on the number of views in each secondary scanning sub-region and the mean variance of the corresponding gaussian mixture model specifically includes:
1401, determining a ranking index of each secondary scanning sub-region based on the number of views in each secondary scanning sub-region and the variance mean of the corresponding Gaussian mixture model;
1402, determining the scanning priority of each secondary scanning sub-area based on the sequencing index of each secondary scanning sub-area;
and the scanning priority of each secondary scanning subarea is positively correlated with the corresponding sorting index.
Specifically, assuming that the number of views C of the current secondary scanning sub-region is C, the calculation formula of the variance mean of the corresponding gaussian mixture model is:
var_mean = (var_X+var_Y)/2；
wherein, var _ mean is the variance mean, X and Y are the central coordinates of the visual fields, var _ X is the variance of the abscissa corresponding to each visual field, and var _ Y is the variance of the ordinate corresponding to each visual field. Respectively carrying out normalization processing on the M and the var _ mean to determine the sorting index of the secondary scanning sub-area, wherein the calculation formula of the sorting index is as follows:
sort_index=(0.5 * M - 0.5 * var_mean)；
the ranking index can reflect the number and concentration of the visual fields of the current secondary scanning area, and the larger the ranking index is, the higher the number and concentration of the visual fields are. Based on this, the scanning priority of each secondary scanning sub-region can be determined based on the ranking index of each secondary scanning sub-region, and the scanning priority of each secondary scanning sub-region is positively correlated with the corresponding ranking index, that is, the larger the ranking index is, the higher the scanning priority is. Based on this, the scanning priority of each secondary scanning sub-region can be rapidly and accurately obtained in the embodiment of the application, and then the second cell image meeting the CAC detection requirement can be efficiently obtained.
According to the method provided by the embodiment of the application, a target scanning area is subjected to pre-scanning to obtain a first cell image; determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of a target sub-region is the largest, and the target sub-region refers to the sub-region with the number of the clustered cells lower than a preset threshold value; determining secondary scanning areas based on cell association characteristics in each target subarea, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas; determining the scanning priority of each secondary scanning subarea based on the number of the visual fields in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model, and scanning the secondary scanning subareas in sequence based on the scanning priority to obtain a second cell image; the cell number in the second cell image is a preset value, so that the area with dense cells and no abnormal conglomerated cells can be quickly determined and preferentially scanned, and the cell image acquisition efficiency meeting the CAC detection requirement is improved.
Based on the above embodiment, the sequentially scanning the sub-areas scanned twice based on the scanning priority includes:
determining the scanning sequence of each secondary scanning sub-area based on the scanning priority, and determining a target secondary scanning sub-area to be scanned currently based on the scanning sequence;
scanning the target secondary scanning sub-area based on a preset route;
and the preset route is a route which takes the central position of the target secondary scanning subarea as a starting point and moves correspondingly in a spiral or itinerant manner.
In particular, it can be understood that the higher the scanning priority, the more forward the scanning order of the corresponding secondary scanning sub-regions. And determining the current target secondary scanning sub-area to be scanned based on the scanning sequence, and further scanning the target secondary scanning sub-area based on a preset route. The preset route refers to a moving route of the pathological section scanner for scanning the target secondary scanning sub-area field by field. The preset route is a route which takes the central position of the target secondary scanning sub-area as a starting point and moves correspondingly in a spiral or circulating mode, and based on the route, the pathological section scanner can perform secondary scanning by the shortest moving route, so that the acquisition efficiency of the second cell image is further improved.
The method provided by the embodiment of the present application, sequentially scanning the sub-areas scanned twice based on the scanning priority, includes: and determining the scanning sequence of each secondary scanning sub-region based on the scanning priority, determining a target secondary scanning sub-region to be scanned currently based on the scanning sequence, and scanning the target secondary scanning sub-region based on a preset route, wherein the preset route is a route which takes the central position of the target secondary scanning sub-region as a starting point and moves correspondingly in a spiral or itinerant manner, so that the acquisition efficiency of the second cell image can be further improved.
Based on any of the above embodiments, fig. 7 is a schematic flow chart of the preset threshold segmentation algorithm provided in the present application, and as shown in fig. 7, for any sub-region, the determining step of the segmentation threshold set is as follows:
1) Initialization parameters a, b, k, n and i. Wherein a and b are pixel value intervals corresponding to the cell image, a =0, b =255, that is, the pixel values of the pixel points in any sub-region are all between 0-255; k is a weight coefficient of the segmentation threshold, which can be set empirically; n is the number of segmentation thresholds, for the present embodiment, n =4; i is used for indicating the execution times of the subsequent segmentation threshold determination step, and the initial value of i is 0;
2) Judging whether i satisfies that i is more than or equal to 0 and less than n/2-1, if so, adding 1 to the value of i, and executing the step 3), otherwise, executing the step 6);
3) Calculating the pixel value mean value mu and the standard deviation sigma of pixel points (namely all pixel points in the subarea) with the pixel values between [ a and b ], and obtaining boundaries T1= mu-k sigma and T2= mu + k sigma of the subintervals;
4) Respectively calculating the pixel value mean values of pixel points positioned in subintervals [ a, T1] (namely [0, mu-k sigma ]) and [ T2, b ] (namely [ mu + k sigma,255 ]) and respectively taking the pixel value mean values as a first segmentation threshold value and a fourth segmentation threshold value;
5) Updating the parameter a = T1+1, b = T2-1, k = k (i + 1), and jumping to execute the step 2); for the embodiment of the present application, since i is 1 at this time, step 6) will be executed after step 2);
6) And calculating the boundary T1= mu and T2= mu +1 of the subintervals, and calculating the pixel value mean values of the pixel points of the subintervals [ a, T1] and [ T2, b ] as a second segmentation threshold and a third segmentation threshold respectively. It is noted that since a and b are updated in step 5), [ a, T1] and [ T2, b ] at this time should be [ mu-k sigma +1, mu ] and [ mu +1, mu k sigma-1], respectively.
Based on the steps 1) to 6), the partition threshold value set of each sub-region can be determined.
Based on any of the above embodiments, the inventors of the present application find, through research, that the fourth segmentation threshold obtained based on the preset threshold segmentation algorithm of the embodiments of the present application can reflect pixel values of a relatively bright portion in a nuclear texture, and the third segmentation threshold can reflect pixel values of a relatively dark portion in a nuclear texture. Therefore, in the case of clear focusing, the difference between the two is the highest, and based on this, in the process of pre-scanning and secondary scanning, the embodiment of the present application further determines the clarity of focusing based on the ratio of the fourth division threshold to the third division threshold. Based on the method, the focusing parameters in the scanning process can be conveniently and timely adjusted to determine the optimal focusing point, and the acquisition efficiency of the first cell image and the second cell image can be further improved.
The following describes an apparatus for acquiring a cell scan image provided in the present application, and the apparatus for acquiring a cell scan image described below and the method for acquiring a cell scan image described above may be referred to in correspondence with each other.
Based on any of the above embodiments, fig. 8 is a schematic structural diagram of a rapid scanning apparatus for cell images provided by the present application, as shown in fig. 8, the apparatus includes:
a first cell image obtaining module 810, configured to perform pre-scanning on a target scanning area to obtain a first cell image;
a region dividing module 820, configured to determine an optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and divide the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of target sub-regions is the largest, and the target sub-regions refer to sub-regions with the number of clustered cells lower than a preset threshold;
a secondary scanning area determining module 830, configured to determine a secondary scanning area based on cell association characteristics in each target sub-area, and cluster the secondary scanning area based on a gaussian mixture model to obtain a plurality of corresponding secondary scanning sub-areas;
the second cell image acquisition module 840 is configured to determine a scanning priority of each secondary scanning sub-region based on the number of fields in each secondary scanning sub-region and a variance mean of a corresponding gaussian mixture model, and sequentially scan the secondary scanning sub-regions based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a predetermined value.
In the apparatus provided in the embodiment of the present application, the first cell image obtaining module 810 performs pre-scanning on a target scanning area to obtain a first cell image; the region dividing module 820 determines the optimal sub-region size corresponding to the first cell image based on a preset threshold segmentation algorithm, and divides the first cell image into a plurality of sub-regions with the same size based on the optimal sub-region size; in the sub-regions, the proportion of a target sub-region is the largest, and the target sub-region refers to the sub-region with the number of the clustered cells lower than a preset threshold value; the secondary scanning area determining module 830 determines a secondary scanning area based on the cell association characteristics in each target sub-area, and clusters the secondary scanning area based on a gaussian mixture model to obtain a plurality of corresponding secondary scanning sub-areas; the second cell image acquisition module 840 determines the scanning priority of each secondary scanning sub-region based on the number of fields of view in each secondary scanning sub-region and the variance-mean of the corresponding gaussian mixture model, and scans the secondary scanning sub-regions in sequence based on the scanning priority to obtain a second cell image; the cell number in the second cell image is a preset value, so that the area with dense cells and no abnormal conglomerated cells can be quickly determined and preferentially scanned, and the cell image acquisition efficiency meeting the CAC detection requirement is improved.
Based on the above embodiment, the determining the scanning priority of each secondary scanning sub-region based on the number of views in each secondary scanning sub-region and the variance-mean of the corresponding gaussian mixture model specifically includes:
determining the sequencing index of each secondary scanning sub-region based on the number of the visual fields in each secondary scanning sub-region and the variance mean value of the corresponding Gaussian mixture model;
determining the scanning priority of each secondary scanning subarea based on the sorting index of each secondary scanning subarea;
and the scanning priority of each secondary scanning subarea is positively correlated with the corresponding sorting index.
Based on any of the above embodiments, the sequentially scanning the secondary scanning sub-regions based on the scanning priority includes:
determining the scanning sequence of each secondary scanning sub-area based on the scanning priority, and determining a target secondary scanning sub-area to be scanned currently based on the scanning sequence;
scanning the target secondary scanning sub-area based on a preset route;
and the preset route is a route which takes the central position of the target secondary scanning subarea as a starting point and moves correspondingly in a spiral or itinerant manner.
Based on any of the embodiments, the determining an optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm specifically includes:
determining a subregion size set corresponding to the first cell image based on the corresponding field sizes of the pre-scanning and the secondary scanning respectively;
dividing the first cell image based on different subregion sizes in the subregion size set to obtain a plurality of subregion sets corresponding to different subregion sizes, and determining a segmentation threshold value set corresponding to each subregion in each subregion set based on a preset threshold value segmentation algorithm;
and determining the proportion of the target sub-regions in different sub-region sets based on the segmentation threshold value set corresponding to each sub-region in each sub-region set, and taking the sub-region size corresponding to the sub-region set with the maximum proportion of the target sub-regions as the optimal sub-region size.
Based on any of the above embodiments, the determining, based on the partition threshold value set corresponding to each sub-region in each sub-region set, the proportion of the target sub-region in different sub-region sets specifically includes:
for any subregion set, determining a first subregion containing cells in the subregion set based on the segmentation threshold value set corresponding to each subregion in the subregion set; the segmentation threshold value set corresponding to the first sub-region comprises first to fourth segmentation threshold values which are sequentially arranged from small to large;
determining the indicator number of the clustered cells of each first subregion based on the second to fourth segmentation threshold values corresponding to each first subregion in the subregion set;
and determining the target sub-regions in the sub-region set based on the indicator number of the clustered cells of each first sub-region, and determining the proportion of the target sub-regions in the sub-region set based on the number of the target sub-regions.
In accordance with any of the embodiments above, the molecules of the clumped cell indicator number of the first subregion are: the first subregion corresponds to a third segmentation threshold squared, and the denominator of the indicator number of the clumped cells in the first subregion is: and correspondingly, the target sub-region in the sub-region set is the first sub-region of which the clumped cell indicator number is in a preset interval.
Based on any one of the embodiments, the determining the secondary scanning region based on the cell association features in each target subregion specifically includes:
determining a first target subarea of which the cell pixel ratio is greater than a first threshold value based on the cell pixel ratio in each target subarea;
a second target sub-region that does not contain abnormal clumping cells and has a cell number greater than a second threshold is determined based on the cell contour in the first target sub-region, and a rescanning region is determined based on the second target sub-region.
Fig. 9 illustrates a physical structure diagram of an electronic device, and as shown in fig. 9, the electronic device may include: a processor (processor) 910, a communication Interface (Communications Interface) 920, a memory (memory) 930, and a communication bus 940, wherein the processor 910, the communication Interface 920, and the memory 930 communicate with each other via the communication bus 940. Processor 910 may invoke logic instructions in memory 930 to perform a fast scan method of cells provided by the above-described methods, the method comprising: pre-scanning a target scanning area to obtain a first cell image; determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of target sub-regions is the largest, and the target sub-regions refer to sub-regions with the number of clustered cells lower than a preset threshold; determining secondary scanning areas based on cell association characteristics in each target subarea, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas; determining the scanning priority of each secondary scanning subarea based on the number of the visual fields in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model, and scanning the secondary scanning subareas in sequence based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a preset value.
Furthermore, the logic instructions in the memory 930 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, the present application further provides a computer program product, the computer program product comprising a computer program, the computer program being stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is capable of executing the method for rapid scanning of cells provided by the above methods, the method comprising: pre-scanning a target scanning area to obtain a first cell image; determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of target sub-regions is the largest, and the target sub-regions refer to sub-regions with the number of clustered cells lower than a preset threshold; determining secondary scanning areas based on cell association characteristics in each target subarea, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas; determining the scanning priority of each secondary scanning subarea based on the number of fields of view in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model, and sequentially scanning the secondary scanning subareas based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a predetermined value.
In yet another aspect, the present application also provides a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method for rapid scanning of cells provided by performing the above methods, the method comprising: pre-scanning a target scanning area to obtain a first cell image; determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of a target sub-region is the largest, and the target sub-region refers to the sub-region with the number of the clustered cells lower than a preset threshold value; determining secondary scanning areas based on cell association characteristics in each target subarea, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas; determining the scanning priority of each secondary scanning subarea based on the number of fields of view in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model, and sequentially scanning the secondary scanning subareas based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a predetermined value.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on the understanding, the above technical solutions substantially or otherwise contributing to the prior art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (7)
1. A method for rapid scanning of cells, comprising:
pre-scanning a target scanning area to obtain a first cell image;
determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm, and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of a target sub-region is the largest, and the target sub-region refers to the sub-region with the number of the clustered cells lower than a preset threshold value;
determining secondary scanning areas based on cell association characteristics in each target subarea, and clustering the secondary scanning areas based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas;
determining the scanning priority of each secondary scanning subarea based on the number of the visual fields in each secondary scanning subarea and the variance mean value of the corresponding Gaussian mixture model, and scanning the secondary scanning subareas in sequence based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a preset value;
the determining the optimal subregion size corresponding to the first cell image based on the preset threshold segmentation algorithm comprises:
determining a subregion size set corresponding to the first cell image based on the corresponding field sizes of the pre-scanning and the secondary scanning;
dividing the first cell image based on different subregion sizes in the subregion size set to obtain a plurality of subregion sets corresponding to different subregion sizes, and determining a segmentation threshold value set corresponding to each subregion in each subregion set based on a preset threshold value segmentation algorithm;
for any subregion set, determining a first subregion containing cells in the subregion set based on a segmentation threshold set corresponding to each subregion in the subregion set; the segmentation threshold value set corresponding to the first sub-region comprises first to fourth segmentation threshold values which are sequentially arranged from small to large;
determining the indicator number of the clustered cells of each first subregion based on the second to fourth segmentation threshold values corresponding to each first subregion in the subregion set;
determining target sub-regions in the sub-region set based on the indication number of the clustered cells of each first sub-region, determining the proportion of the target sub-regions in the sub-region set based on the number of the target sub-regions, and taking the sub-region size corresponding to the sub-region set with the maximum proportion of the target sub-regions as the optimal sub-region size; the clumped cell indicator number molecules of the first subregion are: the square of a third segmentation threshold corresponding to the first subregion, and the denominator of the clumped cell indicator number of the first subregion is: and correspondingly, the target sub-region in the sub-region set is the first sub-region of which the clumped cell indicator number is in a preset interval.
2. The method according to claim 1, wherein the determining the scanning priority of each sub-region based on the number of fields of view in each sub-region and the variance-mean of the corresponding Gaussian mixture model comprises:
determining the sequencing index of each secondary scanning sub-region based on the number of the visual fields in each secondary scanning sub-region and the variance mean value of the corresponding Gaussian mixture model;
determining the scanning priority of each secondary scanning subarea based on the sorting index of each secondary scanning subarea;
and the scanning priority of each secondary scanning subarea is positively correlated with the corresponding sorting index.
3. The method for rapid scanning of cells according to claim 2, wherein the scanning the sub-regions of the secondary scanning in turn based on the scanning priority comprises:
determining the scanning sequence of each secondary scanning sub-area based on the scanning priority, and determining a target secondary scanning sub-area to be scanned currently based on the scanning sequence;
scanning the target secondary scanning sub-area based on a preset route;
and the preset route is a route which takes the central position of the target secondary scanning sub-area as a starting point and moves correspondingly in a spiral or circulating mode.
4. The method according to claim 1, wherein the determining the secondary scanning area based on the cell association features in each target sub-area specifically comprises:
determining a first target subarea of which the cell pixel ratio is greater than a first threshold value based on the cell pixel ratio in each target subarea;
a second target sub-region that does not contain abnormal clumping cells and has a cell number greater than a second threshold is determined based on the cell contour in the first target sub-region, and a rescanning region is determined based on the second target sub-region.
5. An apparatus for rapid scanning of images of cells, comprising:
the first cell image acquisition module is used for pre-scanning a target scanning area to obtain a first cell image;
the region dividing module is used for determining the optimal subregion size corresponding to the first cell image based on a preset threshold segmentation algorithm and dividing the first cell image into a plurality of subregions with the same size based on the optimal subregion size; in the sub-regions, the proportion of a target sub-region is the largest, and the target sub-region refers to the sub-region with the number of the clustered cells lower than a preset threshold value;
the secondary scanning area determining module is used for determining a secondary scanning area based on the cell association characteristics in each target subarea and clustering the secondary scanning area based on a Gaussian mixture model to obtain a plurality of corresponding secondary scanning subareas;
the second cell image acquisition module is used for determining the scanning priority of each secondary scanning sub-region based on the number of the visual fields in each secondary scanning sub-region and the variance mean value of the corresponding Gaussian mixture model, and scanning the secondary scanning sub-regions in sequence based on the scanning priority to obtain a second cell image; the number of cells in the second cell image is a preset value;
the determining the optimal sub-region size corresponding to the first cell image based on the preset threshold segmentation algorithm comprises:
determining a subregion size set corresponding to the first cell image based on the corresponding field sizes of the pre-scanning and the secondary scanning respectively;
dividing the first cell image based on different subregion sizes in the subregion size set to obtain a plurality of subregion sets corresponding to different subregion sizes, and determining a segmentation threshold value set corresponding to each subregion in each subregion set based on a preset threshold value segmentation algorithm;
for any subregion set, determining a first subregion containing cells in the subregion set based on the segmentation threshold value set corresponding to each subregion in the subregion set; the segmentation threshold value set corresponding to the first sub-region comprises first to fourth segmentation threshold values which are sequentially arranged from small to large;
determining the indicator number of the clustered cells of each first subregion based on the second to fourth segmentation threshold values corresponding to each first subregion in the subregion set;
determining target sub-regions in the sub-region set based on the indicator number of the clustered cells of each first sub-region, determining the proportion of the target sub-regions in the sub-region set based on the number of the target sub-regions, and taking the sub-region size corresponding to the sub-region set with the maximum proportion of the target sub-regions as the optimal sub-region size; the molecules of the clumped cell indicator number of the first subregion are: the square of a third segmentation threshold corresponding to the first subregion, and the denominator of the clumped cell indicator number of the first subregion is: and correspondingly, the target sub-region in the sub-region set is the first sub-region of which the clumped cell indicator number is in a preset interval.
6. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program performs the steps of the method for rapid scanning of cells according to any of claims 1 to 4.
7. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of a method for rapid scanning of cells according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211092303.1A CN115165710B (en) | 2022-09-08 | 2022-09-08 | Rapid scanning method and device for cells |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211092303.1A CN115165710B (en) | 2022-09-08 | 2022-09-08 | Rapid scanning method and device for cells |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115165710A CN115165710A (en) | 2022-10-11 |
| CN115165710B true CN115165710B (en) | 2022-11-29 |
Family
ID=83480725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211092303.1A Active CN115165710B (en) | 2022-09-08 | 2022-09-08 | Rapid scanning method and device for cells |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115165710B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117372274A (en) * | 2023-10-31 | 2024-01-09 | 珠海横琴圣澳云智科技有限公司 | Scanned image refocusing method, apparatus, electronic device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110132823A (en) * | 2019-05-15 | 2019-08-16 | 林伟阳 | A kind of method for cell count based on template matching and lookup profile |
| CN110378313A (en) * | 2019-07-26 | 2019-10-25 | 玖壹叁陆零医学科技南京有限公司 | Cell mass recognition methods, device and electronic equipment |
| CN111091571A (en) * | 2019-12-12 | 2020-05-01 | 珠海圣美生物诊断技术有限公司 | Nucleus segmentation method and device, electronic equipment and computer-readable storage medium |
| CN114067114A (en) * | 2021-11-12 | 2022-02-18 | 燕山大学 | Area nucleus segmentation counting method based on algae cell morphology |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003095986A1 (en) * | 2002-05-14 | 2003-11-20 | Amersham Biosciences Niagara, Inc. | System and methods for rapid and automated screening of cells |
-
2022
- 2022-09-08 CN CN202211092303.1A patent/CN115165710B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110132823A (en) * | 2019-05-15 | 2019-08-16 | 林伟阳 | A kind of method for cell count based on template matching and lookup profile |
| CN110378313A (en) * | 2019-07-26 | 2019-10-25 | 玖壹叁陆零医学科技南京有限公司 | Cell mass recognition methods, device and electronic equipment |
| CN111091571A (en) * | 2019-12-12 | 2020-05-01 | 珠海圣美生物诊断技术有限公司 | Nucleus segmentation method and device, electronic equipment and computer-readable storage medium |
| CN114067114A (en) * | 2021-11-12 | 2022-02-18 | 燕山大学 | Area nucleus segmentation counting method based on algae cell morphology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115165710A (en) | 2022-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108961208B (en) | System and method for segmenting and counting aggregated leukocytes | |
| US8331642B2 (en) | Method and device for determining a contour and a center of an object | |
| CN109636824B (en) | Multi-target counting method based on image recognition technology | |
| WO2021139258A1 (en) | Image recognition based cell recognition and counting method and apparatus, and computer device | |
| CN109102506B (en) | Automatic segmentation method for abdominal CT liver lesion image based on three-level cascade network | |
| CN110378313B (en) | Cell cluster identification method and device and electronic equipment | |
| CN112132852B (en) | Automatic image matting method and device based on multi-background color statistics | |
| US8019156B2 (en) | Method and device for segmenting a digital representation of cells | |
| CN112215790A (en) | KI67 index analysis method based on deep learning | |
| CN115165710B (en) | Rapid scanning method and device for cells | |
| CN114240978B (en) | Cell edge segmentation method and device based on adaptive morphology | |
| CN111369523B (en) | Method, system, equipment and medium for detecting cell stack in microscopic image | |
| WO2024051429A1 (en) | Method and device for acquiring cell scanning image | |
| CN112037185A (en) | Chromosome split phase image screening method and device and terminal equipment | |
| CN112927247A (en) | Graph cutting method based on target detection, graph cutting device and storage medium | |
| CN111126162A (en) | Method, device and storage medium for identifying inflammatory cells in image | |
| CN113689431B (en) | Industrial product appearance defect detection method and device | |
| CN114998332B (en) | Method and device for determining karyotype abnormal cells | |
| CN115204323A (en) | Seed multi-feature based clustering and synthesis method, system, device and medium | |
| CN114332027A (en) | Cell uniformity obtaining method and system based on image analysis | |
| CN112529918A (en) | Method, device and equipment for ventricular region segmentation in brain CT image | |
| CN111932499B (en) | Cell diameter calculation method, device and system | |
| Chen et al. | Automated fry counting method based on image processing | |
| CN116912887B (en) | Broiler chicken breeding management method and system | |
| CN117218139A (en) | Method and device for determining cell density of sample |
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 |