CN117575977B - Follicular region enhancement method for ovarian tissue analysis - Google Patents
Follicular region enhancement method for ovarian tissue analysis Download PDFInfo
- Publication number
- CN117575977B CN117575977B CN202410063209.6A CN202410063209A CN117575977B CN 117575977 B CN117575977 B CN 117575977B CN 202410063209 A CN202410063209 A CN 202410063209A CN 117575977 B CN117575977 B CN 117575977B
- Authority
- CN
- China
- Prior art keywords
- cell
- domain
- membrane
- connected domain
- target
- 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
- 230000002611 ovarian Effects 0.000 title claims abstract description 85
- 230000003325 follicular Effects 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000004458 analytical method Methods 0.000 title claims abstract description 16
- 210000004027 cell Anatomy 0.000 claims abstract description 167
- 210000000170 cell membrane Anatomy 0.000 claims abstract description 116
- 210000001519 tissue Anatomy 0.000 claims abstract description 93
- 210000003855 cell nucleus Anatomy 0.000 claims abstract description 49
- 230000009466 transformation Effects 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 9
- 239000012528 membrane Substances 0.000 claims description 79
- 238000004891 communication Methods 0.000 claims description 72
- 210000004940 nucleus Anatomy 0.000 claims description 41
- 238000004422 calculation algorithm Methods 0.000 claims description 37
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 9
- 210000001672 ovary Anatomy 0.000 claims description 7
- 230000000877 morphologic effect Effects 0.000 claims description 5
- 238000003064 k means clustering Methods 0.000 claims description 4
- 230000001413 cellular effect Effects 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 8
- 238000012545 processing Methods 0.000 abstract description 5
- 230000003044 adaptive effect Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000005484 gravity Effects 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000805 cytoplasm Anatomy 0.000 description 1
- 238000005538 encapsulation Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Classifications
-
- 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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image data processing, in particular to a follicle area enhancement method for ovarian tissue analysis, which comprises the following steps: the method comprises the steps of obtaining an ovarian tissue gray level image, dividing the ovarian tissue gray level image into a plurality of super pixel areas, screening a plurality of cell nucleus connected domains from the super pixel areas, obtaining a plurality of cell membrane connected domains from the super pixel areas, obtaining the association degree of each cell membrane connected domain and each cell nucleus connected domain, obtaining the cell area of each cell nucleus connected domain, obtaining the cell possibility of each pixel point in the cell area of each cell nucleus connected domain, and obtaining an enhanced image of the ovarian tissue gray level image. According to the invention, the characteristics of cell nuclei and cell membranes are analyzed to obtain the follicular cell area, and the linear transformation enhancement of the adaptive gain factors is carried out on the pixel points in the follicular cell area, so that the follicular area enhancement effect in the ovarian tissue gray level image is improved.
Description
Technical Field
The invention relates to the technical field of image data processing, in particular to a follicle area enhancement method for ovarian tissue analysis.
Background
In order to observe possible lesions in ovarian tissues, cells in a follicular region need to be observed, and since follicular cells in the ovarian tissues are densely distributed in an image, and boundaries of membranes among cells are fuzzy and are not easy to distinguish, in order to observe follicular cells more accurately, the follicular region in an acquired image needs to be subjected to enhancement processing, so that analysis accuracy of the ovarian tissues is guaranteed, and linear transformation is currently used for image enhancement processing.
The existing problems are as follows: when the gain factors in the linear transformation algorithm are not properly selected, the image enhancement effect is poor, edges among cells in the acquired image are possibly blurred, the follicular region detection is inaccurate, the follicular region cannot be enhanced accurately, and therefore the follicular region enhancement effect in the ovarian tissue image is reduced.
Disclosure of Invention
The invention provides a follicular region enhancement method for ovarian tissue analysis, which aims to solve the existing problems.
The invention discloses a follicular region enhancement method for ovarian tissue analysis, which adopts the following technical scheme:
one embodiment of the present invention provides a follicular zone enhancement method for ovarian tissue analysis, the method comprising the steps of:
collecting an ovarian tissue image, and carrying out graying treatment to obtain an ovarian tissue gray image; dividing an ovarian tissue gray level image into a plurality of super-pixel areas, and screening a plurality of cell nucleus connected areas from all the super-pixel areas;
in the area which is not the cell nucleus connected domain in the ovarian tissue gray level image, according to the pixel point gray level value and the distance between the pixel points, obtaining a plurality of connected domains and the cell membrane possibility of each connected domain, and screening a plurality of cell membrane connected domains from all connected domains;
marking any cell nucleus communicating domain as a reference nucleus communicating domain; marking any one cell membrane communicating domain as a reference membrane communicating domain; obtaining the association degree of the reference membrane connected domain and the reference nuclear connected domain according to the center point of the reference nuclear connected domain, the intersection point of the normals of all pixel points on the skeleton line of the reference membrane connected domain and the cell membrane possibility of the reference membrane connected domain;
obtaining a cell region of the reference nuclear communication domain according to the association degree of all cell membrane communication domains and the reference nuclear communication domain respectively; obtaining the cell probability of each pixel point in the cell region of the reference nuclear communication domain according to the distance from the center point of the reference nuclear communication domain to the center point of the cell region, the association degree of the cell membrane communication domain and the reference nuclear communication domain and the shortest distance from each pixel point to the cell membrane communication domain and the reference nuclear communication domain respectively;
and obtaining an enhanced image of the ovarian tissue gray level image according to the cell possibility of all pixel points in the cell area of all the cell nucleus connected domains.
Further, the method for dividing the gray level image of the ovarian tissue into a plurality of super-pixel areas and screening a plurality of cell nucleus connected areas from all the super-pixel areas comprises the following specific steps:
dividing the ovarian tissue gray level image into a plurality of super pixel areas by using a super pixel dividing algorithm;
any super-pixel area divided by the gray level image of the ovary tissue is marked as a main super-pixel area;
a minimum circumscribed circle algorithm and a maximum inscribed circle algorithm are used for respectively obtaining a minimum circumscribed circle and a maximum inscribed circle of the main super pixel area;
the difference value of the radius of the minimum circumscribing circle of the main super pixel area minus the radius of the maximum inscribing circle is recorded as the circle-like degree of the main communicating area;
and (3) marking the super-pixel region with the degree of quasi-circle larger than a preset nuclear threshold value as a nucleus connected region in all super-pixel regions of the ovarian tissue gray level image segmentation.
Further, in the area which is not a cell nucleus connected domain in the ovarian tissue gray level image, according to the pixel point gray level value and the distance between the pixel points, the cell membrane possibility of a plurality of connected domains and each connected domain is obtained, and a plurality of cell membrane connected domains are screened out from all connected domains, comprising the following specific steps:
in the gray level image of the ovarian tissue, a region which is not a cell nucleus connected domain is marked as a background region;
in the background area, the area formed by all pixel points with gray values larger than a preset gray threshold is marked as a target area;
dividing a target area into a plurality of connected areas by using an area growth algorithm;
any one of the connected domains divided by the target area is marked as a target connected domain;
calculating the distance from each pixel point to the center point of the target connected domain in the target connected domain, and recording the average value of the distances from all the pixel points to the center point of the target connected domain as the continuity degree of the target connected domain;
obtaining the cell membrane possibility of the target connected domain according to the gray values of all pixel points in the target connected domain and the continuity degree of the target connected domain;
among all the connected domains divided in the target region, the connected domain having a membrane potential greater than a preset membrane threshold is designated as a membrane connected domain.
Further, according to the gray values of all the pixels in the target connected domain and the continuity degree of the target connected domain, the specific calculation formula corresponding to the cell membrane probability of the target connected domain is obtained as follows:
；
wherein A is the possibility of cell membrane of the target connected domain, V is the variance of gray values of all pixels in the target connected domain, B is the continuity degree of the target connected domain,k is a preset exponential function adjustment value for an exponential function based on a natural constant.
Further, the correlation degree between the reference membrane connected domain and the reference nuclear connected domain is obtained according to the center point of the reference nuclear connected domain, the intersection point of the normals of all pixel points on the skeleton line of the reference membrane connected domain, and the cell membrane probability of the reference membrane connected domain, and the specific steps are as follows:
obtaining a skeleton line of the reference membrane connected domain by using a morphological refinement algorithm;
calculating intersection points of normal lines of any two pixel points on a skeleton line of a reference film connected domain, and clustering the intersection points of the normal lines of all the pixel points by using a K-means clustering algorithm to obtain a clustering center point in an ovarian tissue gray level image;
calculating the distances from the intersection points of the normals of all the pixel points to the clustering center points respectively, and marking the average value of the distances from the intersection points to the clustering center points respectively as the wrapping degree of the reference film connected domain;
and obtaining the association degree of the reference membrane connected domain and the reference nuclear connected domain according to the possibility of the cell membrane of the reference membrane connected domain, the wrapping degree of the reference membrane connected domain and the distance from the clustering center point to the center point of the reference nuclear connected domain.
Further, the specific calculation formula corresponding to the association degree of the reference membrane connected domain and the reference nuclear connected domain is obtained according to the cell membrane possibility of the reference membrane connected domain, the wrapping degree of the reference membrane connected domain and the distance from the clustering center point to the center point of the reference nuclear connected domain, wherein the specific calculation formula comprises the following steps:
；
wherein S is the association degree of the reference membrane connected domain and the reference nuclear connected domain,for reference to the membrane potential of the membrane-connected domain, -/->For the wrapping degree of the reference membrane connected domain, T is the distance from the clustering center point to the center point of the reference nuclear connected domain, and +.>Is a linear normalization function.
Further, the method for obtaining the cell region of the reference nuclear communication domain according to the association degree of all the cell membrane communication domains and the reference nuclear communication domain respectively comprises the following specific steps:
in the association degrees of all the cell membrane communicating domains with the reference nuclear communicating domain respectively, the cell membrane communicating domain corresponding to the association degree larger than a preset judging threshold value is marked as the cell membrane communicating domain of the reference nuclear communicating domain;
the cell membrane communicating domain of the reference nuclear communicating domain is marked as a target membrane communicating domain;
the region formed by the reference nuclear communication domain and all the target membrane communication domains is marked as a reference region;
obtaining a minimum circumcircle of the reference area by using a minimum circumcircle algorithm;
the region within the smallest circumscribed circle of the reference region is designated as the cell region of the reference nuclear communication domain.
Further, in the cell region of the reference nuclear communication domain, according to the distance from the center point of the reference nuclear communication domain to the center point of the cell region, the association degree of the cell membrane communication domain and the reference nuclear communication domain, and the shortest distance from each pixel point to the cell membrane communication domain and the reference nuclear communication domain, the cell probability of each pixel point is obtained, which comprises the following specific steps:
in the cell area of the reference nucleus communication domain, marking any pixel point as a target point;
and obtaining the cell probability of the target point according to the association degree of all the target membrane connected domains with the reference nuclear connected domain respectively, the shortest distance between the target point and the reference nuclear connected domain and all the target membrane connected domains respectively, and the distance between the center point of the cell region of the reference nuclear connected domain and the center point of the reference nuclear connected domain.
Further, the specific calculation formula corresponding to the cell probability of the target point is obtained according to the association degree of all the target membrane connected domains with the reference nuclear connected domain, the shortest distance between the target point and the reference nuclear connected domain, and the distance between the center point of the cell region of the reference nuclear connected domain and the center point of the reference nuclear connected domain, wherein the specific calculation formula is as follows:
；
where P is the cellular probability of the target site,for the degree of association of the ith target membrane-associated domain with the reference nuclear domain, +.>For the target point to the ith target membrane connected domainC is the shortest distance from the target point to the reference nucleus communication domain, R is the distance from the center point of the cell region of the reference nucleus communication domain to the center point of the reference nucleus communication domain, n is the number of target membrane communication domains, and>is a linear normalization function.
Further, the step of obtaining the enhanced image of the gray level image of the ovarian tissue according to the cell possibility of all pixel points in the cell area of all the cell nucleus connected domain comprises the following specific steps:
in the cell areas of all the cell nucleus communicating domains, pixel points with the cell possibility larger than a preset cell threshold value are marked as cell points;
in the gray level image of the ovary tissue, marking other pixel points which are not cell points as normal points; adding the sum of the cell probability of each cell point to 1, and recording the sum as a gain factor of each cell point; setting the gain factor of each normal point to 1;
and (3) carrying out image enhancement by using a linear transformation algorithm according to gain factors of all pixel points in the ovarian tissue gray level image to obtain an enhanced image of the ovarian tissue gray level image.
The technical scheme of the invention has the beneficial effects that:
in the embodiment of the invention, the gray level image of the ovarian tissue is acquired, the gray level image is divided into a plurality of super-pixel areas, and a plurality of cell nucleus connected areas are screened out from the gray level image, and the accuracy of detecting the follicular cell areas is improved through cell nucleus detection. The cell membrane possibility of a plurality of communicating domains and each communicating domain is obtained, a plurality of cell membrane communicating domains are screened out, and the association degree of each cell membrane communicating domain and each cell nucleus communicating domain is obtained, so that the cell area of each cell nucleus communicating domain is obtained, the accuracy of detecting the follicular cell area is further improved, and the subsequent follicular cell area enhancement is more accurate. The cell possibility of each pixel point in the cell area of each cell nucleus communication domain is obtained, so that an enhanced image of the gray level image of the ovarian tissue is obtained, the gain factors are self-adapted according to the cell possibility, and the enhancement effect of the follicular cell area is improved. The invention obtains the follicular cell area by analyzing the characteristics of the cell nucleus and the cell membrane, and carries out linear transformation enhancement of the adaptive gain factors on the pixel points in the follicular cell area, thereby improving the follicular area enhancement effect in the ovarian tissue gray level image.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the steps of a follicular zone enhancement method for ovarian tissue analysis in accordance with the present invention;
FIG. 2 is a schematic diagram of an ovarian tissue gray scale image according to the present embodiment;
FIG. 3 is a schematic diagram of a minimum circumscribed circle and a maximum inscribed circle of a cell nucleus according to the present embodiment;
FIG. 4 is a schematic diagram showing cell membrane division around a nucleus according to the present embodiment;
fig. 5 is a schematic diagram of a minimum circumscribed circle of a cell according to the present embodiment.
Detailed Description
In order to further describe the technical means and effects of the present invention for achieving the intended purpose, the following detailed description refers to a specific implementation, structure, characteristics and effects of a follicular zone enhancement method for ovarian tissue analysis according to the present invention, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of a follicular region enhancement method for ovarian tissue analysis provided by the present invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps in a method for enhancing a follicular region for ovarian tissue analysis according to an embodiment of the present invention is shown, the method comprising the steps of:
step S001: collecting an ovarian tissue image, and carrying out graying treatment to obtain an ovarian tissue gray image; dividing the gray level image of the ovarian tissue into a plurality of super-pixel areas, and screening a plurality of cell nucleus connected areas from all the super-pixel areas.
The main purpose of this embodiment is to accurately detect the follicular region in the acquired image, and perform adaptive linear transformation increase on the gray value of the pixel point in the follicular region, so as to obtain a high-quality enhanced image, so as to facilitate the subsequent observation of the follicular region in the ovarian tissue.
And acquiring an ovarian tissue image under a microscope, and carrying out gray-scale treatment on the ovarian tissue image to obtain an ovarian tissue gray-scale image. The image graying process is a known technique, and a specific method is not described herein. Fig. 2 is a schematic diagram of an ovarian tissue gray scale image according to the present embodiment.
It is known that when a follicular region in ovarian tissue is observed, cell membrane edges of follicular cells in an image are found to adhere together, boundaries between cells are not obvious, gray scale contrast between adjacent follicular cell regions is poor, and it is difficult to effectively and accurately observe a region to be observed. By observing morphological characteristics of cells in the image, the nuclei of follicular cells are round, the cell membrane is continuous, and the gray value is low. Therefore, in order to accurately observe the cells in the follicular region, it is necessary to divide each follicular cell, and since the difference between the nucleus in the cell and the surrounding region is relatively obvious and the characteristics of the nucleus are obvious, the cell membrane continuity and the cell membrane opening direction of the cell are determined by determining the cell membrane wrapping property of the cell, the probability that a certain cell membrane belongs to a certain cell is calculated, and the positional relationship among the cell membrane, the nucleus and the pixel points is analyzed to obtain the probability that the follicular cell region belongs to, thereby obtaining the gain factor of the image by the probability.
By observing the gray level image of the ovarian tissue, it can be found that cell membrane adhesion exists between cells, the cells are distributed tightly, the gray level values of pixel points in the same cell tissue area are similar, the cell nucleus in the cells is a circular area, and the cell nucleus can be marked by extracting the circular area. For a circle, the smallest circumscribed circle and the largest inscribed circle are both the smallest circumscribed circle and the largest inscribed circle, so that the smaller the difference in radius between the smallest circumscribed circle and the largest inscribed circle of a certain connected domain, the more likely it is a nucleus.
And dividing the ovarian tissue gray level image into a plurality of super pixel areas by using a super pixel division algorithm.
What needs to be described is: the super-pixel segmentation algorithm is a well-known technique, and a specific method is not described here. The super pixel areas divided by the algorithm are connected areas, and the super pixel areas are also called super pixel blocks.
Any one super-pixel area divided by the gray level image of the ovary tissue is marked as a main super-pixel area.
And obtaining the minimum circumcircle of the main super pixel region by using a minimum circumcircle algorithm. And obtaining the maximum inscribed circle of the main super pixel region by using a maximum inscribed circle algorithm. The minimum circumscribed circle algorithm and the maximum inscribed circle algorithm are known techniques, and specific methods are not described herein.
And (3) the difference value obtained by subtracting the maximum inscribed circle radius of the main super pixel region from the minimum circumscribing circle radius of the main super pixel region is recorded as the circle-like degree of the main communication region. Fig. 3 is a schematic diagram of a minimum circumscribed circle and a maximum inscribed circle of a cell nucleus according to the present embodiment. In FIG. 3For the minimum circumscribed circle radius of the nucleus, < > about>Is the maximum inscribed circle radius of the nucleus.
The degree of rounding of each super pixel region is obtained in the above manner.
The core threshold value preset in this embodiment is 5, which is described as an example, and other values may be set in other embodiments, which is not limited in this embodiment.
And (3) marking the super-pixel region with the degree of quasi-circle larger than a preset nuclear threshold value as a nucleus connected region in all super-pixel regions of the ovarian tissue gray level image segmentation.
Step S002: in the region which is not the cell nucleus connected domain in the ovarian tissue gray level image, according to the pixel point gray level value and the distance between the pixel points, a plurality of connected domains and the cell membrane possibility of each connected domain are obtained, and a plurality of cell membrane connected domains are screened out from all connected domains.
It is known that a cell membrane is a membranous structure surrounding a substance inside a cell at the periphery of the cell, and that the cell membrane is a continuous region and can encapsulate the nucleus, and that it is possible to determine the region that is likely to be the cell membrane by analyzing the change in gray scale gradient in the image, and to determine whether there is a continuous edge around the periphery of the nucleus that can encapsulate the nucleus. Extending the normal direction of each pixel point on the continuous edges, and judging the wrapping property of the continuous edges on the cell nucleus according to the aggregation degree of the intersection point of the normal lines in the cell interior intersection point.
By observing the image, the pixel gray value of the cell membrane is higher than the pixel gray values of cytoplasm and cell interstitium at two sides of the cell membrane, and the cell membrane appears as a brighter area, so that the continuity of the cell membrane can be judged by the change direction of the gray gradient. Since there are other highlighted areas around the nucleus in the image, the highlighted pixels in the image include cell membranes and other tissues within the cell.
And because part of the cell membrane in the image is not obvious in the image, a plurality of discontinuous cell membrane connected domains can be obtained in the process of determining the cell membrane in the image, and the possibility that each cell membrane connected domain is a real cell membrane needs to be calculated. FIG. 4 is a schematic diagram showing cell membrane division around a nucleus according to the present embodiment.
In the ovarian tissue gray scale image, a region that is not a nucleus connected domain is referred to as a background region.
The gray threshold value preset in this embodiment is 140, which is described as an example, and other values may be set in other embodiments, which is not limited in this embodiment.
In the background area, an area formed by all pixel points with gray values larger than a preset gray threshold value is recorded as a target area. I.e. the target region comprises a cell membrane region.
Thus, the target region is divided into several connected regions using a region growing algorithm. Thus, there are communicating domains which are cell membranes.
Any one of the connected domains divided by the target region is referred to as a target connected domain.
And obtaining the center point of the target connected domain by using a connected domain gravity center calculation algorithm.
The region growing algorithm and the connected domain gravity center calculating algorithm are known techniques, and specific methods are not described herein.
And in the target connected domain, calculating the distance from each pixel point to the center point of the target connected domain, and recording the average value of the distances from all the pixel points to the center point of the target connected domain as the continuity degree of the target connected domain.
The calculation formula of the cell membrane probability of the target communicating domain is shown as follows:
；
wherein A is the possibility of cell membrane of the target connected domain, V is the variance of gray values of all pixels in the target connected domain, B is the continuity degree of the target connected domain,the present embodiment uses +.>The inverse proportion relation and normalization processing are presented, an implementer can set an inverse proportion function and a normalization function according to actual conditions, k is a preset exponential function adjustment value, and the exponential function is prevented from prematurely tending to 0. The present embodiment sets +.>In the description of this example, other values may be set in other embodiments, and the present example is not limited thereto.
What needs to be described is: the smaller the V is, the more likely the target connected domain is the cell membrane, and the cell membrane is a continuous arc, the length of other tissue connected domains in the cell membrane should be smaller than the cell membrane, so the larger the B is, the longer the target connected domain is, the more likely the target connected domain is the cell membrane. Thus usingThe greater A the probability of membrane of the target communicating domain, the more likely the target communicating domain is the membrane.
In the above manner, the cell membrane potential of each connected domain divided by the target region is obtained.
The preset film threshold value in this embodiment is 0.5, which is described as an example, and other values may be set in other embodiments, and this embodiment is not limited thereto.
Among all the connected domains divided in the target region, the connected domain having a membrane potential greater than a preset membrane threshold is designated as a membrane connected domain.
Step S003: marking any cell nucleus communicating domain as a reference nucleus communicating domain; marking any one cell membrane communicating domain as a reference membrane communicating domain; and obtaining the association degree of the reference membrane connected domain and the reference nuclear connected domain according to the center point of the reference nuclear connected domain, the intersection point of the normals of all pixel points on the skeleton line of the reference membrane connected domain and the cell membrane possibility of the reference membrane connected domain.
From the characteristic of cell membrane encapsulation of the material inside the cell, it is known that the arc opening formed by the cell membrane must be oriented toward the nucleus. I.e. the normals at various points on a cell membrane arc will intersect at a certain point.
In the gray level image of the ovarian tissue, any cell nucleus communicating domain is marked as a reference nucleus communicating domain, and any cell membrane communicating domain is marked as a reference membrane communicating domain.
And obtaining the center point of the reference nuclear connected domain by using a connected domain gravity center calculation algorithm.
And obtaining the skeleton line of the reference membrane connected domain by using a morphological refinement algorithm. Wherein, the morphological refinement algorithm is a known technology, and the specific method is not described here.
And calculating intersection points of normal lines of any two pixel points on a skeleton line of the reference film connected domain, and clustering the intersection points of the normal lines of all the pixel points by using a K-means clustering algorithm to obtain a clustering center point in the ovarian tissue gray level image.
And calculating the distances from the intersection points of the normals of all the pixel points to the clustering center points respectively, and recording the average value of the distances from the intersection points to the clustering center points respectively as the wrapping degree of the reference film connected domain.
What needs to be described is: the K-means clustering algorithm is a well-known technique, and the specific method is not described here. Since the intersection point of the normals of all the pixel points on the skeleton line corresponding to the cell membrane should be a point, the embodiment makes the cluster number K be 1, so that only one cluster and the cluster center point of the cluster are obtained. When two normals have no intersection points or the intersection points are not in the ovarian tissue gray level image, the intersection points are not used as the intersection points in clustering, and the intersection points outside the image are not used in the wrapping degree and subsequent analysis.
Therefore, whether the reference membrane connected domain corresponds to the reference nuclear connected domain or not can be reflected according to the aggregation of the intersection points and the distance between the clustering center point and the center point of the reference nuclear connected domain. Therefore, the calculation formula of the association degree S of the reference membrane connected domain and the reference core connected domain is as follows:
；
wherein S is the association degree of the reference membrane connected domain and the reference nuclear connected domain,for reference to the membrane potential of the membrane-connected domain, -/->And T is the distance from the clustering center point to the center point of the reference nuclear connected domain for the wrapping degree of the reference membrane connected domain. />Normalizing the data values to [0,1 ] as a linear normalization function]Within the interval.
What needs to be described is: the smaller L is, the more concentrated the intersection point of the normals of all pixel points on the skeleton line of the reference membrane connected domain is, namely the more likely that the reference membrane connected domain is a cell membrane,the larger the reference membrane-communicating domain is more likely to be a cell membrane, and the smaller T indicates that the larger the reference membrane-communicating domain is likely to be a cell membrane of the reference nuclear-communicating domain, therefore +.>The normalized value of (2) indicates the degree of association of the reference membrane-connected domain with the reference nuclear-connected domain, the greater the S, the more likely the reference membrane-connected domain is the cell membrane of the reference nuclear-connected domain.
In the above manner, the degree of association of each cell membrane communicating domain and the reference nuclear communicating domain is obtained.
Step S004: obtaining a cell region of the reference nuclear communication domain according to the association degree of all cell membrane communication domains and the reference nuclear communication domain respectively; and in the cell region of the reference nucleus communication domain, obtaining the cell probability of each pixel point according to the distance from the center point of the reference nucleus communication domain to the center point of the cell region, the association degree of the cell membrane communication domain and the reference nucleus communication domain and the shortest distance from each pixel point to the cell membrane communication domain and the reference nucleus communication domain respectively.
The preset determination threshold value in this embodiment is 0.6, which is described as an example, and other values may be set in other embodiments, which is not limited in this embodiment.
And in the association degrees of all the cell membrane communicating domains with the reference nuclear communicating domain respectively, the cell membrane communicating domain corresponding to the association degree larger than the preset judgment threshold value is marked as the cell membrane communicating domain of the reference nuclear communicating domain. Thus obtaining a plurality of cell membrane communicating domains corresponding to the reference nuclear communicating domain.
What needs to be described is: if the cell membrane connected domain of the reference nuclear connected domain does not exist, the reference nuclear connected domain is not analyzed later. And re-selecting one cell nucleus connected domain for analysis.
The cell membrane-connected domain of the reference nuclear connected domain is designated as the target membrane-connected domain.
The region constituted by the reference nuclear communication domain and all the target membrane communication domains is referred to as a reference region.
And obtaining the minimum circumcircle of the reference area by using a minimum circumcircle algorithm. When the reference nucleus communication domain is a true nucleus, the minimum circumcircle of the reference region should be the minimum circumcircle of one cell, thereby detecting the follicular cell region. Fig. 5 is a schematic diagram of a minimum circumscribed circle of a cell according to the present embodiment.
What needs to be described is: the minimum circumscribed circle of the inner nucleus 1 in fig. 5 should be the true nucleus of the cell, while the nucleus 2 and the nucleus 3 are false nuclei obtained according to the degree of rounding, and these false nuclei are eliminated in the process of obtaining the cell membrane connected domain of the cell nucleus connected domain and in the subsequent processes.
Therefore, the belonging relation between all the pixel points and the cells in the minimum circumscribing circle of the reference area is judged according to the distance relation between the cell nucleus, the cell membrane and the pixel points in the minimum circumscribing circle of the reference area. For a pixel point in the minimum circumcircle of the reference area, the closer the pixel point is to the cell nucleus and the farther the pixel point is from the cell membrane, the larger and more important the probability of the pixel point belonging to the cell, and the farther the pixel point is from the cell nucleus and the closer the pixel point is to the cell membrane, the smaller and less important the probability of the pixel point belonging to the cell. Meanwhile, the closer the reference nuclear communication domain is to the center of the minimum circumcircle of the reference region, the greater the probability that the reference nuclear communication domain belongs to the current cell.
The region within the smallest circumscribed circle of the reference region is designated as the cell region of the reference nuclear communication domain.
In the above manner, the cell region of each cell nucleus communication domain can be obtained.
In the cell region of the reference nuclear communication domain, any one pixel point is marked as a target point.
The calculation formula of the cell probability P of the target point is shown as follows:
；
where P is the cellular probability of the target site,for the degree of association of the ith target membrane-associated domain with the reference nuclear domain, +.>For the shortest distance from the target point to the ith target membrane connected domain, c is the shortest distance from the target point to the reference nuclear connected domain, R is the distance from the center point of the cell region of the reference nuclear connected domain to the center point of the reference nuclear connected domain, and n is the number of the target membrane connected domains. />Normalizing the data values to [0,1 ] as a linear normalization function]Within the interval.
What needs to be described is: the center point of the cell region of the reference connected region, that is, the center point of the smallest circumscribed circle of the reference region. When the target point is within the i-th target membrane communication domain,should be 0, c should be 0 when the target point is within the reference kernel connected domain. />The smaller the target point is, the closer the target point is to the ith target membrane-associated domain, i.e. the less likely it is a cell, +.>The larger the target point, the more likely it is that the target point is closer to the reference nuclear communication domain, i.e., the cell, wherein c plus 1 is to prevent the denominator from being 0. But->The larger the i-th target membrane-associated domain, the more likely it is the membrane of the reference nuclear domain, the +.>The more trusted. When R is 0, it is indicated that the cell region center point of the reference nuclear communication domain coincides with the center point of the reference nuclear communication domain, the more likely the cell region is a true cell region, and therefore +.>And (2) represents the cell probability of the target point, wherein R plus 1 is to prevent the denominator from being 0.
In the above manner, the cell probability of each pixel point in the cell region of the reference nuclear communication domain is obtained.
Step S005: and obtaining an enhanced image of the ovarian tissue gray level image according to the cell possibility of all pixel points in the cell area of all the cell nucleus connected domains.
The cell threshold value preset in this example is 0.7, which is described as an example, and other values may be set in other embodiments, and this example is not limited thereto.
And in the cell region of the reference nuclear communication domain, marking the pixel points with the cell probability larger than a preset cell threshold as cell points. Thereby obtaining a plurality of cell points in the cell region of the reference nuclear communication domain.
In the above manner, a plurality of cell points and cell possibilities of the cell points within the cell region of each nucleus communication domain are obtained. The cell probability of a plurality of cell points and cell points in the ovarian tissue gray level image is obtained, namely, the cell points have a high probability of being pixel points of a follicular cell area in the ovarian tissue gray level image, and the larger the cell probability is, the larger the probability of being follicular cells is, and the larger gray level contrast enhancement is required.
What needs to be described is: if a certain cell point in the gray level image of the ovarian tissue is simultaneously located in the cell area of the plurality of cell nucleus connected domains, the cell point has a plurality of cell possibilities, and the maximum value in the cell possibilities is taken as the cell point cell possibility in the embodiment.
In the gray scale image of the ovarian tissue, other pixel points which are not cell points are marked as normal points, the sum of the cell probability of each cell point and 1 is marked as the gain factor of each cell point, and the gain factor of each normal point is set to be 1. Thereby obtaining the gain factor of each pixel point in the ovarian tissue gray level image.
And (3) carrying out image enhancement by using a linear transformation algorithm according to gain factors of all pixel points in the ovarian tissue gray level image to obtain an enhanced image of the ovarian tissue gray level image.
What needs to be described is: the image enhancement linear transformation algorithm is a known technology, the gain factor and the bias factor are main parameters of the linear transformation algorithm, the gain factor is used for adjusting the contrast of the image, the larger gain factor can increase the contrast, so that the image is sharper and brighter, and the smaller gain factor can reduce the contrast, so that the image is softer. The bias factor is used to adjust the brightness of the image, with a larger bias factor causing the image to become brighter and a smaller bias factor causing the image to darken. The present embodiment does not change the brightness of the image, so the preset bias factor b is set to 0, which is described as an example, and other values may be set in other embodiments, which is not limited. In the gray level image of the ovarian tissue, for the pixel points which are not cell points, the gain factor is 1, namely contrast enhancement is not performed, and for the cell points, the gain factor is 1 and the cell probability is increased, namely the cell points with higher cell probability are subjected to larger contrast enhancement, so that the visibility of the follicular region is improved. Gray scale image of ovarian tissueThe linear transformation formula is:wherein b is a predetermined bias factor, +.>Gain factor of jth pixel point in gray level image of ovary tissue, < >>Gray value of jth pixel point in gray image of ovary tissue, < >>And (3) linearly transforming the j-th pixel point in the ovarian tissue gray level image to obtain a gray level value, wherein m is the number of the pixel points in the ovarian tissue gray level image. When->Above 255, it is known to use a gray level cutoff, i.e. to cut off a gray level value exceeding 255 to 255, or to use a gray level normalization, i.e. to linearly scale a gray level value exceeding 255, and map it to a range of 0 to 255.
The present invention has been completed.
In summary, in the embodiment of the present invention, the gray level image of the ovarian tissue is obtained, and is divided into a plurality of super-pixel regions, from which a plurality of nuclear connected regions are selected. The method comprises the steps of obtaining a plurality of connected domains and the possibility of cell membranes of each connected domain, and screening the plurality of cell membrane connected domains from the obtained possibility. The association degree of each cell membrane communicating domain and each cell nucleus communicating domain is obtained, so that the cell area of each cell nucleus communicating domain is obtained, and the cell possibility of each pixel point in the cell area of each cell nucleus communicating domain is obtained, so that the enhanced image of the gray level image of the ovarian tissue is obtained. According to the invention, the characteristics of cell nuclei and cell membranes are analyzed to obtain the follicular cell area, and the linear transformation enhancement of the adaptive gain factors is carried out on the pixel points in the follicular cell area, so that the follicular area enhancement effect in the ovarian tissue gray level image is improved.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (2)
1. A method for enhancing a follicular zone for ovarian tissue analysis, the method comprising the steps of:
collecting an ovarian tissue image, and carrying out graying treatment to obtain an ovarian tissue gray image; dividing an ovarian tissue gray level image into a plurality of super-pixel areas, and screening a plurality of cell nucleus connected areas from all the super-pixel areas;
in the area which is not the cell nucleus connected domain in the ovarian tissue gray level image, according to the pixel point gray level value and the distance between the pixel points, obtaining a plurality of connected domains and the cell membrane possibility of each connected domain, and screening a plurality of cell membrane connected domains from all connected domains;
marking any cell nucleus communicating domain as a reference nucleus communicating domain; marking any one cell membrane communicating domain as a reference membrane communicating domain; obtaining the association degree of the reference membrane connected domain and the reference nuclear connected domain according to the center point of the reference nuclear connected domain, the intersection point of the normals of all pixel points on the skeleton line of the reference membrane connected domain and the cell membrane possibility of the reference membrane connected domain;
obtaining a cell region of the reference nuclear communication domain according to the association degree of all cell membrane communication domains and the reference nuclear communication domain respectively; obtaining the cell probability of each pixel point in the cell region of the reference nuclear communication domain according to the distance from the center point of the reference nuclear communication domain to the center point of the cell region, the association degree of the cell membrane communication domain and the reference nuclear communication domain and the shortest distance from each pixel point to the cell membrane communication domain and the reference nuclear communication domain respectively;
obtaining an enhanced image of the ovarian tissue gray level image according to the cell possibility of all pixel points in the cell area of all cell nucleus connected domains;
in the area which is not a cell nucleus connected domain in the ovarian tissue gray level image, according to the pixel point gray level value and the distance between the pixel points, the cell membrane possibility of a plurality of connected domains and each connected domain is obtained, and a plurality of cell membrane connected domains are screened out from all connected domains, comprising the following specific steps:
in the gray level image of the ovarian tissue, a region which is not a cell nucleus connected domain is marked as a background region;
in the background area, the area formed by all pixel points with gray values larger than a preset gray threshold is marked as a target area;
dividing a target area into a plurality of connected areas by using an area growth algorithm;
any one of the connected domains divided by the target area is marked as a target connected domain;
calculating the distance from each pixel point to the center point of the target connected domain in the target connected domain, and recording the average value of the distances from all the pixel points to the center point of the target connected domain as the continuity degree of the target connected domain;
obtaining the cell membrane possibility of the target connected domain according to the gray values of all pixel points in the target connected domain and the continuity degree of the target connected domain;
among all the connected domains segmented in the target area, the connected domain with the possibility of the cell membrane being larger than a preset membrane threshold value is marked as a cell membrane connected domain;
the specific calculation formula corresponding to the cell membrane probability of the target connected domain is obtained according to the gray values of all the pixel points in the target connected domain and the continuity degree of the target connected domain:
wherein A is the possibility of cell membrane of the target connected domain, V is the variance of gray values of all pixels in the target connected domain, B is the continuity degree of the target connected domain,k is a preset exponential function adjusting value, wherein the exponential function is an exponential function based on a natural constant;
the association degree of the reference membrane connected domain and the reference nuclear connected domain is obtained according to the center point of the reference nuclear connected domain, the intersection point of the normals of all pixel points on the skeleton line of the reference membrane connected domain and the cell membrane possibility of the reference membrane connected domain, and the method comprises the following specific steps:
obtaining a skeleton line of the reference membrane connected domain by using a morphological refinement algorithm;
calculating intersection points of normal lines of any two pixel points on a skeleton line of a reference film connected domain, and clustering the intersection points of the normal lines of all the pixel points by using a K-means clustering algorithm to obtain a clustering center point in an ovarian tissue gray level image;
calculating the distances from the intersection points of the normals of all the pixel points to the clustering center points respectively, and marking the average value of the distances from the intersection points to the clustering center points respectively as the wrapping degree of the reference film connected domain;
obtaining the association degree of the reference membrane connected domain and the reference nuclear connected domain according to the possibility of the cell membrane of the reference membrane connected domain, the wrapping degree of the reference membrane connected domain and the distance from the clustering center point to the center point of the reference nuclear connected domain;
the specific calculation formula corresponding to the association degree of the reference membrane connected domain and the reference nuclear connected domain is obtained according to the cell membrane possibility of the reference membrane connected domain, the wrapping degree of the reference membrane connected domain and the distance from the clustering center point to the center point of the reference nuclear connected domain, wherein the specific calculation formula comprises the following steps:
wherein S is the association degree of the reference membrane connected domain and the reference nuclear connected domain,for reference to the membrane potential of the membrane-connected domain, -/->For the wrapping degree of the reference membrane connected domain, T is the distance from the clustering center point to the reference nuclear connectionDistance of center point of through domain,/>Is a linear normalization function;
obtaining the cell region of the reference nuclear communicating domain according to the association degree of all cell membrane communicating domains and the reference nuclear communicating domain, wherein the cell region comprises the following specific steps:
in the association degrees of all the cell membrane communicating domains with the reference nuclear communicating domain respectively, the cell membrane communicating domain corresponding to the association degree larger than a preset judging threshold value is marked as the cell membrane communicating domain of the reference nuclear communicating domain;
the cell membrane communicating domain of the reference nuclear communicating domain is marked as a target membrane communicating domain;
the region formed by the reference nuclear communication domain and all the target membrane communication domains is marked as a reference region;
obtaining a minimum circumcircle of the reference area by using a minimum circumcircle algorithm;
the area in the smallest circumcircle of the reference area is marked as the cell area of the reference nucleus communication area;
in the cell region of the reference nuclear communication domain, according to the distance from the center point of the reference nuclear communication domain to the center point of the cell region, the association degree of the cell membrane communication domain and the reference nuclear communication domain, and the shortest distance from each pixel point to the cell membrane communication domain and the reference nuclear communication domain, the cell probability of each pixel point is obtained, which comprises the following specific steps:
in the cell area of the reference nucleus communication domain, marking any pixel point as a target point;
obtaining the cell probability of the target point according to the association degree of all the target membrane connected domains with the reference nuclear connected domain respectively, the shortest distance between the target point and the reference nuclear connected domain and all the target membrane connected domains respectively, and the distance between the center point of the cell region of the reference nuclear connected domain and the center point of the reference nuclear connected domain;
the specific calculation formula corresponding to the cell probability of the target point is obtained according to the association degree of all the target membrane connected domains with the reference nuclear connected domain, the shortest distance between the target point and the reference nuclear connected domain and the shortest distance between the cell region center point of the reference nuclear connected domain and the reference nuclear connected domain, and the specific calculation formula corresponding to the cell probability of the target point is:
where P is the cellular probability of the target site,for the degree of association of the ith target membrane-associated domain with the reference nuclear domain, +.>For the shortest distance from the target point to the ith target membrane-connected domain, c is the shortest distance from the target point to the reference nuclear-connected domain, R is the distance from the center point of the cell region of the reference nuclear-connected domain to the center point of the reference nuclear-connected domain, n is the number of target membrane-connected domains,is a linear normalization function;
according to the cell possibility of all pixel points in the cell area of all cell nucleus connected domains, obtaining an enhanced image of an ovarian tissue gray level image, comprising the following specific steps:
in the cell areas of all the cell nucleus communicating domains, pixel points with the cell possibility larger than a preset cell threshold value are marked as cell points;
in the gray level image of the ovary tissue, marking other pixel points which are not cell points as normal points; adding the sum of the cell probability of each cell point to 1, and recording the sum as a gain factor of each cell point; setting the gain factor of each normal point to 1;
and (3) carrying out image enhancement by using a linear transformation algorithm according to gain factors of all pixel points in the ovarian tissue gray level image to obtain an enhanced image of the ovarian tissue gray level image.
2. The method for enhancing a follicle area for ovarian tissue analysis according to claim 1, wherein the steps of dividing the gray level image of the ovarian tissue into a plurality of super-pixel areas and screening a plurality of nuclear connected areas from all the super-pixel areas comprise the following specific steps:
dividing the ovarian tissue gray level image into a plurality of super pixel areas by using a super pixel dividing algorithm;
any super-pixel area divided by the gray level image of the ovary tissue is marked as a main super-pixel area;
a minimum circumscribed circle algorithm and a maximum inscribed circle algorithm are used for respectively obtaining a minimum circumscribed circle and a maximum inscribed circle of the main super pixel area;
the difference value of the radius of the minimum circumscribing circle of the main super pixel area minus the radius of the maximum inscribing circle is recorded as the circle-like degree of the main communicating area;
and (3) marking the super-pixel region with the degree of quasi-circle larger than a preset nuclear threshold value as a nucleus connected region in all super-pixel regions of the ovarian tissue gray level image segmentation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410063209.6A CN117575977B (en) | 2024-01-17 | 2024-01-17 | Follicular region enhancement method for ovarian tissue analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410063209.6A CN117575977B (en) | 2024-01-17 | 2024-01-17 | Follicular region enhancement method for ovarian tissue analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117575977A CN117575977A (en) | 2024-02-20 |
| CN117575977B true CN117575977B (en) | 2024-04-02 |
Family
ID=89862893
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410063209.6A Active CN117575977B (en) | 2024-01-17 | 2024-01-17 | Follicular region enhancement method for ovarian tissue analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117575977B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016091016A1 (en) * | 2014-12-12 | 2016-06-16 | 山东大学 | Nucleus marker watershed transformation-based method for splitting adhered white blood cells |
| CN112823376A (en) * | 2018-10-15 | 2021-05-18 | 文塔纳医疗系统公司 | Image enhancement for improved nuclear detection and segmentation |
| CN115345893A (en) * | 2022-09-15 | 2022-11-15 | 吉林大学中日联谊医院 | Ovarian tissue canceration region segmentation method based on image processing |
| CN116503392A (en) * | 2023-06-26 | 2023-07-28 | 细胞生态海河实验室 | Follicular region segmentation method for ovarian tissue analysis |
| CN116912255A (en) * | 2023-09-14 | 2023-10-20 | 济南宝林信息技术有限公司 | Follicular region segmentation method for ovarian tissue analysis |
| WO2023220913A1 (en) * | 2022-05-17 | 2023-11-23 | 京东方科技集团股份有限公司 | Cell image processing method, electronic device and storage medium |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2001071663A1 (en) * | 2000-03-23 | 2001-09-27 | Japan Science And Technology Corporation | Cell lineage extracting method |
| US20030165263A1 (en) * | 2002-02-19 | 2003-09-04 | Hamer Michael J. | Histological assessment |
| US20050136549A1 (en) * | 2003-10-30 | 2005-06-23 | Bioimagene, Inc. | Method and system for automatically determining diagnostic saliency of digital images |
| US7760927B2 (en) * | 2003-09-10 | 2010-07-20 | Bioimagene, Inc. | Method and system for digital image based tissue independent simultaneous nucleus cytoplasm and membrane quantitation |
| DE102010063965A1 (en) * | 2010-12-22 | 2012-06-28 | Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. | Apparatus and method for determining objects of a color photograph |
| EP3108446B1 (en) * | 2014-02-21 | 2019-03-20 | Ventana Medical Systems, Inc. | Medical image analysis for identifying biomarker-positive tumor cells |
| WO2016075096A1 (en) * | 2014-11-10 | 2016-05-19 | Ventana Medical Systems, Inc. | Classifying nuclei in histology images |
| US20190042826A1 (en) * | 2017-08-04 | 2019-02-07 | Oregon Health & Science University | Automatic nuclei segmentation in histopathology images |
| CN110853022B (en) * | 2019-11-14 | 2020-11-06 | 腾讯科技（深圳）有限公司 | Pathological section image processing method, device and system and storage medium |
-
2024
- 2024-01-17 CN CN202410063209.6A patent/CN117575977B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016091016A1 (en) * | 2014-12-12 | 2016-06-16 | 山东大学 | Nucleus marker watershed transformation-based method for splitting adhered white blood cells |
| CN112823376A (en) * | 2018-10-15 | 2021-05-18 | 文塔纳医疗系统公司 | Image enhancement for improved nuclear detection and segmentation |
| WO2023220913A1 (en) * | 2022-05-17 | 2023-11-23 | 京东方科技集团股份有限公司 | Cell image processing method, electronic device and storage medium |
| CN115345893A (en) * | 2022-09-15 | 2022-11-15 | 吉林大学中日联谊医院 | Ovarian tissue canceration region segmentation method based on image processing |
| CN116503392A (en) * | 2023-06-26 | 2023-07-28 | 细胞生态海河实验室 | Follicular region segmentation method for ovarian tissue analysis |
| CN116912255A (en) * | 2023-09-14 | 2023-10-20 | 济南宝林信息技术有限公司 | Follicular region segmentation method for ovarian tissue analysis |
Non-Patent Citations (1)
| Title |
|---|
| Advanced Multi-Microscopic Views Cell Semi-supervised Segmentation;Fang Hu等;《ARXIV》;20230321;论文第1-23 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117575977A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Song et al. | A deep learning based framework for accurate segmentation of cervical cytoplasm and nuclei | |
| CN108765369B (en) | Method, apparatus, computer device and storage medium for detecting lung nodule | |
| CN111951221B (en) | Glomerular cell image recognition method based on deep neural network | |
| WO2021203795A1 (en) | Pancreas ct automatic segmentation method based on saliency dense connection expansion convolutional network | |
| US10234442B2 (en) | Device and method for finding cell nucleus of target cell from cell image | |
| CN110929728B (en) | Image region-of-interest dividing method, image segmentation method and device | |
| CN112241948A (en) | Lung nodule detection and analysis method and system capable of adapting to thickness of layer | |
| CN116630358B (en) | Threshold segmentation method for brain tumor CT image | |
| CN116912255B (en) | Follicular region segmentation method for ovarian tissue analysis | |
| CN116310882B (en) | Forestry information identification method based on high-resolution remote sensing image | |
| CN114359288B (en) | Medical image cerebral aneurysm detection and positioning method based on artificial intelligence | |
| CN116503392A (en) | Follicular region segmentation method for ovarian tissue analysis | |
| CN109147932B (en) | Cancer cell HER2 gene amplification analysis method and system | |
| CN117252893B (en) | Segmentation processing method for breast cancer pathological image | |
| CN110211117B (en) | Processing system for identifying linear tubular objects in medical image and optimized segmentation method | |
| CN109559321A (en) | A kind of sonar image dividing method and equipment | |
| CN113822279B (en) | Infrared target detection method, device, equipment and medium based on multi-feature fusion | |
| CN116385438A (en) | Nuclear magnetic resonance tumor region extraction method | |
| CN117575977B (en) | Follicular region enhancement method for ovarian tissue analysis | |
| CN116342607B (en) | Power transmission line defect identification method and device, electronic equipment and storage medium | |
| CN101897592A (en) | Method and device for extracting metallic foreign body from X-ray image | |
| CN115131628A (en) | Mammary gland image classification method and equipment based on typing auxiliary information | |
| CN115457549A (en) | Aged cell microscopic image identification method based on deep learning | |
| Parpulov et al. | Convolutional neural network application for cells segmentation in immunocytochemical study | |
| Li et al. | Automatic detection of leukocytes for cytometry with color decomposition |
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 |