CN112419235A - Sperm activity detection system and method - Google Patents
Sperm activity detection system and method Download PDFInfo
- Publication number
- CN112419235A CN112419235A CN202011142158.4A CN202011142158A CN112419235A CN 112419235 A CN112419235 A CN 112419235A CN 202011142158 A CN202011142158 A CN 202011142158A CN 112419235 A CN112419235 A CN 112419235A
- Authority
- CN
- China
- Prior art keywords
- sperm
- image
- segmentation
- sub
- preset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000000694 effects Effects 0.000 title claims abstract description 62
- 238000001514 detection method Methods 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 15
- 230000011218 segmentation Effects 0.000 claims abstract description 80
- 230000019100 sperm motility Effects 0.000 claims description 26
- 238000003709 image segmentation Methods 0.000 claims description 18
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000000926 separation method Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 6
- 230000009087 cell motility Effects 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 230000035899 viability Effects 0.000 claims description 4
- 238000010998 test method Methods 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000004071 biological effect Effects 0.000 abstract description 3
- 238000004364 calculation method Methods 0.000 abstract description 3
- 230000033001 locomotion Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 6
- 238000004806 packaging method and process Methods 0.000 description 6
- 239000000853 adhesive Substances 0.000 description 5
- 230000001070 adhesive effect Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 4
- 230000008602 contraction Effects 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 239000003292 glue Substances 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 239000010931 gold Substances 0.000 description 2
- 229910052737 gold Inorganic materials 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 239000002131 composite material Substances 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06T2207/10—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
- G06T2207/20084—Artificial neural networks [ANN]
Abstract
The invention discloses a sperm activity detection system and a sperm activity detection method, and relates to the field of biological activity detection. The method comprises the following steps: the acquisition module is used for acquiring a long exposure image of a sperm sample; the segmentation module is used for segmenting the long exposure image to obtain a sperm segmentation image; the processing module specifically comprises: the device comprises a storage unit, a calculation unit and a judgment unit. The technical scheme of the invention has the beneficial effects that: utilize active sperm can produce the blurred characteristics in edge under the long exposure image, determine the motion activity degree of sperm through the survey sperm that awaits measuring at the fuzzy degree of long exposure image to further judge the activity of sperm, this technical scheme is applied to the biological activity detection with computer vision technique, only needs a long exposure image, can carry out the active detection of sperm, has practiced thrift the cost of labor, and has ensured the degree of accuracy that the sperm activity detected, makes detection achievement have high-efficient convenient advantage.
Description
Technical Field
The invention relates to the field of biological activity detection, in particular to a sperm activity detection system and a sperm activity detection method.
Background
Nowadays, with the rapid development of artificial intelligence technology, the technology is widely applied to various fields, such as the medical detection field, and there are many applications of artificial intelligence technology. In the face of a large amount of repetitive work, the artificial intelligence technology can often play an efficient role. Therefore, it is expected that combining medical detection images with artificial intelligence technology will be a major trend in the future medical field development.
In the prior art, a microscope is generally used for observing a sample in a sperm activity detection method, and the sperm sample is observed and judged and active sperms are counted manually; however, if there are a large number of samples to be detected, not only the cost of manual work is greatly increased, but also there is a certain influence on the accuracy, thereby causing the detection result to be inaccurate.
Disclosure of Invention
According to the problems in the prior art, the sperm activity detection system and the sperm activity detection method are provided, and the aim is to reduce the labor cost, ensure the accuracy of sperm sample activity detection and ensure the high-efficiency detection.
The technical scheme specifically comprises the following steps:
a sperm motility detection system, comprising:
the acquisition module is used for acquiring a long exposure image of a sperm sample by adopting a long exposure acquisition mode;
the segmentation module is connected to the acquisition module and is used for segmenting the long exposure image to obtain a sperm segmentation image;
the sperm segmentation image comprises a plurality of sub-images, and each sub-image is used for representing one sperm;
the processing module is connected to the segmentation module and specifically comprises:
the storage unit is used for preprocessing to form a preset sperm activity threshold value and storing the sperm activity threshold value;
the computing unit is used for respectively computing and obtaining the edge fuzzy value of each sub-image;
and the judging unit is respectively connected with the storage unit and the calculating unit and is used for respectively comparing the edge fuzzy value of each sub-image with the preset sperm activity threshold value and outputting the sperm activity detection result according to the comparison result.
Preferably, the segmentation module includes:
the first segmentation unit is used for carrying out connected domain segmentation on the long exposure image to obtain a first segmentation image so as to carry out preliminary separation on the image including the sperms and the background part;
and the second segmentation unit is connected to the first segmentation unit and is used for carrying out image segmentation on the first segmentation image by adopting an image segmentation neural network to obtain the sperm segmentation image.
Preferably, the following formula is used as a loss function of the image segmentation neural network:
wherein the content of the first and second substances,
Pρ(X)representing the output probability of a pixel point X in the first segmentation image on a channel where a real label is located;
log(Pρ(X)(X)) represents a cross entropy loss function of the pixel point X;
ωc(X) representing the weight of the balance category proportion of each pixel point X;
d1(X) represents the distance between said pixel point X and the closest pixel point representing the cell;
d2(X) representing the distance between said pixel X and the pixel representing the cell that is the second closest thereto;
ω0and sigma is a constant.
Preferably, the storage unit further includes:
an acquisition means for acquiring in advance a first preset image including only normal viable sperm and a second preset image including only non-viable sperm;
the processing component is used for calculating according to the first preset image and the second preset image to obtain a preset sperm activity threshold value;
and the storage component is connected with the processing component and is used for storing the preset sperm activity threshold value.
Preferably, the processing component calculates the sperm motility threshold according to the formula:
wherein the content of the first and second substances,
t is the sperm motility threshold;
T1the variance value of the first preset image is obtained;
T2and the variance value of the second preset image is obtained.
Preferably, the calculating unit is configured to calculate a variance value of each sub-image and use the variance value as the edge blur value;
the determining unit specifically includes:
the comparison component is used for respectively comparing whether the edge fuzzy value of each sub-image reaches the preset sperm activity threshold value or not and outputting the comparison result;
a marking component connected to the comparing component for marking the corresponding sub-image as normal viable sperm when the comparison result indicates that the edge blur value reaches the sperm viability threshold;
and the counting component is connected to the marking component and used for counting a first total number of the sub-images in the sperm segmentation image and a second total number of the sub-images marked as normal active sperm, and calculating a ratio of the second total number to the first total number to be output as the detection result.
In the technical scheme, the method further comprises the following steps:
a sperm motility detection method is applied to the sperm motility detection system, and comprises the following steps:
s1, acquiring a long exposure image of the sperm sample in a long exposure acquisition mode;
step S2, segmenting the long exposure image to obtain a sperm segmentation image, wherein the sperm segmentation image comprises a plurality of sub-images, and each sub-image is used for representing a sperm;
step S3, respectively calculating and obtaining the edge fuzzy value of each sub-image;
and step S4, comparing the edge fuzzy value of each sub-image with a preset sperm activity threshold value respectively, and outputting a sperm activity detection result according to the comparison result.
Preferably, the step S2 specifically includes:
step S21, carrying out connected domain segmentation on the long exposure image to obtain a first segmentation image so as to carry out preliminary separation on the image including the sperms and the background part;
and step S22, carrying out image segmentation on the first segmentation image through an image segmentation neural network to obtain the sperm segmentation image.
Preferably, the method further comprises a process of presetting the sperm motility threshold, and specifically comprises the following steps:
step A1, acquiring a first preset image only including normal active sperm and a second preset image only including inactive sperm in advance;
step A2, respectively calculating a variance value of the first preset image and a variance value of the second preset image;
step A3, calculating a variance threshold according to the variance value of the first preset image and the variance value of the second preset image, and storing the variance threshold as the sperm motility threshold;
in step S3, respectively calculating a variance value of each sub-image as the edge blur value;
the step S4 specifically includes:
step S41, comparing the variance value of each sub-image with the preset sperm activity threshold value respectively, and outputting a comparison result;
step S42, according to the comparison result, marking the subimage with the variance value reaching the sperm activity threshold value as normal active sperm;
step S43, counting a first total number of the sub-images in the sperm segmentation image and a second total number of the sub-images marked as normal active sperm, and calculating a ratio of the second total number to the first total number to output as the detection result.
The technical scheme of the invention has the beneficial effects that: the invention utilizes the characteristic that the active sperm can generate fuzzy edge under the long exposure image, and judges the motion activity degree of the sperm by measuring the fuzzy degree of the sperm to be detected in the long exposure image, thereby further judging the activity of the sperm.
Drawings
Embodiments of the present invention will now be described more fully hereinafter with reference to the accompanying drawings. The drawings are, however, to be regarded as illustrative and explanatory only and not as restrictive of the scope of the invention.
FIG. 1 is a block diagram of a sperm cell motility detection system according to an embodiment of the present invention;
FIG. 2 is a block diagram of a partitioning module according to an embodiment of the present invention;
FIG. 3 is a block diagram of a processing module according to an embodiment of the present invention;
FIG. 4 is a component diagram of a memory cell according to an embodiment of the present invention;
FIG. 5 is a component diagram of a computing unit of an embodiment of the present invention;
FIG. 6 is a schematic flow chart of a sperm cell motility detection method of embodiments of the present invention;
FIG. 7 is a flowchart illustrating step S2 according to an embodiment of the present invention;
FIG. 8 is a schematic flow chart of a process for obtaining the sperm motility threshold in accordance with embodiments of the present invention;
fig. 9 is a flowchart illustrating step S4 according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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 invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
This technical scheme provides a sperm activity detecting system, its characterized in that includes:
the acquisition module 1 is used for acquiring a long exposure image of a sperm sample by adopting a long exposure acquisition mode;
the segmentation module 2 is connected to the acquisition module 1 and is used for segmenting the long exposure image to obtain a sperm segmentation image;
the sperm segmentation image comprises a plurality of sub-images, each sub-image being used for representing a sperm;
the processing module 3 is connected to the segmentation module 2, and specifically includes:
the storage unit 31 is used for preprocessing and forming a preset sperm activity threshold value and storing the sperm activity threshold value;
a calculating unit 32, configured to calculate and obtain an edge blur value of each sub-image;
and the judging unit 33 is respectively connected with the storage unit 31 and the calculating unit 32, and is used for respectively comparing the edge fuzzy value of each sub-image with a preset sperm activity threshold value and outputting a sperm activity detection result according to the comparison result.
Specifically, the calculation unit 32 performs grayscale image conversion on the sperm segmentation image by using the laplacian operator.
Specifically, through the conversion of the gray level image, the sperm segmentation image becomes a single-channel image.
Specifically, when the edge blur value of the sub-image reaches the sperm motility threshold value, the judgment unit 33 judges the corresponding sperm as a normal viable sperm.
In a preferred embodiment, the acquisition module 1 acquires a long exposure image of a very large field of view of a sperm sample using an image sensor chip.
Specifically, the structure adopted by a single photosensitive pixel in the image sensor chip can be a composite dielectric gate photosensitive detector or a semi-floating gate transistor.
Specifically, the size of a single photosensitive pixel of the image sensor chip is less than or equal to 1 micron multiplied by 1 micron, the number of photosensitive pixels of the whole image sensor chip is more than or equal to 1 hundred million, the smaller the pixel size is, the higher the resolution is determined, the finer sample detail information can be seen, and meanwhile, the pixel size of more than one hundred million ensures that the image sensor chip has a large view field under the condition of high resolution.
Specifically, in this embodiment, the method further includes:
chip photosensitive area and chip packaging glue.
Specifically, the gold wires for packaging and connecting the periphery of the chip photosensitive area A2 are protected by chip packaging glue due to the application of the gold wires to the detection of liquid samples, so that the waterproof effect is achieved.
Specifically, the chip packaging adhesive may be an ultraviolet adhesive or an AB adhesive.
Furthermore, after the chip packaging adhesive is cured, a bulge with the height of about 0.5-1mm is formed on the periphery of the chip packaging adhesive, so that the normal work inside the chip can be effectively protected.
Specifically, the segmentation module 2 includes:
the first segmentation unit 21 is configured to perform connected domain segmentation on the long-exposure image to obtain a first segmented image, so as to perform preliminary separation on an image including sperm and a background portion;
and the second segmentation unit 22 is connected to the first segmentation unit 21 and is used for performing image segmentation on the first segmentation image by using an image segmentation neural network to obtain a sperm segmentation image.
Specifically, the sperm segmentation image may be a set of a plurality of sub-images, or may be a single image.
Specifically, the image segmentation neural network adopts a full convolution network.
Specifically, one connected domain is a pixel set composed of adjacent pixel points with the same pixel value, the connected domain is divided, that is, each connected domain is given a unique identifier to distinguish other connected domains, and the first dividing unit 21 preliminarily distinguishes the sperms and the background in the long-exposure image through the connected domain division.
Specifically, the full convolution network includes:
the contraction unit comprises a first convolution layer and a pooling layer, and the first segmentation image passes through the first convolution layer and the pooling layer to obtain a first sperm image characteristic diagram with reduced size;
the number of channels output by the first segmentation image after passing through a first convolution layer is doubled;
the expansion unit comprises a second convolution layer and a deconvolution layer, and the first segmentation image passes through the second convolution layer and the deconvolution layer to obtain a second sperm image characteristic diagram with expanded size;
the number of channels output by the first segmented image after passing through each second convolution layer is reduced by half.
Specifically, the number of channels of the preliminary sperm image is increased through the first convolution layer, and the number of pixels of the preliminary sperm image is increased at the moment.
Specifically, through the second convolution layer, the number of channels of the preliminary sperm image is reduced by half, and the number of pixels of the preliminary sperm image is reduced at the moment.
Further, the first segmentation image enters a contraction unit to obtain a first primary sperm image characteristic diagram, and the first segmentation image enters an expansion unit to obtain a second primary sperm image characteristic diagram.
Specifically, the shrinking unit and the expanding unit fuse the characteristic information of the two parts in a dimension splicing mode, so that the characteristic information of the sperm image can be better learned, and the finally finely divided second divided sperm image is obtained.
Specifically, the following formula is adopted as a loss function of the image segmentation neural network:
wherein the content of the first and second substances,
Pρ(X)representing the output probability of a pixel point X in the first segmentation image on a channel where the real label is located;
log(Pρ(X)(X)) represents the cross entropy loss function of pixel point X;
ωc(X) representing the weight of the balance category proportion of each pixel point X;
d1(X) the distance between pixel point X and the closest pixel point representing the cell;
d2(X) the distance between pixel X and the pixel representing the cell that is the second closest thereto;
ω0and sigma is a constant.
Specifically, the loss function is a function that maps an event (an element in a sample space) to a real number expressing economic or opportunity cost associated with the event, and is used to measure the degree of disparity between the first segmented image and the actual sample image.
Specifically, the second segmentation unit 22 uses the label of the pixel point X identified as sperm as a real label, and uses the label of the pixel point identified as non-sperm as a false label.
Specifically, in the image segmentation network, the number of final output channels is two, and the labels of the pixel points include two types, namely real labels and false labels.
Specifically, the storage unit 31 further includes:
an acquisition section 311 for acquiring in advance a first preset image including only normal viable sperm and a second preset image including only non-viable sperm;
the processing component 312 is connected to the acquiring component 311, and is configured to calculate according to the first preset image and the second preset image to obtain a preset sperm activity threshold;
the storage unit 313 is connected to the processing unit 312 for storing a preset sperm motility threshold.
Specifically, the processing component 312 calculates the sperm cell motility threshold according to the following equation:
wherein the content of the first and second substances,
t is sperm activity threshold;
T1the variance value of the first preset image is obtained;
T2is the variance value of the second preset image.
Specifically, the storage unit 31 performs grayscale image conversion on the first preset image and the second preset image by using the laplacian operator.
Specifically, through the conversion of the gray level image, both the first preset image and the second preset image become single-channel images.
Specifically, the calculating unit 32 is configured to calculate a variance value of each sub-image and use the variance value as an edge blur value;
the determining unit 33 specifically includes:
a comparing part 331, configured to respectively compare whether the edge blur value of each sub-image reaches a preset sperm activity threshold, and output a comparison result;
a marking part 332 connected to the comparing part 331 for marking the corresponding sub-image as normal viable sperm when the comparison result indicates that the edge blur value reaches the sperm viability threshold;
and a counting part 333 connected to the marking part 332 for counting a first total number of the sub-images in the sperm segmentation image and a second total number of the sub-images marked as normal active sperm, and calculating a ratio of the second total number to the first total number to output as a detection result.
Specifically, if the sperm segmentation image is a set of a plurality of sub-images, the labeling component 332 labels the sub-image whose edge blur value reaches the sperm motility threshold as a normal viable sperm; if the sperm divides the image into one image, the marking component 332 marks the sub-image in the image whose edge blur value reaches the sperm activity threshold, for example, frames the corresponding sub-image.
In another preferred embodiment, the acquisition module 1 acquires long exposure images of the sperm sample using the microscope in a long exposure operating state, without changes in other technical features.
Further, the acquisition module 1 acquires a long exposure image of a sperm sample acquired by a long exposure acquisition mode.
Further, the first segmentation unit 21 segments the long-exposure image by using a connected component segmentation method to obtain a first segmented image.
Further, the second segmentation unit 22 performs image segmentation on the first segmentation image by using an image segmentation neural network to obtain a second segmentation image, where the second segmentation image includes a plurality of sub-images of the sperm, and each sub-image corresponds to one sperm.
Further, the processing component 312 calculates a preset sperm activity threshold according to the first preset image and the second preset image in the obtaining component 311, and stores the threshold in the storage component 313.
Further, the calculation unit 32 calculates and obtains the variance of each sub-image as the edge blur value, respectively.
Further, the comparing part 331 compares whether the edge blur value of each sub-image reaches a preset sperm activity threshold value, and outputs the comparison result to the marking part 332.
Further, the labeling component 332 labels the corresponding sub-image as a normal viable sperm when the comparison result indicates that the edge blur value reaches the sperm viability threshold.
Further, the counting part 333 counts a first total number of the sub-images in the sperm partition image and a second total number of the sub-images labeled as normal viable sperm, and calculates a ratio of the second total number to the first total number to output as a detection result.
Further, the staff judges whether the activity of the sperm sample is normal or not according to a preset standard activity threshold value and the ratio of the second total number to the first total number;
if the ratio reaches a standard activity threshold, the activity of the sperm sample is normal;
if the ratio does not meet the standard activity threshold, the sperm sample is devoid of activity.
In the technical scheme, the method further comprises the following steps:
a sperm motility detection method, applied to the sperm motility detection system, comprises:
s1, acquiring a long exposure image of the sperm sample in a long exposure acquisition mode;
step S2, segmenting the long exposure image to obtain a sperm segmentation image, wherein the sperm segmentation image comprises a plurality of sub-images, and each sub-image is used for representing a sperm;
step S3, respectively calculating and obtaining the edge fuzzy value of each sub-image;
and step S4, comparing the edge fuzzy value of each sub-image with a preset sperm activity threshold value respectively, and outputting a sperm activity detection result according to the comparison result.
Specifically, step S2 specifically includes:
step S21, carrying out connected domain segmentation on the long exposure image to obtain a first segmentation image so as to carry out preliminary separation on the image including the sperms and the background part;
and step S22, carrying out image segmentation on the first segmentation image through an image segmentation neural network to obtain a sperm segmentation image.
Specifically, the method further comprises a process of presetting a sperm activity threshold value, and specifically comprises the following steps:
step A1, acquiring a first preset image only including normal active sperm and a second preset image only including inactive sperm in advance;
step A2, respectively calculating a variance value of a first preset image and a variance value of a second preset image;
step A3, calculating a variance threshold according to the variance value of the first preset image and the variance value of the second preset image, and storing the variance threshold as a sperm motility threshold;
in step S3, the variance value of each sub-image is calculated as the edge blur value;
step S4 specifically includes:
step S41, comparing the variance value of each sub-image with a preset sperm activity threshold value respectively, and outputting a comparison result;
step S42, according to the comparison result, marking the subimage with the variance value reaching the sperm activity threshold as normal active sperm;
and step S43, counting the first total number of the sub-images in the sperm segmentation image and the second total number of the sub-images marked as normal active sperm, and calculating the ratio of the second total number to the first total number to output as a detection result.
The technical scheme of the invention has the beneficial effects that: the invention utilizes the characteristic that the active sperm can generate fuzzy edge under the long exposure image, and judges the motion activity degree of the sperm by measuring the fuzzy degree of the sperm to be detected in the long exposure image, thereby further judging the activity of the sperm.
While the invention has been described with reference to a preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiment, but is intended to cover various modifications, equivalents and obvious changes which may be made therein by those skilled in the art.
Claims (9)
1. A sperm motility detection system, comprising:
the acquisition module is used for acquiring a long exposure image of a sperm sample by adopting a long exposure acquisition mode;
the segmentation module is connected to the acquisition module and is used for segmenting the long exposure image to obtain a sperm segmentation image;
the sperm segmentation image comprises a plurality of sub-images, and each sub-image is used for representing one sperm;
the processing module is connected to the segmentation module and specifically comprises:
the storage unit is used for preprocessing to form a preset sperm activity threshold value and storing the sperm activity threshold value;
the computing unit is used for respectively computing and obtaining the edge fuzzy value of each sub-image;
and the judging unit is respectively connected with the storage unit and the calculating unit and is used for respectively comparing the edge fuzzy value of each sub-image with the preset sperm activity threshold value and outputting the sperm activity detection result according to the comparison result.
2. The sperm motility detection system of claim 1, wherein said segmentation module comprises:
the first segmentation unit is used for carrying out connected domain segmentation on the long exposure image to obtain a first segmentation image so as to carry out preliminary separation on the image including the sperms and the background part;
and the second segmentation unit is connected to the first segmentation unit and is used for carrying out image segmentation on the first segmentation image by adopting an image segmentation neural network to obtain the sperm segmentation image.
3. The sperm motility detection system of claim 2, wherein the following formula is employed as a loss function for said image-segmenting neural network:
wherein the content of the first and second substances,
pρ(X)representing the output probability of a pixel point X in the first segmentation image on a channel where a real label is located;
log(pρ(X)(X)) represents a cross entropy loss function of the pixel point X;
ωC(X) a weight representing the proportion of the balanced categories of each pixel point X;
d1(X) represents the distance between said pixel point X and the closest pixel point representing the cell;
d2(X) representing the distance between said pixel X and the pixel representing the cell that is the second closest thereto;
ω0and sigma is a constant.
4. The sperm motility detection system of claim 1, wherein the storage unit further comprises:
an acquisition means for acquiring in advance a first preset image including only normal viable sperm and a second preset image including only non-viable sperm;
the processing component is used for calculating according to the first preset image and the second preset image to obtain a preset sperm activity threshold value;
and the storage component is connected with the processing component and is used for storing the preset sperm activity threshold value.
5. The sperm motility detection system of claim 4, wherein said processing component calculates said sperm motility threshold according to the formula:
wherein the content of the first and second substances,
t is the sperm motility threshold;
T1the variance value of the first preset image is obtained;
T2and the variance value of the second preset image is obtained.
6. The sperm motility detection system according to claim 1, wherein said computing unit is configured to compute a variance value for each of said sub-images as said edge blur value;
the determining unit specifically includes:
the comparison component is used for respectively comparing whether the edge fuzzy value of each sub-image reaches the preset sperm activity threshold value or not and outputting the comparison result;
a marking component connected to the comparing component for marking the corresponding sub-image as normal viable sperm when the comparison result indicates that the edge blur value reaches the sperm viability threshold;
and the counting component is connected to the marking component and used for counting a first total number of the sub-images in the sperm segmentation image and a second total number of the sub-images marked as normal active sperm, and calculating a ratio of the second total number to the first total number to be output as the detection result.
7. A sperm motility test method applied to the sperm motility test system according to any one of claims 1 to 6, comprising:
s1, acquiring a long exposure image of the sperm sample in a long exposure acquisition mode;
step S2, segmenting the long exposure image to obtain a sperm segmentation image, wherein the sperm segmentation image comprises a plurality of sub-images, and each sub-image is used for representing a sperm;
step S3, respectively calculating and obtaining the edge fuzzy value of each sub-image;
and step S4, comparing the edge fuzzy value of each sub-image with a preset sperm activity threshold value respectively, and outputting a sperm activity detection result according to the comparison result.
8. The sperm cell motility detection method according to claim 7, wherein said step S2 specifically comprises:
step S21, carrying out connected domain segmentation on the long exposure image to obtain a first segmentation image so as to carry out preliminary separation on the image including the sperms and the background part;
and step S22, carrying out image segmentation on the first segmentation image through an image segmentation neural network to obtain the sperm segmentation image.
9. The method of claim 7, further comprising a process of pre-processing to obtain the sperm motility threshold, comprising:
step A1, acquiring a first preset image only including normal active sperm and a second preset image only including inactive sperm in advance;
step A2, respectively calculating a variance value of the first preset image and a variance value of the second preset image;
step A3, calculating a variance threshold according to the variance value of the first preset image and the variance value of the second preset image, and storing the variance threshold as the sperm motility threshold;
in step S3, respectively calculating a variance value of each sub-image as the edge blur value;
the step S4 specifically includes:
step S41, comparing the variance value of each sub-image with the preset sperm activity threshold value respectively, and outputting a comparison result;
step S42, according to the comparison result, marking the subimage with the variance value reaching the sperm activity threshold value as normal active sperm;
step S43, counting a first total number of the sub-images in the sperm segmentation image and a second total number of the sub-images marked as normal active sperm, and calculating a ratio of the second total number to the first total number to output as the detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011142158.4A CN112419235A (en) | 2020-10-22 | 2020-10-22 | Sperm activity detection system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011142158.4A CN112419235A (en) | 2020-10-22 | 2020-10-22 | Sperm activity detection system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112419235A true CN112419235A (en) | 2021-02-26 |
Family
ID=74841089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011142158.4A Pending CN112419235A (en) | 2020-10-22 | 2020-10-22 | Sperm activity detection system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112419235A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1385700A (en) * | 2001-05-10 | 2002-12-18 | 华中科技大学同济医学院 | Sperm fluorescent staining computer detection system |
| EP2293247A1 (en) * | 2009-07-29 | 2011-03-09 | Harman Becker Automotive Systems GmbH | Edge detection with adaptive threshold |
| US20140212959A1 (en) * | 2013-01-28 | 2014-07-31 | National Tsing Hua University | Sperm quality assessment device |
| CN104268515A (en) * | 2014-09-18 | 2015-01-07 | 山东大学 | Sperm morphology anomaly detection method |
| CN108629765A (en) * | 2018-04-20 | 2018-10-09 | 泰山医学院 | Sperm microscopy video sequential filtering assessment method for encoding quality based on queue thresholds difference |
| CN109064469A (en) * | 2018-10-31 | 2018-12-21 | 北京新网视信传媒科技有限公司 | Sperm quality detector and sperm quality detection system |
| CN110211105A (en) * | 2019-05-24 | 2019-09-06 | 厦门美柚信息科技有限公司 | A kind of blurred picture detection method, apparatus and system |
| CN110283888A (en) * | 2013-08-19 | 2019-09-27 | 卓异生物公司 | Measurement and its application for Single Molecule Detection |
| US10426442B1 (en) * | 2019-06-14 | 2019-10-01 | Cycle Clarity, LLC | Adaptive image processing in assisted reproductive imaging modalities |
-
2020
- 2020-10-22 CN CN202011142158.4A patent/CN112419235A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1385700A (en) * | 2001-05-10 | 2002-12-18 | 华中科技大学同济医学院 | Sperm fluorescent staining computer detection system |
| EP2293247A1 (en) * | 2009-07-29 | 2011-03-09 | Harman Becker Automotive Systems GmbH | Edge detection with adaptive threshold |
| US20140212959A1 (en) * | 2013-01-28 | 2014-07-31 | National Tsing Hua University | Sperm quality assessment device |
| CN110283888A (en) * | 2013-08-19 | 2019-09-27 | 卓异生物公司 | Measurement and its application for Single Molecule Detection |
| CN104268515A (en) * | 2014-09-18 | 2015-01-07 | 山东大学 | Sperm morphology anomaly detection method |
| CN108629765A (en) * | 2018-04-20 | 2018-10-09 | 泰山医学院 | Sperm microscopy video sequential filtering assessment method for encoding quality based on queue thresholds difference |
| CN109064469A (en) * | 2018-10-31 | 2018-12-21 | 北京新网视信传媒科技有限公司 | Sperm quality detector and sperm quality detection system |
| CN110211105A (en) * | 2019-05-24 | 2019-09-06 | 厦门美柚信息科技有限公司 | A kind of blurred picture detection method, apparatus and system |
| US10426442B1 (en) * | 2019-06-14 | 2019-10-01 | Cycle Clarity, LLC | Adaptive image processing in assisted reproductive imaging modalities |
Non-Patent Citations (1)
| Title |
|---|
| 冯汉港;张艳青;郑运平;: "精子多目标跟踪的抗遮挡算法的研究与实现", 计算机工程与设计, no. 12, 16 December 2014 (2014-12-16) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111612751B (en) | Lithium battery defect detection method based on Tiny-yolov3 network embedded with grouping attention module | |
| CN111681273B (en) | Image segmentation method and device, electronic equipment and readable storage medium | |
| CN111681240B (en) | Bridge surface crack detection method based on YOLO v3 and attention mechanism | |
| US20160364849A1 (en) | Defect detection method for display panel based on histogram of oriented gradient | |
| CN115829999A (en) | Insulator defect detection model generation method, device, equipment and storage medium | |
| CN111160249A (en) | Multi-class target detection method of optical remote sensing image based on cross-scale feature fusion | |
| CN111274926B (en) | Image data screening method, device, computer equipment and storage medium | |
| CN109409288B (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN115147418B (en) | Compression training method and device for defect detection model | |
| CN111652846B (en) | Semiconductor defect identification method based on characteristic pyramid convolution neural network | |
| CN112396053A (en) | Method for detecting object of all-round fisheye image based on cascade neural network | |
| CN111461101A (en) | Method, device and equipment for identifying work clothes mark and storage medium | |
| Li et al. | A review of deep learning methods for pixel-level crack detection | |
| CN114037907A (en) | Detection method and device for power transmission line, computer equipment and storage medium | |
| CN115496971A (en) | Infrared target detection method and device, electronic equipment and storage medium | |
| CN110008949B (en) | Image target detection method, system, device and storage medium | |
| CN113326846B (en) | Rapid bridge apparent disease detection method based on machine vision | |
| CN111914706B (en) | Method and device for detecting and controlling quality of text detection output result | |
| CN112419235A (en) | Sperm activity detection system and method | |
| CN111311584A (en) | Video quality evaluation method and device, electronic equipment and readable medium | |
| CN115908344A (en) | Micro LED chip defect detection method based on MLCT-YOLO | |
| CN115953371A (en) | Insulator defect detection method, device, equipment and storage medium | |
| CN115240163A (en) | Traffic sign detection method and system based on one-stage detection network | |
| CN114663760A (en) | Model training method, target detection method, storage medium and computing device | |
| CN114998222A (en) | Automobile differential shell surface detection method, electronic equipment and medium |
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 |