Semen evaluation method and device based on lens-free imaging light intensity spatial and temporal distribution analysis

Abstract

The invention discloses a sperm evaluation method and device based on lensless imaging light intensity spatial and temporal distribution analysis, and the method comprises the steps: carrying out lensless imaging continuous collection on a sperm sample, obtaining a lensless imaging time sequence, extracting a single-frame image from the lensless imaging time sequence, reconstructing the single-frame image, and obtaining a sperm evaluation result. Calculating the fractal dimension of a reconstructed image, evaluating the number of sperm individuals according to the fractal dimension, obtaining the sperm density in combination with the sample volume, extracting a pixel light intensity time sequence from the lens-free imaging time sequence, calculating the time correlation and spatial correlation of light intensity fluctuation, and obtaining the sperm density. And evaluating the sperm motility ability based on the temporal correlation and the spatial correlation. According to the invention, accurate, comprehensive, rapid, simple and convenient detection and evaluation on the animal sperm quality in the production field are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of animal semen quality detection, in particular to a semen evaluation method and device based on lensless imaging light intensity spatiotemporal distribution analysis. Background Art

[0002] Semen quality testing is an essential step in the livestock production process. Semen quality analysis generally uses a microscope-based approach. Manual microscope analysis relies on the operator's experience and skills, requiring the operator to observe and count sperm one by one. The analysis results are highly subjective, the process is cumbersome, and time-consuming. Compared to manual microscope analysis, the computer-assisted sperm analysis system (CASA) uses computers and digital image processing technology to process and analyze sperm images taken with a biological microscope. It can perform a series of complex tests and analyses, including detecting sperm count, analyzing sperm morphology, counting movement trajectories, and analyzing sperm dynamic and static parameters. Its detection performance is powerful, with a high degree of automation and high efficiency. However, its equipment is expensive, bulky, difficult to port, and complex to maintain. Its use requires a high level of professional expertise from practitioners, and the professional software is complex to operate. It is only suitable for use in a laboratory environment and cannot complete simple and convenient testing on the production site.

[0003] To reduce the cost of testing equipment, improve portability, and enable rapid testing at the production site, several miniaturized sperm density meters have emerged. Currently, widely used portable sperm density meters are based on colorimetry or spectrophotometry. These meters utilize the absorption characteristics of sperm at specific wavelengths of light, which are closely related to their cellular components (such as proteins, nucleic acids, lipids, etc.) and cell membrane structure. Semen is injected into a cuvette, and sperm density is assessed by comparing the absorbance of the semen sample with that of a reference material. Sperm density meters based on colorimetry or spectrophotometry are small, portable, and easy to operate, allowing for quick and simple testing at the production site. However, these methods essentially measure the biomass of sperm cells and use this biomass to assess sperm count. These methods have low accuracy and are unable to assess sperm motility.

[0004] Other technical solutions utilize lensless imaging technology to image semen diluents. Specifically, the obtained lensless imaging interference pattern is reconstructed and analyzed to achieve sperm count and three-dimensional positioning of individual sperm. The three-dimensional coordinates are connected in series to obtain a trajectory, which is then calculated and extracted to calculate motility. Due to the high density of semen, lensless imaging overlaps significantly, necessitating dilution of the semen before imaging. Furthermore, this process requires lensless imaging reconstruction of each individual sperm. During the reconstruction process, the reconstruction distance parameter must be iteratively calculated and locked. The three-dimensional positioning of the individual sperm is locked based on the geometric relationship of the reconstruction distance. By connecting the three-dimensional coordinates, the individual sperm trajectory is ultimately formed, and motility is calculated based on the trajectory. This process is computationally intensive and requires a high-performance computer and a long analysis time. Summary of the Invention

[0005] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid blurring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0006] Therefore, the purpose of the present invention is to provide a semen evaluation method and device based on lensless imaging light intensity spatiotemporal distribution analysis, so as to realize accurate, comprehensive, rapid and easy detection and evaluation of animal semen quality at the production site.

[0007] To solve the above technical problems, according to one aspect of the present invention, the present invention provides the following technical solutions: A semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis, the steps are as follows: S1, continuously collecting lensless imaging of semen samples to obtain lensless imaging time series; S2. extracting a single-frame image from the lensless imaging time series, reconstructing the single-frame image, calculating the fractal dimension of the reconstructed image, evaluating the number of individual sperm according to the fractal dimension, and obtaining semen density in combination with the sample volume; S3. Extracting a pixel light intensity time series from the lensless imaging time series, calculating the temporal correlation and spatial correlation of light intensity fluctuations, and evaluating sperm motility based on the temporal correlation and spatial correlation.

[0008] As a preferred embodiment of the semen evaluation method based on spatiotemporal distribution analysis of light intensity of lensless imaging according to the present invention, in step S2, the single-frame image is reconstructed, the fractal dimension of the reconstructed image is calculated, and the specific steps of evaluating the number of individual sperm based on the fractal dimension are as follows: Cutting the single-frame image into multiple image blocks, and reconstructing each image block to obtain a reconstructed image block; performing denoising and grayscale normalization processing on the reconstructed image block, and representing the processed reconstructed image block as a two-dimensional surface; For the two-dimensional surface, different step lengths are selected to calculate the variogram; Drawing a logarithmic graph according to the variogram, performing linear fitting to obtain a slope, calculating a Hurst exponent from the slope, and then obtaining a fractal dimension; The fractal dimensions of all image blocks were accumulated to obtain a total fractal dimension, which was then fitted with the total sperm count determined by standard CASA to establish a relationship model, and the number of individual sperm was evaluated based on the relationship model.

[0009] As a preferred embodiment of the semen evaluation method based on spatiotemporal distribution analysis of light intensity of lensless imaging according to the present invention, the specific steps of selecting different step lengths to calculate the variogram for the two-dimensional surface are as follows: For the image block, select different step sizes, use multiple directions to calculate the variation, take the average, and calculate the variation ; Traverse all pixels , calculate the square of the height difference between it and the point at step length 𝑟 and find the mean: Count 𝑉(𝑟) for different 𝑟 values, where 𝑟 is the step size, N(𝑟) is the number of point pairs available when the step size is 𝑟, 𝐼(𝑥,𝑦) is the grayscale value of the pixel in the image, and 𝑟 𝑥 , 𝑟 𝑦 are the step sizes in different directions.

[0010] As a preferred embodiment of the semen evaluation method based on spatiotemporal distribution analysis of light intensity using lensless imaging according to the present invention, in step S3, the temporal correlation and spatial correlation of light intensity fluctuations are calculated, and the specific steps of evaluating sperm motility based on the temporal correlation and spatial correlation are as follows: Performing autocorrelation function analysis on the pixel light intensity time series to obtain a normalized time correlation function, calculating a pixel motion coefficient based on a decay rate of the normalized time correlation function, and accumulating the pixel motion coefficients of all pixels to obtain an overall motion degree; Calculate the gradient synergy of all pixels in the field of view in time and space dimensions to obtain spatial correlation; The overall motility of sperm was obtained by dividing the overall motility by the spatial correlation, and the average motility of individual sperm was obtained by combining the number of sperm.

[0011] As a preferred embodiment of the semen assessment method based on spatiotemporal distribution analysis of light intensity using lensless imaging, the specific steps of performing autocorrelation function analysis on the pixel light intensity time series to obtain a normalized time correlation function, calculating the pixel motion coefficient based on the decay rate of the normalized time correlation function, and accumulating the pixel motion coefficients of all pixels to obtain the overall motion degree are as follows: The time correlation of light intensity fluctuation is calculated based on the time series of light intensity fluctuation of a single pixel. The time correlation of light intensity fluctuation is described by the following autocorrelation function, and its formula is:

[0012]

[0013] in, is the light intensity time series, is the mean, τ is the time delay;

[0014] Normalize the autocorrelation function and get the normalized time correlation function as follows: ;

[0015]

[0016] Among them, β is determined by the optical path geometry of the device, and the decay rate of the autocorrelation function can be used to evaluate the temporal correlation. The slower the decay, the stronger the correlation; the faster the decay, the weaker the correlation.

[0017] The autocorrelation function decays exponentially, and its decay is related to the degree of irregular motion of the particles corresponding to the pixel, that is, the pixel motion coefficient Correlation, pixel motion coefficient The calculation formula is:

[0018]

[0019] in, is the autocorrelation function, α is a constant, and τ is the time delay;

[0020] The pixel motion coefficients of all pixels in the entire field of view By accumulating, you can get the overall exercise level .

[0021] As a preferred embodiment of the semen evaluation method based on spatiotemporal distribution analysis of light intensity of lensless imaging described in the present invention, the gradient synergy of all pixels in the field of view in the temporal and spatial dimensions is calculated to obtain the specific calculation of spatial correlation as follows:

[0022] Spatial correlation is characterized by the gradient synergy of all pixels in the field of view in time and space dimensions. The calculation method is as follows: in, is the spatiotemporal gradient operator, i is the frame number in the video, is the total number of frames; For each grayscale image in the video , calculate its spatiotemporal gradient three-dimensional vector: ; in, and For spatial gradient, use the Sobel operator spatial gradient, and the calculation method is: in, , is the time gradient, which is calculated using the true gradient method. .

[0023] A semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis, comprising: The shell has an opening on one side, and a flip-up closed cover is hingedly connected to the side of the opening, and a sample container carrying platform is provided on the bottom plate of the shell; A sample container, detachably mounted on the sample container carrying platform; a light source, mounted on the top of the inner wall of the housing; a CMOS imaging sensor mounted on the bottom of the inner wall of the housing, with the centers of the sample container, the light source, and the CMOS imaging sensor located on the same plumb line; An operation display screen, mounted on the outer wall of the housing, for controlling the device and displaying test values; The embedded microcomputer is installed in the housing and is electrically connected to the operation display screen and the CMOS imaging sensor to control the operation display screen and the CMOS imaging sensor.

[0024] As a preferred solution of the semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis described in the present invention, the sample container includes a base plate, a rectangular frame arranged on the base plate, and an upper cover plate hinged to the rectangular frame, wherein an open sample cavity is formed at the top of the rectangular frame.

[0025] As a preferred solution of the semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis described in the present invention, the side edges of the sample container supporting platform are provided with sliding grooves that cooperate with the two side edges of the bottom plate.

[0026] As a preferred solution of the semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis described in the present invention, the light source includes a light source sleeve, a single red light-emitting diode arranged on the top of the inner wall of the light source sleeve, a filter arranged on the inner wall of the light source sleeve, and a microplate installed at the bottom of the light source sleeve.

[0027] Compared with the prior art, the present invention has the following beneficial effects: 1. This device can simultaneously detect sperm density and assess sperm motility, providing a more comprehensive assessment of semen quality than other portable sperm density testing instruments. Its test results more accurately reflect the density of active sperm, providing a more accurate and reliable basis for semen dilution.

[0028] 2. The device and method designed in the present invention have the advantages of both absorbance spectrophotometry and microscopy, and have the advantages of high precision, low cost, short detection time and portability.

[0029] 3. The present invention is based on an optical imaging method, which has the advantage of microscopic imaging method in terms of sperm density detection, which can accurately detect individual sperm, and has higher accuracy than the absorbance spectrophotometry method.

[0030] 4. The method of the present invention for evaluating sperm motility is to perform oscillation intensity and energy analysis on the light intensity distribution surface obtained by the diffraction imaging spatial sequence, which can obtain the overall motility performance of the semen sample rather than the analysis and statistics of the individual sperm motility performance. Therefore, the requirements for hardware computing performance are greatly reduced, the detection time is greatly reduced, and the equipment can be miniaturized and highly portable.

[0031] 5. The detection process of the present invention does not require chemical pretreatment of semen such as dyeing, and does not pollute the semen sample. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and detailed embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be derived from these drawings without inventive effort. Among them: Figure 1 Schematic diagram of the overall structure of the semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis of the present invention; Figure 2 This is a partial structural cross-sectional view of a semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis according to the present invention; Figure 3 For the present invention Figure 2A schematic diagram of a partial structure of a semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis after removing the sample container; Figure 4 A schematic structural diagram of a sample container provided by the present invention; Figure 5 A schematic diagram of the cross-sectional structure of the light source provided by the present invention; Figure 6 Flow chart of sperm density detection provided by the present invention; Figure 7 A flow chart of extracting a pixel light intensity time series from the lensless imaging time series provided by the present invention.

[0033] Figure 8 Flow chart of the average sperm motility assessment provided by the present invention. DETAILED DESCRIPTION

[0034] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0035] The present invention provides a semen evaluation method and device based on lensless imaging light intensity spatiotemporal distribution analysis, so as to realize accurate, comprehensive, rapid and convenient detection and evaluation of animal semen quality at the production site.

[0036] Example 1 Figure 1-Figure 5 The figure shows the structure of the semen evaluation device based on the spatiotemporal distribution analysis of light intensity without lens imaging of the present invention. Figure 1-Figure 5 The semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis of this embodiment has a main body including a shell 100, a sample container 200, a light source 300, a CMOS imaging sensor 400, an operation display screen 500 and an embedded microcomputer 600.

[0037] The shell 100 has an opening on the side, and a flip-up closed cover 110 is hinged on the side of the opening. A sample container carrier 120 is provided on the bottom plate. The shell 100 is used to load and remove the sample container 200, ensuring that the semen sample is located in a specific spatial position and that the entire imaging acquisition process is in a sealed optical darkroom.

[0038] The sample container 200 is detachably mounted on the sample container carrier 120. In this embodiment, the sample container 200 includes a bottom plate 210, a rectangular frame 220 disposed on the bottom plate 210, and an upper cover 230 hingedly connected to the rectangular frame 220. An open sample cavity 220a is formed at the top of the rectangular frame 220. The side of the sample container carrier 120 is provided with a slide groove 120a that cooperates with the two sides of the bottom plate 210. The experimenter can insert the bottom plate 210 into the slide groove 120a to fix it, or pull it out from the slide groove 120a to take out the sample container 200. Preferably, the bottom plate 210 and the upper cover plate 230 are both made of pure polystyrene material with high optical transparency, and the surface is polished to maintain the consistency of the light intensity and phase of the outgoing light after passing through the bottom plate 210 and the upper cover plate 230 to the greatest extent. The sample cavity 220a is square with geometric dimensions of 30×28 mm. Its length and width dimensions are consistent with the length and width dimensions of the photosensitive area of ​​the CMOS imaging sensor 400. During the test, a specific volume (1 ml) of semen sample is injected into the sample cavity 220a using a pipette, and the semen depth is about 1.2 mm at this time.

[0039] The light source 300 is mounted on the top of the inner wall of the housing 100. In this embodiment, the light source 300 includes a light source sleeve 310, a single red LED 320 mounted on the top of the inner wall of the light source sleeve 310, a filter 330 mounted on the inner wall of the light source sleeve 310, and a microporous plate 340 mounted on the bottom of the light source sleeve 310. The single red LED 320 can be a 650nm red diode with a longer wavelength. The filter 330 further improves its monochromaticity. The light source sleeve 310 is made of an opaque material, and the diameter of the circular micropore in the center of the microporous plate 340 can be 50μm. After passing through the micropore, the wavefront of the light emitted by the single red LED 320 is split, forming monochromatic partially coherent light with a wavefront that approximates a plane wave.

[0040] The CMOS imaging sensor 400 is mounted on the bottom inner wall of the housing 100, with the sample container 200, the light source 300, and the centers of the CMOS imaging sensor 400 located on the same vertical line. During actual use, the CMOS imaging sensor 400 follows the instructions of the embedded microcomputer 600 to obtain a lensless imaging time series of animal semen at a certain shooting frequency within a certain time period.

[0041] The operation display screen 500 is installed on the outer wall of the housing 100 and is used to control the device and display detection values.

[0042] The embedded microcomputer 600 is installed in the housing 100 and is electrically connected to the operation display screen 500 and the CMOS imaging sensor 400 to control the operation of the operation display screen 500 and the CMOS imaging sensor 400 .

[0043] Combine Figure 1-Figure 5 The semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis of this embodiment is specifically used as follows: After the device is in standby mode, the operating display screen 500 instructs the user to perform a sampling operation. The user opens the sealing cover 110 on the side of the housing 100, removes the sample container 200 from the container carrier 120, opens the upper cover 230 of the sample container 200, and uses a pipette to inject a specified volume of semen into the sample cavity 220a. The upper cover 230 is then replaced, and the sample container 200 is inserted into the sample container carrier 120. The sealing cover 110 is then closed. Pressing the start button on the operating display screen 500 initiates the detection process. The embedded microcomputer 600 turns on the light source 300 and issues instructions to the CMOS imaging sensor 400, performing lensless imaging continuous acquisition. The data preparation process is then executed, and the corresponding data is extracted from the prepared data at once to perform sperm count detection and overall sperm motility performance detection, respectively. Finally, the operating display screen 500 displays the sperm density and average motility index.

[0044] Example 2 The present invention also provides a semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis, the steps of which are as follows: S1, continuously collecting lensless imaging of semen samples to obtain lensless imaging time series; S2. extracting a single-frame image from the lensless imaging time series, reconstructing the single-frame image, calculating the fractal dimension of the reconstructed image, evaluating the number of individual sperm according to the fractal dimension, and obtaining semen density in combination with the sample volume; S3. Extracting a pixel light intensity time series from the lensless imaging time series, calculating the temporal correlation and spatial correlation of light intensity fluctuations, and evaluating sperm motility based on the temporal correlation and spatial correlation.

[0045] The above step S1 is specifically implemented by collecting data using a semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis.

[0046] like Figure 6 As shown, the specific implementation steps of the above step S2 are as follows: Step 1: Extract a frame from the time series of lensless imaging of animal semen, reconstruct the frame image based on the convolution method, and obtain the reconstructed image. During the reconstruction process, the entire image is first cut into small blocks of 256×256 pixels, and the reconstruction object-image distance parameters are set according to the actual object-image distance of the imaging device. The object-image distance parameters are determined by the distance from the bottom of the sample container to the surface of the CMOS sensor, and each small block of the reconstructed image is obtained. Use Gaussian filtering to denoise the image. In order to avoid the influence of uneven light intensity, the image grayscale is normalized and the grayscale value is mapped to [0,1]. The grayscale image is expressed as a two-dimensional surface with pixel grayscale values. Represents the height value of the pixel on the surface.

[0047] Step 2: Calculate the variance function. For each image block, select different step sizes. Here, we take 𝑟=1, 2, 4, 8, 16, 32, and 64 respectively. Calculate the variance in multiple directions (such as horizontal, vertical, and diagonal) and take the average value. , traverse all pixels , calculate the square of the height difference between it and the point at step length 𝑟 and find the mean: Count 𝑉(𝑟) for different 𝑟 values, where 𝑟 is the spatial distance (step length), N(𝑟) is the number of point pairs available when the step length is 𝑟, 𝐼(𝑥,𝑦) is the grayscale value of the pixel in the image (corresponding to the surface height), 𝑟 𝑥 , 𝑟 𝑦 is the step size in different directions (usually the average of 8 directions is chosen).

[0048] Step 3: Calculate the fractal dimension, draw a logarithmic graph, and plot the fractal dimension on the logarithmic axis. and Perform linear fitting, calculate the slope, and calculate the Hurst exponent H from the slope. Calculate the fractal dimension D: D=3-H.

[0049] Step 4: Calculate the fractal dimension of all image blocks and accumulate them to reconstruct the total fractal dimension D of the two-dimensional surface of imaging light intensity distribution 总 To evaluate the number of individual sperm in the sample, before obtaining the sperm count result, the fractal dimension results obtained above should be fitted with the total sperm count determined by standard CASA, a relationship model should be established, and the calibration of the relationship between fractal dimension and sperm count should be completed.

[0050] In the above step S3, two steps are required. The first step is to extract the pixel light intensity time series from the lensless imaging time series, and the second step is to calculate the time correlation and spatial correlation of the light intensity fluctuation, and evaluate the sperm motility based on the time correlation and spatial correlation.

[0051] like Figure 7 As shown, the specific steps of extracting the pixel light intensity time series from the lensless imaging time series are as follows: extract each frame image of the original lensless imaging time series, sort and label them according to the time series. The lensless imaging is essentially a light intensity distribution surface. The pixel grayscale value of the lensless imaging is normalized and recorded as the light intensity at the pixel. Represents the coordinate of the i-th surface in the time series The light intensity value at the position. The time interval (i.e., time difference) Δ𝑡 between two frames of images. The specific value of Δ𝑡 depends on the parameters of the high-speed CMOS camera. To ensure that Δ𝑡 is small, a high-speed camera is used. Specify a set of , extract the pixel grayscale value of the coordinate on each surface in chronological order, connect them in series, add the shooting time interval parameter, and get the light intensity time series of the coordinate Traverse all pixel coordinates in the entire image and obtain the time series of light intensity of all pixels.

[0052] like Figure 8 As shown, the specific steps of calculating the temporal correlation and spatial correlation of light intensity fluctuations and evaluating sperm motility based on the temporal correlation and spatial correlation are as follows: In lensless imaging, when the particles in the sample are not moving, the resulting pixel light intensity fluctuations are determined by the temporal coherence of the light source and are autocorrelated. The change in the temporal correlation of the light intensity fluctuations reflects the degree of irregular motion of the corresponding particles. For each pixel's light intensity time series, the temporal correlation of the light intensity fluctuations is analyzed through the autocorrelation function, and the pixel motion coefficient of the pixel is obtained by fitting. , with pixel motion coefficient Characterizes the degree of irregular movement of particles corresponding to the pixel. Pixel motion coefficient It is the degree of irregular pixel motion described by the temporal autocorrelation of light intensity fluctuations.

[0053] The temporal correlation of light intensity fluctuations is calculated based on the time series of light intensity fluctuations of a single pixel. The temporal correlation of light intensity fluctuations is described by the following autocorrelation function (ACF), whose formula is: in, is the light intensity time series, is the mean, and τ is the time delay.

[0054] The autocorrelation function is normalized as follows: Get the normalized time correlation function . Among them, β is determined by the optical path geometry of the device. The decay rate of the autocorrelation function can be used to evaluate the time correlation. The slower the decay, the stronger the correlation; the faster the decay, the weaker the correlation. The decay curve of the autocorrelation function is obtained as the horizontal axis.

[0055] The autocorrelation function decays exponentially, and its decay is related to the degree of irregular motion of the particles corresponding to the pixel, that is, the pixel motion coefficient Correlation. Pixel motion coefficient The calculation formula is: in, is the autocorrelation function, α is a constant, and τ is the time delay.

[0056] The pixel motion coefficients of all pixels in the entire field of view Accumulate and get the overall exercise level .

[0057] Assessing spatial correlation between pixels: Because the lensless image corresponding to a single sperm individual covers multiple pixels, when all pixels come from a single individual, the pixel variations are spatially correlated. The lensless image obtained is formed by the partial overlap of the patterns corresponding to multiple sperm individuals, resulting in a certain degree of spatial correlation between pixels. Greater spatial correlation indicates that the image was formed by fewer individuals, and vice versa, it indicates that the image was formed by more individuals.

[0058] Spatial correlation is characterized by the gradient synergy of all pixels in the field of view in the temporal and spatial dimensions. The calculation method is shown in the following formula: The more concentrated the gradient direction, the more consistent the change, where is the spatiotemporal gradient operator, i is the frame number in the video, is the total number of frames; For each grayscale image in the video , calculate its spatiotemporal gradient three-dimensional vector .in and For spatial gradient, use the Sobel operator spatial gradient, the calculation method is in, is the time gradient, which is calculated using the true gradient method: Total motion coefficient of field of view divided by the spatial correlation across the field of view The result is the overall sperm motility.

[0059] According to the overall motility obtained above and the number of sperm T obtained above, the average motility is obtained. .

[0060] In other embodiments, the density of highly motile sperm can also be assessed by setting a threshold for the pixel motility coefficient M. Based on the threshold, highly motile pixels whose pixel motility coefficient M exceeds the threshold are screened. Based on the number of highly motile pixels, a percentage k of highly motile pixels is determined. Based on the total number of sperm T obtained previously and the percentage k obtained in this step, the number kT of sperm with motility exceeding the specified standard is calculated.

[0061] Although the present invention has been described above with reference to embodiments, various modifications may be made thereto and equivalent components may be substituted without departing from the scope of the present invention. In particular, as long as there are no structural conflicts, the various features of the embodiments disclosed herein may be combined with each other in any manner, and the omission of an exhaustive description of such combinations in this specification is solely for the sake of space and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis, characterized in that: Here are the steps: S1, continuously collecting lensless imaging of semen samples to obtain lensless imaging time series; S2. extracting a single-frame image from the lensless imaging time series, reconstructing the single-frame image, calculating the fractal dimension of the reconstructed image, evaluating the number of individual sperm according to the fractal dimension, and obtaining semen density in combination with the sample volume; S3. Extracting a pixel light intensity time series from the lensless imaging time series, calculating the temporal correlation and spatial correlation of light intensity fluctuations, and evaluating sperm motility based on the temporal correlation and spatial correlation.

2. The semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 1, characterized in that: In step S2, the single-frame image is reconstructed, the fractal dimension of the reconstructed image is calculated, and the specific steps of evaluating the number of individual sperms based on the fractal dimension are as follows: Cutting the single-frame image into multiple image blocks, and reconstructing each image block to obtain a reconstructed image block; performing denoising and grayscale normalization processing on the reconstructed image block, and representing the processed reconstructed image block as a two-dimensional surface; For the two-dimensional surface, different step lengths are selected to calculate the variogram; Drawing a logarithmic graph according to the variogram, performing linear fitting to obtain a slope, calculating a Hurst exponent from the slope, and then obtaining a fractal dimension; The fractal dimensions of all image blocks were accumulated to obtain a total fractal dimension, which was then fitted with the total sperm count determined by standard CASA to establish a relationship model, and the number of individual sperm was evaluated based on the relationship model.

3. The semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 2, characterized in that: For the two-dimensional surface, the specific steps of selecting different step lengths to calculate the variogram are as follows: For the image block, select different step sizes, calculate the variation in multiple directions, take the average value, and calculate the variation ; Traverse all pixels , calculate the square of the height difference between it and the point at step length 𝑟 and find the mean: Count 𝑉(𝑟) for different 𝑟 values, where 𝑟 is the step size, N(𝑟) is the number of point pairs available when the step size is 𝑟, 𝐼(𝑥,𝑦) is the grayscale value of the pixel in the image, and 𝑟 𝑥 , 𝑟 𝑦 are the step sizes in different directions.

4. The semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 1, characterized in that: In step S3, the time correlation and spatial correlation of light intensity fluctuations are calculated, and the specific steps of evaluating sperm motility based on the time correlation and spatial correlation are as follows: Performing autocorrelation function analysis on the pixel light intensity time series to obtain a normalized time correlation function, calculating a pixel motion coefficient based on a decay rate of the normalized time correlation function, and accumulating the pixel motion coefficients of all pixels to obtain an overall motion degree; Calculate the gradient synergy of all pixels in the field of view in time and space dimensions to obtain spatial correlation; The overall motility of sperm was obtained by dividing the overall motility by the spatial correlation, and the average motility of individual sperm was obtained by combining the number of sperm.

5. The semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 4, characterized in that: The specific steps of performing autocorrelation function analysis on the pixel light intensity time series to obtain a normalized time correlation function, calculating the pixel motion coefficient based on the decay rate of the normalized time correlation function, and accumulating the pixel motion coefficients of all pixels to obtain the overall motion degree are as follows: The time correlation of light intensity fluctuation is calculated based on the time series of light intensity fluctuation of a single pixel. The time correlation of light intensity fluctuation is described by the following autocorrelation function, and its formula is: in, is the light intensity time series, is the mean, τ is the time delay; Normalize the autocorrelation function and get the normalized time correlation function as follows: ; Among them, β is determined by the optical path geometry of the device, and the decay rate of the autocorrelation function can be used to evaluate the temporal correlation. The slower the decay, the stronger the correlation; the faster the decay, the weaker the correlation. The autocorrelation function decays exponentially, and its decay is related to the degree of irregular motion of the particles corresponding to the pixel, that is, the pixel motion coefficient Correlation, pixel motion coefficient The calculation formula is: in, is the autocorrelation function, α is a constant, and τ is the time delay; The pixel motion coefficients of all pixels in the entire field of view By accumulating, you can get the overall exercise level .

6. The semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 1, characterized in that: Calculate the gradient synergy of all pixels in the field of view in time and space dimensions, and obtain the specific calculation of spatial correlation as follows: Spatial correlation is characterized by the gradient synergy of all pixels in the field of view in time and space dimensions. The calculation method is as follows: in, is the spatiotemporal gradient operator, i is the frame number in the video, is the total number of frames; For each grayscale image in the video , calculate its spatiotemporal gradient three-dimensional vector: ； in, and For spatial gradient, use the Sobel operator spatial gradient, and the calculation method is: in, , is the time gradient, which is calculated using the true gradient method. 。 7. A device for implementing the semen evaluation method based on lensless imaging light intensity spatiotemporal distribution analysis as described in any one of claims 1 to 6, characterized in that: include: The housing (100) has an opening on a side, and a flip-up closing cover (110) is hingedly connected to the side of the opening, and a sample container carrying platform (120) is provided on the bottom plate thereof; A sample container (200) is detachably mounted on the sample container carrying platform (120); A light source (300) is mounted on the top of the inner wall of the housing (100); A CMOS imaging sensor (400) is mounted on the bottom of the inner wall of the housing (100), and the centers of the sample container (200), the light source (300), and the CMOS imaging sensor (400) are located on the same vertical line; An operating display screen (500), mounted on the outer wall of the housing (100), for controlling the device and displaying detection values; An embedded microcomputer (600) is installed in the housing (100) and is electrically connected to the operation display screen (500) and the CMOS imaging sensor (400) to control the operation of the operation display screen (500) and the CMOS imaging sensor (400).

8. The semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 7, characterized in that: The sample container (200) comprises a bottom plate (210), a rectangular frame (220) arranged on the bottom plate (210), and an upper cover plate (230) hinged to the rectangular frame (220), wherein an open sample receiving cavity (220a) is formed at the top of the rectangular frame (220).

9. The semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 8, characterized in that: The sides of the sample container carrying platform (120) are provided with sliding grooves (120a) that cooperate with the two side edges of the bottom plate (210).

10. The semen evaluation device based on lensless imaging light intensity spatiotemporal distribution analysis according to claim 8, characterized in that: The light source (300) comprises a light source sleeve (310), a single red light emitting diode (320) arranged on the top of the inner wall of the light source sleeve (310), a filter (330) arranged on the inner wall of the light source sleeve (310), and a microplate (340) mounted on the bottom of the light source sleeve (310).

Images