Sperm flagellum motion parameter calculation method and computer readable storage medium

Abstract

The invention relates to the technical field of computer-assisted semen analysis, and discloses a sperm flagellum motion parameter calculation method and a computer readable storage medium, and the method comprises the steps: extracting multiple frames of sample images from a flagellum motion video of a to-be-analyzed sperm based on a preset sampling frequency; inputting each frame of sample image into the improved U-Net network to obtain a binary flagellum segmentation mask, extracting a flagellum skeleton by using a morphological refinement algorithm, and parameterizing to obtain a parameterized flagellum skeleton curve; a convolution kernel in the improved U-Net network is a convolution kernel in which deformable offset with constraints is introduced; based on the parameterized flagellum skeleton curve, the flagellum length, the flagellum tangential angle, the flagellum propagation speed, the flagellum curvature, the flagellum amplitude, the flagellum frequency, the flagellum wavelength, the flagellum bending amplitude and the flagellum bending index are calculated. According to the method, the receptive field is increased through the convolution kernel with the constrained deformable offset, the segmentation continuity is ensured, and the motion parameters of the flagellum can be quantitatively calculated.

Description

Technical Field

[0001] The present invention relates to the technical field of computer-assisted semen analysis, and in particular to a method for calculating sperm flagella motion parameters and a computer-readable storage medium. Background Art

[0002] In medical and biological research, the motility of sperm is a key indicator for evaluating its functional status and fertility. According to the World Health Organization (WHO), the motility of sperm flagella is an important factor in diagnosing potential male infertility. Quantifying flagellar motion parameters (such as amplitude, frequency, and waveform) not only helps to fundamentally understand the mechanism of sperm motility, but also plays a vital role in diagnosing male infertility and improving artificial reproductive technology.

[0003] Sperm flagellar movement is a highly dynamic process that requires high-throughput data processing. In order to capture the dynamic process of rapid movement of sperm flagella, the imaging frame rate is generally not less than 30 frames per second. In addition, taking clinical infertility diagnosis as an example, the World Health Organization requires the analysis of at least 200 sperm from each patient, that is, each patient generates 6,000 sperm images per second waiting to be processed to quantify flagellar movement parameters. Traditionally, sperm movement analysis relies on manual observation by experienced embryologists under a microscope. This task is extremely time-consuming (each patient requires 2-3 days to annotate images) and is prone to inconsistency and subjectivity. Therefore, in order to fully tap the rich information contained in the movement behavior of sperm flagella, it is urgent to develop automated, high-throughput quantitative computational analysis tools.

[0004] The technical requirement for high-throughput, fully automated quantitative calculation and analysis of sperm flagellar motion parameters is to be able to segment the entire sperm flagellum completely and continuously. Incomplete or broken flagellar segmentation will lead to errors in the quantification of flagellar parameters. However, this task faces the following challenges: (1) The diameter of a sperm flagellum is about 1 μm, which is equivalent to 5-10 pixels in the image (accounting for less than 0.01% of the entire image), and the imaging contrast is low, making it difficult to accurately distinguish it from the background; (2) Sperm flagella swing in three-dimensional space, and some flagella may be outside the depth of field of the microscope, causing the sperm flagellum to partially disappear and become invisible, further increasing the difficulty of segmenting the entire flagellum.

[0005] Existing analysis methods can be divided into two categories: one is to use basic image processing combined with prior knowledge (such as geometric topology) to enhance perception performance by modeling the topological features of the curve structure. However, this method is not only highly sensitive to thresholds, but also heavily dependent on the anisotropy of the background and sperm, resulting in discontinuous or even unsegmentable results for the sperm flagellum segmentation; the other is to use standard convolutional neural networks, such as U-Net, FC-DenseNet, Deeplabv3+, etc., but experiments show that they are difficult to accurately and continuously segment the sperm flagellum; the reason is that the standard convolution has a fixed shape and cannot learn the irregular and time-varying sperm flagellum structure. In order to more accurately learn complex and changeable curve features, deformable convolutions such as DUNet and DCU-Net were introduced, but because their offsets are not constrained, they may over-focus on the background area and ignore the narrow area of ​​the sperm flagellum, resulting in discontinuous and fragmented segmentation results.

[0006] In summary, the existing method of using convolutional networks to segment sperm flagella for quantitative calculation and analysis of sperm flagellar motion parameters is that the fixed kernel shape of standard convolution is difficult to adapt to the irregular shape of flagella, and the arbitrary offset of deformable convolution may cause the segmentation result to be broken. Therefore, it is impossible to accurately distinguish between the background area and the flagella area, resulting in discontinuous segmentation results and low accuracy. Summary of the invention

[0007] To this end, the technical problem to be solved by the present invention is to overcome the problem that the convolution kernel in the prior art is difficult to adapt to the irregular shape of flagella, resulting in the inability to accurately distinguish flagella from the background, and further resulting in the inability to accurately identify flagellar motion parameters.

[0008] In order to solve the above technical problems, the present invention provides a method for calculating sperm flagella motion parameters, comprising: Based on a preset sampling frequency, extract multiple frames of sample images from the flagella movement video of the sperm to be analyzed; For each frame of sample image, input it into the improved U-Net network to obtain the corresponding binary flagella segmentation mask; the convolution kernel in the improved U-Net network is a convolution kernel that introduces a constrained deformable offset; The flagellar skeleton is extracted from the binary flagellar segmentation mask using a morphological thinning algorithm; Parameterize the flagellar skeleton to obtain a parameterized flagellar skeleton curve; Based on the parameterized flagellar skeleton curve, the flagellar length, flagellar tangential angle, flagellar propagation speed and flagellar curvature are calculated; Based on the flagellar curvature, the flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index are calculated.

[0009] Preferably, the acquisition of the improved U-Net network includes: The common convolution kernel in the standard U-Net network is linearly expanded along the x-axis and y-axis directions to obtain the initial state of the continuous convolution kernel; Introduce a constrained deformable offset into the initial continuous convolution kernel to obtain a continuous convolution kernel; The continuous convolution kernel is integrated into the standard U-Net network to obtain an improved U-Net network.

[0010] Preferably, the convolution process of the continuous convolution kernel is expressed as: ; in, Indicates the output feature map at pixel position Where Axis features, Indicates the output feature map at pixel position Where Axis features; Indicates the output feature map at pixel position The characteristics of represents the convolution kernel A grid, represents the convolution kernel The preset weights corresponding to the grids, represents the convolution kernel The preset modulation coefficient corresponding to each grid; , Represents the total number of grids in the convolution kernel; Represents the convolution kernel Axis direction The deformable offset of the mesh is expressed as ; Represents the convolution kernel Axis direction The deformable offset of the mesh is expressed as ; Represents a preset constraint, Represents the convolution kernel Axis direction The offset value of the grid, Represents the convolution kernel Axis direction The offset value of the grid.

[0011] Preferably, improving the training of the U-Net network includes: Live sperm were selected from semen samples using the water bath-swim-up method as sperm samples, and flagellar movement videos of sperm samples were collected using a 40x objective lens; The flagellar motion video is converted into frame sequence images according to the preset sampling frequency, and the outline of the sperm flagella on each image is marked using the polygon tool to obtain the sperm flagella annotation dataset; Each frame of the sperm flagella annotation dataset is converted into a grayscale image and then assigned to three color channels: R, G, and B, to obtain the corresponding three-channel pseudo-color image. The sperm flagella annotation dataset is divided into a training set and a test set, and the training set data is expanded by image rotation, translation and flipping; Based on the expanded data set, the cross entropy loss function and Adam optimizer are used to update the parameters of the improved U-Net network until the preset number of training times is reached to obtain the trained improved U-Net network.

[0012] Preferably, the flagellar skeleton is parameterized to obtain a parameterized flagellar skeleton curve, which is expressed as: ; in, Represents the starting point to point of the parameterized flagellar skeleton curve The arc length.

[0013] Preferably, based on the parameterized flagellar skeleton curve, the flagellar length, flagellar tangential angle, flagellar propagation speed and flagellar curvature are calculated, including: Discretize the parameterized flagellar skeleton curve, obtain multiple sampling points, calculate the Euclidean distance between adjacent sampling points, and sum them to obtain the flagellar length , expressed as: ; , Represents the total number of sampling points, and Respectively represent The x-axis coordinates and y-axis coordinates corresponding to the sampling points; Arc length Find the first-order derivative of the x-axis coordinate and the y-axis coordinate at The first derivative of the x-axis coordinate The inverse tangent function of the ratio , as the flagellar tangential angle , expressed as: ; The flagellar propagation velocity is calculated based on the rate of change of the cumulative length of the flagellar curve at adjacent moments. , expressed as: ; represents the cumulative arc length of the flagellar curve at time t, Indicates a preset time interval; Arc length The x-axis coordinate and y-axis coordinate at the position are used to find the first-order derivative and the second-order derivative, and the curvature of the flagella is calculated. , expressed as: ; and Respectively represent the arc length The second derivative of the x-axis coordinate and the second derivative of the y-axis coordinate at.

[0014] Preferably, based on the flagellar curvature, the flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index are calculated, including: Get multiple pairs of local maxima and local minima of the flagellar curvature as wave peaks With trough ; Calculate the sum of the vertical distances between all adjacent peaks and troughs and average them to obtain the flagellar amplitude , expressed as: ; Represents the vertical distance between adjacent peaks and troughs. It represents the total number of pairs of peaks and troughs; Based on unit time Number of flagellar curvature sign changes at inner arc length s The ratio of the unit time to calculate the flagellar frequency , expressed as: ; Based on the propagation speed The ratio of the flagellar frequency to the flagellar wavelength , expressed as: ; The curvature of the flagella of the parameterized flagellar skeleton curve is integrated along the arc length to obtain the bending amplitude of the flagella , expressed as: ; Based on flagellar curvature The mean With standard deviation , get the flagellar bending index , expressed as: .

[0015] Preferably, it also includes a method based on flagellar amplitude Flagellar wavelength Calculate the flagellar amplitude-wavelength ratio , expressed as: .

[0016] Preferably, it also includes calculating the flagellar tangential angle Arc length The derivative of , expressed as: .

[0017] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for calculating sperm flagella motion parameters as described above are implemented.

[0018] The above technical solution of the present invention has the following beneficial effects compared with the prior art: The method for calculating the motion parameters of sperm flagella described in the present invention introduces a constrained deformable offset into the convolution kernel of a standard U-Net network to obtain an improved U-Net network, and extracts a binary flagella segmentation mask from a sample image, thereby effectively overcoming the limitations of standard convolution and deformable convolution in segmenting sperm flagella. By linearizing the convolution kernel, limiting the offset range, and adopting an offset accumulation strategy, the continuity of the segmentation result is ensured while the receptive field is increased, thereby achieving the accuracy and continuity of flagella segmentation.

[0019] At the same time, the present invention extracts the flagellar skeleton from the binary flagellar segmentation mask, parameterizes it, calculates the curvature of the parameterized flagellar skeleton curve, and uses the curvature as the basic feature to quantitatively calculate multiple parameters including flagellar length, flagellar tangential angle, flagellar propagation speed, flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index, thereby simplifying the calculation process and expanding the scope of use of the calculation method, so that it can analyze non-periodic complex flagellar motion patterns, providing a general tool for characterizing diverse flagellar behaviors, and realizing the quantitative extraction and statistical analysis of motion parameters from single-cell single moment to population-level time series, greatly improving the throughput and efficiency of sperm flagellar motion parameter analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein: Figure 1 It is a flow chart of the steps of the method for calculating the sperm flagella motion parameters provided by the present invention; Figure 2 It is a flow chart of the method for quantitative analysis of sperm flagellar motion parameters; Figure 3 This is the structure diagram of the U-Net model, a semantic segmentation backbone network based on encoder-decoder; Figure 4 This is the working principle diagram of the continuous convolution kernel; Figure 5 This is a schematic diagram of the technical difficulties in accurately and continuously segmenting sperm flagella; Figure 6 This is a diagram showing the limitations of standard convolution and deformable convolution; Figure 7 It is a comparative schematic diagram of the results of sperm flagella segmentation; Figure 8 This is a rendering of the sperm flagella segmentation result; Fig. 9 This is a comparison chart of statistical analysis data of some movement parameters of sperm flagella. DETAILED DESCRIPTION

[0021] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0022] Reference Figure 1 As shown, the flowchart of the method for calculating the sperm flagella motion parameters provided by the present invention includes the following specific steps: S101: extracting multiple frames of sample images from the flagella movement video of the sperm to be analyzed based on a preset sampling frequency; S102: for each frame of sample image, input it into an improved U-Net network to obtain a corresponding binary flagella segmentation mask; the convolution kernel in the improved U-Net network is a convolution kernel that introduces a constrained deformable offset; S103: extracting the flagellar skeleton from the binary flagellar segmentation mask using a morphological thinning algorithm; S104: parameterize the flagellar skeleton and obtain a parameterized flagellar skeleton curve, which is expressed as: ; in, Represents the starting point to point of the parameterized flagellar skeleton curve The arc length.

[0023] S105: Based on the parameterized flagellar skeleton curve, calculate the flagellar length, flagellar tangential angle, flagellar propagation speed and flagellar curvature; S106: Based on the flagellar curvature, the flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index are calculated.

[0024] Specifically, in this embodiment, the acquisition of the U-Net network is improved, including: The common convolution kernel in the standard U-Net network is linearly expanded along the x-axis and y-axis directions to obtain the initial state of the continuous convolution kernel; Introduce a constrained deformable offset into the initial continuous convolution kernel to obtain a continuous convolution kernel; The continuous convolution kernel is integrated into the standard U-Net network to obtain an improved U-Net network.

[0025] Among them, the convolution process of the continuous convolution kernel is expressed as: ; in, Indicates the output feature map at pixel position Where Axis features, Indicates the output feature map at pixel position Where Axis features; Indicates the output feature map at pixel position The characteristics of represents the convolution kernel A grid, represents the convolution kernel The preset weights corresponding to the grids, represents the convolution kernel The preset modulation coefficient corresponding to each grid; , Represents the total number of grids in the convolution kernel; Represents the convolution kernel Axis direction The deformable offset of the mesh is expressed as ; Represents the convolution kernel Axis direction The deformable offset of the mesh is expressed as ; Represents a preset constraint, Represents the convolution kernel Axis direction The offset value of the grid, Represents the convolution kernel Axis direction The offset value of the grid.

[0026] This embodiment uses the trained improved U-Net network to obtain the binary flagella segmentation mask corresponding to the sample image, and improves the training process of the U-Net network, including: Live sperm were selected from semen samples using the water bath-swim-up method as sperm samples, and flagellar movement videos of sperm samples were collected using a 40x objective lens; The flagellar motion video is converted into frame sequence images according to the preset sampling frequency, and the outline of the sperm flagella on each image is marked using the polygon tool to obtain the sperm flagella annotation dataset; Each frame of the sperm flagella annotation dataset is converted into a grayscale image and then assigned to three color channels: R, G, and B, to obtain the corresponding three-channel pseudo-color image. The sperm flagella annotation dataset is divided into a training set and a test set, and the training set data is expanded by image rotation, translation and flipping; Based on the expanded data set, the cross entropy loss function and Adam optimizer are used to update the parameters of the improved U-Net network until the preset number of training times is reached to obtain the trained improved U-Net network.

[0027] The method for calculating the motion parameters of sperm flagella described in the present invention introduces a constrained deformable offset into the convolution kernel of a standard U-Net network to obtain an improved U-Net network, and extracts a binary flagella segmentation mask from a sample image, thereby effectively overcoming the limitations of standard convolution and deformable convolution in segmenting sperm flagella. By linearizing the convolution kernel, limiting the offset range, and adopting an offset accumulation strategy, the continuity of the segmentation result is ensured while the receptive field is increased, thereby achieving the accuracy and continuity of flagella segmentation.

[0028] Specifically, in step S105, based on the parameterized flagella skeleton curve, the flagella length, flagella tangential angle, flagella propagation speed and flagella curvature are calculated, including: S105-1: Discretize the parameterized flagellar skeleton curve, obtain multiple sampling points, calculate the Euclidean distance between adjacent sampling points, and sum them to obtain the flagellar length , expressed as: ; , Represents the total number of sampling points, and Respectively represent The x-axis coordinates and y-axis coordinates corresponding to the sampling points; S105-2: Arc length Find the first-order derivative of the x-axis coordinate and the y-axis coordinate at The first derivative of the x-axis coordinate The inverse tangent function of the ratio , as the flagellar tangential angle , expressed as: ; S105-3: Calculate the flagellar propagation speed based on the rate of change of the cumulative length of the flagellar curve at adjacent moments , expressed as: ; represents the cumulative arc length of the flagellar curve at time t, Indicates a preset time interval; S105-4: Arc length The x-axis coordinate and y-axis coordinate at the position are used to find the first-order derivative and the second-order derivative, and the curvature of the flagella is calculated. , expressed as: ; and Respectively represent the arc length The second derivative of the x-axis coordinate and the second derivative of the y-axis coordinate at.

[0029] Specifically, in step S106, based on the flagellar curvature, the flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index are calculated, including: S106-1: Obtain multiple pairs of local maxima and local minima of the flagellar curvature as wave peaks With trough ; Calculate the sum of the vertical distances between all adjacent peaks and troughs and average them to obtain the flagellar amplitude , expressed as: ; Represents the vertical distance between adjacent peaks and troughs. It represents the total number of pairs of peaks and troughs; S106-2: Based on unit time Number of flagellar curvature sign changes at inner arc length s The ratio of the unit time to calculate the flagellar frequency , expressed as: ; S106-3: Based on propagation speed The ratio of the flagellar frequency to the flagellar wavelength , expressed as: ; S106-4: Integrate the flagellar curvature of the parameterized flagellar skeleton curve along the arc length to obtain the flagellar bending amplitude , expressed as: ; S106-5: Based on flagellar curvature The mean With standard deviation , get the flagellar bending index , expressed as: .

[0030] Based on the above embodiment, in this embodiment, based on the flagellar amplitude Flagellar wavelength Calculate the flagellar amplitude-wavelength ratio , expressed as: .

[0031] Based on the above embodiment, in this embodiment, the flagellar tangential angle is calculated Arc length The derivative of , expressed as: .

[0032] At the same time, the present invention extracts the flagellar skeleton from the binary flagellar segmentation mask, parameterizes it, calculates the curvature of the parameterized flagellar skeleton curve, and uses the curvature as the basic feature to quantitatively calculate multiple parameters including flagellar length, flagellar tangential angle, flagellar propagation speed, flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index, thereby simplifying the calculation process and expanding the scope of use of the calculation method, so that it can analyze non-periodic complex flagellar motion patterns, providing a general tool for characterizing diverse flagellar behaviors, and realizing the quantitative extraction and statistical analysis of motion parameters from single-cell single moment to population-level time series, greatly improving the throughput and efficiency of sperm flagellar motion parameter analysis.

[0033] Based on the above embodiments, in the embodiments of the present invention, the improved U-Net network of the present invention is used as the sperm flagella segmentation network model, and the sperm flagella motion parameter calculation method provided by the present invention is used to calculate the flagella motion parameters of the sperm sample. Figure 2 As shown, it is a flow chart of the method for quantitative analysis of sperm flagellar motion parameters, which specifically includes: S201: Obtain sperm movement video and establish sperm flagella dataset; According to the guidance of the World Health Organization Laboratory Manual for the Examination and Processing of Human Semen, sperm samples were subjected to the "water bath-swim-up method" to screen out live sperm, and sperm movement videos were collected under a 40x objective lens of the microscope; The sperm motion video is converted into a frame sequence image according to the preset sampling frequency, and the outline of the sperm flagellum is marked using the polygon tool to form a sperm flagellum annotation dataset; Using frame sequence images as input for the network model training process; S202: Preprocess and expand the sperm flagella annotation dataset; For each sperm flagella image in the sperm flagella annotation dataset, convert it into a grayscale image and then assign it to three color channels: red (R), green (G), and blue (B) to form a three-channel pseudo-color image to eliminate color errors caused by different microscope settings, lighting conditions, and white balance during microscopic imaging; The sperm flagella annotation dataset is divided into a training set and a test set. The training set is subjected to data augmentation operations such as image rotation, translation, and flipping to increase data diversity, simulate different observation conditions, and normalize all images. S203: Constructing a semantic segmentation network based on an encoder-decoder structure; Reference Figure 3As shown, it is a structural diagram of the U-Net model of the semantic segmentation backbone network based on encoder-decoder; The encoder extracts image features through multi-layer downsampling, and the decoder restores the original resolution through multi-layer upsampling, segments the sperm flagella contour, and finally outputs the sperm flagella segmentation result; The encoder and decoder adopt a symmetrical multi-layer structure; S204: construct a continuous convolution kernel and integrate it into the above semantic segmentation network; The common convolution kernel is linearly expanded along the x-axis and y-axis directions to form the initial state of the continuous convolution kernel. The receptive field is expanded through linearization operation and adapted to different flagellar orientations. Introducing Deformable Offset and , respectively represent the offset of the kth convolution kernel grid in the x-axis and y-axis directions, , is the total number of convolution kernel grids in the x-axis and y-axis directions; In order to maintain the connectivity of the segmentation result, a constraint is imposed on the deformable offset, and the offset value Limiting to the range of [-1,1], ensuring that the offset difference between adjacent grids does not exceed 1 pixel, thus maintaining the connectivity of the segmentation results in space; Through the offset accumulation method, the offset of each grid is recursively calculated. The offset of the grid far away from the center is recursively obtained from the offset of the previous grid, so that the continuous convolution kernel can adapt to local curvature and orientation changes, which can be expressed as: ; ; The continuous convolution kernel is integrated into the encoder-decoder network to obtain the sperm flagellum segmentation model; Reference Figure 4 As shown in the figure, it is the working principle diagram of the continuous convolution kernel; the calculation process of the continuous convolution is expressed as: ; in, and The output and input feature maps are at positions The characteristics of For the convolution kernel grid coordinates, is the corresponding weight, is the deformable offset in the y-axis direction, is the corresponding modulation coefficient; Reference Figure 5 The figure shows the technical difficulties in accurately and continuously segmenting sperm flagella; Figure 6The figure shows the limitations of standard convolution and deformable convolution. The present invention proposes a continuous convolution module, which effectively overcomes the limitations of standard convolution and deformable convolution in segmenting sperm flagella by introducing a constrained deformable offset. The fixed kernel shape of the standard convolution is difficult to adapt to the irregular shape of the flagellum, and the arbitrary offset of the deformable convolution may cause the segmentation result to be broken. The continuous convolution kernel linearizes the convolution kernel, limits the offset range, and adopts an offset accumulation strategy, which ensures the continuity of the segmentation result while increasing the receptive field, thereby achieving the accuracy and continuity of flagella segmentation.

[0034] S205: Using the training set to train the sperm flagella segmentation model, using the cross entropy loss function and the Adam optimizer, to obtain a trained sperm flagella segmentation network model; using the test set to evaluate the model performance and calculate various accuracy indicators; S206: using the trained sperm flagellum segmentation network model, processing the sperm image sequence to be segmented frame by frame to obtain a binary flagellum segmentation mask sequence; S207: extracting the skeleton of each frame of flagella segmentation mask, and using a morphological thinning algorithm to obtain a flagella skeleton with a single pixel width; The skeleton curve is parameterized as , is the normalized arc length along the skeleton curve; S208: By parameterizing the skeleton curve coordinates , calculate the curvature of the sperm flagellum , the calculation formula is: ; in, , and , Respectively , about The first and second derivatives of .

[0035] S209: Based on the curvature of the sperm flagellum, the sperm flagellum motion parameters are calculated, including flagellum length, flagellum amplitude, flagellum frequency, flagellum wavelength, flagellum tangential angle, flagellum propagation speed, flagellum bending arc, flagellum amplitude-to-wavelength ratio, flagellum waveform index, and flagellum bending index, which are expressed as: ① Flagella length : , discretize the skeleton curve, calculate the Euclidean distance between adjacent sampling points and sum them up, is the number of sampling points on the skeleton curve.

[0036] ② Flagellar amplitude : , find the curvature The local maximum (peak) and minimum (trough) points of the wave are calculated and averaged. For adjacent peaks and valleys (curvature The vertical distance between local extreme points, Represents the total number of pairs of peaks and troughs.

[0037] ③Flagellar frequency : , select the arc length position, and count the unit time The number of times the curvature sign changes (turns) at a specific arc length position .

[0038] ④Flagellar wavelength : , where the propagation wave speed The calculation is estimated by tracking the movement of the extreme points of curvature over time.

[0039] ⑤ Flagellar tangential angle :

[0040] ⑥Flagellar propagation speed : ,in is the cumulative arc length of the flagellar curve at time t, is the time interval.

[0041] ⑦Flagellar curvature : , for curvature Integrate along the arc length.

[0042] ⑧ Flagellar amplitude to wavelength ratio :

[0043] ⑨Flagellar waveform index : , the derivative of the tangent angle with respect to the arc length.

[0044] ⑩Flagellar bending index : ,in, , The curvature The mean and standard deviation of .

[0045] S210: Repeat steps S206-S209 to calculate the motion parameters of each sperm flagellum in each frame of the frame sequence image of the sperm motion video, and then analyze the change pattern of the parameters over time.

[0046] Reference Figure 7 The following is a comparative diagram of the results of sperm flagella segmentation; Figure 8 The following is a rendering of the sperm flagellum segmentation result; see Fig. 9 As shown, it is a comparison chart of the statistical analysis data of some motion parameters of sperm flagella. After obtaining the flagellar motion parameters using the method of the present invention, the flagellar motion parameters of different groups (for example, normal or infertile) can be statistically tested and data mined in combination with clinical pathological data to reveal the potential association between abnormal flagellar motion and disease. At the same time, the present invention organically combines the links of sperm flagella segmentation, skeleton extraction, parameter calculation, etc., forming a set of fully automated analysis processes, and successfully applied to the analysis of large-scale sperm samples, realizing the quantitative extraction and statistical analysis of motion parameters from single cell single moment to population-level time series, greatly improving the throughput and efficiency of sperm flagellar motion parameter analysis, and having important scientific significance and application value in basic research on reproductive biology, diagnosis of male infertility, assisted reproduction, etc., providing a quantitative analysis tool for further studying the mechanism of sperm motion and assisting the clinical diagnosis of male infertility, and providing technical support for the diagnosis and treatment of sperm motility disorders.

[0047] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for calculating sperm flagella motion parameters as described above are implemented.

[0048] The method for calculating the motion parameters of sperm flagella described in the present invention introduces a constrained deformable offset into the convolution kernel of a standard U-Net network to obtain an improved U-Net network, and extracts a binary flagella segmentation mask from a sample image, thereby effectively overcoming the limitations of standard convolution and deformable convolution in segmenting sperm flagella. By linearizing the convolution kernel, limiting the offset range, and adopting an offset accumulation strategy, the continuity of the segmentation result is ensured while the receptive field is increased, thereby achieving the accuracy and continuity of flagella segmentation. At the same time, the present invention extracts the flagellar skeleton from the binary flagellar segmentation mask, parameterizes it, calculates the curvature of the parameterized flagellar skeleton curve, and uses the curvature as the basic feature to quantitatively calculate multiple parameters including flagellar length, flagellar tangential angle, flagellar propagation speed, flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index, thereby simplifying the calculation process and expanding the scope of use of the calculation method, so that it can analyze non-periodic complex flagellar motion patterns, providing a general tool for characterizing diverse flagellar behaviors, and realizing the quantitative extraction and statistical analysis of motion parameters from single-cell single moment to population-level time series, greatly improving the throughput and efficiency of sperm flagellar motion parameter analysis.

[0049] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0050] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0051] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0052] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0053] Obviously, the above embodiments are merely examples for the purpose of clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.

Claims

1. A method for calculating sperm flagella motion parameters, characterized in that: include: Based on a preset sampling frequency, extract multiple frames of sample images from the flagella movement video of the sperm to be analyzed; For each frame of sample image, input it into the improved U-Net network to obtain the corresponding binary flagella segmentation mask; The convolution kernel in the improved U-Net network is a convolution kernel that introduces a constrained deformable offset; The flagellar skeleton is extracted from the binary flagellar segmentation mask using a morphological thinning algorithm; Parameterize the flagellar skeleton to obtain a parameterized flagellar skeleton curve; Based on the parameterized flagellar skeleton curve, the flagellar length, flagellar tangential angle, flagellar propagation speed and flagellar curvature are calculated; Based on the flagellar curvature, the flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index are calculated.

2. The method for calculating sperm flagella motion parameters according to claim 1, wherein: The acquisition of the improved U-Net network includes: The common convolution kernel in the standard U-Net network is linearly expanded along the x-axis and y-axis directions to obtain the initial state of the continuous convolution kernel; Introduce a constrained deformable offset into the initial continuous convolution kernel to obtain a continuous convolution kernel; The continuous convolution kernel is integrated into the standard U-Net network to obtain an improved U-Net network.

3. The method for calculating sperm flagella motion parameters according to claim 2, wherein: The convolution process of the continuous convolution kernel is expressed as: ； in, Indicates the output feature map at pixel position Where Axis features, Indicates the output feature map at pixel position Where Axis features; Indicates the output feature map at pixel position The characteristics of represents the convolution kernel A grid, represents the convolution kernel The preset weights corresponding to the grids, represents the convolution kernel The preset modulation coefficient corresponding to each grid; , Represents the total number of grids in the convolution kernel; Represents the convolution kernel Axis direction The deformable offset of the mesh is expressed as ； Represents the convolution kernel Axis direction The deformable offset of the mesh is expressed as ； Represents a preset constraint, Represents the convolution kernel Axis direction The offset value of the grid, Represents the convolution kernel Axis direction The offset value of the grid.

4. The method for calculating sperm flagella motion parameters according to claim 2, wherein: Improvements to U-Net network training, including: Live sperm were selected from semen samples using the water bath-swim-up method as sperm samples, and flagellar movement videos of sperm samples were collected using a 40x objective lens; The flagellar motion video is converted into frame sequence images according to the preset sampling frequency, and the outline of the sperm flagella on each image is marked using the polygon tool to obtain the sperm flagella annotation dataset; Each frame of the sperm flagella annotation dataset is converted into a grayscale image and then assigned to three color channels: R, G, and B, to obtain the corresponding three-channel pseudo-color image. The sperm flagella annotation dataset is divided into a training set and a test set, and the training set data is expanded by image rotation, translation and flipping; Based on the expanded data set, the cross entropy loss function and Adam optimizer are used to update the parameters of the improved U-Net network until the preset number of training times is reached to obtain the trained improved U-Net network.

5. The method for calculating sperm flagella motion parameters according to claim 1, wherein: The flagellar skeleton is parameterized to obtain the parameterized flagellar skeleton curve, which is expressed as: ； in, Represents the starting point to point of the parameterized flagellar skeleton curve The arc length.

6. The method for calculating sperm flagella motion parameters according to claim 5, wherein: Based on the parameterized flagellar skeleton curve, the flagellar length, flagellar tangential angle, flagellar propagation speed and flagellar curvature are calculated, including: Discretize the parameterized flagellar skeleton curve, obtain multiple sampling points, calculate the Euclidean distance between adjacent sampling points, and sum them to obtain the flagellar length , expressed as: ； , Represents the total number of sampling points, and Respectively represent The x-axis coordinates and y-axis coordinates corresponding to the sampling points; Arc length Find the first-order derivative of the x-axis coordinate and the y-axis coordinate at The first derivative of the x-axis coordinate The inverse tangent function of the ratio , as the flagellar tangential angle , expressed as: ; The flagellar propagation velocity is calculated based on the rate of change of the cumulative length of the flagellar curve at adjacent moments. , expressed as: ; represents the cumulative arc length of the flagellar curve at time t, Indicates a preset time interval; Arc length The x-axis coordinate and y-axis coordinate at the position are used to find the first-order derivative and the second-order derivative, and the curvature of the flagella is calculated. , expressed as: ; and Respectively represent the arc length The second derivative of the x-axis coordinate and the second derivative of the y-axis coordinate at.

7. The method for calculating sperm flagella motion parameters according to claim 6, wherein: Based on the flagellar curvature, the flagellar amplitude, flagellar frequency, flagellar wavelength, flagellar bending amplitude and flagellar bending index are calculated, including: Get multiple pairs of local maxima and local minima of the flagellar curvature as wave peaks With trough ; Calculate the sum of the vertical distances between all adjacent peaks and troughs and average them to obtain the flagellar amplitude , expressed as: ; Represents the vertical distance between adjacent peaks and troughs. It represents the total number of pairs of peaks and troughs; Based on unit time Number of flagellar curvature sign changes at inner arc length s The ratio of the unit time to calculate the flagellar frequency , expressed as: ; Based on the propagation speed Ratio of the frequency to the flagellar wavelength , expressed as: ; The curvature of the flagella of the parameterized flagellar skeleton curve is integrated along the arc length to obtain the bending amplitude of the flagella , expressed as: ; Based on flagellar curvature The mean With standard deviation , get the flagellar bending index , expressed as: .

8. The method for calculating sperm flagella motion parameters according to claim 7, wherein: It also includes flagellar amplitude-based Flagellar wavelength Calculate the flagellar amplitude-wavelength ratio , expressed as: .

9. The method for calculating sperm flagella motion parameters according to claim 7, wherein: It also includes the calculation of the flagellar tangential angle Arc length The derivative of , expressed as: .

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for calculating sperm flagellar motion parameters as described in any one of claims 1 to 9 are implemented.

Images