Automatic sperm screening system and method based on photoelectric tweezers

Abstract

The invention provides an automatic sperm screening system and method based on photoelectric tweezers, and belongs to the technical field of sperm screening. According to the method, the advantages of high throughput, no damage and no contact of photoelectric tweezers are utilized, and a new platform is provided for sperm screening by building a photoelectric tweezers sperm control screening hardware system. And a sperm detection framework is constructed by combining a computer vision algorithm, so that high-precision detection of sperm positions, forms and speeds is realized. Based on the software and hardware platform, rapid and automatic sperm screening based on different biological characteristics is realized. And dynamic population control of multiple sperms is realized, and the sperm control number and efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of sperm screening, and particularly relates to a sperm automatic screening system and method based on an optoelectronic tweezer. BACKGROUND

[0002] According to statistics of the World Health Organization (WHO), about 17.5% of adults worldwide are affected by infertility, and nearly 50% of them are caused by male factors. In the past 45 years, the quality of male sperm has shown a significant downward trend, especially the significant reduction in sperm quantity and concentration, which has become a major challenge in the global reproductive health field. This trend has driven the rapid development of assisted reproductive technology (ART), such as in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI), and has highlighted the urgent need for efficient and precise sperm screening technology.

[0003] The current commonly used sperm screening methods in clinics and laboratories mainly include the following three types, but all have significant defects:

[0004] (1) Microscopic visual assessment: professional personnel observe the semen sample through a microscope and subjectively assess the number, motility and morphology of sperm. This method requires a high degree of dependence on operator experience, requires months of professional training, and has strong subjectivity and low efficiency, making it difficult to handle large quantities of samples and achieve precise sperm manipulation, and can only screen sperm that cannot be directly used for ICSI.

[0005] (2) Computer-aided sperm analysis (CASA): automatically assesses sperm movement trajectory, morphology and other parameters through image analysis algorithms. Although this method improves the degree of automation, it still lacks effective sperm manipulation means, and requires other techniques (such as microsuction) for ICSI operation after screening, and has insufficient accuracy in identifying high-concentration or low-motility samples.

[0006] (3) Optical tweezer (Optical Tweezers) technology: uses a high-focus laser beam to form an optical potential well to non-contact manipulate sperm. This method uses high laser power that may damage sperm DNA or cell structure, affecting subsequent fertilization potential; and the operation is complex, requiring frequent adjustment of the optical well position, low efficiency, high equipment cost, and difficult to popularize.

[0007] Existing technologies cannot simultaneously meet the three core needs of high-precision screening, low-damage manipulation, and fully automated process. SUMMARY

[0008] In view of the above technical problems, the present application provides a sperm automatic screening system and method based on an optoelectronic tweezer.

[0009] The first aspect of the present application discloses a sperm automatic screening method based on an optoelectronic tweezer, the method comprising:

[0010] S1, acquiring a plurality of biological characteristics of each sperm in a sperm sample on a photoelectric tweezers chip, image position information of each sperm, and image position information of each target point; the plurality of biological characteristics include sperm morphological characteristics, sperm motility speed characteristics, and fluorescent vitality characteristics of each sperm after staining;

[0011] S2, selecting a plurality of target sperms according to the plurality of biological characteristics based on a preset screening standard;

[0012] S3, calculating an optimal moving path of each target sperm to a corresponding target point based on the image position information of the target sperm and the image position information of each target point through a path matching algorithm, and then generating a dynamically adjustable light pattern template sequence according to the optimal moving path, and projecting the light pattern template sequence onto the photoelectric tweezers chip to synchronously control the target sperm to move to the target point position.

[0013] Optionally, step S1 specifically comprises:

[0014] S11, incubating the sperm sample with a live-dead staining reagent to obtain a stained sample, and observing the stained sample under a fluorescence microscope to obtain a fluorescent signal intensity as the fluorescent vitality characteristics of each sperm after staining; wherein the live-dead staining reagent is configured by mixing 2 mM calcein solution, 1.5 mM propidium iodide solution, and pure water according to a preset volume ratio;

[0015] S12, acquiring an image of the stained sample on the photoelectric tweezers chip, and identifying the image position information of each sperm and the image position information of each target point based on a target detection model, and outputting a recognition box of each sperm;

[0016] S13, performing background threshold segmentation on an image region in the recognition box to calculate a number of pixels occupied by a sperm cell head as the sperm morphological characteristics;

[0017] S14, assigning a sperm ID to each recognition box through a target tracking model, and associating the sperm ID in consecutive frames to calculate an average motion speed of the sperm within a preset number of frames as the sperm motility speed characteristics.

[0018] Optionally, before the sperm sample is incubated with the live-dead staining reagent in step S11, it further comprises:

[0019] After the cryopreserved sperm sample is thawed, it is centrifugally washed with a low-conductivity solution; the low-conductivity solution is a 5% glucose solution.

[0020] Optionally, in step S2, the preset screening standard satisfies the following conditions simultaneously:

[0021] the fluorescent signal intensity is greater than a preset intensity;

[0022] The number of pixels occupied by the sperm cell head is greater than a preset number;

[0023] The average movement speed of the sperm is greater than a preset speed.

[0024] Optionally, in step S3, the path matching algorithm is the Hungarian algorithm, and the optimal moving path of each target sperm to the corresponding target point is calculated, specifically including:

[0025] A cost matrix of the target sperm and the target point is constructed, wherein the cost is the Euclidean distance between the target sperm and the target point;

[0026] The cost matrix is solved by the Hungarian algorithm to obtain a target sperm-target point matching pair with the minimum total moving distance;

[0027] When the number of target points is less than the number of target sperms, the nearest target sperm-target point pair is preferentially matched, and the remaining target sperms are marked as unmatching state.

[0028] Optionally, in step S3, the sequence of the light pattern templates is dynamically adjusted in at least one of the following parameters:

[0029] Shape, size, line width, filling mode, brightness and rotation angle.

[0030] Optionally, in step S3, the sequence of the light pattern templates satisfies the following conditions:

[0031] The shape of each light pattern is circular, square or cross-shaped, and the diameter matches the size of the target sperm;

[0032] The displacement step Δx, Δy of the light pattern is dynamically adjusted according to the difference between the current position of the target sperm and the coordinate of the target point.

[0033] The second aspect of the present application discloses a sperm automatic screening system based on an optoelectronic tweezer, which comprises:

[0034] The first processing module is configured to acquire a plurality of biological characteristics of each sperm in the sperm sample on the optoelectronic tweezer chip, image position information of each sperm and image position information of each target point; the plurality of biological characteristics include sperm morphological characteristics, sperm swimming speed characteristics and sperm fluorescence activity characteristics after staining;

[0035] The second processing module is configured to select a plurality of target sperms according to the plurality of biological characteristics based on a preset screening standard.

[0036] The third processing module is configured to calculate, based on the image position information of the target sperm and the image position information of each target point, an optimal moving path of each target sperm to a corresponding target point through a path matching algorithm, to generate a dynamic adjustable light pattern template sequence according to the optimal moving path, and to project the light pattern template sequence onto the optoelectronic tweezer chip to synchronously control the target sperm to move to the target point position.

[0037] The third aspect of the present application discloses an electronic device. The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of the sperm automatic screening method based on the optoelectronic tweezer according to the first aspect of the present application when executing the computer program.

[0038] The fourth aspect of the present application discloses a computer readable storage medium. The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the sperm automatic screening method based on the optoelectronic tweezer according to the first aspect of the present application.

[0039] In summary, the scheme provided by the present application has the following technical effects: compared with the prior art, the present application uses the advantages of high throughput, non-invasive and non-contact of the optoelectronic tweezer, and provides a new platform for sperm screening by building an optoelectronic tweezer sperm manipulation and screening hardware system. Combined with computer vision algorithm, a sperm detection framework is constructed to realize high-precision detection of sperm position, shape and speed. Based on the above hardware and software platform, rapid and automatic screening of sperm based on different biological characteristics is realized. And dynamic population manipulation of multiple sperm is realized, and the number and efficiency of sperm manipulation are improved. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0041] Figure 1 The method flowchart of the embodiment of the present application is shown in the figure;

[0042] Figure 2 The light pattern projection process diagram of the embodiment of the present application is shown in the figure;

[0043] Figure 3 The sperm detection and tracking algorithm schematic diagram of the embodiment of the present application is shown in the figure;

[0044] Figure 4 The three sperm screening methods of the optoelectronic tweezer system of the embodiment of the present application are shown in the figure;

[0045] Figure 5 Figure 1 is a schematic diagram of a sperm dynamic manipulation method of an optical tweezers system according to an embodiment of the present application;

[0046] Figure 6 Figure 1 is a schematic diagram of a sperm dynamic manipulation method of an optical tweezers system according to an embodiment of the present application; DETAILED DESCRIPTION

[0047] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0048] The first aspect of the present application discloses a sperm automatic screening method based on optical tweezers, please refer to Figure 1 , the method comprises:

[0049] S1, obtaining a plurality of biological characteristics of each sperm in the sperm sample on the optical tweezers chip, image position information of each sperm and image position information of each target point; the plurality of biological characteristics include sperm morphological characteristics, sperm swimming speed characteristics and fluorescent vitality characteristics of each sperm after staining;

[0050] Optionally, step S1 specifically comprises:

[0051] S11, mixing and incubating the sperm sample with a live-dead staining reagent to obtain a stained sample, and observing the stained sample under a fluorescence microscope to obtain a fluorescent signal intensity as the fluorescent vitality characteristics of each sperm after staining; wherein the configuration method of the live-dead staining reagent is: mixing 2mM calcein solution, 1.5mM propidium iodide solution and pure water according to a preset volume ratio;

[0052] Optionally, before the sperm sample is mixed and incubated with the live-dead staining reagent in step S11, it further comprises:

[0053] After thawing the frozen sperm sample, centrifugal cleaning is performed with a low-conductivity solution; the low-conductivity solution is a 5% glucose solution.

[0054] S12, collecting an image of the stained sample on the optical tweezers chip, and identifying the image position information of each sperm and the image position information of each target point based on a target detection model, and outputting a recognition box of each sperm;

[0055] S13, performing background threshold segmentation on the image region in the recognition box to calculate the number of pixels occupied by the sperm cell head as the sperm morphological characteristics;

[0056] S14, assigning sperm ID to each recognition box through the target tracking model, and associating the sperm ID in the continuous frames to calculate the average movement speed of the sperm within the preset frame number as the sperm swimming speed feature.

[0057] As an optional implementation, in order to develop the sperm detection algorithm under the optoelectronic tweezers system, three modules of sperm detection, target tracking and speed calculation are divided. At present, the deep learning technology develops rapidly, and its application in target detection, image classification and target tracking fields is more and more extensive, which can significantly improve the accuracy and efficiency of image recognition. Therefore, the detection algorithm based on deep learning is applied to realize the rapid screening of sperm under the optoelectronic tweezers system.

[0058] In this embodiment, it is necessary to identify the sperm cells under the optoelectronic tweezers system in real time and quickly, obtain the image position information of the target sperm, and then project the corresponding light pattern to realize the control of the sperm cells. At present, the target detection algorithm based on deep learning includes R-CNN series and YOLO series. These two are the pioneers of double-stage target detection and single-stage target detection. Compared with double-stage target detection, single-stage target detection directly calculates the image to generate detection results, and the whole process only needs one step, which is very fast and has high detection accuracy. Therefore, the YOLO series algorithm is selected as the sperm detection module framework of the optoelectronic tweezers system to realize the real-time detection of sperm cells under the optoelectronic tweezers system.

[0059] The data set mainly consists of three parts: sperm cell images photographed in real time under the optoelectronic tweezers system; sperm cell images photographed under the laboratory biological microscope and metallographic microscope; and open source sperm cell data set to enhance data diversity and algorithm generalization ability.

[0060] Please refer to Figure 2 , the optoelectronic tweezers system includes a computer 101, a bright field light source 102, a first imaging lens 103, a signal generator 104, a DMD spatial light modulator 105, a second imaging lens 106, a third imaging lens 107, a first filter 108, a CCD camera 109, a second filter 110, a fourth imaging lens 111, a light splitting plate 112, a fifth imaging lens 113, an optoelectronic tweezers chip 114, a three-dimensional micro stage 115, an objective lens 116, and a mirror 117.

[0061] The overall workflow is as follows: 114 The real-time image on the optoelectronic tweezer chip is transmitted through 113 the fifth imaging lens, 112 the light splitting sheet, and 111 the fourth imaging lens, and then filtered by 110 the second filter to remove red light, and finally collected and imaged by 109 the CCD camera. Then the image data is transmitted to 101 the computer through the USB interface, and 101 the computer generates a black background white circle light pattern template combined with advanced vision algorithms. The light pattern template is transmitted to 105 the DMD spatial light modulator for projection through the HDMI interface, and then sequentially passes through 106 the second imaging lens and 107 the third imaging lens to reduce the size of the light pattern, and then passes through 108 the first filter to remove blue light, and then passes through 117 the mirror to transmit the image to 116 the objective lens, and finally projects into 114 the optoelectronic tweezer chip to generate a non-uniform electric field, realizing non-contact manipulation of sperm. In addition, 102 the bright field light source becomes parallel light through 103 the first imaging lens, and then passes through 112 the light splitting sheet and 113 the fifth imaging lens, and then is used to illuminate the overall field of view of 114 the optoelectronic tweezer chip. The 104 signal generator is directly connected to 114 the optoelectronic tweezer chip on the 115 three-dimensional microstage through positive and negative leads to provide an alternating electric field.

[0062] The main detection of this embodiment is sperm cells under the optoelectronic tweezer system, but due to the limitations of the resolution and magnification of the existing system, it is not possible to make a detailed division of the morphology of sperm, so only one type of target object, sperm, is temporarily set.

[0063] Then use the labelImg data labeling tool to label the data set. After all the pictures are labeled, the label file is stored in.txt format to match the YOLO series algorithm, and the format of each target labeling box is as follows:

[0064] Table 1 Target labeling box format

[0065]

[0066] Among them, the category is only one, represented by the number 0, and the last four values represent the normalized target labeling box center point x, y coordinates and size width, length. After the data set is collected and labeled, the selection and training of the detection model can begin.

[0067] Please refer to Figure 3 , the target detection module uses the YOLOv8 algorithm, which has excellent real-time performance and high precision in image processing. In this module, the YOLOv8 algorithm is used to analyze sperm images in real time, automatically detect and recognize sperm cells, and generate sperm recognition boxes that contain the location and ID information of sperm. The location information of sperm is a key reference for the optoelectronic tweezer system to manipulate sperm, ensuring that the position of the target sperm can be accurately locked when the light pattern is projected.

[0068] The target tracking module selects the ByteTrack algorithm, which has a significant advantage in dynamic video analysis due to its efficient and stable tracking capability. ByteTrack maintains the consistency of sperm recognition boxes by combining target detection results and target position information from the previous frame, ensuring that the sperm tracking process is not interrupted in complex environments.

[0069] When performing image detection on the first frame, the algorithm constructs a trajectory object for all recognized target boxes and stores them, marking them as tracked trajectories. After the second frame, the algorithm classifies trajectories and target boxes. Trajectories are classified as active or inactive based on whether they have been continuously tracked for more than two frames. Target boxes are classified as high confidence and low confidence based on their confidence levels.

[0070] The first tracking of trajectories is then performed, focusing only on matching high-confidence target boxes for active trajectories. This step primarily uses the Kalman filter algorithm to predict the possible position and size of the next frame's target box for the preliminary tracking trajectory. Then, the Hungarian algorithm is used to match the preliminary tracking trajectory and high-confidence target boxes based on the IoU loss matrix. Successfully matched groups replace the preliminary trajectory's prediction box with the current frame's high-confidence target box, maintaining the same ID and marking it as a tracked trajectory.

[0071] The second tracking of trajectories is then performed, matching the predicted next frame's target box for trajectories that did not match high-confidence target boxes in the first match. If this fails, the trajectory is marked as lost and left for the next frame to match and track.

[0072] Subsequently, inactive trajectories are matched with high-confidence target boxes that have not been matched before. If the match fails, the inactive trajectory is marked as deleted, and the high-confidence target box that has not been matched is marked as a new inactive trajectory.

[0073] Finally, all tracked trajectories are returned for image post-processing and display.

[0074] The speed calculation module is responsible for calculating the motion speed of the target sperm, reflecting its vitality. This module calculates the average speed of sperm by tracking the displacement between images at specified intervals and using the standard physical formula (distance / time). Sperm speed is an important indicator of its vitality, with high speed generally indicating strong sperm vitality.

[0075] The sperm motility can be reflected by the speed of sperm movement to some extent. The target tracking module provides stable and reliable data support for the speed calculation module. Since the tracking module has stored the position information of each detection target in multiple consecutive frames, the speed calculation module can specify the average speed of the target in a certain number of frames to reflect the movement of the target. The main parameters are shown in Table 2:

[0076] Table 2 Main parameter settings of speed calculation module

[0077]

[0078] The specific calculation process of the speed calculation module is as follows:

[0079] First, specify the total number of frames for speed calculation;

[0080] Then, for all detected and tracked targets, take out their position coordinates in each frame of the image;

[0081] Then, according to the existing image position coordinates, calculate the absolute distance between the two frames, and accumulate the total distance traveled by the detection target in the corresponding frame number, which is px here;

[0082] Then, divide the total distance traveled by the detection target by the time spent in the corresponding frame number to get the average speed of the target movement, which is px / s here.

[0083] If the actual movement speed of the target needs to be calculated later, the image coordinate speed obtained can be multiplied by the ratio of the actual size of the target to the pixels occupied by the image.

[0084] S2, selecting multiple target sperms based on a plurality of biological characteristics according to a preset screening standard;

[0085] Optionally, in step S2, the preset screening standard satisfies the following conditions simultaneously:

[0086] The fluorescence signal intensity is greater than a preset intensity;

[0087] The number of pixels occupied by the sperm cell head is greater than a preset number;

[0088] The average movement speed of the sperm is greater than a preset speed.

[0089] Figure 4 Three sperm screening methods of the optoelectronic tweezers system are shown, which are morphological screening, speed screening and fluorescence screening.

[0090] The morphological screening method identifies sperm cells in the field of view in real time through a sperm detection algorithm. According to the area and size of the sperm head, sperm that meet the preset standards are screened. The sperm screening criteria in this process include the size, shape and other related geometric characteristics of the sperm head. Subsequently, the system generates an optical circle through the optoelectronic tweezers technology to separate and manipulate the target sperm from the rest of the sperm population, achieving sperm screening.

[0091] The speed screening method combines sperm detection, target tracking and speed calculation modules. First, the sperm detection algorithm accurately identifies and tracks individual sperm in the field of view, and the target tracking module obtains the motion trajectory of the sperm in real time. Then, based on the speed calculation algorithm, the motion speed of each sperm is calculated. During the screening process, the system selects the sperm with the fastest speed based on the set speed threshold, and finally separates and screens the target sperm cells through the optoelectronic tweezers.

[0092] The fluorescence screening method is based on cell viability staining reagents. First, a specific viability staining reagent is added to the sperm sample, which can distinguish between active and dead sperm. The sperm cells are screened by the difference in fluorescence intensity of active and dead sperm under different wavebands. The fluorescence intensity is used to judge the vitality level, and finally the system selects the active sperm with better performance for subsequent screening and separation operation.

[0093] The core task of this step is to screen sperm cells based on the optoelectronic tweezers system, using different biological characteristics of sperm cells such as sperm morphology, sperm swimming speed and fluorescence intensity after sperm staining, to select the best single sperm for separation and culture.

[0094] First, the morphological screening is verified. The specific experimental process is as follows: the sperm detection algorithm developed in the above is used to identify sperm cells in the field of view, and the best single sperm is selected according to the morphology, and then the target sperm is separated and screened by projecting a light circle. However, due to the limitations of lens magnification and optical system, it is not possible to clearly distinguish the good and bad sperm morphology, so the strategy is changed to select sperm based on the area and size of the sperm cell head in the sperm recognition box. The detected sperm cells are all framed by green rectangular recognition boxes. Then the algorithm crops the corresponding image according to the size and position of the green recognition box, and then performs background threshold segmentation to obtain the number of pixels occupied by the sperm cells in the recognition box, and then displays it above the green recognition box. At the same time, the algorithm records the position of the green rectangular box with the most pixels of sperm cells in real time, preparing for the subsequent light pattern projection and sperm cell path planning.

[0095] Then the speed screening of sperm cells is carried out, which needs to comprehensively use the three modules in the sperm detection algorithm, i.e. sperm detection, target tracking and speed calculation module, so as to pick out the fastest sperm and separate and screen it out by using the light ring. The specific experimental speed measurement process is as follows: first, the algorithm uses the sperm detection module to process the real-time transmission of the sperm image of the CCD camera, identifies the sperm cells in the field of view and frames them with a red rectangular identification box; then the target tracking module is used to assign an ID to each rectangular identification box and ensure the consistency of the ID; then the speed calculation module is used to calculate the average speed of sperm movement every 5 frames, and then the pixel and actual size are converted to obtain the actual speed of sperm movement, and the speed size is displayed above each red rectangular identification box. At the same time, the system also records the position and ID of the red rectangular identification box with the fastest sperm swimming speed in real time, in order to prepare for the next light pattern screening to separate the sperm with the fastest swimming speed.

[0096] The last one in the sperm screening experiment is the vitality screening of sperm cells, which is screened according to the degree of vitality of sperm cells. It was originally intended to use different vitality of sperm to attract and capture different forces under the same light ring to screen live and dead sperm, but it was found during the experiment that it was difficult to realize the stratified screening of sperm with different vitality by setting the voltage amplitude, frequency and light ring parameters. Therefore, the common live and dead staining method is used to replace it to realize the fluorescence screening of sperm with different vitality.

[0097] The steps of live and dead staining fluorescence experiment are as follows:

[0098] Staining solution configuration: the prepared cell live and dead staining reagents Calcein-AM Solution (2mM) and Propidium Iodide PI Solution (1.5mM), 10x Assay Buffer buffer and pure water are configured in a ratio of 1:1:1:7 to obtain 1ml of staining solution. "10x" means that the buffer is a 10-fold concentrated stock solution, which needs to be diluted to 1x working concentration in proportion when used.

[0099] Centrifugal washing of sperm cells: the frozen sperm tube is taken out from the liquid nitrogen tank for water bath thawing, then added into the centrifugal tube, washed with 1x Assay Buffer buffer, and the sperm cells are centrifuged twice, and then diluted with 5% glucose solution to obtain the sperm solution.

[0100] Sperm cell staining incubation: mix the staining solution and the sperm cell solution in a ratio of 1:2 to obtain 900μl of stained cell solution, and incubate in the dark for 15 minutes to allow the staining reagent to fully contact with the cells.

[0101] Fluorescence microscopy: under the fluorescence microscope, the stained cells were observed with green and red light bands, respectively, to obtain the staining conditions of live and dead sperm.

[0102] Then, by combining the fluorescence microscopy and the optoelectronic tweezers system, the position of the best motility sperm cell can be obtained under the same field of view, and then the light pattern projection is switched back to the bright field to realize the separation and screening of the best motility sperm.

[0103] Since the existing device has combined the fluorescence microscope and the optoelectronic tweezers system, direct experiments can be performed. However, during the experiment, it was found that no matter how the aperture parameters, voltage amplitude and frequency were adjusted, the attraction and manipulation of the stained sperm cells could not be realized. Therefore, it is suspected that there are too many ionic substances in the staining solution reagent, which makes the conductivity of the stained sperm cell solution too high, and the attraction and manipulation of the sperm cells on the original amorphous silicon optoelectronic chip cannot be realized.

[0104] Then, the conductivity of the prepared solution was measured by a conductivity meter, and the results are shown in the following table:

[0105] Table 3 Conductivity of different solutions μS / cm

[0106]

[0107] It can be found that the conductivity of the Assay Buffer buffer solution is much higher than that of the glucose solution and pure water. Therefore, in the staining solution preparation step, the Assay Buffer buffer solution is replaced with pure water, and in the centrifugal washing process of the sperm cells, the 5% glucose solution is used instead. After replacing the Assay Buffer buffer solution, it is found that the attraction and manipulation of the sperm cells under the optoelectronic tweezers system can be realized, and the fluorescence phenomenon under the fluorescence field can also be observed.

[0108] S3, based on the image position information of the target sperm and the image position information of each target point, calculating the optimal moving path of each target sperm to the corresponding target point through a path matching algorithm, and then generating a dynamically adjustable light pattern template sequence according to the optimal moving path, and projecting the light pattern template sequence onto the optoelectronic tweezers chip to synchronously manipulate the target sperm to move to the target point position.

[0109] Optionally, in step S3, the path matching algorithm is the Hungarian algorithm, which calculates the optimal moving path of each target sperm to the corresponding target point, specifically including:

[0110] Constructing a cost matrix of the target sperm and the target point, wherein the cost is the Euclidean distance between the target sperm and the target point;

[0111] Solving the cost matrix by the Hungarian algorithm to obtain a target sperm-target point matching pair with the minimum total moving distance.

[0112] When the number of target points is less than the target sperm, the closest target sperm-target point pair is matched preferentially, and the remaining target sperm is marked as unmatching.

[0113] Optionally, in step S3, the sequence of light pattern templates is dynamically adjusted in at least one of the following parameters:

[0114] shape, size, line width, filling mode, brightness, and rotation angle.

[0115] Optionally, in step S3, the sequence of light pattern templates satisfies the following conditions:

[0116] each light pattern is circular, square, or cross-shaped, and the diameter matches the size of the target sperm;

[0117] The displacement step size Δx, Δy of the light pattern is dynamically adjusted according to the difference between the current position of the target sperm and the coordinates of the target point.

[0118] The current light pattern control logic used in laboratory equipment is to change the relative position of the preset light pattern by moving the micro-stage to achieve control of the microsphere cells suspended in the microfluidic channel. However, this control is not flexible and does not make good use of the programmable characteristics of the light pattern in the optoelectronic tweezers technology. Therefore, the light pattern control logic is improved in this application, which can flexibly change the shape, size, line width, filling mode, and brightness of the light pattern by keyboard input, and add an automatic path planning function for the subsequent sperm screening experiment.

[0119] The projection of the light pattern is actually to transmit the image of the black background white circle to the light machine for projection, similar to the operation of the projector projecting PPT, and the specific implementation logic is as follows: the CCD camera collects the cell image in the optoelectronic tweezers chip in real time, then transmits it to the PC through the USB connection, and then uses the opencv function library to process the image to generate a light pattern template of black background white circle, and then transmits it to the projector containing a digital micro-reflector device (DMD) for projection, and then through a series of lens combinations, it is projected into the optoelectronic tweezers chip to control the microsphere cells.

[0120] In order to facilitate convenient and intuitive light pattern control changes during the experiment, the light pattern adjustment control strategy is designed and improved in this application, which can allow the size, shape, position, filling mode, rotation angle, and brightness of the light pattern to be adjusted by keyboard input during the experiment. Through specific keys, such as number 1-4 keys for adjusting the shape and size of the light pattern, WASD keys for moving the light pattern, L and K keys for adjusting the background brightness, etc., so that the light pattern can be observed and modified in real time during the optoelectronic tweezers experiment, and the required operation is realized.

[0121] The specific keyboard control logic is as follows:

[0122] Position adjustment: through the letter keys WASD, the light pattern is moved up, down, left and right step by step.

[0123] Shape switching: through the number keys 1, the preset shapes are switched in a cycle: circle, square and cross.

[0124] Fill mode switching: through the number keys 2, the solid and hollow states of the light pattern are switched.

[0125] Size adjustment: through the number keys 3 and 4, the size of the light pattern is flexibly adjusted.

[0126] Speed adjustment: through the number keys 5 and 6, the speed of the light pattern movement is flexibly changed.

[0127] Line width adjustment: through the number keys 7 and 8, the line thickness of the hollow light pattern is freely changed.

[0128] Brightness control: through the number keys 9 and 0, the brightness of the light pattern itself is adjusted; and through the letter keys L and K, the brightness of the light pattern background is adjusted.

[0129] Rotation adjustment: through the letter keys P and O, the direction and angle of the light pattern rotation are flexibly controlled.

[0130] Number adjustment: through the letter keys U and I, the light pattern can be flexibly added and deleted.

[0131] Light pattern switching: through the letter key Y, the light pattern to be adjusted can be selected from multiple light patterns.

[0132] Of course, in addition to precise light pattern control through keyboard input, the program also adds the function of directly dragging the light pattern for movement through the mouse. The specific implementation logic is as follows: when the mouse arrow appears on the light pattern template created by Opencv, a small cross light pattern is automatically projected to represent the position of the mouse arrow. When the mouse arrow moves to the light pattern to be adjusted, click the left mouse button, at this time the coordinates of the center of the light pattern will coincide with the center of the mouse arrow; then move the mouse while holding the left button, the light pattern will also move synchronously, making the movement of the light pattern more flexible. During the mouse control process, the shape, size, fill mode, etc. of the light pattern can also be adjusted synchronously, which is convenient for more flexible manipulation of cell particles in subsequent experiments.

[0133] The light pattern movement controlled by keyboard and mouse is a manual form, mainly used for the manipulation of microspheres, sperm and other objects in pre-experiments to verify the feasibility of the experimental scheme. When the final goal is to realize an automatic sperm screening process, after detecting the sperm cells under the photoelectric tweezers system, the data of the identification box where the target sperm cell is located are obtained, then the corresponding light pattern of black background and white circle can be created by using the Opencv function for projection.

[0134] After that, the system needs to plan the trajectory of the light pattern movement and move the required sperm cells to the corresponding position for collection. The planned scheme at present is a step-by-step path planning scheme. That is, when the system detects and screens, it identifies the sperm with the best performance indicators, and then automatically plans the trajectory of the light circle movement, generating a specific path from the sperm position coordinates to the target position. After that, the system will stop detecting, and according to the path generated, it will gradually drive the projection light pattern to move until it reaches the specified position and stops.

[0135] Alternatively, after morphological screening, the single sperm cell with the most pixels is transported and separated, wherein at 0s, the algorithm obtains the position of the green rectangular frame with the most sperm cell pixels, and then plans the trajectory for single sperm transport and separation. Then at 1s, the system projects a light circle at the position of the green rectangular frame, and the light circle moves according to the planned trajectory, attracting the target sperm to move horizontally and then vertically to the specified position. Finally, at 16s, the projection light circle is removed, completing the screening and separation of the single sperm.

[0136] Alternatively, after speed calculation, the single sperm cell with the fastest swimming speed is separated and screened in the following process: at 0s, the algorithm obtains the speed information of all sperm cells in the current field of view, sorts them by size, obtains the red rectangular frame position and ID of the sperm cell with the fastest swimming speed, and plans the trajectory of the light circle movement according to the identification frame position and the target transport position. At 1s, the system projects a light circle at the position of the sperm cell with a swimming speed of 22.72 μm / s, and the light circle starts to attract the sperm to move according to the planned trajectory. At 23s, the sperm reaches the horizontal position of the target transport, and then moves along the vertical direction. At 64s, it can be seen that the sperm in the red small ring has reached the target position, and the system has removed the projection light circle, successfully completing the speed screening and separation experiment of the single sperm.

[0137] Optionally, switch back to the bright field after fluorescence observation, and project a light circle at the position of the sperm showing fluorescence under green light band to separate and screen the sperm with better vitality. At time 0, the sperm cell in the red small ring is the position of the sperm showing fluorescence under green light band. Then draw a light pattern near the red small ring sperm position and project it at time 3s, then manually control the light pattern to continuously attract the sperm cell to move, at time 15s, the sperm cell is attracted to the upper right corner of the original position, then moves along the horizontal direction, at time 32s, the sperm cell is attracted to the upper left corner of the original position, finally at time 33s, the light circle is removed, and the screening and separation experiment of the sperm with better vitality is successfully completed.

[0138] Figure 5 The method for dynamic manipulation of multiple sperm cells in the photoelectric tweezers system is shown. It includes parallel manipulation of multiple sperm cells, sperm enrichment and concentration, and group sperm control.

[0139] The method for parallel manipulation of multiple sperm cells is based on sperm detection algorithm to identify sperm in CCD image, then according to the set number of sperm, select multiple sperm for parallel manipulation. For each selected sperm, the system calculates the shortest path of the sperm to reach the target position by path planning algorithm, and guides it to the preset equidistant arrangement point one by one.

[0140] Sperm enrichment and concentration is to use photoelectric tweezers technology to efficiently capture and concentrate sperm cells through a gradually shrinking light circle. In this process, the size of the light circle gradually decreases over time, so that the randomly distributed sperm cells are effectively attracted and concentrated in the center area of the light circle.

[0141] The method for group sperm control uses photoelectric tweezers technology to collect and manipulate high-concentration sperm solution at a fixed point. During the whole movement of the light pattern, the sperm cells always remain within the control range of the light circle and do not leave the control area. It can effectively realize the accurate positioning and dynamic control of the sperm group.

[0142] In addition to moving a single sperm cell to a specific location, a multi-target parallel trajectory planning scheme is added on this basis. Here, the polystyrene microspheres and groove images taken under the photoelectric tweezers are used to verify the feasibility of the algorithm. The image on the top left shows the image after the algorithm detects and identifies the target object. As can be seen, all the grooves and microspheres are correctly framed. Next, the algorithm performs one-to-one matching between the microspheres and the grooves, and generates the corresponding path. Here, since the number of grooves is less than the number of microspheres, one microsphere is not planned a trajectory. Then, the image on the top right shows the black and white circle light pattern template projected by the system at a certain moment in the planned trajectory. This template is the object used for DMD digital micromirror projection in actual photoelectric tweezers experiments. The image on the bottom left shows the path that the microsphere has moved on the image, which is convenient for observation. The image on the bottom right is the projection of the simulated light pattern on the actual image, which finally verifies the feasibility of the algorithm scheme and lays the foundation for subsequent sperm parallel control experiments.

[0143] The groove is a micro-recess structure (matching the size of the microsphere) on the surface or inside of the photoelectric tweezers chip, which is usually processed by photolithography or etching process, and is used to receive or accommodate the microsphere (or sperm) to be controlled, that is, the end point position (i.e. "target position") to which the microsphere (or sperm) needs to be moved.

[0144] The second aspect of the present application discloses a sperm automatic screening system based on photoelectric tweezers, which comprises:

[0145] The first processing module is configured to obtain a plurality of biological characteristics of each sperm in the sperm sample on the photoelectric tweezers chip, image position information of each sperm, and image position information of each target point; the plurality of biological characteristics include sperm morphological characteristics, sperm swimming speed characteristics, and fluorescent vitality characteristics after sperm staining;

[0146] The second processing module is configured to select a plurality of target sperms according to the plurality of biological characteristics based on a preset screening standard;

[0147] The third processing module is configured to calculate an optimal moving path of each target sperm to the corresponding target point by a path matching algorithm based on the image position information of the target sperm and the image position information of each target point, and then generate a sequence of dynamically adjustable light pattern templates according to the optimal moving path, and project the sequence of light pattern templates onto the photoelectric tweezers chip to synchronously control the target sperm to move to the target point position.

[0148] The third aspect of the present application discloses an electronic device. The electronic device comprises a memory and a processor, and the memory stores a computer program. When the processor executes the computer program, the steps of the sperm automatic screening method based on photoelectric tweezers described in the first aspect of the present application are implemented.

[0149] Figure 6 FIG. 1 is a structural diagram of an electronic device according to an embodiment of the present invention. Figure 6 As shown, the electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the electronic device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, near field communication (NFC) or other technologies. The display screen of the electronic device can be a liquid crystal display or an electronic ink display screen, and the input device of the electronic device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the electronic device housing, or an external keyboard, touchpad or mouse.

[0150] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a structural diagram of the part related to the technical solution of the present disclosure, and does not constitute a limitation on the electronic device to which the solution of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0151] A fourth aspect of the present invention discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the automated sperm screening method based on photoelectric tweezers described in the first aspect of the present invention.

[0152] The above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may be modified or some or all of the technical features thereof may be replaced with equivalents, and such modifications or replacements do not deviate from the essence of the corresponding technical solutions within the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automated sperm screening based on photoelectric tweezers, characterized in that: The method comprises: S1, acquiring multiple biological characteristics of each sperm in the sperm sample on the photoelectric tweezers chip, image position information of each sperm, and image position information of each target point; the multiple biological characteristics include: sperm morphology characteristics, sperm swimming speed characteristics, and fluorescence activity characteristics after sperm staining; S2, based on preset screening criteria, multiple target sperm are selected according to multiple biological characteristics; S3, based on the image position information of the target sperm and the image position information of each target point, calculate the optimal movement path of each target sperm to the corresponding target point through a path matching algorithm, and then generate a dynamically adjustable light pattern template sequence according to the optimal movement path, and project the light pattern template sequence onto the photoelectric tweezers chip to synchronously control the target sperm to move to the target point position.

2. The method according to claim 1, characterized in that Step S1 specifically includes: S11, mixing and incubating the sperm sample with a live-dead staining reagent to obtain a stained sample, and observing the stained sample under a fluorescence microscope to obtain a fluorescence signal intensity as a fluorescence activity characteristic of each sperm after staining; wherein the live-dead staining reagent is prepared by mixing 2 mM calcein solution, 1.5 mM propidium iodide solution, and purified water in a preset volume ratio; S12, collecting an image of the stained sample on the photoelectric tweezers chip, identifying the image position information of each sperm and the image position information of each target point based on the target detection model, and outputting an identification frame of each sperm; S13, performing background threshold segmentation on the image area within the identification frame to calculate the number of pixels occupied by the sperm cell head as a sperm morphological feature; S14, assigning a sperm ID to each identification frame through the target tracking model, and associating the sperm IDs in consecutive frames to calculate the average movement speed of the sperm within a preset number of frames as the sperm swimming speed feature.

3. The method according to claim 2, characterized in that In step S11, before the sperm sample is mixed and incubated with the live-dead staining reagent, the following steps are also included: After the frozen sperm sample is thawed, it is centrifuged and washed with a low conductivity solution; the low conductivity solution is a 5% glucose solution.

4. The method according to claim 2, characterized in that In step S2, the preset screening criteria are to simultaneously meet the following conditions: The fluorescence signal intensity is greater than the preset intensity; The number of pixels occupied by the sperm cell head is greater than the preset number; The average sperm movement speed is greater than the preset speed.

5. The method according to claim 1, wherein In step S3, the path matching algorithm is the Hungarian algorithm, which calculates the optimal movement path of each target sperm to the corresponding target point, specifically including: Construct a cost matrix between the target sperm and the target point, where the cost is the Euclidean distance between the target sperm and the target point; Solving the cost matrix using the Hungarian algorithm to obtain a target sperm-target point matching pair that minimizes the total moving distance; When the number of target points is less than the number of target sperm, the closest target sperm-target point pair is matched first, and the remaining target sperm are marked as unmatched.

6. The method according to claim 1, characterized in that In step S3, at least one of the following parameters of the light pattern template sequence is dynamically adjusted: Shape, size, line width, fill type, brightness and rotation angle.

7. The method according to claim 6, characterized in that In step S3, the light pattern template sequence meets the following conditions: Each light pattern is shaped like a circle, square, or cross, and its diameter matches the target sperm size; The displacement steps Δx and Δy of the light pattern are dynamically adjusted according to the difference between the current position of the target sperm and the coordinates of the target point.

8. An automated sperm screening system based on photoelectric tweezers, characterized in that: The system comprises: The first processing module is configured to obtain multiple biological characteristics of each sperm in the sperm sample on the photoelectric tweezers chip, image position information of each sperm, and image position information of each target point; the multiple biological characteristics include: sperm morphology characteristics, sperm swimming speed characteristics, and fluorescence activity characteristics after sperm staining; a second processing module configured to select a plurality of target sperms according to a plurality of biological characteristics based on a preset screening criterion; The third processing module is configured to calculate the optimal movement path of each target sperm to the corresponding target point based on the image position information of the target sperm and the image position information of each target point through a path matching algorithm, and then generate a dynamically adjustable light pattern template sequence according to the optimal movement path, and project the light pattern template sequence onto the photoelectric tweezers chip to synchronously control the target sperm to move to the target point position.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the automated sperm screening method based on photoelectric tweezers according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the automated sperm screening method based on photoelectric tweezers according to any one of claims 1 to 7 are implemented.

Images