A method and related apparatus for DNA integrity assessment based on single sperm Raman spectroscopy

Abstract

This application provides a method and related equipment for DNA integrity assessment based on single sperm Raman spectroscopy, belonging to the field of sperm quality assessment technology in assisted reproductive technology. This method, based on a machine learning algorithm, directly assesses DNA integrity using the Raman spectra of single sperm, thereby achieving non-destructive screening of individual high-quality sperm in assisted reproductive technology. The process of constructing the algorithm training sample dataset includes a step-by-step fixation and analysis method for sample single sperm: first, Raman spectroscopy testing is performed, followed by fluorescence staining analysis, reducing the impact of sperm position changes during detection, analysis, or staining. Then, optical and fluorescence images of single sperm within preset shape units are matched to determine the validity of the sample single sperm. Finally, a training sample dataset is constructed based on the relevant data of valid single sperm, establishing a correlation between the Raman spectrum of the same single sperm and its sperm chromatin structure analysis results.

Description

Technical Field

[0001] This application relates to the field of sperm quality assessment technology, and in particular to a method and related equipment for assessing DNA integrity based on single sperm Raman spectroscopy. Background Technology

[0002] In related technologies, Raman spectroscopy has been introduced into the field of sperm quality assessment due to its non-destructive, label-free, and single-cell resolution characteristics, in order to overcome the limitations of detecting sperm DNA integrity defects. Raman spectroscopy uses laser excitation to excite the vibration of sample molecules, generating inelastic scattering. It collects characteristic scattered spectra with frequencies different from the incident light, forming a "fingerprint spectrum" that reflects the vibrational modes of molecular bonds, thereby achieving label-free analysis of sperm chemical components.

[0003] Current Raman spectroscopy studies mostly employ population spectral analysis of semen smears, which involves collecting the average spectrum of a large number of sperm to represent the overall quality of the semen sample. However, in assisted reproductive applications, it is only necessary to select high-quality individual sperm. If the sperm evaluation results of a sample are characterized by the average level of the population, it cannot reflect the DNA integrity of individual sperm and cannot meet the sperm selection requirements of assisted reproductive technology.

[0004] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0005] The main objective of this application is to propose a DNA integrity assessment method and related equipment based on single sperm Raman spectroscopy. This method can directly assess the DNA integrity of a single sperm based on the Raman spectrum obtained through training, thereby achieving DNA integrity assessment of a single sperm in a non-destructive manner at the single sperm scale, thus meeting the optimal selection requirements for single sperm in assisted reproductive technology.

[0006] To achieve the above objectives, one aspect of this application proposes a method for assessing DNA integrity based on single sperm Raman spectroscopy, the method comprising: Obtain the Raman spectrum of the single sperm sample to be tested; The Raman spectrum of the sperm to be tested is input into a trained DNA integrity assessment model for calculation, and the DNA integrity assessment result of the sperm to be tested is determined based on the calculation result. The training sample dataset used to train the DNA integrity assessment model is constructed in the following manner: An optical image of a detection chip with a preset shape unit is acquired; a sperm sample solution prepared based on a sample semen is added to the preset shape unit; several sample sperm cells are distributed within the preset shape unit; Raman spectra of single sperm samples within the preset shape unit are collected; the Raman spectra are obtained after preliminary fixation of single sperm samples in the sperm sample solution within the preset shape unit. The fluorescence image of the detection chip of the preset shape unit is acquired and the DNA fragmentation index of each sperm sample in the preset shape unit is calculated; the fluorescence image is acquired by performing in-situ secondary fixation and staining on each sperm sample in the preset shape unit. The optical and fluorescence images of the detection chip based on the preset shape unit are compared and matched, and the distribution of effective single sperm samples within the preset shape unit is determined based on the comparison and matching results. The training sample dataset is constructed based on the Raman spectra of single sperm samples within the preset shape unit and the DNA fragmentation index.

[0007] In some embodiments, the Raman spectrum of each single sperm sample within the preset shape unit is acquired in the following manner: The distribution of single sperm samples within the preset shape unit is determined based on the optical image of the detection chip of the preset shape unit; the distribution of single sperm samples within the preset shape unit includes the relative position information of each single sperm sample within the preset shape unit; After initial fixation of the sperm sample solution in the preset shape unit by removing liquid and moisturizing, the Raman spectrum of each sperm sample is collected based on the relative position information of each sperm sample in the preset shape unit.

[0008] In some embodiments, the DNA fragmentation index of each sperm sample within the preset shape unit is calculated in the following manner: The fluorescence image of each sperm sample within the preset shape unit is determined based on the fluorescence image of the detection chip of the preset shape unit. The green fluorescence intensity and red fluorescence intensity of each sperm sample were determined by analyzing the fluorescence image of each sperm sample. The DNA fragmentation index of each sperm sample is calculated based on a preset formula and the green and red fluorescence intensities of each sperm sample.

[0009] In some embodiments, the step of comparing and matching the optical image and fluorescence image of the detection chip of the preset shape unit, and determining the distribution of effective single sperm samples within the preset shape unit based on the comparison and matching results, includes: The distribution of sperm cells in the sample within the preset shape unit before staining is determined based on the optical image of the detection chip of the preset shape unit; the distribution of sperm cells in the preset shape unit includes the morphology and relative position of each sperm cell in the detection chip of the preset shape unit. The distribution of single sperm cells in the detection chip of the preset shape unit after staining is determined based on the fluorescence image of the detection chip of the preset shape unit. The distribution of single sperm samples within the detection chip of the preset shape unit before and after staining is compared and matched, and the effective distribution of single sperm samples within the detection chip of the preset shape unit is determined based on the comparison and matching results.

[0010] In some embodiments, before inputting the Raman spectra of the effective single sperm sample into the trained DNA integrity assessment model for calculation, the Raman spectra of the effective single sperm sample need to be preprocessed; the preprocessing is performed in the following manner: The Raman spectra of the effective single sperm samples were filtered based on a preset filtering algorithm to obtain filtered data. Based on preset parameters and the asymmetric least squares method, baseline correction is performed on the filtered data to complete the preprocessing of the Raman spectra of the effective sample single sperm.

[0011] In some embodiments, constructing the training sample dataset based on the Raman spectra of effective single sperm samples within the preset shape unit and the DNA fragmentation index includes: The DNA fragmentation index corresponding to each valid single sperm cell within the preset shape unit is statistically analyzed, and the DNA fragmentation index classification threshold is determined based on the statistical results. The DNA fragmentation index is classified according to the classification threshold and the DNA fragmentation index corresponding to each valid sperm sample. The classification label corresponding to each valid sperm sample is determined according to the classification results. The training sample dataset is constructed based on the Raman spectrum and classification label of each valid single sperm sample.

[0012] In some embodiments, the sperm DNA integrity assessment model is trained in the following manner: Obtain the training sample dataset and randomly divide the training sample dataset to obtain several training sample subsets; Select several subsets of sample data from several training sample subsets as the test set, and use the remaining subsets of sample data as the training set; The sperm DNA integrity assessment pre-model is trained based on the training set. The preset category weights of the sperm DNA integrity assessment pre-model are adjusted according to the training results to obtain the adjusted sperm DNA integrity assessment pre-model. The test set is input into the adjusted sperm DNA integrity assessment pre-model for verification to obtain the classification accuracy of the adjusted sperm DNA integrity assessment pre-model. The sperm DNA integrity assessment model is trained, adjusted, and validated by repeatedly selecting training and testing sets until the classification accuracy of the adjusted sperm DNA integrity assessment pre-model meets the preset requirements. The trained sperm DNA integrity assessment model is obtained based on the adjustment results.

[0013] To achieve the above objectives, another aspect of this application proposes a DNA integrity assessment system based on single sperm Raman spectroscopy, the system comprising: The prediction module is used to input the Raman spectrum of the sperm to be tested into a trained DNA integrity assessment model for calculation, and determine the DNA integrity assessment result of the sperm to be tested based on the calculation result. The training sample dataset used to train the sperm DNA integrity assessment model is constructed in the following manner: An optical image of a detection chip with a preset shape unit is acquired; a sperm sample solution prepared based on a sample semen is added to the preset shape unit; several sample sperm cells are distributed within the preset shape unit; Collect the Raman spectra of each single sperm sample within the preset shape unit; The fluorescence image of the detection chip of the preset shape unit is acquired and the DNA fragmentation index of each sperm sample in the preset shape unit is calculated; the fluorescence image is acquired by in-situ fixation and staining of the sperm sample solution in the preset shape unit. The optical and fluorescence images of the detection chip based on the preset shape unit are compared and matched, and the distribution of effective single sperm samples within the preset shape unit is determined based on the comparison and matching results. The training sample dataset is constructed based on the Raman spectra of single sperm samples within the preset shape unit and the DNA fragmentation index.

[0014] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0015] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0016] The embodiments of this application include at least the following beneficial effects: This application provides a method, system, electronic device, and storage medium for DNA integrity assessment based on single sperm Raman spectroscopy; on the one hand, this method can directly assess the single sperm based on the Raman spectrum corresponding to the single sperm through a trained model, and achieve DNA integrity assessment of single sperm at the single sperm scale in a non-destructive manner, thereby meeting the preferred requirements of single sperm in assisted reproductive technology. Secondly, in constructing the training sample dataset, the method of this invention places the sperm sample solution in a detection chip with a preset shape unit and performs preliminary fixation by removing liquid and moisturizing. This allows for the localization of the sperm based on the relative position information between the sperm and the preset shape unit. After acquiring the Raman spectrum of the sperm, an in-situ secondary fixation process is performed to reduce the impact of changes in sperm position caused by staining on the reliability of the training samples, thereby improving the quality of model training and effectively increasing the accuracy of the model in assessing DNA integrity. Finally, in constructing the training sample dataset, the method of this invention establishes a correlation between the Raman spectrum of the same sperm and its sperm chromatin structure analysis results through a detection process of "Raman spectroscopy test first, followed by fluorescence staining analysis" and the validity verification of the sperm. This achieves an accurate correlation between Raman spectroscopy and clinical gold standard detection results, thereby improving the reliability of the sperm quality assessment results. Attached Figure Description

[0017] Figure 1 This is a flowchart of a DNA integrity assessment method based on single sperm Raman spectroscopy provided in an embodiment of this application; Figure 2 This is a flowchart of a method for constructing a single sperm DNA integrity assessment model provided in an embodiment of this application; Figure 3 This is a schematic diagram of a grid pattern provided in an embodiment of this application; Figure 4 This is an example optical image of a square unit provided in an embodiment of this application; Figure 5 This is a schematic optical image of the square unit and the single sperm inside it provided in an embodiment of this application; Figure 6 This is a Raman spectroscopy test result of a sperm head provided in an embodiment of this application; Figure 7 This is a diagram showing the differences in Raman spectra of single sperm in different scattering media, provided in an embodiment of this application. Figure 8 This is a flowchart illustrating an example of a method for in-situ sperm fixation and sperm DFI tagging after single sperm Raman spectroscopy acquisition, provided in an embodiment of this application. Figure 9 This is an example diagram comparing the optical image of the same square unit during the Raman spectroscopy acquisition process with the fluorescence image acquired by the SCSA test in an embodiment of this application; Figure 10 This is a flowchart illustrating an example of a spectral preprocessing and LDA model training method provided in this application. Figure 11 This is an example diagram of the heatmap and AUC curve of the LDA model prediction results in the embodiments of this application; Figure 12 This is an example diagram of the feature importance results of the LDA model in the embodiments of this application; Figure 13 This is a structural block diagram of a DNA integrity assessment system based on single sperm Raman spectroscopy provided in an embodiment of this application; Figure 14 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0021] 1) Raman spectroscopy: an analytical technique that obtains information such as molecular structure and crystal orientation of a substance by analyzing the inelastic scattered light produced after the interaction of a laser with the substance.

[0022] 2) Sperm chromatin structure assay (SCSA): Acridine orange (AO) staining combined with flow cytometry is used to analyze the different colors of fluorescence emitted by AO when binding intact DNA and fragmented DNA (AO emits green fluorescence when binding intact DNA and red fluorescence when binding fragmented DNA). The DNA fragmentation index (DFI) is obtained based on the ratio of red and green fluorescence intensity, which is used to assess the DNA integrity of sperm.

[0023] 3) DNA fragmentation index (DFI): This refers to the percentage of sperm with DNA breaks in semen, as detected by techniques such as flow cytometry. It is a key indicator for measuring sperm DNA integrity and male fertility potential.

[0024] 4) Savitzky-Golay filtering algorithm: A convolution algorithm that effectively smooths noise while preserving the original form of the signal to the maximum extent by performing local polynomial least squares fitting on the data within a sliding window.

[0025] 5) Asymmetric Least Squares Method: This method introduces an asymmetric weighting function to impose different penalties on the positive and negative values ​​of the residuals during the fitting process, thereby effectively separating the baseline and sharp peaks in the signal.

[0026] 6) LDA model (Linear Discriminant Analysis): A supervised learning method for dimensionality reduction and classification. Its core idea is to find a projection direction for the data so that the projection of different classes of data in this direction can maximize the inter-class variance and minimize the intra-class variance, thereby achieving the best classification effect.

[0027] 7) Cross-validation: A statistical method that divides the dataset into training and validation sets and alternates the division multiple times to fully evaluate the model’s generalization ability and prevent overfitting.

[0028] In related technologies, Raman spectroscopy has been introduced into the field of sperm quality assessment due to its non-destructive, label-free, and single-cell resolution characteristics, in order to overcome the limitations of detecting sperm DNA integrity defects. Raman spectroscopy uses laser excitation to excite the vibration of sample molecules, generating inelastic scattering. It collects characteristic scattered spectra with frequencies different from the incident light, forming a "fingerprint spectrum" that reflects the vibrational modes of molecular bonds, thereby achieving label-free analysis of sperm chemical components.

[0029] Current work has used Raman spectroscopy to analyze the DNA integrity of sperm: (1) Raman spectroscopy was used to analyze the differences in Raman spectra between artificially damaged sperm and sperm in natural conditions, and it was found that the ratio of Raman characteristic peak intensities related to the DNA phosphate backbone was positively correlated with the overall degree of genetic material damage in sperm samples; (2) Raman microscopy combined with principal component analysis was used to study the ultraviolet damage model of sperm DNA, and it was observed that the main peak of the DNA backbone was oriented towards 1042 cm⁻¹. -1 (3) Focusing on the Raman spectra of sperm with abnormal heads and normal sperm, by comparing the intensity of Raman peaks related to DNA and protein, information on the packaging of DNA-protein complexes in sperm heads is provided. In addition, by comparing the relative intensity of two other Raman peaks related to DNA and protein, it is confirmed that there is a correlation between the shape of sperm cell head and Raman spectrum. It is also found that sperm with normal heads may still have abnormal DNA encapsulation, which further provides a theoretical basis for Raman spectroscopy to assess the integrity of sperm DNA.

[0030] Because Raman spectroscopy eliminates the need for staining, fixation, or cell lysis, and water exhibits extremely low Raman signals, it effectively avoids sperm damage caused by traditional detection methods. Secondly, Raman spectroscopy is typically combined with confocal microscopy, allowing for in-situ analysis of individual sperm cells and achieving single-cell resolution. Furthermore, a single Raman spectroscopy scan can capture structural changes in multiple biomolecules, including DNA, proteins, and lipids, enabling simultaneous detection of multiple components. In conclusion, Raman spectroscopy has become an ideal tool for assessing sperm DNA integrity and holds great potential in the selection of single sperm cells for assisted reproduction.

[0031] Despite the feasible technical principles, current Raman spectroscopy research faces two major obstacles to clinical translation. First, current work on Raman spectroscopy for sperm assessment typically focuses on the correlation between Raman spectra and conventional sperm parameters (motility, morphology), or on validating feasibility through in vitro induced damage models, lacking a direct link to clinical gold standard testing methods. Second, existing Raman spectroscopy studies often employ population spectral analysis of semen smears, collecting the average spectrum of a large number of sperm to represent the overall quality of the semen sample. However, average spectra cannot reflect signal deviations and abnormalities in individual sperm, cannot identify DNA-damaged sperm mixed in with the sperm population, and lack the technical design for locating and tracking individual sperm, thus hindering the realization of a Raman spectroscopy-clinical gold standard linkage.

[0032] In view of this, this application provides a method, system, electronic device, and storage medium for DNA integrity assessment based on single sperm Raman spectroscopy. On the one hand, this method can directly assess the DNA integrity of a single sperm based on the Raman spectrum corresponding to the single sperm through a trained model, thereby achieving DNA integrity assessment of a single sperm at the single sperm scale in a non-destructive manner, thus meeting the optimal selection requirements of single sperm in assisted reproductive technology. Secondly, in constructing the training sample dataset, the method of this invention places the sperm sample solution in a detection chip with a preset shape unit and performs preliminary fixation. Simultaneously, based on the relative positional information between the sperm and the preset shape unit, the sperm can be located. After acquiring the Raman spectra of the sperm, in-situ secondary fixation with agar gel reduces the impact of staining treatment on the reliability of the training samples, thereby improving the quality of model training and effectively increasing the accuracy of the model in assessing DNA integrity. Finally, in constructing the training sample dataset, the method of this invention establishes a correlation between the Raman spectrum of the same sperm and its sperm chromatin structure analysis results through a detection process of "Raman spectroscopy testing followed by fluorescence staining analysis" and the validity verification of the sperm. This achieves an accurate correlation between Raman spectroscopy and clinical gold standard detection results, thereby improving the reliability of the sperm quality assessment results.

[0033] The DNA integrity assessment method based on single sperm Raman spectroscopy provided in this application relates to the field of information technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the DNA integrity assessment method based on single sperm Raman spectroscopy, but is not limited to the above forms.

[0034] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0035] Figure 1 This is an optional flowchart of the DNA integrity assessment method based on single sperm Raman spectroscopy provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S100 to S200.

[0036] Step S100: Obtain the Raman spectrum of the single sperm to be tested.

[0037] Raman spectra of the single sperm samples to be tested were obtained to prepare data for subsequent DNA integrity assessment.

[0038] Step S200: Input the Raman spectrum of the sperm to be tested into the DNA integrity assessment model for calculation, and determine the DNA integrity assessment result of the sperm to be tested based on the calculation result.

[0039] In response, the trained model can directly evaluate the single sperm based on the Raman spectrum corresponding to the single sperm, and achieve DNA integrity assessment of the single sperm in a non-destructive manner at the single sperm scale, thereby meeting the single sperm selection requirements in assisted reproductive technology.

[0040] In this embodiment of the application, the training sample dataset used to train the DNA integrity assessment model is constructed in the following manner: Step S310: Obtain an optical image of the detection chip with a preset shape unit; a sperm sample solution prepared based on the sample semen is dropped into the preset shape unit of the detection chip with the preset shape unit; several sample sperm are distributed in the preset shape unit of the detection chip with the preset shape unit. The optical image of the detection within the preset shape unit is acquired to prepare for subsequent Raman spectroscopy acquisition; the location of the single sperm can be achieved based on the relative positional relationship between the single sperm and the shape boundary of the preset shape unit.

[0041] Step S320: Collect the Raman spectrum of each single sperm sample within the preset shape unit; the Raman spectrum is obtained after preliminary fixation of the single sperm sample in the sperm sample solution within the preset shape unit.

[0042] Raman spectra of single sperm samples are collected to prepare for the subsequent construction of a training sample dataset. The method in this application improves the stability of Raman spectroscopy acquisition by initially fixing single sperm samples in a sperm sample solution within a preset shape unit.

[0043] Step S330: Obtain the fluorescence image of the detection chip of the preset shape unit and calculate the DNA fragmentation index of each sperm sample in the preset shape unit; the fluorescence image is obtained by performing in-situ secondary fixation and staining on each sperm sample in the preset shape unit.

[0044] In-situ secondary fixation of the sperm samples, followed by staining, combined with the aforementioned initial fixation, can further reduce changes in sperm position before and after staining, thus minimizing their impact on the reliability of the evaluation results. Specifically, in-situ secondary fixation of the sperm samples can be performed using agar gel, and staining can be done using SCSA staining.

[0045] Step S340: Compare and match the optical image and fluorescence image of the detection chip based on the preset shape unit, and determine the distribution of effective single sperm samples within the preset shape unit based on the comparison and matching results.

[0046] By comparing and matching the optical and fluorescence images of the detection chip of the preset shape unit, the relative position information between the single sperm sample and the preset shape unit before and after staining can be analyzed, thereby verifying the validity of the single sperm sample within the shape unit.

[0047] Step S350: Construct a training sample dataset based on the Raman spectra of single sperm cells in the effective sample within the preset shape unit and the DNA fragmentation index.

[0048] A training sample dataset is constructed based on the Raman spectra and DNA fragmentation index of valid single sperm samples. By analyzing the DNA fragmentation index of the sample single sperm, the results of sperm chromatin structure analysis are determined, thereby establishing the correlation between the Raman spectra of the same single sperm and the results of sperm chromatin structure analysis. This achieves an accurate correlation between Raman spectra and clinical gold standard detection results, providing a high-quality sample dataset for model training.

[0049] In some embodiments, in step S320, the Raman spectrum of each single sperm sample within the preset shape unit is acquired in the following manner: Step S321: Determine the distribution of single sperm samples within the preset shape unit based on the optical image of the detection chip of the preset shape unit; the distribution of single sperm samples within the preset shape unit includes the relative position information of each single sperm sample within the preset shape unit.

[0050] The position of each sperm sample within the shape unit relative to the shape unit can be determined based on the optical image of the detection chip of the preset shape unit, which serves as a reference for subsequent Raman spectroscopy acquisition at the corresponding position of the sperm sample.

[0051] Step S322: After preliminarily fixing the sperm samples in the sperm sample solution within the preset shape unit by removing the liquid and keeping them moist, the Raman spectrum of each sperm sample is acquired based on the relative position information of each sperm sample within the preset shape unit.

[0052] After initial fixation of the single sperm samples in the sperm sample solution by removing the liquid and keeping them moist, and determining the relative position information of each single sperm sample, Raman spectra were collected for each target sperm.

[0053] In some embodiments, in step S330, the DNA fragmentation index of each sperm sample within the preset shape unit is calculated in the following manner: Step S331: Determine the fluorescence image of each single sperm sample within the preset shape unit based on the fluorescence image of the detection chip of the preset shape unit.

[0054] Based on the fluorescence image of the preset shape unit, the fluorescence image of the location of each sperm cell in the sample within the shape unit can be analyzed, which serves as the basis for subsequent analysis.

[0055] Step S332: Analyze the fluorescence image of each sperm sample to determine the green fluorescence intensity and red fluorescence intensity of each sperm sample.

[0056] The green fluorescence intensity and red fluorescence intensity of each sperm sample were determined as the data basis for judging whether the sperm was intact. Green fluorescence corresponds to intact DNA, while red fluorescence corresponds to fragmented DNA.

[0057] Step S333: Based on the preset formula and the green fluorescence intensity and red fluorescence intensity of each sperm sample, the DNA fragmentation index of each sperm sample is calculated.

[0058] We calculate the DNA fragmentation index (DFI) of each sperm sample using equation (1): (1) DFI stands for DNA fragmentation index of a single sperm sample. The intensity of red fluorescence in the detected single sperm cells of the sample. This indicates the intensity of green fluorescence in the detected single sperm cells of the sample.

[0059] In some embodiments, step S340 involves comparing and matching the optical image and fluorescence image of the detection chip of the preset shape unit, and determining the distribution of effective single sperm samples within the preset shape unit based on the comparison and matching results. This process includes, but is not limited to, steps S341 to S343: Step S341: Determine the distribution of single sperm samples within the preset shape unit based on the optical image of the detection chip of the preset shape unit; the distribution of single sperm samples within the preset shape unit includes the morphology and relative position of each single sperm sample within the preset shape unit.

[0060] By analyzing the optical image of the detection chip of the preset shape unit, the boundary of the preset shape unit and the position information of the single sperm sample within the preset shape unit can be determined. Furthermore, after determining the relative position information of each single sperm sample, the morphology of each single sperm sample can be determined based on the optical image of the preset shape unit, and the distribution of the single sperm sample within the preset shape unit can be determined by combining the results.

[0061] Step S342: Determine the distribution of single sperm cells in the preset shape unit after staining based on the fluorescence image of the detection chip of the preset shape unit.

[0062] Similar to the processing method of optical images of detection chips for preset shape units, the distribution of single sperm samples within preset shape units after staining can be determined based on the fluorescence images of detection chips for preset shape units.

[0063] Step S343: Compare and match the distribution of single sperm samples within the preset shape unit before and after staining treatment, and determine the effective distribution of single sperm samples within the preset shape unit based on the comparison and matching results.

[0064] By comparing whether there are sperm samples with the same morphology at the same position within the preset shape unit before and after staining, and whether the same sperm sample has deviated at that position, it can be determined whether the sperm sample is a valid sperm sample (i.e., a sperm sample that has not been damaged or changed).

[0065] In some embodiments, before inputting the Raman spectrum of the effective single sperm sample into the trained DNA integrity assessment model for calculation in step S200, the Raman spectrum of the effective single sperm sample needs to be preprocessed; the preprocessing is performed in the following manner: The Raman spectra of single sperm from valid samples are filtered using a preset filtering algorithm to obtain filtered data. Baseline correction is then performed on the filtered data using preset parameters and the asymmetric least squares method to complete the preprocessing of the Raman spectra of single sperm from valid samples.

[0066] The Savitzky-Golay filtering algorithm can be used to reduce noise in Raman spectra, and then the asymmetric least squares method can be used for baseline correction, thereby achieving preprocessing of the single sperm Raman spectra of the sample and improving the reliability of Raman spectral data.

[0067] In some embodiments, the DNA integrity assessment model is trained in the following manner: Step S410: Obtain the sample dataset and randomly divide the sample dataset to obtain several sample data subsets.

[0068] The model training method of this invention is based on the idea of ​​cross-validation. First, the sample dataset is randomly divided to obtain several sample data subsets (n folds).

[0069] Step S420: Select one subset of sample data from several subsets of sample data as the test set, and use the remaining subsets of sample data as the training set.

[0070] Furthermore, based on the idea of ​​cross-validation, one set is selected as the test set and the others are selected as the training set, serving as the data basis for training and validation.

[0071] Step S430: Train the DNA integrity assessment pre-model based on the training set, adjust the preset class weights of the DNA integrity assessment pre-model according to the training results, and obtain the adjusted DNA integrity assessment pre-model; input the test set into the adjusted DNA integrity assessment pre-model for verification, and obtain the classification accuracy of the adjusted DNA integrity assessment pre-model.

[0072] The evaluation model of this invention is essentially a linear discriminant analysis model. It determines whether a sperm is intact by inputting the Raman spectrum and DNA fragmentation index of a single sperm. The model is trained using a training set, and the class weights of the model are adjusted based on the training results. The classification accuracy of the adjusted model is then calculated using a validation set and used as an evaluation metric for the training structure.

[0073] Step S440: Repeatedly select the training set and test set to train, adjust and validate the DNA integrity assessment model until the classification accuracy of the adjusted DNA integrity assessment pre-model meets the preset requirements, and obtain the trained DNA integrity assessment model based on the adjustment results.

[0074] Based on the idea of ​​cross-validation, the training set and test set are extracted repeatedly to train the model until the model class weight corresponding to the highest classification accuracy is found, thus completing the training of the model and obtaining the trained DNA integrity assessment model.

[0075] In summary, the embodiments of this method include, but are not limited to, the following beneficial effects: (1) In the construction of the training sample dataset for the single sperm DNA integrity assessment model, the method of the present invention places the sperm sample solution in the detection chip of the preset shape unit and achieves preliminary fixation by dehydration and moisturization treatment. At the same time, the single sperm is located according to the relative position information between the single sperm and the boundary of the shape unit. After collecting the Raman spectrum of the single sperm, the in-situ secondary fixation treatment of agar gel is performed to reduce the impact of the change in sperm position caused by staining treatment on the reliability of the sample sperm assessment results.

[0076] (2) In the construction of the training sample dataset for the single sperm DNA integrity assessment model, the method of the present invention establishes the correlation between the Raman spectrum of the same single sperm and the analysis results of its sperm chromatin structure through the detection process of “Raman spectroscopy test first, followed by fluorescence staining analysis” and the validity verification of single sperm. This achieves accurate correlation between Raman spectroscopy and clinical gold standard detection results, thereby improving the credibility of sperm quality assessment sample results.

[0077] (3) The method of the present invention is based on machine learning algorithm to realize the classification and prediction of single sperm Raman spectra, and establishes a non-destructive and accurate single sperm DNA integrity assessment and optimization method. It achieves single sperm DNA integrity assessment in a non-destructive manner at the single sperm scale, thereby meeting the single sperm optimization requirements in assisted reproductive technology.

[0078] Figure 2 This is a flowchart illustrating the steps involved in constructing the single sperm DNA integrity assessment model provided in this application embodiment. The method includes: 1. Preparation of sperm sample solution Liquefaction treatment: Take a semen sample and liquefy it in a 37°C water bath for 30 minutes; Centrifugation and washing: (1) Take 1 mL of liquefied sperm, centrifuge at 200 g for 5 minutes, and discard the supernatant; (2) Add 1 mL of LTNE solution (Tris-NaCl-EDTA buffer) to resuspend the sperm, centrifuge and discard the supernatant, repeat twice; (3) Add 1 mL of deionized water to resuspend the sperm, centrifuge and discard the supernatant, repeat twice; Concentration adjustment: Resuspend sperm in deionized water and adjust sperm concentration using a cell counting chamber. .

[0079] 2. Fabrication of detection chips for preset shape units Substrate treatment: Take a 2.5cm×7.5cm single-sided polished stainless steel plate (5mm thick), and ultrasonically clean it with acetone, ethanol and deionized water for 15 minutes in sequence, and then blow it dry with nitrogen for later use.

[0080] Laser engraving: Using an ultraviolet laser engraving machine (laser wavelength 325nm, power 1.6W), a preset shape unit, such as a grid pattern, is engraved on the substrate surface at an engraving line speed of 100mm / s. (The grid pattern consists of a total of 7×7 square units, with each square unit having a side length of 400μm and a spacing of 200μm.)

[0081] Secondary cleaning: After engraving, repeat the acetone-ethanol-deionized water ultrasonic cleaning process, and then blow dry with nitrogen for later use.

[0082] In some embodiments, the fabrication process of the detection chip for the preset shape unit includes: like Figure 3 As shown, Figure 3 This is a schematic diagram of a grid pattern detection chip provided in an embodiment of this application. The overall size of the grid pattern is 4.8mm × 4.8mm, composed of 7 × 7 square units with a side length of 400μm and a spacing of 200μm. Each square unit has a positioning number in the grid pattern. Then, an ultraviolet laser engraving machine (laser wavelength 325nm, power 1.6W) is used to engrave the grid pattern on the surface of a single-sided polished stainless steel substrate at an engraving line speed of 100mm / s. Figure 4 As shown, Figure 4 This is an example optical image of a square unit detection chip provided in an embodiment of this application. Under a microscope, the laser markings are clearly visible, and the square unit is flat and clean. The background of the preset shape unit can be used to obtain the morphology and relative position of each sperm being tested, providing a reference for the subsequent determination of sperm Raman spectral labels.

[0083] 3. Single sperm Raman spectroscopy acquisition Sample loading: Take 20 μL of the sperm suspension prepared above and drop it onto the grid pattern area of ​​the preset shape unit detection chip, and remove the liquid and keep it moist by air drying at room temperature.

[0084] Spectroscopic detection: Spectroscopic testing was performed using a confocal Raman microscope (in Via Qontor, Renishaw), equipped with a 50x objective lens (working distance 0.21mm), a 633nm 17mW laser, a spot diameter of 1.2μm, and a diffraction grating with 2400 grooves / mm. During the spectral detection process, optical images of all sperm within a single square cell of the grating pattern were first acquired under low magnification, followed by optical images of each individual sperm within the square cell under high magnification. Raman spectra were then acquired for the target sperm (integration time 10 seconds per iteration, totaling 2 integrations).

[0085] In some embodiments, the process of acquiring single sperm Raman spectra may include: Figure 5 This is a schematic optical image of a square unit and its internal single sperm cells provided in an embodiment of this application; wherein, Figure 5 The middle image (a) is an optical image of a single square unit of sperm suspension after air drying under low magnification. Figure 5 Figure (b) shows an optical image of a single sperm cell within a unit after the sperm suspension has been air-dried under high magnification, clearly demonstrating the sperm morphology. By comparing the optical image of the square unit with the optical image of a single sperm cell within the unit, the relative position of each sperm cell within the square unit can be determined. Then, the Raman spectra of the single sperm cells are acquired and analyzed to obtain the biomolecules and vibrational modes corresponding to the Raman spectral vibrational peaks, such as... Figure 6 As shown, Figure 6 This is a Raman spectroscopy result image of a sperm head provided in an embodiment of this application, wherein 728 This represents the cyclic respiratory vibration peak of adenine in DNA, 785. The peaks represent the cyclic respiratory vibrations of thymine, guanine, and cytosine, as well as the peaks of the DNA phosphate backbone, 827. 1096 The Raman vibration peak at 1060 represents the symmetrical vibrational stretching of PO2- in the DNA backbone, while the peak at 1060 represents the symmetrical vibrational stretching of PO2- in the DNA backbone. The Raman vibration peak at 1253 is associated with the oxidatively damaged DNA phosphate backbone. This represents the adenine, thymine, and cytosine cycle respiration pattern, 1335 This represents the circular respiration pattern of adenine and guanine in DNA, 1373 This illustrates the circular respiration patterns of adenine, thymine, and guanine in DNA, 1483 This represents the circumspiratory pattern of adenine and guanine and purine bases, 1583 The Raman vibration peaks at this location represent adenine and guanine peaks. Furthermore, the Raman spectral peaks at the sperm nucleus also show peaks related to proteins and lipids, with 728 being a notable example. The peak at 1003 also indicates the vibration of CS bonds in proteins and CC bonds in lipids. The Raman peak at 1447 reflects the symmetrical cyclic respiration pattern of phenylalanine. The peak at 1660 represents the deformation of methyl, methylene, and ethyl groups associated with proteins, as well as the vibration of CH bonds in protein lipids. This primarily represents the C=C vibration in lipids and the amide I vibration in proteins. The above analysis confirms the advantages of Raman spectroscopy in simultaneous multi-component analysis for sperm DNA integrity assessment.

[0086] Furthermore, the following experimental examples validate the feasibility of applying Raman spectroscopy in the clinical assessment of sperm DNA integrity: In the process of establishing the Raman spectroscopy dataset of this invention, sperm were subjected to Raman spectroscopy testing after the dispersion medium was removed. However, in clinical assisted reproductive applications, clinicians perform sperm selection and manipulation in solution. Therefore, the differences in the average Raman spectra of single sperm in different dispersion media (dispersed in solution and air) were evaluated. The Raman spectroscopy acquisition conditions were the same as the large-scale acquisition conditions for single sperm Raman spectra. The results are as follows: Figure 7 As shown, Figure 7 This is a Raman spectral difference result diagram of single sperm in different scattering media provided in the embodiments of this application. It can be seen that: only at 1044 A significant difference in intensity was observed at this point. Comparison with multiple studies on the substances corresponding to the characteristic peaks of sperm Raman spectra did not reveal the presence of 1044. The differences in the substances corresponding to the characteristic peaks are presumably caused by the dispersion medium and signal processing. Therefore, it can be assumed that there are no differences in the peaks or shapes of sperm in the two states, which confirms the clinical feasibility of the present invention.

[0087] 4. Secondary fixation of sperm in situ In-situ secondary fixation ensures that the sperm's position remains unchanged after Raman spectroscopy testing, thus guaranteeing that the Raman spectrum and SCSA results originate from the same sperm. Without in-situ secondary fixation, the liquid reagent added to the SCSA will alter the sperm's position, making it impossible to determine whether the Raman spectroscopy data and SCSA results truly originate from the same sperm.

[0088] In some embodiments, in-situ sperm fixation includes the following steps: Preparation of PDMS confined rings: The PDMS prepolymer (liquid A:liquid B = 10:1) was mixed and degassed, cured at 70°C for 2 hours, and cut into ring-shaped columns with an inner diameter of 10 mm and a thickness of 10 mm.

[0089] Low-melting-point agarose solidification: Place the PDMS ring in the grid pattern area of ​​the detection chip, add 20 μL of 0.75% low-melting-point agarose solution to the hollow area of ​​the ring column, and solidify at 4°C for 10 minutes to lock the sperm position.

[0090] In some embodiments, the flowchart of sperm in-situ secondary fixation and sperm DFI tagging after single sperm Raman spectroscopy acquisition can be illustrated as follows: Figure 8 As shown.

[0091] 5. Sperm DFI tagging (SCSA staining) Lysis and staining: (1) Add 40 μL of lysis buffer (0.08 M) to the PDMS ring. 150mM 0.1% (pH 1.2), act for 30 seconds.

[0092] (2) Add 120 μL of staining solution (6 mg / L acridine orange, 0.1 M citric acid, 0.2 M... , Stain at pH 6.0 for 20 minutes in the dark.

[0093] Fluorescence imaging: A laser confocal microscope (40x objective lens) was used to acquire fluorescence images of the entire square cell. Excitation light 485nm (power 3mW) → Green fluorescence (intact DNA) Excitation light 565nm (power 1mW) → red fluorescence (fragmented DNA) DFI Calculation and Label Determination: (1) Use ImageJ to analyze fluorescence images and calculate the single sperm DFI according to the formula:

[0094] Among them, DFI is the DNA fragmentation index of single sperm. The intensity of red fluorescence indicating the detected single sperm cell. This indicates the intensity of green fluorescence in the detected single sperm.

[0095] (2) Location association: By comparing the optical and fluorescence images of sperm within the square cell during Raman spectroscopy acquisition, and matching the sperm morphology and position before and after staining, only sperm data in which neither morphology nor position changed were selected, and DFI was used as the label for sperm Raman spectra.

[0096] (3) Classification: The critical DFI is calculated by comparing the sperm DFI with the %DFI of the semen sample given clinically. Based on the sperm DFI, sperm are divided into two groups: DNA-intact and DNA-damaged.

[0097] In some embodiments, the comparison between the sperm optical image acquired in the same square cell and the sperm fluorescence image after fixation and staining can be as follows: Figure 9 As shown, by comparing and matching the morphology and position of sperm before and after staining, only sperm data in which neither the morphology nor the position changed were selected, and DFI was used as the label for sperm Raman spectra.

[0098] 6. Raman spectroscopy preprocessing and LDA model training and prediction (1) Raman spectroscopy preprocessing: Noise reduction: The Savitzky-Golay filtering algorithm is used for noise reduction (window length 9, polynomial order 2).

[0099] Baseline correction: Baseline correction was performed using asymmetric least squares (smoothing factor 10). 5 Asymmetry factor 0.1, maximum number of iterations 200, convergence threshold 10 -6 ) (2) LDA model training and prediction Dataset split: 70% training set, 30% test set.

[0100] Category weight adjustment: The prior probability of DNA-intact samples is set to 0.01, and the prior probability of DNA-damaged samples is set to 0.99.

[0101] The training process includes data preprocessing and model selection (LDA). Based on the classification results, the prior probabilities (parameters) of the two classes are adjusted through 5-fold cross-validation. Finally, the parameter with the highest DNA-intact class prediction accuracy obtained through 5-fold cross-validation is selected, and the accuracy and precision of different models are output.

[0102] In LDA, the prior probability is an initial estimate or basic assumption we make about the likelihood of a sample belonging to each category before any feature data of the sample has been observed. For binary classification tasks, the prior probabilities of the two categories are combined to 1. Setting a higher prior for DNA-damaged aims to make the model more cautious in predicting the data as DNA-intact, thereby improving the prediction accuracy of DNA-intact, i.e., sperm of the DNA intact category.

[0103] Five-fold cross-validation involves dividing the dataset into five mutually exclusive subsets (or "folds"), using one subset as the test set and the other four as the training set, and repeating this process five times. The average of the five evaluation results is then used as the performance metric for the model.

[0104] Biological interpretability verification: Calculate the absolute value of the weight of each feature in the discrimination direction in the LDA model and arrange them from largest to smallest. The top 100 features with the highest absolute values ​​are directly represented on the sperm Raman spectrum.

[0105] In some embodiments, Figure 10 This is a schematic diagram illustrating the process of Raman spectroscopy preprocessing, LDA model training, and prediction. Figure 11 A binary classification model was constructed for LDA, introducing a prior probability parameter, and the average classification heatmap and ROC curve obtained using 5-fold cross-validation were plotted. Figure 11Figure a in the figure is the average classification heatmap. Figure 11 Figure b in the figure shows the ROC curve. It achieved an overall prediction accuracy of 95.87%, a DNA-intact category prediction accuracy of 97.44%, and a mean AUC of 0.9545, confirming the good performance of this invention in assessing single sperm DNA integrity. Furthermore, to verify the biological interpretability of the LDA-based binary classification model, feature importance checks were performed on the model, such as... Figure 12 As shown, the absolute values ​​of the weights of each feature in the discrimination direction were calculated, and the top 100 features by absolute value were directly represented on the Raman spectrum of sperm. The results showed 785, 1096, 1036, 1335, and 1373. Raman characteristic peaks associated with DNA played a major role in classification, confirming the biological interpretability of the LDA binary classification model.

[0106] In summary, the embodiments of this method include, but are not limited to, the following beneficial effects: (1) Establishment of Raman Spectroscopy-Sperm DFI Correlation: This invention successfully established the correlation between the Raman spectrum of the same sperm and the sperm chromatin structure analysis results through the detection process of "Raman spectroscopy test first, followed by fluorescence staining analysis" and the introduction of patterned low background detection chip and low melting point agarose dual fixation technology during the detection process, thus realizing the accurate correlation between Raman spectroscopy and clinical gold standard detection results.

[0107] Specifically, to achieve a one-to-one mapping between Raman data of a single sperm cell and its corresponding staining results, and to obtain sperm Raman spectra with less interference, the method of this application adopts a scheme of first removing the liquid and moisturizing the fixation for Raman spectroscopy testing, and then fixing with agarose gel for fluorescence staining analysis. Compared with first encapsulating sperm in agarose gel and then performing Raman spectroscopy detection, this invention can directly obtain sperm Raman spectra without the need to subtract the Raman scattering spectrum of agarose gel, whose intensity is much higher than that of sperm, through an algorithm. This avoids the loss of original data due to data processing and achieves a one-to-one correspondence between sperm Raman spectral data and sperm chromatin structure analysis data.

[0108] (2) Machine learning algorithm to achieve Raman spectroscopy classification prediction: This invention uses the LDA model and introduces prior probability to achieve careful prediction of DNA-intact category, and establishes a non-destructive and accurate preferred method for assessing the integrity of single sperm DNA.

[0109] Please see Figure 13 This application also provides a DNA integrity assessment system based on single sperm Raman spectroscopy, which can implement the above-described method. The system includes: The data acquisition module is used to acquire the Raman spectra of the single sperm samples to be tested. The prediction module is used to input the Raman spectrum of the sperm to be tested into the DNA integrity assessment model for calculation, and determine the DNA integrity assessment result of the sperm to be tested based on the calculation results. The training sample dataset used to train the DNA integrity assessment model was constructed in the following way: Obtain an optical image of the detection chip with a preset shape unit; a sperm sample solution prepared based on the sample semen is added to the detection chip with the preset shape unit; several sample sperm cells are distributed within the preset shape unit; Collect the Raman spectra of each single sperm sample within a pre-defined shape unit; The fluorescence image of the detection chip of the preset shape unit is obtained and the DNA fragmentation index of each sperm sample in the preset shape unit is calculated; the fluorescence image is obtained by in-situ fixation and staining of the sperm sample solution in the preset shape unit. The optical and fluorescence images of the detection chip based on the preset shape unit are compared and matched, and the distribution of single sperm in the effective sample within the preset shape unit is determined based on the comparison and matching results. A training sample dataset is constructed based on the Raman spectra of single sperm samples within a predefined shape unit and the DNA fragmentation index.

[0110] It is understood that the content of the above method embodiments is applicable to the system embodiments. The specific functions implemented by the system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0111] Please see Figure 14 This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0112] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0113] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0114] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0115] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0116] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0117] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0118] This application provides a method, system, electronic device, and storage medium for DNA integrity assessment based on single sperm Raman spectroscopy. On the one hand, this method can directly assess the DNA integrity of a single sperm based on the Raman spectrum obtained through training, thereby achieving DNA integrity assessment of a single sperm at the single sperm scale in a non-destructive manner, thus meeting the optimal selection requirements of single sperm in assisted reproductive technology. Secondly, in constructing the training sample dataset, the method of this invention places the sperm sample solution in a detection chip with a preset shape unit for dehydration and moisturization treatment to achieve preliminary fixation of the single sperm in the solution. Simultaneously, the single sperm can be located based on the relative position information between the single sperm and the preset shape unit. After acquiring the Raman spectra of the single sperm, in-situ secondary fixation with agar gel is performed to reduce the impact of changes in sperm position caused by staining treatment on the reliability of the training samples, thereby improving the quality of model training and effectively increasing the accuracy of the model in assessing DNA integrity. Finally, in constructing the training sample dataset, the method of this invention establishes a correlation between the Raman spectrum of the same single sperm and its sperm chromatin structure analysis results through a detection process of "Raman spectroscopy testing followed by fluorescence staining analysis" and single sperm validity verification. This achieves accurate correlation between Raman spectroscopy and clinical gold standard detection results, thereby improving the reliability of the sample sperm quality assessment results.

[0119] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in this application are also applicable to similar technical problems.

[0120] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0121] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0122] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0123] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0124] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0125] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0126] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0127] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0128] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for assessing DNA integrity based on single sperm Raman spectroscopy, characterized in that, The method includes the following steps: Obtain the Raman spectrum of the single sperm sample to be tested; The Raman spectrum of the sperm to be tested is input into a trained DNA integrity assessment model for calculation, and the DNA integrity assessment result of the sperm to be tested is determined based on the calculation result. The training sample dataset used to train the DNA integrity assessment model is constructed in the following manner: An optical image of a detection chip with a preset shape unit is acquired; a sperm sample solution prepared based on a sample semen is added to the preset shape unit; several sample sperm cells are distributed within the preset shape unit; Raman spectra of single sperm samples within the preset shape unit are collected; the Raman spectra are obtained after preliminary fixation of single sperm samples in the sperm sample solution within the preset shape unit. The fluorescence image of the detection chip of the preset shape unit is acquired and the DNA fragmentation index of each sperm sample in the preset shape unit is calculated; the fluorescence image is acquired by performing in-situ secondary fixation and staining on each sperm sample in the preset shape unit. The optical and fluorescence images of the detection chip based on the preset shape unit are compared and matched, and the distribution of effective single sperm samples within the preset shape unit is determined based on the comparison and matching results. The training sample dataset is constructed based on the Raman spectra of single sperm samples within the preset shape unit and the DNA fragmentation index.

2. The method according to claim 1, characterized in that, The Raman spectra of each single sperm sample within the preset shape unit were acquired using the following method: The distribution of single sperm samples within the preset shape unit is determined based on the optical image of the detection chip of the preset shape unit; the distribution of single sperm samples within the preset shape unit includes the relative position information of each single sperm sample within the preset shape unit; After initial fixation of the sperm sample solution in the preset shape unit by removing liquid and moisturizing, the Raman spectrum of each sperm sample is collected based on the relative position information of each sperm sample in the preset shape unit.

3. The method according to claim 1, characterized in that, The DNA fragmentation index of each sperm sample within the preset shape unit is calculated in the following way: The fluorescence image of each sperm sample within the preset shape unit is determined based on the fluorescence image of the detection chip of the preset shape unit. The green fluorescence intensity and red fluorescence intensity of each sperm sample were determined by analyzing the fluorescence image of each sperm sample. The DNA fragmentation index of each sperm sample is calculated based on a preset formula and the green and red fluorescence intensities of each sperm sample.

4. The method according to claim 1, characterized in that, The step of comparing and matching the optical image and fluorescence image of the detection chip of the preset shape unit, and determining the distribution of effective single sperm samples within the preset shape unit based on the comparison and matching results, includes: The distribution of sperm cells in the sample within the preset shape unit before staining is determined based on the optical image of the detection chip of the preset shape unit; the distribution of sperm cells in the preset shape unit includes the morphology and relative position of each sperm cell in the detection chip of the preset shape unit. The distribution of single sperm cells in the detection chip of the preset shape unit after staining is determined based on the fluorescence image of the detection chip of the preset shape unit. The distribution of single sperm samples within the detection chip of the preset shape unit before and after staining is compared and matched, and the effective distribution of single sperm samples within the detection chip of the preset shape unit is determined based on the comparison and matching results.

5. The method according to claim 1, characterized in that, Before inputting the Raman spectra of the valid sperm samples into the trained DNA integrity assessment model for calculation, the Raman spectra of the valid sperm samples need to be preprocessed; the preprocessing is performed in the following manner: The Raman spectra of the effective single sperm samples were filtered based on a preset filtering algorithm to obtain filtered data. Based on preset parameters and the asymmetric least squares method, baseline correction is performed on the filtered data to complete the preprocessing of the Raman spectra of the effective sample single sperm.

6. The method according to claim 1, characterized in that, The construction of the training sample dataset based on the Raman spectra of effective single sperm samples within the preset shape unit and the DNA fragmentation index includes: The DNA fragmentation index corresponding to each valid single sperm cell within the preset shape unit is statistically analyzed, and the DNA fragmentation index classification threshold is determined based on the statistical results. The DNA fragmentation index is classified according to the classification threshold and the DNA fragmentation index corresponding to each valid sperm sample. The classification label corresponding to each valid sperm sample is determined according to the classification results. The training sample dataset is constructed based on the Raman spectrum and classification label of each valid single sperm sample.

7. The method according to claim 1, characterized in that, The sperm DNA integrity assessment model was trained in the following way: Obtain the training sample dataset and randomly divide the training sample dataset to obtain several training sample subsets; Select several subsets of sample data from several training sample subsets as the test set, and use the remaining subsets of sample data as the training set; The sperm DNA integrity assessment pre-model is trained based on the training set. The preset category weights of the sperm DNA integrity assessment pre-model are adjusted according to the training results to obtain the adjusted sperm DNA integrity assessment pre-model. The test set is input into the adjusted sperm DNA integrity assessment pre-model for verification to obtain the classification accuracy of the adjusted sperm DNA integrity assessment pre-model. The sperm DNA integrity assessment model is trained, adjusted, and validated by repeatedly selecting training and testing sets until the classification accuracy of the adjusted sperm DNA integrity assessment pre-model meets the preset requirements. The trained sperm DNA integrity assessment model is obtained based on the adjustment results.

8. A DNA integrity assessment system based on single sperm Raman spectroscopy, characterized in that, The system includes: The data acquisition module is used to acquire the Raman spectra of the single sperm samples to be tested. The prediction module is used to input the Raman spectrum of the sperm to be tested into a trained DNA integrity assessment model for calculation, and determine the DNA integrity assessment result of the sperm to be tested based on the calculation result. The training sample dataset used to train the sperm DNA integrity assessment model is constructed in the following manner: An optical image of a detection chip with a preset shape unit is acquired; a sperm sample solution prepared based on a sample semen is added to the preset shape unit; several sample sperm cells are distributed within the preset shape unit; Collect the Raman spectra of each single sperm sample within the preset shape unit; The fluorescence image of the detection chip of the preset shape unit is acquired and the DNA fragmentation index of each sperm sample in the preset shape unit is calculated; the fluorescence image is acquired by in-situ fixation and staining of the sperm sample solution in the preset shape unit. The optical and fluorescence images of the detection chip based on the preset shape unit are compared and matched, and the distribution of effective single sperm samples within the preset shape unit is determined based on the comparison and matching results. The training sample dataset is constructed based on the Raman spectra of single sperm samples within the preset shape unit and the DNA fragmentation index.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images