Multi-dimensional evaluation method and system for spermatogenic function

Abstract

The invention provides a spermatogenic function multi-dimensional evaluation method and system, and the method comprises the steps: obtaining abnormal parameters from spermatogenic problem positioning, determining the deviation degree through comparison with a preset normal range, and obtaining a male evaluation score; if the type is female, extracting an ovarian reserve mark from the preliminary classification result, processing reserve parameters by adopting a support vector machine model, judging a reserve level, and obtaining reserve problem positioning; obtaining a level index from reserve problem positioning, determining a reserve deviation by calculating a difference with a preset reference, and obtaining a female evaluation score; integrating the deviation data from the male assessment score or the female assessment score, generating a comprehensive feedback report by adopting a fusion algorithm, and determining an overall fertility level; and grade information is extracted from the comprehensive feedback report, a clinical result is generated through a rapid output module, and a final evaluation conclusion is obtained.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method and system for multidimensional evaluation of spermatogenic function. Background Technology

[0002] In research on human health and fertility, fertility assessment is a crucial element in ensuring family well-being and population health, possessing undeniable value. This field directly relates to the accurate assessment of an individual's fertility potential, profoundly impacting solutions to infertility and improvements in quality of life. However, despite continuous technological advancements, numerous challenges remain, necessitating innovative breakthroughs to address complex practical needs.

[0003] Existing fertility assessment methods often fail to take into account the physiological differences and individualized needs of different sexes, generally suffering from insufficient testing coverage and low efficiency. Many programs are designed to ignore the fundamental differences in male and female reproductive mechanisms, resulting in assessment results that do not accurately reflect individual circumstances. Furthermore, the testing process is time-consuming, leading to a poor patient experience, and there is also a certain risk of misdiagnosis. These problems limit the effectiveness of existing methods in practical applications, making it difficult to meet the dual expectations of modern medicine for precision and efficiency.

[0004] A deeper technical challenge lies in designing appropriate testing protocols for gender differences and achieving a balance between speed and accuracy. The different physiological mechanisms resulting from gender differences require testing methods to capture the unique characteristics of male spermatogenesis and female ovarian reserve. This difference often makes it difficult for universal protocols to delve into the root causes of specific problems. Furthermore, this difference leads to a conflict between testing time and accuracy, especially in clinical scenarios requiring rapid results, where traditional lengthy testing procedures cannot meet the demands. For example, in male fertility assessment, relying solely on surface indicators is insufficient to pinpoint the exact problem; patients may require multiple tests to determine the cause, consuming significant time and effort.

[0005] Therefore, designing an assessment system that can accurately identify problematic areas while significantly shortening testing time, while respecting gender physiological differences, has become a critical issue that urgently needs to be addressed. Solving this problem not only involves technological breakthroughs but also directly impacts the patient's actual experience during the medical process and the reliability of the final results. Summary of the Invention

[0006] This invention provides a multi-dimensional assessment method for spermatogenesis, mainly including: Obtain patient physiological data and gender identifiers, extract spermatogenic function indicators and ovarian reserve markers from the data, filter out differential features through preset thresholds, and obtain preliminary classification results; Based on the preliminary classification results, the gender type is determined. If the type is male, the support vector machine model is used to analyze the spermatogenic function parameters, identify abnormal links, and locate the spermatogenic problem. Abnormal parameters are obtained from the identification of spermatogenesis problems, and the degree of deviation is determined by comparing them with the preset normal range to obtain a male assessment score; If the type is female, then ovarian reserve markers are extracted from the preliminary classification results, and a support vector machine model is used to process the reserve parameters, determine the reserve level, and obtain the location of the reserve problem. Level indicators are obtained from the positioning of reserve issues, and reserve deviations are determined by calculating the differences from preset benchmarks to obtain women's assessment scores; By integrating bias data from male or female assessment scores, a fusion algorithm is used to generate a comprehensive feedback report to determine the overall fertility level. The grade information is extracted from the comprehensive feedback report, and the clinical results are generated through the rapid output module to obtain the final evaluation conclusion.

[0007] This invention provides a multi-dimensional assessment system for spermatogenesis function, mainly comprising: The data acquisition and feature extraction module is used to acquire patient physiological data and gender identification, extract spermatogenic function indicators and ovarian reserve markers from the data, filter out differential features through preset thresholds, and obtain preliminary classification results. The preliminary classification and gender determination module is used to determine the gender type based on the preliminary classification results. If the type is male, the support vector machine model is used to analyze the spermatogenic function parameters, identify abnormal links, and locate the spermatogenic problem. The male spermatogenesis analysis module is used to obtain abnormal parameters from the spermatogenesis problem location, determine the degree of deviation by comparing with the preset normal range, and obtain the male assessment score; The male assessment score generation module is used to extract ovarian reserve markers from the preliminary classification results if the type is female, process the reserve parameters using a support vector machine model, determine the reserve level, and obtain the reserve problem location. The female ovarian reserve analysis module is used to obtain level indicators from the positioning of reserve problems, determine the reserve deviation by calculating the difference from the preset benchmark, and obtain the female assessment score. The female assessment score generation module is used to integrate bias data from male or female assessment scores, and use a fusion algorithm to generate a comprehensive feedback report to determine the overall fertility level. The comprehensive report generation and output module is used to extract grade information from the comprehensive feedback report, generate clinical results through the rapid output module, and obtain the final evaluation conclusion. The technical solution provided by this invention embodiment can include the following beneficial effects: This invention discloses a method for addressing the challenges of gender-differentiated analysis and comprehensive feedback in fertility assessment. It proposes a logically coherent solution: by classifying and filtering patient physiological data by gender, it accurately identifies male spermatogenic dysfunction or female ovarian reserve issues, and integrates the biased data to generate a comprehensive fertility assessment report. First, the data is preliminarily classified based on preset thresholds to determine gender type. For males, a support vector machine (SVM) model is used to analyze spermatogenic parameters, locate abnormalities, calculate the degree of deviation, and derive an assessment score. For females, ovarian reserve parameters are processed to determine reserve levels and identify deviations, generating an assessment score. Finally, a fusion algorithm integrates the data from both sexes to output a comprehensive feedback report and clinical conclusions. This invention achieves accuracy and comprehensiveness in fertility assessment through differentiated modeling and comprehensive analysis. Its core lies in the synergistic application of SVM and the fusion algorithm, effectively improving assessment efficiency and result reliability, and providing a scientific basis for clinical decision-making. Attached Figure Description

[0008] Figure 1 This is a flowchart of a multi-dimensional evaluation method for spermatogenesis function according to the present invention. Detailed Implementation

[0009] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0010] like Figure 1 This embodiment discloses a multi-dimensional assessment method for spermatogenesis function, which may specifically include: S101. Obtain patient physiological data and gender identifier, extract spermatogenic function indicators and ovarian reserve markers from the data, filter out differential features through preset thresholds, and obtain preliminary classification results.

[0011] In this embodiment, as the first step in an auxiliary diagnostic or health assessment system, the system first needs to establish a data foundation. Physiological data can be obtained through various means, such as direct retrieval from a hospital's electronic medical record system, data from a physical examination report manually entered by the user through a terminal device, or data collected in real time by connected medical testing equipment. The physiological data includes the patient's basic physical parameters and specific reproductive system test data. Simultaneously, gender identification, as a key index distinguishing subsequent data processing logic, is usually determined when the patient's file is created or data is initially entered. This identification is used to indicate which feature extraction model the system should subsequently call.

[0012] In its implementation, the system adaptively activates different data parsing channels based on gender identification when extracting spermatogenic function indicators and ovarian reserve markers from the data. If the gender identifier indicates male, the system focuses on parsing spermatogenic function indicators from physiological data. These indicators include, but are not limited to, sperm concentration, total sperm count, percentage of progressively motile sperm, sperm morphology analysis results, and reproductive hormone levels (such as testosterone and follicle-stimulating hormone). If the gender identifier indicates female, the system extracts ovarian reserve markers from physiological data. These markers primarily cover key parameters such as anti-Müllerian hormone (AMH) levels, basal follicle-stimulating hormone (FSH) levels, antral follicle count (AFC), and estradiol (E2) levels. This gender-identified targeted extraction mechanism effectively filters out irrelevant data, improving the efficiency and accuracy of data processing.

[0013] Furthermore, after extracting the aforementioned key indicators, this embodiment of the application filters out differential features using preset thresholds. These preset thresholds are reference ranges set based on large-scale clinical medical sample data or internationally accepted medical diagnostic standards (such as the semen analysis standards published by the World Health Organization or the ovarian reserve assessment guidelines published by the Chinese Society for Reproductive Medicine). The system compares the actual values ​​of the extracted spermatogenic function indicators or ovarian reserve markers with the corresponding preset thresholds item by item. When the value of a certain indicator exceeds or falls below the preset normal range, that indicator is marked as a differential feature. For example, if a man's sperm concentration is lower than the preset oligospermia diagnostic threshold, or a woman's AMH value is lower than the preset ovarian reserve decline threshold, these abnormal data points will be retained and weighted as differential features.

[0014] Ultimately, the system obtains preliminary classification results based on the selected differential features. This process is essentially an initial screening of the patient's reproductive health status. If all indicators are within the normal threshold range, the preliminary classification result is "normal physiological function"; if differential features exist, the system categorizes the patient into different risk categories based on the type and degree of deviation of the differential features, such as "risk of spermatogenesis dysfunction," "risk of diminished ovarian reserve," or "risk of endocrine disorders." This preliminary classification result will serve as input for subsequent in-depth analysis, treatment plan recommendations, or further examination suggestions. Through the embodiments of this application, automated cleaning and structuring of massive amounts of physiological data are achieved, enabling rapid identification of the patient's main health problems and providing doctors with objective auxiliary diagnostic references, thereby significantly improving the intelligence level of the diagnosis and treatment process.

[0015] S102. Determine the gender type based on the preliminary classification results. If the type is male, use a support vector machine model to analyze the spermatogenesis function parameters, identify abnormal links, and locate the spermatogenesis problem.

[0016] In this embodiment, the system first analyzes the preliminary classification results obtained in step S101 to determine the patient's gender type. When the gender type is identified as male, the system automatically activates the deep analysis module for the male reproductive system. This module uses a support vector machine model as a classification and regression tool, aiming to perform multi-dimensional analysis of complex spermatogenic parameters, thereby accurately identifying the specific factors leading to abnormal spermatogenic function and achieving precise localization of spermatogenic problems.

[0017] In the first possible implementation of step S102, to ensure that the data input to the model has sufficient representativeness and clinical value, the data acquisition process follows specific timing rules, specifically including: S1021. Semen samples are obtained from male patients according to preset collection rules. Samples are collected at a testing frequency of once a day and once every three days to obtain an initial sample dataset.

[0018] Specifically, semen assessment for male patients is not based on a single sampling, but rather employs a strategy combining high-frequency and low-frequency sampling. The system guides patients or healthcare professionals to collect samples according to pre-defined collection rules. One set of samples is collected continuously, once a day, to capture short-term fluctuations in semen quality. The other set of samples is collected every three days to reflect the regular cyclical changes in semen production. These two sets of samples, collected at different frequencies, together constitute the initial sample dataset, providing rich data support for subsequent differential analysis.

[0019] S1022. For the initial sample dataset, a unified sample processing procedure is adopted to perform standardized testing on the samples and obtain physiological state-related data for each sample.

[0020] To eliminate the interference of external environmental factors on the test results, this application implements a unified processing procedure for all collected samples. This includes, but is not limited to, sample liquefaction, standardization of counting chamber specifications, and standardization of microscopic imaging parameters. Through this standardized testing, the system can accurately extract physiological state-related data from each sample, such as sperm motility, viability, and morphological parameters, ensuring the consistency and comparability of data sources.

[0021] S1023. Based on physiological state-related data, the support vector machine algorithm is used to classify the sample data once a day and once every three days, and to determine the feature differences between the two sets of data at different detection frequencies.

[0022] In this step, the Support Vector Machine (SVM) algorithm is used to construct the classification hyperplane. The system labels data collected daily as class 1 and data collected every three days as class 2, and inputs the extracted physiological state-related data into the SVM model for training and classification. By analyzing the support vectors and classification boundaries in the model, the system can quantitatively determine the feature differences between the two sets of data at different detection frequencies. This difference reflects the specific impact of sampling frequency on the evaluation results.

[0023] S1024. If the feature difference exceeds the preset threshold, further time interval analysis is performed on the two sets of data to obtain the time correlation distribution of the difference formation, and a personalized evaluation strategy is generated accordingly.

[0024] When the calculated characteristic difference exceeds the system's preset threshold, it indicates that the different sampling frequencies significantly affect the assessment of the patient's spermatogenic function. At this point, the system will initiate in-depth time interval analysis to explore the pattern of difference changes over time, thereby obtaining the time-related distribution of the difference formation. Combined with the pre-set detection cycle, the system will perform horizontal and vertical data comparisons on the assessment results to determine the stability of semen assessment at the two different frequencies.

[0025] Ultimately, based on the aforementioned stability performance, the system will automatically generate an adjustment plan for the analysis method tailored to this patient. This may include adjusting subsequent sampling time points, increasing the sampling weight for specific time periods, or modifying the parameters of the evaluation algorithm, thereby obtaining a final personalized evaluation strategy. Using this strategy, subsequent sample collection plans and semen evaluation procedures can be specifically optimized, and the most suitable long-term testing frequency configuration for this patient can be determined, ensuring monitoring accuracy while avoiding unnecessary excessive consumption of medical resources.

[0026] S103. Obtain abnormal parameters from the spermatogenesis problem localization, determine the degree of deviation by comparing with the preset normal range, and obtain the male assessment score.

[0027] After determining the specific location of the spermatogenesis problem in step S102, this embodiment of the application further quantifies and evaluates these located abnormalities. The system first extracts specific abnormal parameters from the results of spermatogenesis problem localization. These abnormal parameters may involve multiple dimensions such as sperm concentration, motility, and morphology, depending on the analysis results of the support vector machine model in the preceding steps. For example, if the problem localization points to insufficient sperm motility, the extracted abnormal parameters may include the percentage of progressively motile sperm (PR) and the percentage of non-progressively motile sperm (NP).

[0028] After acquiring abnormal parameters, the system compares the actual values ​​of these parameters with preset normal ranges. These preset normal ranges are typically based on the lower reference limits in the latest edition of the World Health Organization's (WHO) *Laboratory Manual for the Examination and Processing of Human Semen* or statistical results from large clinical datasets. Through comparison, the system calculates the degree of deviation of each abnormal parameter from the lower limit of the normal range. The degree of deviation can be expressed as a percentage, a multiple of the standard deviation, or an absolute difference, aiming to objectively reflect the severity of the patient's indicators deviating from normal levels.

[0029] Finally, the system calculates a male assessment score based on the calculated degree of deviation and the weight of each parameter in the spermatogenesis function evaluation system. This score is a quantitative numerical indicator that directly reflects the current spermatogenesis function status of the male patient. The lower the score, the more severe the spermatogenesis function impairment, requiring more aggressive medical intervention; the higher the score, the closer the spermatogenesis function is to a relatively normal level. This assessment score will serve as an important reference for subsequently developing personalized treatment plans or assisted reproductive strategies.

[0030] S104. If the type is female, then extract ovarian reserve markers from the preliminary classification results, use a support vector machine model to process the reserve parameters, determine the reserve level, and obtain the reserve problem location.

[0031] In this embodiment, when the preliminary classification result obtained in step S101 indicates that the sex type is female, the system switches to an analysis path specifically designed for women's reproductive health. First, relevant ovarian reserve markers are extracted from the preliminary classification result. These markers typically include key indicators such as anti-Müllerian hormone levels, basal follicle-stimulating hormone concentration, and antral follicle count, serving as the basic data for subsequent model input. After extracting the ovarian reserve markers, the system uses a support vector machine (SVM) model to process these reserve parameters. This model constructs a classification boundary in a high-dimensional feature space to classify and regress the multidimensional data of the reserve parameters, thereby determining the patient's ovarian reserve level. Specifically, the SVM algorithm optimizes the decision function based on a preset training dataset to identify whether the reserve parameters deviate from the normal physiological range.

[0032] Finally, based on the output of the support vector machine model, the system determines the level of ovarian reserve and thus identifies the problem. For example, if the model detects that the overall reserve parameters are low, it identifies the problem as diminished ovarian reserve. This identification of the reserve problem provides accurate input for subsequent calculations of female assessment scores, ensuring the relevance and accuracy of the entire assessment process.

[0033] S105. Obtain level indicators from the positioning of reserve issues, determine the reserve deviation by calculating the difference from the preset benchmark, and obtain the female assessment score.

[0034] In this embodiment, based on the location of the ovarian reserve problem determined in step S104, the system further performs a refined quantitative assessment of the ovarian reserve function of female patients. First, the system extracts specific level indicators from the analysis results of the reserve problem location. These level indicators are key biochemical and imaging parameters reflecting ovarian function, specifically including but not limited to serum anti-Müllerian hormone concentration, basal follicle-stimulating hormone levels, and the total antral follicle count of both ovaries. These indicators are directly related to the amount of primordial follicles stored in the ovaries and are core data support for assessing female fertility potential.

[0035] After obtaining the level indicators, the system calls a preset benchmark for comparative analysis. It is worth noting that a woman's ovarian reserve function has a significant age-related correlation, naturally declining with age. Therefore, the preset benchmark in this embodiment is not a fixed, static value, but a dynamic reference model based on age stratification. This model is typically built on a large-scale database of healthy women's fertility, containing the median values ​​and normal fluctuation ranges of various indicators corresponding to different age groups (e.g., every two years as a stratification interval). The system automatically matches the corresponding benchmark interval based on the patient's actual age and calculates the numerical distance between the level indicator and the median value of that benchmark interval, thereby determining the reserve deviation. This reserve deviation objectively reflects the difference between the patient's current ovarian biological age and actual physiological age, such as whether there is premature ovarian failure or a significantly lower reserve function compared to peers.

[0036] Ultimately, the system weights and aggregates the reserve deviations of various indicators to obtain a female assessment score. During the calculation process, considering the differences in accuracy and stability of different indicators in reflecting ovarian reserve function, the system assigns different weight coefficients based on the sensitivity and specificity of each indicator in clinical diagnosis. For example, anti-Müllerian hormone (AMH) is usually given a higher weight because it is not affected by the menstrual cycle and can reflect the decline in reserve earlier, while basal follicle-stimulating hormone (FSH) may be given a relatively lower weight. The calculated female assessment score is a standardized value that not only intuitively displays the patient's current fertility potential level but also provides scientific and objective data support for doctors to develop personalized ovulation induction protocols, predict the success rate of assisted reproductive technologies, and assess the probability of natural conception.

[0037] S106. Integrate the deviation data from the male or female assessment scores, use a fusion algorithm to generate a comprehensive feedback report, and determine the overall fertility level.

[0038] In this embodiment, after completing the independent assessment process for men or women, the system enters the final comprehensive analysis stage. First, the system extracts core quantitative results from the male assessment score obtained in step S103 or the female assessment score obtained in step S105. Simultaneously, the system retrospectively integrates all deviation data generated during the assessment process, including the degree of abnormal deviation in male spermatogenic function parameters and the difference between female ovarian reserve indicators and benchmark values. This deviation data is not only a component of the score but also a key clue revealing specific pathological mechanisms.

[0039] After obtaining the scores and deviation data, the system employs a fusion algorithm for in-depth processing. This fusion algorithm is a multimodal data processing logic designed to organically combine single-dimensional scores with multi-dimensional deviation features. The algorithm first normalizes data from different sources to eliminate the influence of dimensions. Then, using a pre-defined weighted model, it uses the evaluation score as the primary factor and the deviation data as a correction factor to generate a detailed comprehensive feedback report. This report not only includes the final score but also details the specific physiological indicators that led to the high or low scores and the degree to which they deviated from the normal range, providing an evidence-based explanation.

[0040] Ultimately, the system determines the patient's overall fertility level based on a weighted average of the comprehensive feedback report. This level is typically divided into several tiers, such as "Excellent Fertility," "Good Fertility," "Decreased Fertility," and "Low Fertility." Each tier corresponds to specific clinical recommendations, such as guidance on natural conception, lifestyle interventions, drug treatment, or recommendations for assisted reproductive technologies. Through this tiered mechanism, the system provides doctors and patients with an intuitive and actionable basis for decision-making, thereby achieving intelligent assistance throughout the entire process from data collection to clinical decision-making.

[0041] S107. Extract grade information from the comprehensive feedback report, generate clinical results through the rapid output module, and obtain the final evaluation conclusion.

[0042] In this embodiment, after the comprehensive feedback report is generated by the fusion algorithm, the system enters the final information output and conclusion solidification stage. To transform the complex algorithmic analysis results into standardized diagnostic language easily understood by clinicians and patients, the system activates a rapid output module to perform in-depth analysis of the report content. First, the system accurately extracts the core hierarchical information from the structured data stream of the comprehensive feedback report. This hierarchical information is the final product of the comprehensive assessment based on male spermatogenesis or female ovarian reserve function in the preceding steps. It highly summarizes the patient's current reproductive system health status and is usually presented in the form of standardized medical grading codes or quantitative scores.

[0043] After successfully extracting the grading information, the rapid output module immediately performs data mapping and formatting operations to generate clinical results. This module integrates a professional clinical terminology database and diagnostic templates, automatically matching corresponding clinical descriptive text based on the extracted grading information. For example, if the grading information indicates decreased fertility, the module will automatically retrieve relevant pathological explanations, potential risk factor alerts, and standard treatment recommendation templates. Simultaneously, the module will embed key abnormal parameters and deviation data identified in previous steps into the results in the form of charts or highlighted text, thus forming a logically rigorous and well-supported clinical outcome document.

[0044] Ultimately, based on the above processing steps, the system arrives at a final evaluation conclusion. This conclusion is not merely a simple diagnostic label, but a complete electronic report with legal validity and medical reference value. The report details the patient's current fertility status, key issues, and corresponding personalized recommendations. The system can present this conclusion to the attending physician in real time via a display terminal, assisting in final diagnosis and treatment plan formulation; alternatively, it can synchronize the conclusion to the hospital's electronic medical record system or the patient's personal health record via an encrypted channel, achieving closed-loop management of diagnostic and treatment information. This application embodiment significantly shortens the time cycle from data analysis to clinical decision-making, improving diagnostic efficiency and standardization.

[0045] If the technical solution of this application involves the collection, processing, or application of personal information, the relevant products have, before implementing any personal information processing activities, fully and clearly informed individuals of the processing rules in accordance with the "Personal Information Protection Law of the People's Republic of China" and other current laws and regulations, and obtained their voluntary and explicit consent. If sensitive personal information is involved, the product has obtained the individual's separate consent before processing, and such consent is given in an explicit manner. For example, prominent signs are set up in the area where information collection devices such as cameras are located, clearly indicating "Entering is considered as consent to the collection of personal information"; or through pop-ups, checkboxes, user-initiated uploads, etc., under the premise of clearly listing the processor's identity, processing purpose, processing method, and information type, the user actively completes the authorization operation. The above mechanisms ensure that all personal information processing activities are based on legal authorization and fully comply with national compliance requirements regarding personal information protection.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-dimensional assessment method for spermatogenic function, characterized in that, The method includes: The process involves acquiring patient physiological data and gender identifiers, extracting spermatogenic function indicators and ovarian reserve markers from the data, filtering for discrepancies using preset thresholds to obtain preliminary classification results, determining gender type based on the preliminary classification results, and using a support vector machine (SVM) model to analyze spermatogenic function parameters, identify abnormalities, and pinpoint the spermatogenic problem. Abnormal parameters are then extracted from the spermatogenic problem location, and the degree of deviation is determined by comparing them to preset normal ranges to obtain a male assessment score. If the type is female, ovarian reserve markers are extracted from the preliminary classification results, and a SVM model is used to process reserve parameters, determine the reserve level, and pinpoint the reserve problem. Level indicators are then extracted from the reserve problem location, and the reserve deviation is determined by calculating the difference from a preset benchmark to obtain a female assessment score. Deviation data is integrated from either the male or female assessment scores, and a fusion algorithm is used to generate a comprehensive feedback report to determine the overall fertility level. Level information is extracted from the comprehensive feedback report, and clinical results are generated through a rapid output module to obtain the final assessment conclusion.

2. The method for multidimensional assessment of spermatogenic function according to claim 1, characterized in that, The process involves acquiring patient physiological data and gender identifiers, extracting spermatogenic function indicators and ovarian reserve markers from the data, filtering out differential features through preset thresholds, and obtaining preliminary classification results.

3. The method for multidimensional assessment of spermatogenic function according to claim 1, characterized in that, The sex type is determined based on the preliminary classification results. If the type is male, a support vector machine model is used to analyze the spermatogenesis function parameters, identify abnormal links, and locate the spermatogenesis problem.

4. The method and system for multidimensional assessment of spermatogenic function according to claim 3, characterized in that, Also includes: Semen analysis of male patients was performed once daily and once every three days, specifically including: Semen samples were obtained from male patients according to pre-defined collection rules, with samples collected daily and every three days to obtain initial sample datasets. A standardized sample processing procedure was used to standardize the samples and obtain physiological state-related data for each sample. Based on this data, a support vector machine algorithm was used to classify the daily and every-three-day sample data, identifying the characteristic differences between the two sets of data at different detection frequencies. If the characteristic differences exceeded a pre-defined threshold, further time interval analysis was performed on the two sets of data to obtain the temporal correlation distribution of the differences. Using the temporal correlation distribution and the set detection cycle, the evaluation results were compared to determine the stability of semen assessment at the two frequencies. Based on the stability performance, an adjustment plan for the analysis method at different detection frequencies was generated, resulting in a final personalized evaluation strategy. This final personalized evaluation strategy was used to optimize subsequent sample collection and semen assessment, determining the frequency configuration for long-term testing.

5. The method for multidimensional assessment of spermatogenic function according to claim 1, characterized in that, Abnormal parameters are obtained from the spermatogenesis problem localization, and the degree of deviation is determined by comparing them with a preset normal range to obtain a male assessment score.

6. The method for multidimensional assessment of spermatogenic function according to claim 1, characterized in that, If the type is female, then ovarian reserve markers are extracted from the preliminary classification results, and a support vector machine model is used to process the reserve parameters, determine the reserve level, and obtain the reserve problem location.

7. The method for multidimensional assessment of spermatogenic function according to claim 1, characterized in that, The process involves obtaining level indicators from the positioning of reserve issues, determining reserve deviations by calculating the differences from preset benchmarks, and obtaining female assessment scores.

8. The method for multidimensional assessment of spermatogenic function according to claim 1, characterized in that, The process involves integrating deviation data from male or female assessment scores, using a fusion algorithm to generate a comprehensive feedback report, and determining the overall fertility level.

9. The method for multidimensional assessment of spermatogenic function according to claim 1, characterized in that, The system extracts grade information from the comprehensive feedback report, generates clinical results through the rapid output module, and obtains the final evaluation conclusion.

10. A multi-dimensional assessment system for spermatogenesis function, characterized in that, The system includes: a data acquisition and feature extraction module, used to acquire patient physiological data and gender identifiers, extract spermatogenic function indicators and ovarian reserve markers from the data, filter differential features through preset thresholds, and obtain preliminary classification results; a preliminary classification and gender determination module, used to determine the gender type based on the preliminary classification results, and if the type is male, then a support vector machine model is used to analyze spermatogenic function parameters, identify abnormal links, and obtain the location of spermatogenic problems; a male spermatogenic analysis module, used to obtain abnormal parameters from the spermatogenic problem location, determine the degree of deviation by comparing with preset normal ranges, and obtain a male assessment score; and a male assessment score generation module, used to generate a score if the type is male. If the type is female, ovarian reserve markers are extracted from the preliminary classification results, and a support vector machine model is used to process reserve parameters, determine the reserve level, and identify the reserve problem. The female ovarian reserve analysis module is used to obtain level indicators from the reserve problem identification, determine the reserve deviation by calculating the difference with the preset benchmark, and obtain the female assessment score. The female assessment score generation module is used to integrate deviation data from male or female assessment scores, use a fusion algorithm to generate a comprehensive feedback report, and determine the overall fertility level. The comprehensive report generation and output module is used to extract level information from the comprehensive feedback report, generate clinical results through the rapid output module, and obtain the final assessment conclusion.

Images