Stem cell culture information management method and system

Abstract

The invention relates to the technical field of biological information, in particular to a stem cell culture information management method and system. The method comprises the following steps: collecting culture process basic data in a stem cell culture process, wherein the culture process basic data comprises cell counting data, culture condition parameter data and pollution condition data; performing growth kinetic analysis on the basic data in the culture process to obtain cell growth characteristic data; establishing a cell growth prediction model based on the cell growth characteristic data, and generating cell growth early warning data; and monitoring the culture process in real time according to the cell growth early warning data, triggering an early warning signal when abnormal fluctuation is detected, and generating a culture abnormality processing scheme. The growth prediction model is established based on the cell growth characteristic data, the culture dynamic state can be monitored in real time, an early warning signal is provided, the detection and response speed of abnormal fluctuation is remarkably improved, and culture failure is avoided.

Description

Technical Field

[0001] The present invention relates to the field of biological information technology, and in particular to a stem cell culture information management method and system. Background Art

[0002] Stem cells are a type of cell with the ability to self-replicate and multidirectional differentiation potential. They play an important role in the development and tissue repair of organisms. During the culture process, stem cells need to be separated and passaged regularly to maintain the vitality and stability of the cell population. Cell separation usually uses enzymatic methods, by adding enzymes such as trypsin to disperse the cells, and then obtain cell precipitates through operations such as centrifugation for cell passage. There are three main methods for stem cell culture: traditional feeder layer culture, feeder-free culture, and direct stem cell culture.

[0003] However, current methods for managing stem cell culture information often have the following problems: Traditional methods usually discover problems only after stem cell culture fails, lack real-time early warning and dynamic control methods, and abnormal fluctuations cannot be identified and handled in a timely manner, resulting in unstable success rates for stem cell culture, affecting research progress and clinical applications. Optimization of culture conditions is often achieved through repeated trial and error, which is inefficient and lacks specificity, especially in large-scale culture or industrial production. This method leads to waste of resources and rising costs. Summary of the invention

[0004] Based on this, it is necessary for the present invention to provide a stem cell culture information management method and system to solve at least one of the above technical problems.

[0005] To achieve the above object, a stem cell culture information management method comprises the following steps:

[0006] Step S1: collecting basic data of the culture process during stem cell culture, wherein the basic data of the culture process includes cell counting data, culture condition parameter data and contamination status data;

[0007] Step S2: Perform growth dynamics analysis on the basic data of the culture process to obtain cell growth characteristic data; establish a cell growth prediction model based on the cell growth characteristic data to generate cell growth warning data; monitor the culture process in real time according to the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan;

[0008] Step S3: dynamically regulating the culture condition parameters according to the culture abnormality processing scheme, and recording the regulation effect in real time to generate parameter regulation response data; correlating the parameter regulation response data with the cell growth characteristic data to establish a culture condition optimization model;

[0009] Step S4: Based on the culture condition optimization model, the basic data of the culture process is mined and analyzed to generate the optimal culture parameter combination; a standardized culture program is formulated according to the optimal culture parameter combination, and a culture quality assessment is performed to generate a culture quality assessment report.

[0010] The present invention ensures the comprehensiveness and accuracy of the data by comprehensively collecting cell counting data, culture condition parameter data and pollution status data during the culture process, and provides a solid foundation for subsequent analysis. The multi-source data integration method improves the coverage of information, so that every key factor in the culture process is monitored, especially through automated equipment such as automatic cell counters and microscope image acquisition systems, reducing the error of manual operation and improving the real-time and accuracy of data. This high-quality data acquisition helps to understand the relationship between the growth dynamics of cells and the culture environment in a deeper way. By analyzing the growth dynamics of the basic data of the culture process, the core characteristic data of cell growth can be extracted, which provides a scientific basis for further establishing a cell growth prediction model. The establishment of the prediction model enables the future growth trend of cells and possible abnormal conditions to be predicted in advance, so that preventive measures can be taken before problems occur. The real-time monitoring and early warning functions significantly improve the control of the culture process, reduce the risk of cell quality decline or failure due to abnormal fluctuations, and the generated culture abnormality handling scheme provides specific guidance measures for the operator, enhancing the emergency response capability of the system. By dynamically regulating the culture condition parameters, the system can optimize the culture environment in real time according to the actual situation, thereby ensuring that the growth conditions of the cells are always in the optimal state. The real-time recording and analysis of the regulation effect not only improves the ability to evaluate the effectiveness of the regulation measures, but also provides reference experience for similar problems in the future. Through correlation analysis, the regulation response data is associated with the cell growth characteristic data, and the culture condition optimization model is further improved. This closed-loop dynamic regulation and optimization mechanism significantly improves the stability and success rate of the culture, while accumulating valuable data resources, which is conducive to the continuous improvement of the system. Through data mining analysis, the optimal combination of culture parameters is generated, which provides a scientific basis for the formulation of standardized culture programs. This process ensures the optimal configuration of culture conditions, which can greatly improve the culture efficiency and the quality of stem cells. The formulation of standardized programs makes the culture process more repeatable and reliable, and is convenient for large-scale promotion and application. At the same time, the generation of culture quality assessment reports further verifies the optimization effect, helping researchers or operators to more intuitively understand the success rate and quality control level of the culture, thereby improving the standardization and efficiency of the overall operation process.

[0011] The present invention also provides a stem cell culture information management system for executing the above-mentioned stem cell culture information management method, wherein the stem cell culture information management system comprises:

[0012] A data acquisition module is used to collect basic data of the culture process during stem cell culture, wherein the basic data of the culture process includes cell counting data, culture condition parameter data and contamination status data;

[0013] The growth dynamic monitoring module is used to perform growth dynamics analysis on the basic data of the culture process to obtain cell growth characteristic data; establish a cell growth prediction model based on the cell growth characteristic data to generate cell growth warning data; monitor the culture process in real time based on the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan;

[0014] The parameter control optimization module is used to dynamically control the culture condition parameters according to the culture abnormality processing plan, and to record the control effect in real time to generate parameter control response data; the parameter control response data is correlated with the cell growth characteristic data to establish a culture condition optimization model;

[0015] The training program generation module is used to mine and analyze the basic data of the training process based on the training condition optimization model to generate the optimal training parameter combination; formulate a standardized training program based on the optimal training parameter combination, conduct training quality assessment, and generate a training quality assessment report.

[0016] The data acquisition module in the present invention can systematically collect key data in the stem cell culture process and provide basic data support for subsequent analysis. By acquiring cell counting data, culture condition data and pollution status data in real time, the key changes in the culture process can be fully monitored to ensure the transparency and accuracy of the experimental process. In particular, the collection of pollution status data can timely discover potential pollution risks and ensure the effectiveness and repeatability of the experiment. This helps to optimize the culture scheme in the future and reduce the culture failure caused by environmental changes or pollution. The growth dynamic monitoring module uses the basic data collected in real time to analyze the cell growth dynamics, which can identify the trend and law of cell growth, and then establish an accurate growth prediction model. This enables dynamic monitoring to be achieved during the culture process and predict possible abnormal situations in advance. When the growth data fluctuates abnormally, the system can quickly trigger an early warning signal, generate a culture abnormality processing plan, and adjust the culture plan in time to prevent problems in the culture process from causing irreversible damage to cell growth. Through real-time monitoring and early warning of the cell growth process, the culture efficiency can be greatly improved, and unnecessary waste of resources and cell quality degradation can be avoided. The function of the parameter control optimization module is to adjust the culture condition parameters in real time according to the abnormalities found during the culture process. Through dynamic regulation, such as changing the culture environment factors such as temperature, humidity, and gas composition, it is possible to respond to different stages of cell growth in a timely manner. This module can not only record the regulation effect in real time, but also combine the parameter regulation response data with the cell growth characteristic data through correlation analysis, so as to establish an accurate culture condition optimization model. This allows the optimization of culture conditions to be data-driven, which not only improves the culture efficiency but also ensures the stability of cell quality. Ultimately, through regulation and optimization, the culture success rate and the predictability of cell growth can be significantly improved. The culture scheme generation module can comprehensively analyze the basic data of the culture process and combine it with the culture condition optimization model to accurately mine the laws behind the data and generate the optimal culture parameter combination. Through the formulation of standardized culture schemes, various conditions for cell culture can be unified, the consistency and repeatability of experiments can be improved, and the cell growth conditions during the culture process can be fully optimized. In addition, culture quality assessment and generation of culture quality assessment reports can help evaluate the overall effect of culture, such as cell consistency, morphological uniformity, and genetic stability, thereby providing reliable quality assurance for subsequent research and application. This process helps to maintain the stability of cell quality in multiple experiments, enhance the controllability of the culture process, and reduce the risk of experimental failure. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0018] Figure 1 This is a schematic diagram of the steps of the stem cell culture information management method of the present invention;

[0019] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0020] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0021] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0022] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0023] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0024] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for managing stem cell culture information, the method comprising the following steps:

[0025] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of a process flow of a stem cell culture information management method according to the present invention. In this example, the stem cell culture information management method comprises the following steps:

[0026] Step S1: collecting basic data of the culture process during stem cell culture, wherein the basic data of the culture process includes cell counting data, culture condition parameter data and contamination status data;

[0027] The embodiment of the present invention uses automated equipment to collect basic data of the culture process. The number of live cells and the number of dead cells are recorded every 6 hours by an automatic cell counter, and cell morphology parameters are obtained in combination with microscope imaging to form cell counting data; the culture environment parameters are collected in real time by the built-in sensor of the incubator, including temperature (maintained at 37°C ± 0.1°C), humidity (relative humidity above 95%), CO2 concentration (5% ± 0.1%) and culture medium pH value (controlled at 7.2-7.4); the image data of the contamination status that may occur during the culture process are recorded by the microscope image acquisition system, including bacterial contamination and fungal contamination characteristics, and the culture operation record data (such as the liquid change time is set to 48 hours, the liquid change volume is 80% of the culture medium volume, and the subculture ratio is controlled to 1:3) are uniformly cleaned and standardized to generate basic data of the culture process including cell counting data, culture condition parameter data and contamination status data.

[0028] Step S2: Perform growth dynamics analysis on the basic data of the culture process to obtain cell growth characteristic data; establish a cell growth prediction model based on the cell growth characteristic data to generate cell growth warning data; monitor the culture process in real time according to the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan;

[0029] The embodiment of the present invention analyzes the collected basic data of the culture process, first extracts the key data points in the logarithmic growth period of the cell activity data for 72 consecutive hours, uses the nonlinear least squares method to fit the exponential growth model to obtain the growth fitting curve model, and calculates the model fitting goodness data and the cell doubling time; by scoring and analyzing the dynamic stability of the doubling time, the cell growth characteristic data is generated; based on these characteristic data, a multidimensional cell growth prediction model including growth rate prediction, survival rate prediction and growth trend prediction is constructed, and after cross-validation and parameter calibration to optimize the model performance, the confidence interval of the prediction result data is calculated; when the prediction result shows that the growth trend deviates from the preset range, by generating an early warning signal, combined with a historical abnormal event library, a culture abnormality processing plan is automatically formulated.

[0030] Step S3: dynamically regulating the culture condition parameters according to the culture abnormality processing scheme, and recording the regulation effect in real time to generate parameter regulation response data; correlating the parameter regulation response data with the cell growth characteristic data to establish a culture condition optimization model;

[0031] The embodiment of the present invention adjusts the culture condition parameters in real time according to the culture abnormality processing scheme. For example, when an abnormal fluctuation in CO2 concentration is detected (deviating from the set value by more than ±0.2%), the gas supply ratio is adjusted immediately; for abnormal culture medium pH value (such as lower than 7.2), NaHCO3 solution is added to slowly adjust it to the target range, and cell growth response changes are monitored at the same time; the parameter changes of each control operation are correlated with the corresponding cell growth characteristic data, and a quantitative mapping relationship between culture conditions and cell growth indicators is constructed. The optimization model is subjected to sensitivity analysis and iterative adjustment in multiple experimental verifications, thereby gradually improving the culture condition optimization model.

[0032] Step S4: Based on the culture condition optimization model, the basic data of the culture process is mined and analyzed to generate the optimal culture parameter combination; a standardized culture program is formulated according to the optimal culture parameter combination, and a culture quality assessment is performed to generate a culture quality assessment report.

[0033] The embodiment of the present invention uses a culture condition optimization model to perform multi-dimensional data mining on the basic data of the historical culture process, screen out the key parameters that have the greatest impact on cell growth (for example, the influence weights of temperature, CO2 concentration, and passage ratio are 0.35, 0.30, and 0.25, respectively), and generate the optimal culture parameter combination based on this; formulate a standardized culture plan according to the optimal parameters, including setting the cell seeding density to 1×10 4 Pieces / cm 2 The culture medium formula is DMEM+10% FBS, and the fluid change cycle is 48 hours; multiple repeated experiments are carried out under the standardized scheme, and by recording key culture quality indicators such as cell growth consistency index (target value ≥ 0.90) and morphological uniformity score (target value ≥ 85 points), a culture quality assessment report is finally generated to provide a scientific reference for subsequent large-scale applications.

[0034] The present invention ensures the comprehensiveness and accuracy of the data by comprehensively collecting cell counting data, culture condition parameter data and pollution status data during the culture process, and provides a solid foundation for subsequent analysis. The multi-source data integration method improves the coverage of information, so that every key factor in the culture process is monitored, especially through automated equipment such as automatic cell counters and microscope image acquisition systems, reducing the error of manual operation and improving the real-time and accuracy of data. This high-quality data acquisition helps to understand the relationship between the growth dynamics of cells and the culture environment in a deeper way. By analyzing the growth dynamics of the basic data of the culture process, the core characteristic data of cell growth can be extracted, which provides a scientific basis for further establishing a cell growth prediction model. The establishment of the prediction model enables the future growth trend of cells and possible abnormal conditions to be predicted in advance, so that preventive measures can be taken before problems occur. The real-time monitoring and early warning functions significantly improve the control of the culture process, reduce the risk of cell quality decline or failure due to abnormal fluctuations, and the generated culture abnormality handling scheme provides specific guidance measures for the operator, enhancing the emergency response capability of the system. By dynamically regulating the culture condition parameters, the system can optimize the culture environment in real time according to the actual situation, thereby ensuring that the growth conditions of the cells are always in the optimal state. The real-time recording and analysis of the regulation effect not only improves the ability to evaluate the effectiveness of the regulation measures, but also provides reference experience for similar problems in the future. Through correlation analysis, the regulation response data is associated with the cell growth characteristic data, and the culture condition optimization model is further improved. This closed-loop dynamic regulation and optimization mechanism significantly improves the stability and success rate of the culture, while accumulating valuable data resources, which is conducive to the continuous improvement of the system. Through data mining analysis, the optimal combination of culture parameters is generated, which provides a scientific basis for the formulation of standardized culture programs. This process ensures the optimal configuration of culture conditions, which can greatly improve the culture efficiency and the quality of stem cells. The formulation of standardized programs makes the culture process more repeatable and reliable, and is convenient for large-scale promotion and application. At the same time, the generation of culture quality assessment reports further verifies the optimization effect, helping researchers or operators to more intuitively understand the success rate and quality control level of the culture, thereby improving the standardization and efficiency of the overall operation process.

[0035] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0036] Step S11: recording the number of live cells, the number of dead cells and cell morphology parameters every 6 hours using an automatic cell counter to obtain cell counting data;

[0037] The embodiment of the present invention uses an automatic cell counter (such as Countess II FL automatic cell counter) to collect data from the cultured cell samples every 6 hours. The specific operation includes extracting 100 μL of cell suspension from the culture bottle, adding an equal volume of trypan blue dye solution and mixing, taking 10 μL of the mixture and placing it on the cell counter slide for counting; recording the number of live cells and the number of dead cells, and extracting cell morphological parameters (such as average diameter, aspect ratio, circularity, etc.) through the system's built-in image recognition algorithm, and the output data format includes the live cell / dead cell ratio corresponding to the time point, cell density and morphological characteristics table.

[0038] Step S12: collecting real-time data including temperature, humidity, CO2 concentration and culture medium pH value through the incubator sensor to obtain culture environment parameters;

[0039] The embodiment of the present invention uses the environment monitoring sensor provided by the incubator to automatically record the environmental parameters every 10 minutes. This includes the culture temperature data recorded by the temperature sensor (Pt100 thermal resistor), the relative humidity data recorded by the humidity sensor (capacitive sensor), and the CO2 concentration data recorded by the CO2 infrared analyzer; in addition, the pH value is monitored in real time by the culture medium pH probe (the target range is 7.2-7.4). All collected data are automatically stored in the electronic recording system of the incubator and output as an environmental parameter curve file indexed by time.

[0040] Step S13: recording pollution status data through a microscope image acquisition system, including bacterial pollution image data and fungal pollution image data;

[0041] The embodiment of the present invention uses an image acquisition system equipped with a high-resolution microscope (such as Nikon Eclipse Ti2) to record pollution status image data at regular intervals. The operation steps include extracting a small amount of culture medium sample from the culture bottle, preparing a slide sample and then performing microscopic observation to obtain bacterial pollution characteristics (such as bacillus and cocci morphology) and fungal pollution characteristics (such as branched hyphae structure). The collected images are processed and annotated by pollution analysis software (such as ImageJ), and pollution image files and quantitative evaluation data (such as pollution area ratio, pollution type ratio) are output.

[0042] Step S14: collecting culture operation record data through an electronic record system, including liquid replacement time, liquid replacement volume, and subculture ratio; merging the culture environment parameters and culture operation record data into culture condition parameter data;

[0043] The embodiment of the present invention enters culture operation data through a laboratory electronic recording system (such as LabWare LIMS), mainly including liquid replacement operation and subculture operation. The specific process is to record the liquid replacement time (such as every 48 hours) and the liquid replacement volume (such as 80% of the total culture medium) during the liquid replacement operation, and record the subculture ratio (such as 1:3) during the subculture operation; the system automatically associates and integrates these data with the incubator environmental parameter data to generate culture condition parameter data in a time series format.

[0044] Step S15: cleaning and standardizing the multi-source monitoring data to obtain basic data of the culture process, wherein the multi-source monitoring data includes cell counting data, contamination status data and culture condition parameter data.

[0045] The embodiment of the present invention uses data processing software (such as a data cleaning script written in Python) to clean the multi-source monitoring data collected above, including removing outliers (such as outliers are defined as values ​​exceeding the mean ± 3 times the standard deviation) and processing missing values ​​(filled by interpolation). The cleaned data are uniformly standardized (such as normalized to the [0,1] range or Z-score standardization). The final output of the culture process basic data includes time-series cell counting data, culture condition parameter data and contamination status data, providing a data basis for consistency and reliability for subsequent analysis steps.

[0046] The present invention uses an automatic cell counter to regularly record the number of live cells and dead cells and morphological parameters, which not only reduces the errors caused by manual operation, but also provides high-frequency time series data. This high-precision and real-time monitoring method helps to capture the dynamic changes in cell growth, timely discover abnormal growth conditions, and provide reliable data to support subsequent analysis and model building. This method significantly improves the quality and research efficiency of cell counting data. The key environmental parameters are collected in real time by the incubator sensor, which can ensure accurate monitoring of culture conditions. These parameters such as temperature, humidity, CO2 concentration and pH value directly affect the survival and growth state of stem cells. Therefore, accurate collection and recording of these data helps to quickly identify deviations in the culture environment and adjust the culture conditions in time to maintain the optimal growth environment of cells. This highly sensitive monitoring provides a reliable guarantee for improving culture efficiency and success rate. The microscope image acquisition system can accurately record the relevant image data of bacterial and fungal contamination, providing a key basis for early detection and accurate treatment of contamination problems. Image data has the advantages of being more intuitive and comprehensive than traditional quantitative data, and can assist in the rapid diagnosis and tracing of pollution sources. Through this real-time image monitoring method, contamination events can be handled more quickly, thereby reducing the negative impact on cell culture and improving the reliability and safety of culture. The culture operation record data is collected through the electronic record system to ensure the integrity and traceability of the operation information. Information such as the time of medium change, the amount of medium change and the ratio of subculture are critical control variables in the culture process. After combining with the culture environment parameters, the culture condition parameter data generated can more comprehensively reflect the overall status of the culture conditions. This data integration method provides multi-dimensional data support for subsequent analysis, which helps to optimize the culture conditions and achieve standardized operating procedures. Data cleaning and standardization are key steps to ensure the consistency and availability of multi-source data. Through this process, redundancy, errors and missing values ​​in the collection process can be eliminated, and data from different sources and units can be converted into a unified format, thereby significantly improving the accuracy of data analysis and the reliability of model training. The basic data of the culture process finally generated covers key elements such as cell counts, contamination conditions and culture conditions, providing high-quality input data for subsequent modeling, prediction and optimization.

[0047] Preferably, step S15 comprises the following steps:

[0048] Step S151: performing outlier detection and elimination on the cell counting data, eliminating data points that deviate from the mean by more than three times the standard deviation, thereby obtaining cell activity counting data;

[0049] The embodiment of the present invention uses a data cleaning script written in Python to detect and remove outliers in cell counting data. The specific operation is to first calculate the mean and standard deviation of the live cell count, and mark the data points that deviate from the mean by more than three times the standard deviation as outliers; these outliers are removed by setting a threshold range (such as mean ± 3 times the standard deviation), and replaced with the mean of the adjacent values ​​before and after the time series to maintain data continuity. Taking 48 hours of continuous culture data as an example, after cleaning, a data table without abnormalities containing the number of live cells at each time point is output, providing highly reliable cell activity count data for subsequent analysis.

[0050] Step S152: performing image quality assessment on the pollution status data, eliminating unqualified images such as blur and underexposure, so as to obtain high-confidence pollution feature data;

[0051] The embodiment of the present invention uses image processing software (such as ImageJ) to evaluate the image quality of the pollution status data. The specific operation is to perform clarity detection on the collected microscope images one by one, use the Laplace operator to calculate the image edge sharpness, and set the sharpness threshold (such as 20 or more for qualified images); at the same time, perform a histogram analysis on the exposure status to eliminate images with severe grayscale distribution deviation or overexposure / underexposure. After screening, high-confidence pollution feature data is output, including clear images and quantitative analysis results of bacterial and fungal pollution, which are used for subsequent pollution feature extraction and evaluation.

[0052] Step S153: performing parameter correction processing on the culture condition parameter data to obtain culture condition parameter correction data, wherein the parameter correction processing specifically uses an interpolation algorithm to process the missing culture environment parameters caused by sensor failure, and unifies the time format and operation terms of the culture operation record data;

[0053] When the embodiment of the present invention corrects the culture condition parameter data, firstly, linear interpolation is used to fill in the missing data caused by sensor failure. For example, if a breakpoint occurs in a certain period of time in the CO2 concentration sensor, interpolation repair is performed based on the values ​​of adjacent time points; secondly, the culture operation record data is processed in a unified time format, and inconsistent time records (such as "3PM" and "15:00") are standardized to a 24-hour format, and operation terms (such as "medium change" and "change liquid") are translated for consistency. The corrected data is output in the form of a standardized parameter file for subsequent data integration.

[0054] Step S154: standardize the cell activity count data, high confidence contamination feature data, and culture condition parameter correction data, and perform time series alignment processing to obtain basic culture process data.

[0055] The embodiment of the present invention performs standardization processing on cell activity count data, high confidence contamination feature data and culture condition parameter correction data respectively. The standardization method includes normalizing the numerical data to the range of [0,1], and converting the categorical variables into numerical values ​​by one-hot encoding; then, the various types of data are aligned in time series based on the timestamp to ensure that the cell counts, contamination features and culture condition parameters at the same time point remain consistent. The final output data file is saved in CSV format, which contains multi-dimensional, time series aligned basic data of the culture process, laying the foundation for further growth dynamics analysis.

[0056] Eliminating data points that deviate from the mean by more than three times the standard deviation in the present invention is a classic method for eliminating outliers, which can effectively reduce the impact of instrument errors or environmental interference on data. After such processing, the generated cell activity counting data is more accurate and reliable, providing higher quality basic data for subsequent analysis. By eliminating outliers, they are prevented from causing deviations in modeling, and the credibility of cell growth feature analysis and prediction models is improved. By eliminating blurred or underexposed images through image quality assessment, high confidence in pollution status data is ensured. High-quality images can more clearly reflect pollution characteristics, which is convenient for accurate identification and subsequent treatment of pollution. This screening mechanism reduces the risk of misdiagnosis and misjudgment, provides a solid basis for pollution source analysis and solution formulation, and improves the overall reliability of culture process monitoring. The missing of culture environment parameters caused by sensor failure is repaired by interpolation algorithm, which can supplement the integrity of data and reduce the analysis deviation caused by data missing. At the same time, the culture operation record data is standardized in a unified time format and terminology, which improves the consistency and availability of data. This correction process ensures the accuracy and compatibility of culture condition parameter data, and provides more reliable input data for subsequent model training and optimization. Standardization can eliminate the dimensional differences between different data sources, making cell activity counts, high-confidence contamination features, and culture condition parameters comparable, which is convenient for subsequent model calculation and analysis. Time series alignment further unifies multi-source data into the same time dimension, enhancing the temporal consistency of the data. The basic culture process data ultimately generated by these processing steps has high integrity, consistency, and analytical value, laying a solid foundation for accurate cell growth prediction and culture optimization.

[0057] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment of the present invention, step S2 includes the following steps:

[0058] Step S21: extracting the number of live cells and the number of dead cells from the cell activity counting data for 72 consecutive hours, thereby obtaining time-series cell activity data;

[0059] The embodiment of the present invention uses the Pandas library in Python to load 72 hours of continuous cell counting data, which is divided into two columns of live cell count and dead cell count according to the timestamp to construct time series cell activity data. During extraction, the integrity of the time series is ensured, and the unevenly spaced data points are supplemented using linear interpolation. For example, in a certain experiment, the data recorded every 6 hours are accurately aligned, and the final generated time series cell activity data is expressed in the number of live cells and dead cells per hour, providing accurate input data for subsequent analysis.

[0060] Step S22: selecting data points of the logarithmic growth period of the time-series cell activity data, and fitting an exponential growth model based on the nonlinear least squares method, thereby obtaining a growth fitting curve model;

[0061] The embodiment of the present invention extracts the logarithmic growth period data points identified in the time series cell activity data, usually performs segmented analysis on the growth curve, and selects the stage with significantly increased growth rate as the logarithmic growth period by calculating the logarithmic change rate of the number of living cells. Then, the curve_fit function in the Scipy library is used to fit the exponential growth model to the selected data points based on the nonlinear least squares method, and the fitting formula is N(t)=N0·e r·t , where N0 is the initial number of cells and r is the growth rate. In a set of experimental data, data points from 12 to 36 hours were extracted, and the fitting results showed that the curve of the exponential growth model can better describe the changes in cell activity.

[0062] Step S23: calculating the slope and determination coefficient of the fitting curve of the growth fitting curve model, and evaluating the model fitting goodness, thereby obtaining model fitting goodness data;

[0063] The embodiment of the present invention uses the growth curve model obtained by fitting to calculate the slope of the curve (i.e., the growth rate parameter r) and the determination coefficient R 2 The specific method is to use NumPy to calculate the gradient to obtain the slope, and call the r2_score function in the Scikit-learn library to calculate the coefficient of determination. In the experiment, the fitting slope of a cell culture data is 0.025 and the coefficient of determination is 0.98, indicating that the goodness of fit is high. The result data is stored as a file containing the slope and R 2 The structured table provides reliable support for subsequent doubling time calculations.

[0064] Step S24: calculating the cell population doubling time according to the model fitting goodness data, and evaluating the doubling time stability, thereby generating doubling time dynamic stability data;

[0065] The embodiment of the present invention uses the formula The doubling time of the cell population is calculated, where r is the slope of the fitted curve. The volatility of the doubling time is evaluated for stability. Specifically, the standard deviation of the doubling time is calculated using the sliding window method to quantify the amplitude of the fluctuation. For example, in one experiment, the doubling time was calculated to be 27.7 hours, and the standard deviation was 2.3 hours, indicating that the doubling time remained stable during the logarithmic growth period. The stability data is recorded as a doubling time dynamic stability curve graph to intuitively display the dynamic change trend of the doubling time.

[0066] Step S25: extracting cell growth characteristic parameters from the doubling time dynamic stability data, thereby obtaining cell growth characteristic data;

[0067] The embodiment of the present invention extracts parameters of the dynamic stability data of the doubling time, including the mean, standard deviation and extreme value range of the doubling time, and generates a cell growth characteristic data table. For example, in a certain experimental data, the doubling time mean is 28 hours, the standard deviation is 2 hours, and the maximum and minimum values ​​are 30 hours and 26 hours, respectively. The extracted growth characteristic parameters are stored in JSON format to provide support for the input of the cell growth prediction model.

[0068] Step S26: establishing a cell growth prediction model based on the cell growth characteristic data to generate cell growth early warning data;

[0069] The embodiment of the present invention constructs a cell growth prediction model based on cell growth feature data, uses LSTM (Long Short-Term Memory Network) to model time series data, and the input data includes the number of living cells and the dynamic characteristics of doubling time. The target output is the predicted value of the number of living cells in the next 12 hours. In model training, Adam optimizer and mean square error (MSE) are selected as optimization targets and loss functions. Through training and verification on multiple groups of experimental data, the average prediction error of the prediction model is controlled within ±5%, and finally cell growth warning data containing risk factors are generated.

[0070] Step S27: Monitor the culture process in real time according to the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan.

[0071] The embodiment of the present invention uses cell growth warning data to monitor the culture process in real time. When the number of live cells deviates from the predicted value by more than ±10%, a warning signal is triggered, and a culture abnormality treatment plan is generated based on the abnormal characteristics. For example, the abnormality may be caused by insufficient CO2 concentration or contamination. The treatment plan will recommend increasing the ventilation frequency of the incubator or replacing the culture medium. In a certain experiment, real-time monitoring detected that the growth rate of live cells dropped sharply in the 45th hour. The system triggered a contamination warning and generated a treatment plan that recommended thorough cleaning of the culture container, providing timely support for the recovery of experimental results.

[0072] The present invention extracts the time series of the number of living cells and dead cells through continuous data recording for 72 hours, and provides a detailed dynamic change trend for analyzing the cell growth state. Long-term continuous monitoring can capture the key processes of cell proliferation and apoptosis, ensure the comprehensiveness and timeliness of data, and provide a reliable basis for subsequent model fitting and growth feature extraction. Selecting the data points of the logarithmic growth period can focus on the key stage of rapid cell growth, so as to more accurately reflect the reproductive capacity of the cell population. The nonlinear least squares method is used for fitting, which can effectively deal with data noise, generate an exponential growth model that conforms to the actual growth dynamics, and provide accurate numerical expression for further analysis of cell growth trends. Calculating the slope of the fitting curve can quantify the cell growth rate, and the determination coefficient can evaluate the degree of interpretation of the fitting model to the data. Through the goodness of fit evaluation, the accuracy and reliability of the model can be verified, so as to ensure that the generated growth model can truly reflect the cell growth characteristics, and provide strong theoretical support for subsequent prediction and early warning. Doubling time is an important indicator for evaluating the proliferation capacity of cell populations, and its stability can reflect whether the culture conditions are uniform and suitable. By calculating and evaluating the dynamic stability of the doubling time, potential abnormal fluctuations in the culture process can be discovered in time, providing a scientific basis for optimizing the culture conditions and ensuring the efficiency and consistency of cell culture. Extracting cell growth characteristic parameters, such as growth rate and doubling time fluctuation amplitude, can comprehensively characterize the growth behavior characteristics of cells. The generated cell growth characteristic data is the core input for the establishment of subsequent prediction models, which can enhance the pertinence and practicality of dynamic monitoring of the culture process. By building a prediction model based on cell growth characteristic data, quantitative prediction of cell growth trends can be achieved. The generated early warning data can identify potential abnormalities in advance and avoid cell culture failure. The model improves the intelligence level of the culture process, reduces the need for manual intervention, and significantly improves the culture efficiency and success rate. By real-time monitoring of the cell growth process and using early warning data to quickly detect and respond to abnormal fluctuations, the occurrence of culture accidents can be effectively avoided. Triggering early warning signals and generating abnormal handling plans enable operators to quickly take targeted measures to ensure that the culture conditions are restored to the optimal state. This mechanism enhances the stability and safety of the culture system and reduces the risk of cell contamination and proliferation failure.

[0073] Preferably, step S24 comprises the following steps:

[0074] Step S241: Calculating the cell population doubling time according to the model fit goodness data, thereby generating doubling time trend data;

[0075] The embodiment of the present invention uses the doubling time calculation formula Calculate the doubling time of the cell population, where r is the growth rate obtained by fitting. Organize the doubling time at different time points into trend data, and generate a line chart using the Matplotlib library to display the trend. For example, in an experiment, the doubling time of cells cultured from the 1st to the 5th generation was 28.2 hours, 27.5 hours, 28.0 hours, 27.8 hours, and 27.6 hours, respectively. The generated doubling time trend data showed that the range of cell doubling time was stable within ±0.5 hours.

[0076] Step S242: comparing and analyzing the doubling times of different passage times according to the doubling time trend data, establishing a doubling time stability scoring system to evaluate the stability of the doubling time, thereby obtaining doubling time stability scoring data, wherein the doubling time stability scoring system is specifically that when the coefficient of variation of the doubling time of cells for three consecutive generations is less than 10%, the score is the first level, when the coefficient of variation of the doubling time of cells for three consecutive generations is between 10% and 20%, the score is the second level, and when the coefficient of variation of the doubling time of cells for three consecutive generations is greater than 20%, the score is the third level;

[0077] The present invention calculates the coefficient of variation (CV) between different passage times based on the doubling time trend data, and the formula is: Where σ is the standard deviation of doubling time and μ is the mean value. The doubling time stability is graded according to the scoring system, and the scoring data is stored as a structured table. For example, in a certain experiment, the coefficient of variation of cells was 7.8%, 11.3% and 22.5%, and the corresponding scores were first, second and third level, respectively, providing data support for culture stability analysis.

[0078] Step S243: when the doubling time stability score data is the first level, the current culture conditions are recorded as high-quality culture condition data;

[0079] In the embodiment of the present invention, when the doubling time stability score is the first level, the key parameters in the current culture environment, including temperature, humidity, CO2 concentration, and culture medium components, are recorded in the database to generate high-quality culture condition data. For example, in an experiment, when the score is the first level, the recorded high-quality culture condition data include 37°C, 90% humidity, 5% CO2 concentration, and a formula containing 10% fetal bovine serum in the culture medium.

[0080] Step S244: when the doubling time stability score data is the second level, the monitoring frequency is increased, and the change trend of the doubling time stability is analyzed; when the doubling time stability tends to the third level, the concentration change analysis of the main growth factor is performed, and the culture condition optimization data is generated; when the doubling time stability tends to the first level, the current culture condition is recorded as the high-quality culture condition data; when the doubling time stability remains unchanged, the non-main growth factor regulation processing is performed, and the key regulation parameters are recorded as the non-main growth factor adjustment data;

[0081] In the embodiment of the present invention, when the doubling time stability score is the second level, the system increases the monitoring frequency to once an hour and analyzes the stability change trend at the same time. If the doubling time tends to the third level, the concentration change of the main growth factor in the culture medium is detected by ELISA to generate optimization suggestion data; if the doubling time tends to the first level, the current culture conditions are recorded as high-quality culture conditions data; if the doubling time remains unchanged, the concentration of non-main growth factors (such as amino acids or trace elements) is adjusted, and the adjustment parameters are recorded as non-main growth factor adjustment data. For example, in a certain experiment, after the concentration of the main growth factor was adjusted from 50ng / mL to 75ng / mL, the doubling time tended to be stable.

[0082] Step S245: When the doubling time stability score data is the third level, extract the temperature and CO2 concentration fluctuation records of the culture environment, analyze the concentration changes of the main growth factors in the culture medium, and generate culture condition optimization data;

[0083] In the embodiment of the present invention, when the doubling time stability score is the third level, the data of abnormal fluctuations in the culture environment record are extracted, such as the temperature fluctuation range of 36.5°C to 37.8°C and the CO2 concentration fluctuation range of 4.7% to 5.2%. Subsequently, HPLC is used to detect the concentration changes of the main growth factors in the culture medium to generate culture condition optimization data. For example, in a certain experiment, after the VEGF concentration is increased from 20ng / mL to 40ng / mL, the stability score is increased to the second level.

[0084] Step S246: Merge the high-quality culture condition data, the non-essential growth factor adjustment data, and the culture condition optimization data into the doubling time dynamic stability data.

[0085] The embodiment of the present invention integrates the high-quality culture condition data, non-major growth factor adjustment data, and culture condition optimization data into dynamic stability data of doubling time, and stores it as a database file for subsequent optimization and analysis. For example, in a certain experiment, the final integrated dynamic stability data includes the optimal culture temperature of 37°C, CO2 concentration of 5%, and main growth factor concentration of 50ng / mL, and records the situation where the stability score remains unchanged after adjusting the non-major growth factor, providing a reliable parameter basis for subsequent experiments.

[0086] The present invention calculates the cell doubling time through the model fitting goodness data, can efficiently and accurately quantify the cell growth cycle characteristics, and convert it into doubling time trend data that is easy to analyze. The generation of trend data provides a comprehensive view of cell growth stability on a long time scale, laying a foundation for further analysis and optimization of culture conditions. By comparing and analyzing the doubling times of different generations, the trend of cell stability changes during the generation process can be identified. The doubling time stability scoring system is divided into grades according to the coefficient of variation, making the evaluation of cell culture conditions more intuitive and standardized. The scoring system provides a clear stability reference, which helps to timely identify potential problems in the cell proliferation process and ensure the controllability and consistency of the culture process. The first-level doubling time stability score indicates that the culture conditions are optimal. At this time, recording the current culture conditions as high-quality condition data can not only provide successful experience accumulation in the culture process, but also provide a reliable reference for subsequent experiments and large-scale cell culture, reducing the trial and error cost required for optimization. By increasing the monitoring frequency, the doubling time change trend can be captured in time. When the doubling time stability tends to the third level: the concentration change analysis of the main growth factor is performed and the culture condition optimization data is generated, and the key factors are quickly located and adjusted to avoid further deterioration of the cell state. When the doubling time stability tends to the first level: record the current culture conditions as high-quality culture condition data and summarize the effects of culture optimization measures. When the doubling time stability remains unchanged: by regulating non-major growth factors and fine-tuning the culture environment, ensure that cell culture remains within an acceptable range, generate key control parameters, and accumulate adjustment experience for non-major growth factors. This process ensures the flexibility and continuity of dynamic optimization of culture conditions and improves the overall stability of cell culture. When the cell doubling time stability score drops to the third level, it indicates that there are significant problems with the culture conditions. By extracting the temperature and CO2 concentration fluctuation records, potential external environmental interference sources can be quickly located. At the same time, combined with the analysis of changes in the concentration of major growth factors, comprehensive diagnosis and generation of optimization data can be performed to quickly adjust the culture conditions. This measure effectively avoids the risk of culture failure and improves the scientificity and efficiency of problem handling. Integrating culture condition data from different sources to form doubling time dynamic stability data can comprehensively and systematically reflect the adjustment history and optimization results of the culture process. These data can not only serve as a basis for decision-making in the cell culture process, but can also be used for subsequent model training and experience accumulation, significantly improving the level of intelligent management of cell culture.

[0087] Preferably, step S25 comprises the following steps:

[0088] Step S251: Calculate the cell survival rate at each time point according to the dynamic stability data of the doubling time, thereby drawing the survival rate change curve data;

[0089] The present invention uses the cell survival rate calculation formula based on the dynamic stability data of the doubling time Calculate the cell survival rate at each time point and plot the survival rate data into a curve. For example, in a certain experiment, the number of live cells recorded was 5 million, 5.2 million, and 5.3 million, and the total number of cells was 5.5 million. The calculated survival rates were 90.9%, 94.5%, and 96.3%, respectively. Use Matplotlib in Python to plot the survival rate change curve and analyze its dynamic change trend over time.

[0090] Step S252: extracting key growth kinetic parameters according to the survival rate change curve data and the doubling time trend data, thereby obtaining growth kinetic parameter data, wherein the key growth kinetic parameters include growth rate, exponential growth coefficient and maximum growth rate;

[0091] The embodiment of the present invention extracts key growth dynamics parameters through curve fitting based on the survival rate change curve data and doubling time trend data. Using the exponential growth model formula N(t) = N0·e r·t , where r is the growth rate, and the exponential growth coefficient and maximum growth rate are obtained by combining the nonlinear least squares fitting method. In a certain experiment, the growth rate fitting value is 0.03, the exponential growth coefficient is 2.1, and the maximum growth rate is 0.05. The results are stored as structured growth dynamics parameter data.

[0092] Step S253: segmenting the survival rate change curve data, dividing the culture period into an adaptation period, a logarithmic growth period and a plateau period, thereby obtaining segmented survival curve data;

[0093] The embodiment of the present invention performs segmented processing on the survival rate change curve data, setting the adaptation period as the stage where the survival rate is lower than 80%, the logarithmic growth period as the stage where the survival rate rises rapidly and the growth slope is the largest, and the plateau period as the stage where the survival rate tends to be stable. For example, in a certain experiment, the adaptation period is 0-12 hours, the logarithmic growth period is 12-48 hours, and the plateau period is after 48 hours. Each segment of data is extracted and marked and stored as survival curve segmented data.

[0094] Step S254: Calculate the deviation between the mean survival rate of each period and the reference curve according to the survival curve segmentation data and the preset survival rate reference curve, so as to obtain the survival rate deviation value data;

[0095] The embodiment of the present invention compares the survival curve segmentation data with the preset survival rate reference curve to calculate the survival rate mean of each stage, and calculates the deviation value from the reference curve, and the formula is deviation value = actual mean value - reference mean value. For example, in an experiment, the actual survival rate mean value is 85%, the reference mean value is 90%, the deviation value is -5%, and the result is stored as the survival rate deviation value data.

[0096] Step S255: identifying abnormal fluctuations of the survival rate on the survival rate deviation value data according to a preset deviation threshold, thereby obtaining abnormal survival rate data;

[0097] The embodiment of the present invention identifies abnormal fluctuations in the survival rate deviation value data according to a preset deviation threshold (e.g., ±5%), and marks the time period that exceeds the threshold. For example, in an experiment, it is found that the deviation value of a certain time period is -7%, which exceeds the threshold, and it is recorded as abnormal survival rate data, and the abnormal time period is marked as 36-48 hours.

[0098] Step S256: extracting the culture operations within 24 hours before the abnormal time period of the abnormal survival rate data, and matching the culture operations with the preset historical abnormal event library to determine the cause of the abnormal survival rate, thereby obtaining deviation tracing data;

[0099] The embodiment of the present invention extracts the culture operation records of the 24 hours before the abnormal period corresponding to the abnormal survival rate data, including the liquid replacement time, the culture medium component adjustment records, etc. The natural language processing algorithm is used to perform pattern matching with the historical abnormal event library to identify the causes that may cause the abnormality. For example, in a certain experiment, the abnormal traceability results show that there is a record of not replacing the culture medium according to the standard procedure 24 hours before the abnormal period, and the deviation traceability data is stored as "the culture medium is not replaced in time."

[0100] Step S257: The growth dynamics parameter data and the deviation tracing data are combined into cell growth characteristic data.

[0101] The embodiment of the present invention integrates the extracted growth kinetic parameter data with the deviation traceability data to form complete cell growth characteristic data. For example, in a certain experiment, the cell growth characteristic data generated by combining the growth rate of 0.03, the maximum growth rate of 0.05 and the deviation traceability result "the culture medium is not replaced in time" is standardized tabular data for subsequent analysis and prediction model use.

[0102] The present invention can draw a survival rate change curve reflecting cell growth by combining the doubling time dynamic stability data and the cell survival rate. This curve helps to provide the survival status of each time point in the cell growth process, facilitates tracking and evaluating the growth state of the cell, and provides basic data support for subsequent growth dynamics analysis and abnormal detection. This step improves the real-time monitoring capability of growth stability and health during cell culture. By combining the survival rate change curve data and the doubling time trend data, key growth dynamics parameters in the cell growth process, such as growth rate, exponential growth coefficient and maximum growth rate, can be extracted. These parameters can help to deeply understand the basic characteristics of cell growth, and then provide accurate basis for optimizing culture conditions. Key growth dynamics parameters help to establish a cell growth model and improve the predictability and accuracy of the culture process. The survival rate change curve is segmented into an adaptation period, a logarithmic growth period and a plateau period, and the different stages of cell growth can be analyzed in detail. Each stage represents a different growth characteristic of the cell, and segmented processing makes the analysis more accurate. For example, in the logarithmic growth period, the cell growth rate is the fastest, while the plateau period may represent that the cell reaches the maximum growth capacity. The segmented data provides a clear reference for subsequent growth dynamics analysis and abnormal fluctuation identification, ensuring that each key stage of the culture process is reasonably monitored and optimized. By comparing the survival rate curve with the preset survival rate baseline curve, the deviation value of the survival rate at each stage can be calculated. The deviation value reflects whether there is an abnormal fluctuation that deviates from the expected during the culture process. If the deviation between the survival rate and the baseline curve is large, it may indicate that there is a problem with the culture environment or operation. This step helps detect possible abnormal phenomena in the culture process and provides strong data support for subsequent abnormal fluctuation identification and correction. By setting a deviation threshold to identify abnormal fluctuations in the survival rate deviation data, the abnormal fluctuation of the survival rate can be captured in time. This method can effectively and automatically identify abnormal fluctuations and give early warnings, ensuring that potential problems in the cell culture process can be discovered as early as possible and handled in a timely manner. This step plays an important role in improving the stability of the culture and reducing the failure rate, especially in large-scale cell culture or key experiments. By tracing back the culture operation 24 hours before the abnormal survival rate occurs, combined with the historical abnormal event library for pattern matching, the specific cause of the abnormal survival rate can be accurately found. This process ensures timely tracing of problems, avoids pure speculation, and provides a scientific basis for adjusting culture conditions. This analysis based on historical data helps to identify potential regularities and systemic problems, thereby optimizing future cell culture processes. The growth kinetic parameter data is integrated with the deviation traceability data to ultimately form cell growth characteristic data. This data contains key parameters for cell growth and the specific causes of abnormal fluctuations, and is a comprehensive record and analysis of the cell growth process.By integrating these data, a comprehensive assessment of the health of cell culture can be conducted, potential risks can be identified, culture protocols can be improved, and support can be provided for further automated culture management. These cell growth characteristic data provide a reliable basis for optimization, quality control, and management of the culture process.

[0103] Preferably, step S26 comprises the following steps:

[0104] Step S261: constructing a multidimensional cell growth prediction model based on the cell growth characteristic data, wherein the multidimensional cell growth prediction model includes a cell growth rate prediction model, a survival rate prediction model, and a growth trend prediction model;

[0105] The embodiment of the present invention uses the Scikit-learn framework in Python to construct a multi-dimensional cell growth prediction model based on cell growth characteristic data, including a cell growth rate prediction model, a survival rate prediction model, and a growth trend prediction model. In the cell growth rate prediction model, a random forest regression algorithm is used to perform regression analysis on key growth dynamics parameters (such as growth rate, exponential growth coefficient); the survival rate prediction model uses a support vector machine classification algorithm to predict future survival rate intervals; the growth trend prediction model uses a long short-term memory (LSTM) neural network to predict the survival rate and growth trend curve of future time periods. In a certain experiment, the characteristic data of the last three generations (such as a growth rate of 0.03, an exponential growth coefficient of 2.1, and a survival rate of 95%) were selected for model training, and the accuracy of the obtained model was initially evaluated to be 85%.

[0106] Step S262: calibrating and cross-validating the parameters of the multidimensional cell growth prediction model, and determining the prediction accuracy of the model by calculating the predicted confidence interval, thereby obtaining model prediction accuracy data; optimizing the multidimensional cell growth prediction model according to the model prediction accuracy data, and performing multidimensional prediction, thereby obtaining prediction result data;

[0107] The embodiment of the present invention performs parameter calibration on the constructed multi-dimensional cell growth prediction model, and uses a grid search algorithm to adjust key hyperparameters, such as setting the number of trees in the random forest to 100 and the maximum depth to 10. The prediction error (such as the mean square error is 0.02) and the confidence interval (95% confidence interval is ±0.01) are then calculated through five-fold cross validation to evaluate the prediction accuracy of the model. For example, in a certain experiment, the accuracy of the survival rate prediction model reached 90%, but the confidence interval of the growth trend prediction model was wide. The learning rate of the LSTM model was further adjusted to 0.001 and the number of training rounds was increased. Finally, the overall prediction accuracy of the model was improved to 92%. The optimized model is used to generate cell growth trend prediction result data for the next 48 hours.

[0108] Step S263: setting key cell growth indicator thresholds, including a minimum acceptable survival rate, a critical growth rate, and a maximum growth rate;

[0109] The embodiment of the present invention sets the thresholds of key cell growth indicators according to the requirements of cell culture, including the minimum acceptable survival rate of 80%, the critical growth rate of 0.02 / h, and the maximum growth rate of 0.05 / h. In a certain experimental scenario, when the survival rate is lower than 80%, it may indicate that the culture environment is polluted, the growth rate is lower than 0.02 / h, which may indicate a decrease in cell activity, and the maximum growth rate exceeds 0.05 / h, there may be unnatural external interference. These thresholds are stored in the system configuration file as a benchmark for subsequent comparisons.

[0110] Step S264: compare the prediction result data with the key indicator threshold of cell growth in real time. When the prediction result exceeds the preset threshold range, a cell growth abnormality warning signal is generated, thereby obtaining cell growth warning data.

[0111] The embodiment of the present invention compares the generated prediction result data with the set cell growth key indicator threshold in real time. For example, in a certain experiment, the prediction results show that the survival rate may drop to 78% in the next 24 hours, and the growth rate is as low as 0.015 / h, triggering a warning signal for abnormal cell growth. The system records abnormal information based on the warning signal and generates a culture abnormality processing plan, such as recommending immediate replacement of the culture medium and adjusting the temperature and CO2 concentration of the culture environment. The warning data is also stored in the log file for subsequent analysis.

[0112] The present invention can predict and analyze the cell growth process from multiple angles by constructing a multidimensional cell growth prediction model. The cell growth rate prediction model can predict the speed of cell division and proliferation, the survival rate prediction model can reflect the health status of cells, and the growth trend prediction model helps to infer the growth trend of cells in the future. This multidimensional prediction model provides comprehensive dynamic monitoring and prediction capabilities for cell culture, helping researchers to deeply understand the cell growth process and make optimization adjustments. Parameter calibration and cross-validation help to improve the accuracy and reliability of the prediction model. By optimizing and verifying the accuracy of the model, it can be ensured that the model can provide high-quality prediction results in practical applications. Calculating the confidence interval of the prediction helps to measure the uncertainty of the prediction results and ensure that the error range is taken into account when making decisions. This process enhances the prediction ability of the model, enables it to operate stably under different culture conditions and environments, and further improves the controllability and prediction accuracy of cell growth. Setting the thresholds of key indicators of cell growth can provide clear boundaries for the judgment criteria in the cell culture process. These thresholds include the minimum acceptable survival rate, the critical growth rate, and the maximum growth rate, which can help to quickly identify abnormal conditions during the culture process. For example, if the survival rate is lower than the minimum acceptable survival rate, or the growth rate exceeds the critical value, it may mean that there is a problem with the culture environment or operation. By setting these key indicators, more refined control can be provided for the cell culture process to ensure that cells grow within a reasonable range. By real-time monitoring and comparing the predicted results with the key indicator thresholds, an early warning signal can be issued as soon as abnormal cell growth occurs. This real-time early warning mechanism can effectively reduce unexpected situations in the culture process, such as cell decay, contamination or other abnormal phenomena, thereby avoiding experimental failures or production problems. This mechanism provides an intelligent means for the refined management of cell culture, which can adjust the culture conditions in time to ensure the stability and consistency of the culture results.

[0113] Preferably, step S3 comprises the following steps:

[0114] Step S31: Accurately control the culture condition parameter data according to the culture abnormality processing scheme, and continuously track and record the cell growth response of each control operation to generate parameter control response data;'

[0115] The embodiment of the present invention accurately regulates the culture condition parameter data according to the culture abnormality processing scheme. For example, during the culture condition optimization process, when the cell survival rate is detected to be lower than the set threshold, the CO2 concentration in the incubator is adjusted (from 5% to 6%), the temperature (from 37°C to 36.5°C) and the pH value of the culture medium (adjusted to 7.4). After each regulation, an automatic cell counter is used to continuously track and record cell growth, including cell activity, mortality rate, and morphological changes, to generate parameter regulation response data. After each operation, the recorded cell growth response data will include changes in cell activity before and after regulation (for example, the number of live cells increased by 20% after regulation), and these data will serve as the basis for subsequent analysis.

[0116] Step S32: performing correlation analysis on the parameter control response data and the cell growth characteristic data, and constructing a quantitative mapping relationship between the culture conditions and the cell growth indicators, thereby establishing an initial culture condition optimization model;

[0117] The embodiment of the present invention performs correlation analysis based on each parameter regulation response data and cell growth characteristic data. The multivariate linear regression analysis method is used to associate culture conditions (such as temperature, humidity, CO2 concentration, etc.) with cell growth indicators (such as cell proliferation rate, survival rate, doubling time, etc.). Through analysis, it is found that the culture temperature has a positive effect on cell survival rate, and the CO2 concentration has a negative effect on cell doubling time. For example, it was found through analysis that when the CO2 concentration increased from 5% to 6%, the cell survival rate increased significantly by 5%, while the doubling time was shortened by about 10%. Based on this result, an initial culture condition optimization model was constructed, which can predict cell growth performance according to changes in culture conditions.

[0118] Step S33: cross-validate the initial culture condition optimization model and perform parameter sensitivity analysis to evaluate the prediction accuracy and stability of the model, and iteratively optimize the model parameters to obtain the culture condition optimization model.

[0119] The embodiment of the present invention performs cross-validation and parameter sensitivity analysis on the initial culture condition optimization model. First, the initial model is verified using different data sets, and the stability and prediction accuracy of the model are evaluated by 10-fold cross-validation. During the verification process, the prediction error of the model for different cell populations (such as the mean square error MSE of 0.03) is recorded, showing that the model has good adaptability. Subsequently, a sensitivity analysis is performed on the key parameters of the model to check the effects of factors such as culture temperature and CO2 concentration on cell growth prediction, and it is found that temperature has a high sensitivity to the model prediction accuracy. Based on these analysis results, the model parameters are adjusted, the algorithm structure and parameter weights of the model are optimized, and finally the prediction accuracy of the optimized model is increased to 95%, and a stable culture condition optimization model is obtained.

[0120] The present invention can capture the response of cells to each adjustment in real time by accurately regulating the culture conditions and continuously tracking and recording the cell growth response of each operation. This continuous tracking provides key data for researchers to gain a deeper understanding of the relationship between cell growth and environmental conditions. The effect of each regulation can be directly fed back into the model optimization to ensure that the adjustment process can gradually improve the culture conditions. This data management helps to improve the controllability and optimizability of the culture process and provide a detailed basis for subsequent optimization of culture conditions. Correlation analysis of parameter regulation response data and cell growth characteristic data can reveal the quantitative relationship between culture conditions (such as temperature, pH value, nutrients, etc.) and cell growth indicators (such as cell proliferation rate, survival rate, etc.). The construction of this quantitative mapping relationship provides a more accurate adjustment tool for the cell culture process, which can achieve fine regulation from empirical regulation to data-based regulation. After establishing the initial culture condition optimization model, researchers can quickly adjust the culture parameters according to actual growth feedback to optimize the cell growth process. Cross-validation and parameter sensitivity analysis help to evaluate the prediction accuracy and stability of the initial model. Through cross-validation, the generalization ability of the model can be verified to ensure that cell growth responses can be reliably predicted under different experimental conditions. Parameter sensitivity analysis helps identify the culture parameters that have the greatest impact on model results, allowing for more effective adjustments. Through these evaluation and optimization steps, the initial model is continuously improved, gradually improving prediction accuracy and providing more reliable optimization suggestions for cell culture. The resulting culture condition optimization model can accurately guide the cell culture process and achieve efficient and stable culture results.

[0121] Preferably, step S4 comprises the following steps:

[0122] Step S41: performing multi-dimensional data mining on the basic data of the culture process based on the culture condition optimization model to obtain associated result data, wherein the multi-dimensional data mining specifically includes performing correlation analysis and pattern recognition on the cell counting data, culture condition parameter data, and growth dynamics characteristic data;

[0123] The embodiment of the present invention performs multi-dimensional data mining on the basic data of the culture process based on the culture condition optimization model. For example, by combining cell counting data, culture condition parameter data and growth dynamics characteristic data, data mining methods (such as principal component analysis PCA and cluster analysis) are used to perform correlation analysis and pattern recognition on these data. In a specific application scenario, by analyzing the relationship between cell counting data and CO2 concentration and temperature changes, it is found that the increase in temperature is positively correlated with the cell growth rate, while the increase in CO2 concentration may inhibit cell proliferation. Through cluster analysis, typical patterns of cell growth under different conditions are identified. For example, under high temperature and high CO2 concentration, the proliferation rate of cells shows a significantly decreased pattern. This finding provides a basis for the next step of optimizing culture conditions.

[0124] Step S42: performing feature weight evaluation on the correlation result data, screening the key parameters that have the most significant impact on cell growth, and thus generating an optimal culture parameter combination;

[0125] The embodiment of the present invention performs feature weight evaluation on the associated result data. Regression analysis and feature selection algorithms (such as LASSO regression) are used to evaluate the weight of each feature parameter based on the relationship between cell growth indicators (such as cell proliferation rate, survival rate) and different culture conditions (such as temperature, humidity, pH value, etc.). For example, the analysis results show that temperature and culture medium pH value have the most significant effects on cell proliferation rate, with temperature contributing 40% to cell proliferation rate and pH value 30%, and the rest, such as CO2 concentration, contributing less to the proliferation rate. Based on these weight evaluation results, the most influential parameters, such as temperature, pH value and culture medium composition, are screened out to generate the optimal culture parameter combination.

[0126] Step S43: formulating a standardized culture program according to the optimal culture parameter combination, wherein the standardized culture program includes specific parameters of cell seeding density, culture medium formula, liquid replacement cycle, passage ratio, and culture environment parameters;

[0127] The embodiment of the present invention formulates a standardized culture scheme based on the optimal culture parameter combination. For example, under the conditions of determining the temperature to be 37°C, the pH value to be 7.4, and the CO2 concentration to be 5%, a suitable cell seeding density (e.g., 5×10 4 cells / mL), medium formula (e.g. DMEM medium with 10% fetal bovine serum), medium change cycle every 48 hours, and subculture ratio set to 1:3. At the same time, according to the experimental data, specific parameters of the culture environment are formulated, such as maintaining the culture temperature at 37°C ± 0.5°C, humidity at above 95%, and ensuring that the CO2 concentration is controlled at 5%. This standardized protocol provides a unified operating guide for subsequent cell culture experiments.

[0128] Step S44: performing repeated experiments to verify the standardized culture scheme, and recording the cell growth index of each experiment, thereby obtaining experimental result data;

[0129] The embodiments of the present invention conduct repeated experiments to verify the standardized culture scheme. Multiple cell lines are selected for culture, and experiments are performed according to the standardized culture scheme. After each experiment, an automatic cell counter and a microscope are used to evaluate the proliferation of cells, and growth indicators such as cell proliferation rate, morphological changes, and survival rate are recorded. For example, after 5 experiments, the average cell proliferation rate was 30% / day, and the survival rate was stable at more than 90%. By comparing the results of different experiments, it is ensured that the culture scheme can stably produce consistent cell growth performance in multiple experiments, and the experimental results data are recorded, including the cell proliferation curve, cell morphological distribution, and survival rate data of each experiment.

[0130] Step S45: Calculate the key indicators of culture quality according to the experimental result data to generate a culture quality assessment report, wherein the key indicators of culture quality include cell growth consistency index, morphological uniformity score, gene stability rating and overall culture success rate.

[0131] The embodiment of the present invention calculates the key indicators of culture quality based on the experimental result data. In combination with the experimental data, the formula is used to calculate the cell growth consistency index (such as the ratio of the standard deviation of the cell proliferation rate in each experiment to the mean), and the value is 0.05, indicating that the consistency of cell growth between different experiments is high. At the same time, the morphological uniformity score (such as the uniformity of cell morphology, the standard deviation of morphological distribution is calculated using image analysis software, and the score is 95 points) and the gene stability rating (the consistency of gene expression after each culture is analyzed by gene sequencing, and a rating of A is given). Finally, the overall culture success rate is calculated based on the success rate of the experiment (such as 4 successes in 5 experiments, that is, the success rate is 80%). These key indicators will be summarized into a culture quality assessment report to guide subsequent culture optimization.

[0132] The present invention can deeply explore the potential associations between various types of data in the culture process through multi-dimensional data mining, revealing the mutual influence between cell count, culture conditions and growth dynamics characteristics. This step helps to identify the key factors and potential optimization space that affect cell growth by performing correlation analysis and pattern recognition on these data. Data mining not only helps us understand the effects of existing culture conditions, but also provides data support for the improvement of future culture schemes, thereby improving the predictability and controllability of culture efficiency and cell quality. By performing feature weight evaluation on the associated result data, the key parameters that affect cell growth can be effectively screened out. This process helps identify those factors that have a significant impact on cell growth, so as to give priority to and optimize these key parameters. By optimizing these parameters, the cell growth effect can be maximized, resource waste can be reduced, and the efficiency of the culture process can be ensured. The optimal culture parameter combination finally generated provides a theoretical basis for formulating a standardized culture scheme, making the culture process more accurate and controllable, and helping to improve the success rate of cell culture. The standardized culture scheme formulated based on the optimal culture parameter combination can unify and standardize every link of cell culture. Through standardization, the differences in human operations can be eliminated, ensuring the comparability and consistency between different experiments. Standardized culture protocols can ensure the stability of culture conditions, contribute to the healthy growth of cells, and lay the foundation for the scale-up and automation of cell culture. In addition, precise parameters such as culture medium formula, passage ratio, and inoculation density can optimize the cell growth environment and improve the proliferation rate and quality of cells. By verifying the standardized culture protocol through repeated experiments, its stability and feasibility under different conditions can be ensured. The cell growth index records of each experiment provide reliable data for further analysis to ensure that the standardized protocol can achieve the expected results in various practical operations. This verification process can detect potential deviations or problems, so as to optimize the protocol in a timely manner. Through the accumulation of experimental data, the accuracy and reliability of the culture protocol can be further improved to ensure that the culture conditions have a wide range of adaptability in practical applications. By calculating key indicators of culture quality, such as cell growth consistency index, morphological uniformity score, gene stability rating, etc., the success of the culture process can be comprehensively evaluated. These indicators can reflect the impact of culture conditions on cell quality and help identify problems or optimization space in culture. The generated culture quality assessment report not only provides a basis for process control of cell culture, but also provides key feedback for subsequent research and application to ensure the stability and repeatability of cell culture in quality. Through these assessments, researchers can optimize culture protocols in a targeted manner to improve the biological characteristics, morphological consistency, and genetic stability of cells, thereby providing higher quality cell populations for cell applications.

[0133] The present invention also provides a stem cell culture information management system for executing the above-mentioned stem cell culture information management method, wherein the stem cell culture information management system comprises:

[0134] A data acquisition module is used to collect basic data of the culture process during stem cell culture, wherein the basic data of the culture process includes cell counting data, culture condition parameter data and contamination status data;

[0135] The growth dynamic monitoring module is used to perform growth dynamics analysis on the basic data of the culture process to obtain cell growth characteristic data; establish a cell growth prediction model based on the cell growth characteristic data to generate cell growth warning data; monitor the culture process in real time based on the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan;

[0136] The parameter control optimization module is used to dynamically control the culture condition parameters according to the culture abnormality processing plan, and to record the control effect in real time to generate parameter control response data; the parameter control response data is correlated with the cell growth characteristic data to establish a culture condition optimization model;

[0137] The training program generation module is used to mine and analyze the basic data of the training process based on the training condition optimization model to generate the optimal training parameter combination; formulate a standardized training program based on the optimal training parameter combination, conduct training quality assessment, and generate a training quality assessment report.

[0138] The data acquisition module in the present invention can systematically collect key data in the stem cell culture process and provide basic data support for subsequent analysis. By acquiring cell counting data, culture condition data and pollution status data in real time, the key changes in the culture process can be fully monitored to ensure the transparency and accuracy of the experimental process. In particular, the collection of pollution status data can timely discover potential pollution risks and ensure the effectiveness and repeatability of the experiment. This helps to optimize the culture scheme in the future and reduce the culture failure caused by environmental changes or pollution. The growth dynamic monitoring module uses the basic data collected in real time to analyze the cell growth dynamics, which can identify the trend and law of cell growth, and then establish an accurate growth prediction model. This enables dynamic monitoring to be achieved during the culture process and predict possible abnormal situations in advance. When the growth data fluctuates abnormally, the system can quickly trigger an early warning signal, generate a culture abnormality processing plan, and adjust the culture plan in time to prevent problems in the culture process from causing irreversible damage to cell growth. Through real-time monitoring and early warning of the cell growth process, the culture efficiency can be greatly improved, and unnecessary waste of resources and cell quality degradation can be avoided. The function of the parameter control optimization module is to adjust the culture condition parameters in real time according to the abnormalities found during the culture process. Through dynamic regulation, such as changing the culture environment factors such as temperature, humidity, and gas composition, it is possible to respond to different stages of cell growth in a timely manner. This module can not only record the regulation effect in real time, but also combine the parameter regulation response data with the cell growth characteristic data through correlation analysis, so as to establish an accurate culture condition optimization model. This allows the optimization of culture conditions to be data-driven, which not only improves the culture efficiency but also ensures the stability of cell quality. Ultimately, through regulation and optimization, the culture success rate and the predictability of cell growth can be significantly improved. The culture scheme generation module can comprehensively analyze the basic data of the culture process and combine it with the culture condition optimization model to accurately mine the laws behind the data and generate the optimal culture parameter combination. Through the formulation of standardized culture schemes, various conditions for cell culture can be unified, the consistency and repeatability of experiments can be improved, and the cell growth conditions during the culture process can be fully optimized. In addition, culture quality assessment and generation of culture quality assessment reports can help evaluate the overall effect of culture, such as cell consistency, morphological uniformity, and genetic stability, thereby providing reliable quality assurance for subsequent research and application. This process helps to maintain the stability of cell quality in multiple experiments, enhance the controllability of the culture process, and reduce the risk of experimental failure.

[0139] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0140] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for managing stem cell culture information, characterized in that: The following steps are involved: Step S1: collecting basic data of the culture process during stem cell culture, wherein the basic data of the culture process includes cell counting data, culture condition parameter data and contamination status data; Step S2: Perform growth dynamics analysis on the basic data of the culture process to obtain cell growth characteristic data; establish a cell growth prediction model based on the cell growth characteristic data to generate cell growth warning data; monitor the culture process in real time according to the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan; Step S3: dynamically regulating the culture condition parameters according to the culture abnormality processing scheme, and recording the regulation effect in real time to generate parameter regulation response data; correlating the parameter regulation response data with the cell growth characteristic data to establish a culture condition optimization model; Step S4: Based on the culture condition optimization model, the basic data of the culture process is mined and analyzed to generate the optimal culture parameter combination; a standardized culture program is formulated according to the optimal culture parameter combination, and a culture quality assessment is performed to generate a culture quality assessment report.

2. The stem cell culture information management method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: recording the number of live cells, the number of dead cells and cell morphology parameters every 6 hours using an automatic cell counter to obtain cell counting data; Step S12: collecting real-time data including temperature, humidity, CO2 concentration and culture medium pH value through the incubator sensor to obtain culture environment parameters; Step S13: recording pollution status data through a microscope image acquisition system, including bacterial pollution image data and fungal pollution image data; Step S14: collecting culture operation record data through an electronic record system, including liquid replacement time, liquid replacement volume, and subculture ratio; merging the culture environment parameters and culture operation record data into culture condition parameter data; Step S15: cleaning and standardizing the multi-source monitoring data to obtain basic data of the culture process, wherein the multi-source monitoring data includes cell counting data, contamination status data and culture condition parameter data.

3. The stem cell culture information management method according to claim 2, characterized in that: Step S15 includes the following steps: Step S151: performing outlier detection and elimination on the cell counting data, eliminating data points that deviate from the mean by more than three times the standard deviation, thereby obtaining cell activity counting data; Step S152: performing image quality assessment on the pollution status data, eliminating unqualified images such as blur and underexposure, so as to obtain high-confidence pollution feature data; Step S153: performing parameter correction processing on the culture condition parameter data to obtain culture condition parameter correction data, wherein the parameter correction processing specifically uses an interpolation algorithm to process the missing culture environment parameters caused by sensor failure, and unifies the time format and operation terms of the culture operation record data; Step S154: standardize the cell activity count data, high confidence contamination feature data, and culture condition parameter correction data, and perform time series alignment processing to obtain basic culture process data.

4. The stem cell culture information management method according to claim 3, characterized in that: Step S2 includes the following steps: Step S21: extracting the number of live cells and the number of dead cells from the cell activity counting data for 72 consecutive hours, thereby obtaining time-series cell activity data; Step S22: selecting data points of the logarithmic growth period of the time-series cell activity data, and fitting an exponential growth model based on the nonlinear least squares method, thereby obtaining a growth fitting curve model; Step S23: calculating the slope and determination coefficient of the fitting curve of the growth fitting curve model, and evaluating the model fitting goodness, thereby obtaining model fitting goodness data; Step S24: calculating the cell population doubling time according to the model fit goodness data, and evaluating the doubling time stability, thereby generating doubling time dynamic stability data; Step S25: extracting cell growth characteristic parameters from the doubling time dynamic stability data, thereby obtaining cell growth characteristic data; Step S26: establishing a cell growth prediction model based on the cell growth characteristic data to generate cell growth early warning data; Step S27: Monitor the culture process in real time according to the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan.

5. The stem cell culture information management method according to claim 4, characterized in that: Step S24 includes the following steps: Step S241: Calculating the cell population doubling time according to the model fit goodness data, thereby generating doubling time trend data; Step S242: comparing and analyzing the doubling times of different passage times according to the doubling time trend data, establishing a doubling time stability scoring system to evaluate the stability of the doubling time, thereby obtaining doubling time stability scoring data, wherein the doubling time stability scoring system is specifically that when the coefficient of variation of the doubling time of cells for three consecutive generations is less than 10%, the score is the first level, when the coefficient of variation of the doubling time of cells for three consecutive generations is between 10% and 20%, the score is the second level, and when the coefficient of variation of the doubling time of cells for three consecutive generations is greater than 20%, the score is the third level; Step S243: when the doubling time stability score data is the first level, the current culture conditions are recorded as high-quality culture condition data; Step S244: when the doubling time stability score data is the second level, the monitoring frequency is increased, and the change trend of the doubling time stability is analyzed; when the doubling time stability tends to the third level, the concentration change analysis of the main growth factor is performed, and the culture condition optimization data is generated; when the doubling time stability tends to the first level, the current culture condition is recorded as the high-quality culture condition data; when the doubling time stability remains unchanged, the non-main growth factor regulation processing is performed, and the key regulation parameters are recorded as the non-main growth factor adjustment data; Step S245: When the doubling time stability score data is the third level, extract the temperature and CO2 concentration fluctuation records of the culture environment, analyze the concentration changes of the main growth factors in the culture medium, and generate culture condition optimization data; Step S246: Merge the high-quality culture condition data, the non-major growth factor adjustment data, and the culture condition optimization data into the doubling time dynamic stability data.

6. The stem cell culture information management method according to claim 5, characterized in that: Step S25 includes the following steps: Step S251: Calculate the cell survival rate at each time point according to the dynamic stability data of the doubling time, thereby drawing the survival rate change curve data; Step S252: extracting key growth kinetic parameters according to the survival rate change curve data and the doubling time trend data, thereby obtaining growth kinetic parameter data, wherein the key growth kinetic parameters include growth rate, exponential growth coefficient and maximum growth rate; Step S253: segmenting the survival rate change curve data, dividing the culture period into an adaptation period, a logarithmic growth period and a plateau period, thereby obtaining segmented survival curve data; Step S254: Calculate the deviation between the mean survival rate of each period and the reference curve according to the survival curve segmentation data and the preset survival rate reference curve, so as to obtain the survival rate deviation value data; Step S254: identifying abnormal fluctuations in survival rate on the survival rate deviation value data according to a preset deviation threshold, thereby obtaining abnormal survival rate data; Step S256: extracting the culture operations within 24 hours before the abnormal time period of the abnormal survival rate data, and matching the culture operations with the preset historical abnormal event library to determine the cause of the abnormal survival rate, thereby obtaining deviation tracing data; Step S257: The growth dynamics parameter data and the deviation tracing data are combined into cell growth characteristic data.

7. The stem cell culture information management method according to claim 6, characterized in that: Step S26 includes the following steps: Step S261: constructing a multidimensional cell growth prediction model based on the cell growth characteristic data, wherein the multidimensional cell growth prediction model includes a cell growth rate prediction model, a survival rate prediction model, and a growth trend prediction model; Step S262: calibrating and cross-validating the parameters of the multidimensional cell growth prediction model, and determining the prediction accuracy of the model by calculating the predicted confidence interval, thereby obtaining model prediction accuracy data; optimizing the multidimensional cell growth prediction model according to the model prediction accuracy data, and performing multidimensional prediction, thereby obtaining prediction result data; Step S263: setting key cell growth indicator thresholds, including a minimum acceptable survival rate, a critical growth rate, and a maximum growth rate; Step S264: compare the prediction result data with the key indicator threshold of cell growth in real time. When the prediction result exceeds the preset threshold range, a cell growth abnormality warning signal is generated, thereby obtaining cell growth warning data.

8. The stem cell culture information management method according to claim 7, characterized in that: Step S3 includes the following steps: Step S31: accurately regulating the culture condition parameter data according to the culture abnormality processing scheme, and continuously tracking and recording the cell growth response of each regulation operation to generate parameter regulation response data; Step S32: performing correlation analysis on the parameter control response data and the cell growth characteristic data, and constructing a quantitative mapping relationship between the culture conditions and the cell growth indicators, thereby establishing an initial culture condition optimization model; Step S33: cross-validate the initial culture condition optimization model and perform parameter sensitivity analysis to evaluate the prediction accuracy and stability of the model, and iteratively optimize the model parameters to obtain the culture condition optimization model.

9. The stem cell culture information management method according to claim 8, characterized in that: Step S4 includes the following steps: Step S41: performing multi-dimensional data mining on the basic data of the culture process based on the culture condition optimization model to obtain associated result data, wherein the multi-dimensional data mining specifically includes performing correlation analysis and pattern recognition on the cell counting data, culture condition parameter data, and growth dynamics characteristic data; Step S42: performing feature weight evaluation on the correlation result data, screening the key parameters that have the most significant impact on cell growth, and thus generating an optimal culture parameter combination; Step S43: formulating a standardized culture program according to the optimal culture parameter combination, wherein the standardized culture program includes specific parameters of cell seeding density, culture medium formula, liquid replacement cycle, passage ratio, and culture environment parameters; Step S44: performing repeated experiments to verify the standardized culture scheme, and recording the cell growth index of each experiment, thereby obtaining experimental result data; Step S45: Calculate the key indicators of culture quality according to the experimental result data to generate a culture quality assessment report, wherein the key indicators of culture quality include cell growth consistency index, morphological uniformity score, gene stability rating and overall culture success rate.

10. A stem cell culture information management system, characterized in that: Used to execute the stem cell culture information management method according to claim 1, the stem cell culture information management system comprising: A data acquisition module is used to collect basic data of the culture process during stem cell culture, wherein the basic data of the culture process includes cell counting data, culture condition parameter data and contamination status data; The growth dynamic monitoring module is used to perform growth dynamics analysis on the basic data of the culture process to obtain cell growth characteristic data; establish a cell growth prediction model based on the cell growth characteristic data to generate cell growth warning data; monitor the culture process in real time based on the cell growth warning data, trigger a warning signal when abnormal fluctuations are detected, and generate a culture abnormality processing plan; The parameter control optimization module is used to dynamically control the culture condition parameters according to the culture abnormality processing plan, and to record the control effect in real time to generate parameter control response data; the parameter control response data is correlated with the cell growth characteristic data to establish a culture condition optimization model; The training program generation module is used to mine and analyze the basic data of the training process based on the training condition optimization model to generate the optimal training parameter combination; formulate a standardized training program based on the optimal training parameter combination, conduct training quality assessment, and generate a training quality assessment report.

Images