A high-reliability digital management platform for cell preparation and storage

Abstract

The present invention relates to the field of cell preparation data processing technology, and specifically to a high-reliability digital management platform for cell preparation and storage. The present invention collects and analyzes the premise factor data and quality inspection result data of the production and environment in the cell preparation process, obtains the influence coefficient of the premise factor data of the real-time cell preparation process on the quality inspection result data, and the intergenerational influence factor between different culture generations; combines and compares the correlation between the premise factor data and the quality inspection result data between the real-time cell preparation and the historical different batches of cell preparation; monitors the influence of subtle changes in the premise factor data on the quality inspection result data, and performs digital management such as traceability, monitoring, and early warning on the real-time cell preparation process, while providing data reference for subsequent process improvements. The present invention improves the management effect of cell preparation-related data by obtaining the accurate influence of the premise factor data on the quality inspection result data in the cell preparation process.

Description

Technical Field

[0001] The present invention relates to the technical field of cell preparation data processing, and in particular to a high-reliability digital management platform for cell preparation and storage. Background Art

[0002] The quality of cell preparation is related to the safety and effectiveness of treatment. With the development of big data and artificial intelligence technologies, these technologies are increasingly used in the field of cell preparation, providing new technical paths and analytical tools for cell preparation. Considering that in the cell preparation process, when different culture methods or culture systems are used, the choice of different culture systems will lead to certain differences between the prepared cells. At the same time, when the culture environment and factors are fixed, the performance of cells cultured at different generations is theoretically similar, so it is necessary to analyze the data during the cell preparation process.

[0003] In the existing technology, big data analysis technology is used to conduct in-depth analysis of historical data on cell preparation; however, in actual situations, there will be certain differences between the prepared cells and the ideal state. Especially when the differences are small, the data performance detected will be overwhelmed by the reasonable data changes of the prerequisite factors caused by different culture methods, and thus it is impossible to reflect the quality changes of the cell preparation process. The digital management platform for cell preparation is less effective. Summary of the Invention

[0004] In order to solve the technical problem that changes in the prerequisite factor data cannot reflect the quality changes in the cell preparation process, and the digital management platform for cell preparation is poor, the purpose of the present invention is to provide a high-reliability digital management platform for cell preparation and storage. The technical solutions adopted are as follows:

[0005] The present invention proposes a high-reliability digital management platform for cell preparation and storage, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented:

[0006] Obtain multiple prerequisite factor data and quality inspection result data for multiple generations of cell preparation in a time series;

[0007] Based on the distribution of prerequisite factor data and quality inspection result data at different culture generations in real-time cell preparation, the influence coefficient of each prerequisite factor data in the real-time cell preparation process on each quality inspection result data, as well as the intergenerational influence factor between different culture generations, are obtained;

[0008] Based on the differences in each prerequisite factor data between different batches of historical cell preparations at the same culture passage, as well as the generational influencing factors and influence coefficients of each quality inspection result data corresponding to real-time cell preparation, highly similar historical cells corresponding to real-time cell preparation are screened out; based on the correlation between the quality inspection result data of real-time cell preparation and the corresponding different highly similar historical cell preparations at the same culture passage, quality inspection difference result data of different generations are screened out;

[0009] Based on the quality inspection difference result data of the different generations between the real-time cell preparation and different highly similar historical cell preparations, as well as the data of each prerequisite factor of each other generation before the different generation, combined with the generational impact factor and impact coefficient of each prerequisite factor data on the quality inspection difference result data, the degree of difference in the impact of each prerequisite factor data before the different generation of the real-time cell preparation on the quality inspection difference result data is obtained;

[0010] Digital management of real-time cell preparation is performed based on the degree of influence of the prerequisite factor data on the quality inspection difference result data.

[0011] Furthermore, the method for obtaining the influence coefficient includes:

[0012] For each prerequisite factor data or quality inspection result data, obtain the prerequisite factor data sequence of the real-time cell preparation and the corresponding historical cell preparation of the same type in each generation of culture, as well as the quality inspection result data sequence in accordance with the time sequence;

[0013] Obtaining the ratio of the standard deviations between the premise factor data series and the quality inspection result data series of different generations as the first ratio; obtaining the covariance between the premise factor data series and the quality inspection result data series as the influence weight;

[0014] The product between the first ratio and the influence weight is calculated and normalized as the influence coefficient of each prerequisite factor data of the real-time cell preparation process on each quality inspection result data.

[0015] Furthermore, the method for obtaining the intergenerational influence factor includes:

[0016] Obtain the mean influence coefficient of all prerequisite factor data on each quality inspection result data in real-time cell preparation at each generation of culture as the average influence level;

[0017] Within the neighborhood of each generation of culture, the maximum value of the influence coefficient of each prerequisite factor data on each quality inspection result data in all generations of culture of real-time cell preparation is selected;

[0018] According to the influence coefficient of each prerequisite factor data in each generation on each quality inspection result data in different generations within the neighborhood, combined with the average influence level and the maximum influence coefficient, the intergenerational influence factor of each prerequisite factor data on each quality inspection result data in each generation of real-time cell preparation was obtained. The influence coefficient was positively correlated with the intergenerational influence factor, and the average influence level and the maximum influence coefficient were negatively correlated with the intergenerational influence factor.

[0019] Furthermore, the method for obtaining highly similar historical cells includes:

[0020] Based on the differences in each prerequisite factor data between different batches of historical cell preparations at the same culture generation, as well as the intergenerational influencing factors and influencing coefficients of each quality inspection result data corresponding to real-time cell preparations, the similarity of each quality inspection result data between real-time cell preparations and different batches of historical cell preparations was obtained;

[0021] For any quality inspection result data, if the result similarity between the quality inspection result data of the real-time cell preparation and any batch of historical cell preparation is greater than the preset similarity threshold, the corresponding batch of historical cells will be regarded as high-similar historical cells.

[0022] Furthermore, the method for obtaining the result similarity includes:

[0023] The result similarity is obtained according to the formula for obtaining the result similarity. The formula for obtaining the result similarity is:

[0024] ;in, Represents real-time cell preparation and Batch history between cell preparations The similarity of the quality inspection result data; Indicates the number of types of premise factor data; represents the number of passages cultured in real-time cell preparation; Indicates real-time cell preparation Daixiadi The data of the prerequisite factors Intergenerational impact factors of seed quality inspection result data; Indicates real-time cell preparation Daixiadi The data of the prerequisite factors The influence coefficient of seed quality inspection result data; Real-time cell preparation In the Next generation Precondition data; Indicates the Batch history cell preparation Daixiadi Precondition data; Indicates taking the absolute value; Indicates the adjustment parameter.

[0025] Furthermore, the method for obtaining the quality inspection difference result data of the difference generation includes:

[0026] Based on the differences in quality inspection results between real-time cell preparation and corresponding different highly similar historical cell preparations at the same culture passage, as well as the similarity of the results, the deviation of the culture results of each quality inspection result data of the real-time cell preparation at each passage is obtained;

[0027] The deviation of the culture results of each quality inspection result data of real-time cell preparation at each generation is normalized. If the normalized result is greater than the preset deviation threshold, the quality inspection result data of the corresponding generation is used as the quality inspection difference result data of the difference generation.

[0028] Furthermore, the method for obtaining the deviation of the culture result includes:

[0029] Obtain the squared difference between the real-time cell preparation and each highly similar historical cell preparation for each quality inspection result data at each generation, as well as the result similarity;

[0030] The mean of the product of the similarity and squared difference between the real-time cell preparation and the corresponding different highly similar historical cell preparations was obtained, and the square root of the mean result was calculated as the culture result deviation of each quality inspection result data of the real-time cell preparation at each generation.

[0031] Furthermore, the method for obtaining the difference impact degree includes:

[0032] According to the quality inspection difference result data of the different generations between the real-time cell preparation and different highly similar historical cell preparations, and the data of each prerequisite factor of each other generation of the highly similar historical cell preparation before the different generation, the prediction data of each prerequisite factor corresponding to each other generation are obtained;

[0033] Obtain the difference between each prerequisite factor data and the prerequisite factor prediction data in each other generation of real-time cell preparation before the corresponding difference generation as the prerequisite factor difference value;

[0034] The generational impact factor and impact coefficient of each prerequisite factor data on the quality inspection difference result data in each other generation, as well as the product between the prerequisite factor difference values, are obtained and normalized as the degree of difference in influence of each prerequisite factor data on the quality inspection difference result data before the real-time cell preparation difference generation.

[0035] Furthermore, the method for obtaining the premise factor prediction data includes:

[0036] Obtaining the result similarity of the quality inspection difference result data of the difference generation between the real-time cell preparation and each highly similar historical cell preparation, and the product of each prerequisite factor data of each other generation of the corresponding highly similar historical cell preparation before the difference generation, as a first product;

[0037] The mean of the first product between the real-time cell preparation and different highly similar historical cell preparations was obtained as the prediction data for each antecedent factor corresponding to each other generation.

[0038] Furthermore, the preset similarity threshold is 0.7.

[0039] The present invention has the following beneficial effects:

[0040] The present invention takes into account that in the actual cell preparation process, different premise factor data have different effects on different qualities, and the more similar the premise factor data are, the more similar the quality inspection result data are. According to the distribution of premise factor data and quality inspection result data at different culture generations in real-time cell preparation, the influence coefficient of each premise factor data in the real-time cell preparation process on each quality inspection result data, as well as the intergenerational influence factors between different culture generations are obtained, reflecting the stability and transfer efficiency of the premise factor data in the cell passage process; according to the differences in each premise factor data between the same culture generations of different historical batches of cell preparations, and the intergenerational influence factors and influence coefficients of each quality inspection result data corresponding to real-time cell preparation, highly similar historical cells corresponding to real-time cell preparation are screened out, which is helpful to find the state that is most similar to the real-time cell preparation. For historical cell preparation that is close; according to the correlation between the quality inspection result data of real-time cell preparation and the corresponding different highly similar historical cell preparations at the same culture generations, the quality inspection difference result data of the different generations are screened out, and the cell quality inspection result data with difference problems are discovered in time, thereby enhancing the traceability of the cell preparation process; according to the quality inspection difference result data of the different generations between real-time cell preparation and different highly similar historical cell preparations, as well as each of the premise factor data of each other generation before the different generation, combined with the generational impact factor and impact coefficient of each premise factor data on the quality inspection difference result data, the difference influence degree of each premise factor data before the real-time cell preparation different generation on the quality inspection difference result data is obtained, which helps to deeply understand which premise factors have a significant impact on abnormal quality inspection result data; digital management of real-time cell preparation is performed. The present invention improves the management effect of cell preparation related data by obtaining the accurate influence of premise factor data on quality inspection result data during cell preparation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 A flowchart of an implementation method of a high-reliability digital management platform for cell preparation and storage provided by one embodiment of the present invention;

[0043] Figure 2 A flow chart of a method for obtaining an intergenerational influence factor according to one embodiment of the present invention is provided;

[0044] Figure 3 A flow chart of a method for obtaining the degree of difference impact provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0045] To further illustrate the technical means and effects employed by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a high-reliability digital management platform for cell preparation and storage proposed by the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The following describes in detail a high-reliability digital management platform for cell preparation and storage provided by the present invention with reference to the accompanying drawings.

[0048] See also Figure 1 , which shows a flow chart of an implementation method of a high-reliability digital management platform for cell preparation and storage provided by one embodiment of the present invention, the method specifically comprising:

[0049] Step S1: Acquire multiple prerequisite factor data and quality inspection result data of multiple cell preparation batches and multiple generations of culture in a time sequence.

[0050] In an embodiment of the present invention, cell preparation is a process of obtaining cells by collecting original samples for cell proliferation. In order to ensure the quality and efficiency of cell preparation, the cell preparation process is traced and data analyzed based on a digital management platform; wherein, the digital management platform includes main functions such as data acquisition and storage, data preservation and display, big data analysis, differential data traceability and authority management. The data acquisition and storage function can automatically collect data from various links such as cell preparation, culture and storage, including sample information, culture environment parameters, test results, etc., to ensure the integrity and accuracy of the data, and support manual uploading or modification of collected data; the data preservation and display function has efficient data storage capabilities, and supports long-term preservation of large amounts of data through cloud databases, which facilitates subsequent data tracing and analysis; at the same time, the system also has powerful data analysis and visualization capabilities, and can convert complex cell data into intuitive charts and images, such as bar charts, line charts, scatter plots, etc., through data Visualization allows users to see the trends, distribution and correlations of data more clearly, facilitating data analysis and decision-making; the big data analysis function uses advanced big data analysis technology to deeply mine and analyze massive data in the cell preparation and storage process, compare real-time data with historical data, and quickly identify possible process deviations, providing data support for process optimization and improving the quality and efficiency of cell preparation; the differential data traceability function enables the system to automatically trace the relevant data when deviations in cell quality are detected, and analyze the prerequisite factors that may cause the differences, such as culture medium components, added factors, environmental parameters, etc., to provide a clear direction for quality improvement; through permission management, enterprises can ensure that only authorized personnel can access specific data and functional modules. For example, financial personnel can only access modules related to financial management, and R&D personnel can only access modules related to R&D data, which helps protect the company's sensitive information and intellectual property rights and prevent data leakage and abuse.

[0051] In each batch of cell preparation and each generation of culture, the quality inspection result data of the final cell output is obtained for analysis, and a variety of prerequisite factor data and quality inspection result data of each batch of cell preparation and each generation of culture are obtained in time sequence; among which, the prerequisite factor data include culture medium, added factors, cell type, sample collection time, transportation method, transportation duration, etc.; the quality inspection result data include stage inventory rate, cell crawling speed, expression data of various markers and other detection data.

[0052] Step S2: Based on the distribution of prerequisite factor data and quality inspection result data at different culture generations in real-time cell preparation, the influence coefficient of each prerequisite factor data in the real-time cell preparation process on each quality inspection result data, as well as the intergenerational influence factor between different culture generations, are obtained.

[0053] In the actual cell preparation process, different premise factor data have different impacts on different qualities. The more similar the premise factor data, the more similar the quality inspection result data. By analyzing the distribution of the data, the degree of influence of the premise factor data on the quality inspection result data is evaluated; based on the distribution of premise factor data and quality inspection result data at different culture generations in real-time cell preparation, the influence coefficient of each premise factor data on each quality inspection result data in the real-time cell preparation process is obtained.

[0054] Preferably, in one embodiment of the present invention, the method for obtaining the influence coefficient includes:

[0055] For each prerequisite factor data or quality inspection result data, obtain the prerequisite factor data sequence of the real-time cell preparation and the corresponding historical cell preparation of the same type in each generation of culture, as well as the quality inspection result data sequence in accordance with the time sequence;

[0056] Obtaining the ratio of the standard deviations between the premise factor data series and the quality inspection result data series of different generations as the first ratio; obtaining the covariance between the premise factor data series and the quality inspection result data series as the influence weight;

[0057] The product between the first ratio and the influence weight is calculated and normalized as the influence coefficient of each prerequisite factor data of the real-time cell preparation process on each quality inspection result data.

[0058] Based on this, the influence coefficient of each prerequisite factor data on each quality inspection result data in multiple generations of culture during the real-time cell preparation process is obtained, that is, the influence coefficient of each prerequisite factor data on each quality inspection result data in each generation of real-time cell preparation is obtained.

[0059] It should be noted that, in one embodiment of the present invention, in order to avoid data overlap and confusion due to different cell types, it is necessary to select cells of the same type for analysis. Cells can be selected from areas such as the umbilical cord, placenta, umbilical cord blood, or peripheral blood. If the cells are in the same area, they are considered to be the same type of cells for analysis; historical cell preparation refers to the selection of previous cell preparations based on real-time cell preparation.

[0060] In one embodiment of the present invention, the influence coefficient is expressed as follows:

[0061] ;

[0062] in, Indicates the Next generation culture The data of the prerequisite factors The influence coefficient of seed quality inspection result data; Indicates the Next generation culture The standard deviation of the quality inspection result data series; Indicates the Next generation culture The standard deviation of the data series of the antecedent factors; represents the covariance function; Indicates the real-time cell preparation and the corresponding historical cell preparation of the same type in the Next generation culture A sequence of premise factor data; Indicates the real-time cell preparation and the corresponding historical cell preparation of the same type in the Next generation culture A sequence consisting of quality inspection result data; represents the normalization function; Indicates the adjustment parameter.

[0063] In the formula of influence coefficient, In addition This is to avoid the formula denominator being 0, which would make the formula meaningless, and the empirical value is 0.01. The larger the standard deviation of the quality inspection result data, the more dispersed the quality inspection result data distribution is, and the greater the data difference is. The smaller the standard deviation of the premise factor data, the smaller the premise factor data distribution is, the more concentrated the data is, and the smaller the data difference is. In other words, the smaller the standard deviation of the premise factor data, the larger the standard deviation of the quality inspection result data, the smaller the change in the premise factor data, the greater the change in the quality inspection result data, and the larger the impact coefficient. It represents the covariance between the premise factor data sequence and the quality inspection result data sequence, that is, the influence weight. The larger the covariance, the greater the difference between the data, the greater the influence weight, and the greater the influence coefficient.

[0064] During cell preparation, in order to obtain a sufficient number of cells and allow the cells to be adequately acclimated, cells will be cultured for multiple generations. Under different generations of culture, changes in prerequisite factors will cause different effects on the cells in subsequent generations of culture. By considering the influence coefficient of different prerequisite factor data on each quality inspection result data, the influence on the transmission of quality inspection result data between different generations is reflected, and the intergenerational influence factor of each prerequisite factor data on each quality inspection result data in each generation of culture in real-time cell preparation is quantified.

[0065] Preferably, in one embodiment of the present invention, the method for obtaining the intergenerational influence factor can be found in Figure 2 , which shows a flow chart of a method for obtaining an intergenerational influence factor, including:

[0066] Step S201: Obtain the average of the influence coefficients of all prerequisite factor data on each quality inspection result data of real-time cell preparation in each generation of culture as the average influence level.

[0067] The overall impact of the prerequisite factor data on the quality inspection result data in each generation of cultivation is quantified by calculating the mean.

[0068] Step S202: within the neighborhood of each generation of culture, selecting the maximum value of the influence coefficient of each prerequisite factor data of the real-time cell preparation on each quality inspection result data in all generations of culture.

[0069] The maximum value reflects the maximum change or impact that the premise factor may cause within the neighborhood range of each generation of cultivation, which helps to identify key influencing factors.

[0070] It should be noted that, in one embodiment of the present invention, the neighborhood range of each generation of culture is based on each generation of culture and is composed of the neighborhood range of all other subsequent generations of culture. In other embodiments of the present invention, the size of the neighborhood range can be set according to specific circumstances, and is not limited or elaborated here.

[0071] Step S203: Based on the influence coefficient of each prerequisite factor data in each generation on each quality inspection result data in different generations within the neighborhood, combined with the average influence level and the maximum influence coefficient, the intergenerational influence factor of each prerequisite factor data on each quality inspection result data in each generation of real-time cell preparation is obtained. The influence coefficient is positively correlated with the intergenerational influence factor, and the average influence level and the maximum influence coefficient are both negatively correlated with the intergenerational influence factor.

[0072] The average impact level and the maximum impact coefficient are combined to comprehensively evaluate the overall impact of the prerequisite factors on cell culture results. The impact coefficient provides a local perspective on the impact of the prerequisite factors on cell culture results at different generations of culture, which helps to understand the timeliness and stability of the impact of the prerequisite factors.

[0073] It should be noted that, in one embodiment of the present invention, the method for obtaining the influence coefficient of each prerequisite factor data in each generation on each quality inspection result data in each generation within the neighborhood range is: by obtaining each prerequisite factor data sequence of real-time cell preparation and corresponding historical cell preparation of the same type in each generation of culture, as well as each quality inspection result data sequence of each generation of culture within the neighborhood range in time sequence, and performing corresponding calculations on the subsequent influence coefficients.

[0074] In one embodiment of the present invention, the formula for the intergenerational influence factor is expressed as:

[0075] ;

[0076] in, Indicates real-time cell preparation Next generation culture The data of the prerequisite factors Intergenerational impact factors of seed quality inspection result data; Indicates the The number of other generations within the neighborhood of the generation culture; Indicates real-time cell preparation Next generation culture The data of the prerequisite factors Next generation culture The influence coefficient of seed quality inspection result data; Indicates the All the prerequisite factors data under the first generation culture The average impact level of seed quality inspection result data; Indicates in Within the neighborhood of the generation culture, the maximum value of the influence coefficient of each prerequisite factor data on each quality inspection result data in all generation cultures of real-time cell preparation is selected; Represents the normalization function.

[0077] In the formula of intergenerational impact factor, It represents the product of the maximum value of the influence coefficient and the average influence level. The larger the maximum value of the influence coefficient, the greater the average influence level, the greater the possibility of being affected by the same generation culture, and the smaller the intergenerational influence factor, which is negatively correlated. Next generation culture The data of the prerequisite factors Next generation culture The greater the influence coefficient of the seed quality inspection result data, the greater the degree of influence between different generations of cultivation, and the greater the intergenerational influence factor, which is positively correlated.

[0078] Step S3: Based on the differences in each prerequisite factor data between different batches of historical cell preparations at the same culture generation, and the generational influencing factors and influence coefficients of each quality inspection result data corresponding to the real-time cell preparation, the highly similar historical cells corresponding to the real-time cell preparation are screened out; based on the correlation between the quality inspection result data of the real-time cell preparation and the corresponding different highly similar historical cell preparations at the same culture generation, the quality inspection difference result data of different generations are screened out.

[0079] By analyzing the differences in the data of each prerequisite factor between each historical cell preparation and the real-time cell preparation at the same culture generation, the similarity between cells can be reflected. The greater the difference, the smaller the similarity. Generation culture is one of the important factors affecting cell quality. With the increase of culture generations, cells may experience problems such as slowed proliferation rate, decreased genetic stability, loss of differentiation ability and cell aging. The impact coefficient reflects the specific impact of the prerequisite factor on cell quality. By introducing two coefficients, the similarity and difference between different cell preparations can be more comprehensively evaluated, which helps to screen out historical cell preparations with higher similarity. According to the differences in the data of each prerequisite factor between different batches of historical cell preparations at the same culture generation, as well as the generational impact factor and impact coefficient of each quality inspection result data corresponding to the real-time cell preparation, highly similar historical cells corresponding to the real-time cell preparation are screened out.

[0080] Preferably, in one embodiment of the present invention, the method for obtaining highly similar historical cells includes:

[0081] Based on the differences in each prerequisite factor data between different batches of historical cell preparations at the same culture generation, as well as the intergenerational influencing factors and influencing coefficients of each quality inspection result data corresponding to real-time cell preparations, the similarity of each quality inspection result data between real-time cell preparations and different batches of historical cell preparations was obtained;

[0082] For any quality inspection result data, if the result similarity between the quality inspection result data of the real-time cell preparation and any batch of historical cell preparation is greater than the preset similarity threshold, the corresponding batch of historical cells will be regarded as high-similar historical cells.

[0083] It should be noted that, in one embodiment of the present invention, each quality inspection result data is analyzed to obtain corresponding relevant high-similarity historical cells for data analysis; the preset similarity threshold is set to 0.7; in other embodiments of the present invention, the size of the preset similarity threshold can be set according to the specific situation, and is not limited or elaborated here.

[0084] Preferably, in one embodiment of the present invention, the method for obtaining result similarity includes:

[0085] The result similarity is obtained according to the formula for obtaining the result similarity. The formula for obtaining the result similarity is:

[0086] ;

[0087] in, Represents real-time cell preparation and Batch history between cell preparations The similarity of the quality inspection result data; Indicates the number of types of premise factor data; represents the number of passages cultured in real-time cell preparation; Indicates real-time cell preparation Next generation The data of the prerequisite factors Intergenerational impact factors of seed quality inspection result data; Indicates real-time cell preparation Daixiadi The data of the prerequisite factors The influence coefficient of seed quality inspection result data; Real-time cell preparation In the Daixiadi Precondition data; Indicates the Batch history cell preparation Daixiadi Precondition data; Indicates taking the absolute value; Indicates the adjustment parameter.

[0088] In the formula for obtaining the result similarity, Plus In order to avoid the denominator of the formula being 0, the formula becomes meaningless, and the empirical value is taken as 0.01; Real-time cell preparation Hedi Batch history between cell preparations Daixiadi The greater the difference in the premise factor data, the greater the difference, the greater the difference in the premise factor data, the impact on the quality inspection result data, and the smaller the similarity of the quality inspection result data; the larger the intergenerational influence factor, the larger the influence coefficient, indicating that the change in the premise factor data under the corresponding training generation will be more transmitted to the subsequent generations, and the impact on the quality inspection result data is similar, and the greater the importance of the premise factor data change to the similarity of the quality inspection result data.

[0089] Result similarity is an important indicator to measure the degree of similarity between historical cell preparation and real-time cell preparation. The higher the result similarity, the higher the consistency between the cell preparations in terms of culture conditions, cell status, etc., which helps to discover potential culture problems. Based on the correlation between the quality inspection result data of the real-time cell preparation and the corresponding different highly similar historical cell preparations at the same culture generation, the quality inspection difference result data of different generations are screened out.

[0090] Preferably, in one embodiment of the present invention, the method for obtaining quality inspection difference result data of the difference generation includes:

[0091] Based on the differences in quality inspection results between real-time cell preparation and corresponding different highly similar historical cell preparations at the same culture passage, as well as the similarity of the results, the deviation of the culture results of each quality inspection result data of the real-time cell preparation at each passage is obtained;

[0092] It should be noted that, in one embodiment of the present invention, by analyzing the differences and similarities between the quality inspection result data, the correlation between the quality inspection result data can be analyzed. The greater the similarity of the results, the greater the analytical credibility of the correlation between the quality inspection result data. Therefore, the greater the difference, the smaller the correlation between the quality inspection result data, and the greater the culture deviation. The method for obtaining the culture result deviation includes:

[0093] Obtain the squared difference between the real-time cell preparation and each highly similar historical cell preparation for each quality inspection result data at each generation, as well as the result similarity;

[0094] The mean of the product of the similarity and squared difference between the real-time cell preparation and the corresponding different highly similar historical cell preparations was obtained, and the square root of the mean result was calculated as the culture result deviation of each quality inspection result data of the real-time cell preparation at each generation.

[0095] In one embodiment of the present invention, the formula for the deviation of the culture result is expressed as:

[0096] ;

[0097] in, Indicates real-time cell preparation Daixiadi Deviation of culture results of seed quality inspection data; Real-time cell preparation In the Daixiadi Seed quality inspection result data; Indicates the Batch high similarity historical cell preparation Daixiadi Seed quality inspection result data; Represents real-time cell preparation and Highly similar historical cell preparations between batches The similarity of the quality inspection result data; Indicates the number of highly similar historical cells corresponding to the quality inspection result data; Indicates square root.

[0098] In the formula for the deviation of culture results, the greater the similarity of the results, the more credible the analysis of the squared difference value is; Represents real-time cell preparation and Highly similar historical cell preparations between batches Daixiadi The larger the squared difference value is, the greater the difference between the quality inspection result data is, and the more likely there is a quality problem.

[0099] Based on this, targeted analysis of different quality inspection result data can be performed to obtain the culture result deviation of various quality inspection result data of real-time cell preparation at different culture generations.

[0100] The deviation of the culture results of each quality inspection result data of real-time cell preparation at each generation is normalized. If the normalized result is greater than the preset deviation threshold, the quality inspection result data of the corresponding generation is used as the quality inspection difference result data of the difference generation.

[0101] It should be noted that, in one embodiment of the present invention, the preset deviation threshold is set to 0.7; in other embodiments of the present invention, the size of the preset deviation threshold can be set according to specific circumstances, which is not limited or elaborated here.

[0102] Step S4: Based on the quality inspection difference result data of the different generations between the real-time cell preparation and the different highly similar historical cell preparations, as well as each premise factor data of each other generation before the different generation, combined with the generational impact factor and impact coefficient of each premise factor data on the quality inspection difference result data, the degree of difference in influence of each premise factor data on the quality inspection difference result data before the different generation of real-time cell preparation is obtained.

[0103] Result similarity is an important indicator to measure the degree of similarity of quality inspection difference result data between different cell preparations. The greater the similarity, the more consistent the prerequisite factors between cell preparations. Taking into account the generational influence factor and influence coefficient can help to accurately assess the degree of difference impact.

[0104] Preferably, in one embodiment of the present invention, the method for obtaining the difference impact degree can be found in Figure 3 , which shows a flow chart of a method for obtaining the degree of difference impact, including:

[0105] Step S301: Based on the quality inspection difference result data of the different generations between the real-time cell preparation and different highly similar historical cell preparations, and the data of each prerequisite factor of each other generation before the different generation corresponding to the highly similar historical cell preparation, obtain the prediction data of each prerequisite factor corresponding to each other generation.

[0106] Preferably, in one embodiment of the present invention, the method for obtaining the premise factor prediction data includes:

[0107] Obtaining the result similarity of the quality inspection difference result data of the difference generation between the real-time cell preparation and each highly similar historical cell preparation, and the product of each prerequisite factor data of each other generation of the corresponding highly similar historical cell preparation before the difference generation, as a first product;

[0108] The mean of the first product between the real-time cell preparation and different highly similar historical cell preparations was obtained as the prediction data for each antecedent factor corresponding to each other generation.

[0109] In one embodiment of the present invention, the formula for the premise factor prediction data is expressed as:

[0110] ;

[0111] in, Indicates the quality inspection difference result data , difference generation Other generations before Next Prediction data of antecedent factors; Represents real-time cell preparation and Corresponding quality inspection difference results data between batches of highly similar historical cell preparations The similarity of the results; Indicates the Batch high similarity historical cell preparation in different generations Other generations before Next Precondition data; Indicates the number of cells with high similarity history.

[0112] In the formula for the premise factor prediction data, the Corresponding quality inspection difference result data between batch-highly similar historical cell preparation and real-time cell preparation The similarity of the results The larger the value, the more consistent the prerequisite factors between cell preparations are, and the more consistent the quality inspection result data are. , No. Batch high similarity to historical cell preparation in other generations Next The higher the credibility of the premise factor data, the Batch high similarity history cell preparation before the difference generation other generations Next Precondition data The larger the value, the greater the predicted value of the corresponding prerequisite factor at each other generation when performing real-time cell preparation analysis.

[0113] Step S302: Obtain the difference between each prerequisite factor data and the prerequisite factor prediction data in each other generation of real-time cell preparation before the corresponding difference generation as the prerequisite factor difference value.

[0114] During the real-time cell preparation process, the actual data of each prerequisite factor may change due to operational differences, equipment status, environmental factors, etc. By comparing the acquired prerequisite factor data with the predicted data based on historical cell preparation data, the prerequisite factor difference value can be calculated to monitor the process stability of the cell preparation process. If the difference value is large, it may mean that there are unstable factors in the preparation process.

[0115] Step S303: Obtain the generational impact factor and impact coefficient of each prerequisite factor data on the quality inspection difference result data in each other generation, as well as the product between the prerequisite factor difference values, and normalize them as the degree of difference in influence of each prerequisite factor data on the quality inspection difference result data before the real-time cell preparation difference generation.

[0116] In one embodiment of the present invention, the formula for the difference impact degree is expressed as:

[0117] ;

[0118] in, Indicates the real-time cell preparation corresponding to the difference generation Other generations before Next Precondition factor data and quality inspection difference result data the degree of differential impact; Indicates the corresponding other generations Next The data of the prerequisite factors The influence coefficient of the seed quality inspection difference result data; Indicates that real-time cell preparation is Next The data of the prerequisite factors Intergenerational impact factors of seed quality inspection difference results data; Real-time cell preparation Other generations before the difference generation Next Precondition data; Indicates the quality inspection difference result data , difference generation Other generations before Next Prediction data of antecedent factors; Represents the normalization function.

[0119] In the formula for the degree of difference impact, Real-time cell preparation In the difference generation Other generations before Next The difference between the premise factor data and the premise factor prediction data, the greater the difference, the greater the impact on the quality inspection difference result data, the greater the difference between the training process and the prediction, the greater the generational influence factor, the greater the influence coefficient, the greater the credibility of the difference, the corresponding generation next first The greater the impact of the premise factor data on the quality inspection difference result data.

[0120] Based on this, the quality inspection difference result data of each differential generation were analyzed in turn to obtain the degree of difference in the influence of each prerequisite factor data on the quality inspection difference result data in each other generation before the corresponding differential generation of real-time cell preparation.

[0121] Step S5: Digitally manage the real-time cell preparation based on the degree of influence of the premise factor data on the quality inspection difference result data.

[0122] By analyzing the degree of differential impact, we can more accurately understand the changing patterns and influencing factors in the cell preparation process, thereby improving the accuracy and reliability of the cell preparation process.

[0123] It should be noted that, in another embodiment of the present invention, after obtaining the degree of difference influence, the real-time cell preparation is digitally managed, including: for the quality inspection difference result data of each difference generation, selecting the three with the largest values ​​of the degree of difference influence of different premise factor data of all generations before the corresponding difference generation on the quality inspection difference result data, and using the premise factor data of the corresponding generation as the source of the quality inspection difference result data; and then, the premise factor data of the corresponding generation culture in the cell culture process can be targeted analyzed, which helps to provide data guidance for the optimization of the subsequent cell preparation evaluation system and improve the efficiency and quality of digital management of cell preparation.

[0124] In summary, the present invention collects and analyzes the premise factor data and quality inspection result data such as production and environment in the cell preparation process, obtains the influence coefficient of the premise factor data of the real-time cell preparation process on the quality inspection result data, and the intergenerational influence factor between different culture generations; combines and compares the correlation between the premise factor data and the quality inspection result data between real-time cell preparation and historical different batches of cell preparation; monitors the influence of subtle changes in the premise factor data on the quality inspection result data, and performs digital management such as traceability, monitoring, and early warning on the real-time cell preparation process, while providing data reference for subsequent process improvements. The present invention improves the management effect of cell preparation-related data by obtaining the accurate influence of the premise factor data on the quality inspection result data in the cell preparation process.

[0125] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0126] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A high-reliability digital management platform for cell preparation and storage, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the following steps are implemented: Obtain multiple prerequisite factor data and quality inspection result data for multiple generations of cell preparation in a time series; Based on the distribution of prerequisite factor data and quality inspection result data at different culture generations in real-time cell preparation, the influence coefficient of each prerequisite factor data in the real-time cell preparation process on each quality inspection result data, as well as the intergenerational influence factor between different culture generations, are obtained; Based on the differences in each prerequisite factor data between different batches of historical cell preparations at the same culture passage, as well as the generational influencing factors and influence coefficients of each quality inspection result data corresponding to real-time cell preparation, highly similar historical cells corresponding to real-time cell preparation are screened out; based on the correlation between the quality inspection result data of real-time cell preparation and the corresponding different highly similar historical cell preparations at the same culture passage, quality inspection difference result data of different generations are screened out; Based on the quality inspection difference result data of the different generations between the real-time cell preparation and different highly similar historical cell preparations, as well as the data of each prerequisite factor of each other generation before the different generation, combined with the generational impact factor and impact coefficient of each prerequisite factor data on the quality inspection difference result data, the degree of difference in the impact of each prerequisite factor data before the different generation of the real-time cell preparation on the quality inspection difference result data is obtained; Digital management of real-time cell preparation is performed based on the degree of influence of the prerequisite factor data on the quality inspection difference result data.

2. A high-reliability digital management platform for cell preparation and storage according to claim 1, characterized in that: The method for obtaining the influence coefficient includes: For each prerequisite factor data or quality inspection result data, obtain the prerequisite factor data sequence of the real-time cell preparation and the corresponding historical cell preparation of the same type in each generation of culture, as well as the quality inspection result data sequence in accordance with the time sequence; Obtaining the ratio of the standard deviations between the premise factor data series and the quality inspection result data series of different generations as the first ratio; obtaining the covariance between the premise factor data series and the quality inspection result data series as the influence weight; The product between the first ratio and the influence weight is calculated and normalized as the influence coefficient of each prerequisite factor data of the real-time cell preparation process on each quality inspection result data.

3. A high-reliability digital management platform for cell preparation and storage according to claim 2, characterized in that: The method for obtaining the intergenerational influence factor includes: Obtain the mean influence coefficient of all prerequisite factor data on each quality inspection result data in real-time cell preparation at each generation of culture as the average influence level; Within the neighborhood of each generation of culture, the maximum value of the influence coefficient of each prerequisite factor data on each quality inspection result data in all generations of culture of real-time cell preparation is selected; According to the influence coefficient of each prerequisite factor data in each generation on each quality inspection result data in different generations within the neighborhood, combined with the average influence level and the maximum influence coefficient, the intergenerational influence factor of each prerequisite factor data on each quality inspection result data in each generation of real-time cell preparation was obtained. The influence coefficient was positively correlated with the intergenerational influence factor, and the average influence level and the maximum influence coefficient were negatively correlated with the intergenerational influence factor.

4. A high-reliability digital management platform for cell preparation and storage according to claim 1, characterized in that: The method for obtaining the highly similar historical cells comprises: Based on the differences in each prerequisite factor data between different batches of historical cell preparations at the same culture generation, as well as the intergenerational influencing factors and influencing coefficients of each quality inspection result data corresponding to real-time cell preparations, the similarity of each quality inspection result data between real-time cell preparations and different batches of historical cell preparations was obtained; For any quality inspection result data, if the result similarity between the quality inspection result data of the real-time cell preparation and any batch of historical cell preparation is greater than the preset similarity threshold, the corresponding batch of historical cells will be regarded as high-similar historical cells.

5. A high-reliability digital management platform for cell preparation and storage according to claim 4, characterized in that: The method for obtaining the result similarity includes: The result similarity is obtained according to the formula for obtaining the result similarity. The formula for obtaining the result similarity is: ;in, Represents real-time cell preparation and Batch history between cell preparations The similarity of the quality inspection result data; Indicates the number of types of premise factor data; represents the number of passages cultured in real-time cell preparation; Indicates real-time cell preparation Daixiadi The data of the prerequisite factors Intergenerational impact factors of seed quality inspection result data; Indicates real-time cell preparation Daixiadi The data of the prerequisite factors The influence coefficient of seed quality inspection result data; Real-time cell preparation In the Daixiadi Precondition data; Indicates the Batch history cell preparation Daixiadi Precondition data; Indicates taking the absolute value; Indicates the adjustment parameter.

6. A high-reliability digital management platform for cell preparation and storage according to claim 5, characterized in that: The method for obtaining the quality inspection difference result data of the difference generation includes: Based on the differences in quality inspection results between real-time cell preparation and corresponding different highly similar historical cell preparations at the same culture passage, as well as the similarity of the results, the deviation of the culture results of each quality inspection result data of the real-time cell preparation at each passage is obtained; The deviation of the culture results of each quality inspection result data of real-time cell preparation at each generation is normalized. If the normalized result is greater than the preset deviation threshold, the quality inspection result data of the corresponding generation is used as the quality inspection difference result data of the difference generation.

7. A high-reliability digital management platform for cell preparation and storage according to claim 6, characterized in that: The method for obtaining the culture result deviation includes: Obtain the squared difference between the real-time cell preparation and each highly similar historical cell preparation for each quality inspection result data at each generation, as well as the result similarity; The mean of the product of the similarity and squared difference between the real-time cell preparation and the corresponding different highly similar historical cell preparations was obtained, and the square root of the mean result was calculated as the culture result deviation of each quality inspection result data of the real-time cell preparation at each generation.

8. A high-reliability digital management platform for cell preparation and storage according to claim 6, characterized in that: The method for obtaining the difference impact degree includes: According to the quality inspection difference result data of the different generations between the real-time cell preparation and different highly similar historical cell preparations, and the data of each prerequisite factor of each other generation of the highly similar historical cell preparation before the different generation, the prediction data of each prerequisite factor corresponding to each other generation are obtained; Obtain the difference between each prerequisite factor data and the prerequisite factor prediction data in each other generation of real-time cell preparation before the corresponding difference generation as the prerequisite factor difference value; The generational impact factor and impact coefficient of each prerequisite factor data on the quality inspection difference result data in each other generation, as well as the product between the prerequisite factor difference values, are obtained and normalized as the degree of difference in influence of each prerequisite factor data on the quality inspection difference result data before the real-time cell preparation difference generation.

9. A high-reliability digital management platform for cell preparation and storage according to claim 8, characterized in that: The method for obtaining the premise factor prediction data includes: Obtaining the result similarity of the quality inspection difference result data of the difference generation between the real-time cell preparation and each highly similar historical cell preparation, and the product of each prerequisite factor data of each other generation of the corresponding highly similar historical cell preparation before the difference generation, as a first product; The mean of the first product between the real-time cell preparation and different highly similar historical cell preparations was obtained as the prediction data for each antecedent factor corresponding to each other generation.

10. A high-reliability digital management platform for cell preparation and storage according to claim 4, characterized in that: The preset similarity threshold is 0.7.

Images