Drug production quality identification method and platform based on big data

Abstract

The present invention discloses a drug production quality identification method and platform based on big data, and relates to the field of data processing technology. The method includes: obtaining the target raw material and target solvent for producing the target nano drug, constructing the production control optimization constraint, optimizing the production control index, and obtaining the optimal control decision; activating the chromatographic probe, performing chromatographic detection on the pre-finished nano drug produced according to the optimal control decision, and comparing the chromatogram of the reference product, analyzing and generating the pre-finished product impurity index; if the pre-finished product impurity index is within the predetermined impurity index threshold, drying and post-processing the pre-finished nano drug. The present invention solves the technical problem in the prior art that the quality of the final product is affected by the poor control of the raw material particle size and distribution of the nano drug production, and achieves the technical effect of optimizing the raw material particle size control through big data analysis, optimizing the particle size and particle size distribution, and thus improving the product quality.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a drug production quality identification method and platform based on big data. Background Art

[0002] Quality control and impurity identification have always been a complex and challenging area in the drug production process, especially in the production of nanomedicines. Due to their characteristics, nanomedicines often have extremely high requirements for the purity, particle size and control accuracy of raw materials during production. However, multivariate factors in drug production, such as raw material quality, solvent characteristics, temperature, humidity, etc., will affect the particle size and distribution of raw materials. Particle size and distribution is a key parameter of nanomedicines and will affect the quality of the final product, including affecting the bioavailability, stability and efficacy of the drug. Summary of the invention

[0003] The present application provides a drug production quality identification method and platform based on big data, which is used to solve the technical problem in the prior art that the quality of the final product is affected by the poor control of the particle size and distribution of raw materials in the production of nanomedicines.

[0004] In a first aspect of the present application, a method for identifying drug production quality based on big data is provided, the method comprising: obtaining a target drug material, the target drug material referring to a target raw material and a target solvent for producing a target nano drug; using the target material characteristic information formed by the raw material characteristic information of the target raw material and the solvent characteristic information of the target solvent as a production control optimization constraint; using a predetermined particle size evaluation function as a production control optimization evaluation, optimizing the production control index of the target nano drug under the production control optimization constraint, and obtaining an optimal control decision; activating a chromatographic probe, performing chromatographic detection on a pre-finished nano drug produced according to the optimal control decision, and obtaining a pre-finished chromatogram; comparing the pre-finished chromatogram with a reference product chromatogram to obtain a chromatogram comparison result, the chromatogram comparison result including a plurality of differential chromatographic segments; calling an impurity prediction model to analyze a first segment characteristic set of a first differential chromatographic segment among the plurality of differential chromatographic segments, and obtaining a first impurity index; determining whether the pre-finished impurity index of the pre-finished nano drug obtained by adding the first impurity index is within a predetermined impurity index threshold; if so, issuing a drug forming instruction, and performing drying post-processing on the pre-finished nano drug based on the drug forming instruction.

[0005] The second aspect of the present application provides a drug production quality identification platform based on big data, the platform comprising: a target drug material acquisition module, the target drug material acquisition module is used to acquire target drug materials, the target drug materials refer to target raw materials and target solvents for producing target nano-drugs; a production control optimization constraint setting module, the production control optimization constraint setting module is used to use the target material characteristic information composed of the raw material characteristic information of the target raw material and the solvent characteristic information of the target solvent as the production control optimization constraint; a production control index optimization module, the production control index optimization module is used to use a predetermined particle size evaluation function as the production control optimization evaluation, optimize the production control index of the target nano-drug under the production control optimization constraint, and obtain the optimal control decision; a pre-finished product chromatogram generation module, the pre-finished product chromatogram generation module is used to activate the chromatographic probe to produce the target nano-drug according to the optimal control decision The pre-finished product nano-drug is subjected to chromatographic detection to obtain a pre-finished product chromatogram; a chromatogram comparison module, the chromatogram comparison module is used to compare the pre-finished product chromatogram with the reference product chromatogram to obtain a chromatogram comparison result, wherein the chromatogram comparison result includes multiple differential chromatographic segments; an impurity index prediction module, the impurity index prediction module is used to call an impurity prediction model to analyze a first segment feature set of a first differential chromatographic segment among the multiple differential chromatographic segments to obtain a first impurity index; a pre-finished product impurity index judgment module, the pre-finished product impurity index judgment module is used to judge whether the pre-finished product impurity index of the pre-finished product nano-drug obtained by adding the first impurity index is within a predetermined impurity index threshold; a post-drying processing module, the post-drying processing module is used to issue a drug forming instruction if the pre-finished product impurity index is within the predetermined impurity index threshold, and perform post-drying processing on the pre-finished product nano-drug based on the drug forming instruction.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The drug production quality identification method and platform based on big data provided by the present application relate to the field of data processing technology. By obtaining the target raw materials and target solvents for producing the target nano-drug, the production control indicators are optimized, pre-production is carried out according to the optimal control decision, and the chromatographic probe is activated to perform chromatographic detection on the pre-finished nano-drug, and the chromatogram of the benchmark product is compared to analyze the impurity index of the pre-finished nano-drug. If it is within the predetermined impurity index threshold, the pre-finished nano-drug is dried and post-processed, which solves the technical problem in the prior art that the quality of the final product is affected by the poor control of the raw material particle size and distribution of the nano-drug production, and realizes the technical effect of optimizing the raw material particle size control through big data analysis, optimizing the particle size and particle size distribution, and thus improving the product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0009] Figure 1 A schematic diagram of a process for identifying drug production quality based on big data provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of a process for obtaining an optimal control decision in a drug production quality identification method based on big data provided in an embodiment of the present application;

[0011] Figure 3 A schematic diagram of the structure of a drug production quality identification platform based on big data provided in an embodiment of the present application.

[0012] Explanation of the accompanying drawings: target drug material acquisition module 11, production control optimization constraint setting module 12, production control index optimization module 13, pre-finished product chromatogram generation module 14, chromatogram comparison module 15, impurity index prediction module 16, pre-finished product impurity index judgment module 17, post-drying processing module 18. DETAILED DESCRIPTION

[0013] The present application provides a drug production quality identification method and platform based on big data, which is used to solve the technical problem in the prior art that the quality of the final product is affected by the poor control of the particle size and distribution of raw materials in the production of nanomedicines.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0016] Embodiment 1

[0017] like Figure 1 As shown, the present application provides a method for identifying drug production quality based on big data, the method comprising:

[0018] P10: Obtain target drug materials, where the target drug materials refer to target raw materials and target solvents for producing target nanomedicines.

[0019] Specifically, in the initial stage of drug production quality control, the primary task is to accurately obtain the target drug material. The target drug materials referred to here are the two core elements that constitute the cornerstone of future nanomedicines: target raw materials and target solvents. Among them, the target raw materials are the direct source of active ingredients of the drug, and their purity, stability and physicochemical properties play a decisive role in the final efficacy of the drug. In order to ensure the quality of raw materials, advanced material identification technologies such as near-infrared spectroscopy and nuclear magnetic resonance (NMR) can be used. These technologies can quickly and accurately identify the composition and characteristics of raw materials to ensure that each batch of raw materials meets the established quality standards.

[0020] At the same time, the selection of the target solvent is also crucial. The solvent not only participates in the drug synthesis process as a reaction medium, but may also affect the crystal structure, solubility and stability of the drug. Therefore, when selecting a solvent, factors such as its solubility, toxicity, volatility and compatibility with raw materials need to be considered comprehensively. With the help of computer simulation technology and solvent screening database, the optimal solvent combination is screened out, laying a solid foundation for the preparation of nanomedicines.

[0021] P20: The target material characteristic information formed by the raw material characteristic information of the target raw material and the solvent characteristic information of the target solvent is used as the production control optimization constraint. The raw material characteristic information includes raw material purity, raw material particle size and raw material crystallinity, and the solvent characteristic information includes solvent type, solvent polarity and solvent concentration.

[0022] It should be understood that after the target drug material is determined, the key link of production control optimization is to build production control constraints based on the target material characteristic information. The core of this process is to fully integrate the raw material characteristic information of the target raw material with the solvent characteristic information of the target solvent to form a complete set of target material characteristic information system as the cornerstone of subsequent production control optimization.

[0023] Specifically, the raw material characteristic information covers three key dimensions: raw material purity, raw material particle size and raw material crystallinity. Raw material purity refers to the percentage of active ingredients in the raw material, which directly affects the efficacy and safety of the drug. High-precision analytical instruments, such as high-performance liquid chromatography (HPLC) and gas chromatography (GC), can be used to conduct strict purity tests on raw materials. Raw material particle size refers to the size distribution of raw material particles, which is related to the dissolution rate and bioavailability of the drug in the body. Advanced equipment such as laser particle size analyzers can accurately measure the particle size distribution of raw materials. The raw material crystallinity reflects the degree of perfection of the raw material crystal structure, which is crucial to the stability and long-term preservation of the drug. X-ray diffraction (XRD) technology is a powerful tool for evaluating the crystallinity of raw materials.

[0024] On the other hand, the solvent characteristic information includes solvent type, solvent polarity and solvent concentration. The selection of solvent type needs to be based on comprehensive consideration of factors such as its compatibility with raw materials, the effect on reaction rate and yield. Solvent polarity determines the size of its dissolving capacity, which in turn affects the solubility and stability of the drug. Solvent concentration is an adjustable parameter. By optimizing the solvent concentration, the reaction rate and the purity of the product can be controlled. In order to accurately grasp the solvent characteristic information, solvent databases and computer simulation technology can be used to conduct detailed analysis and prediction of the various properties of the solvent.

[0025] By combining the raw material characteristic information and the solvent characteristic information, a complete target material characteristic information can be constructed. This characteristic information not only covers the key properties of the target raw materials and solvents, but is also used to guide the subsequent production control process. Specifically, during the production process, the optimization of production control (i.e., determination of the optimal parameters in the production process) is carried out based on the target material characteristic information. This means that by constraining these characteristics, the system can ensure the optimization of parameters at each step in the production process and maximize the quality and stability of nanomedicines.

[0026] P30: Using a predetermined particle size evaluation function as a production control optimization evaluation, optimizing the production control index of the target nanomedicine under the production control optimization constraints, and obtaining the optimal control decision.

[0027] Further, such as Figure 2 As shown, step P30 of the embodiment of the present application also includes:

[0028] P31: Extract the first indicator among the production control indicators, and match the first effective control threshold of the first indicator; P32: Form a first control decision based on the first indicator value in the first effective control threshold randomly extracted; P33: Obtain a first control simulation particle size value sequence, the first control simulation particle size value sequence refers to a particle size value sequence of multiple production stages obtained by simulating production control of the target drug material based on the first control decision; P34: Analyze the particle size value comparison result between the first control simulation particle size value sequence and the reference particle size value sequence according to the predetermined particle size evaluation function, and obtain a first particle size fitness; P35: Continue to iterate until the first particle size fitness meets the predetermined fitness target, and use the first control decision at that time as the optimal control decision.

[0029] Optionally, after the constraints of production control optimization are established, a predetermined particle size evaluation function is used as the core evaluation criterion to perform refined optimization of the production control indicators of the target nano-drug. The predetermined particle size evaluation function can be constructed by machine learning or statistical analysis tools, and the ideal state of particle size can be predicted based on a large amount of historical data and experimental results. This process aims to find the optimal control decision that can maximize the requirements of product quality and production efficiency by continuously optimizing production control parameters.

[0030] To achieve this goal, the production control indicators are first broken down and the key indicators, namely the first indicators, are extracted one by one. This indicator is one of the key parameters affecting the quality of nanomedicines, such as temperature, pressure or stirring speed. For each first indicator, the corresponding effective control threshold range is pre-set. These thresholds are based on historical data, expert experience and theoretical calculations to ensure the stability and controllability of the production process.

[0031] Next, we enter the iterative optimization phase. In each iteration, a value within the first effective control threshold is randomly selected as the target value of the first indicator, and a preliminary production control decision, i.e., the first control decision, is constructed based on this value. This decision contains the specific set values ​​of all production control indicators and is an important basis for guiding actual production.

[0032] In order to verify the effectiveness of the first control decision, simulation technology was used to simulate production. By simulating each link in the production process, the particle size value sequence of the target drug material at different production stages based on the first control decision was obtained, that is, the first control simulation particle size value sequence. This sequence intuitively reflects the impact of production control decisions on drug particle size.

[0033] Subsequently, the first control simulation particle size value sequence and the reference particle size value sequence are compared and analyzed using a predetermined particle size evaluation function. The reference particle size value sequence represents the particle size distribution of the drug under an ideal state and is an important reference for evaluating the actual production effect. By calculating the particle size value comparison results between the two, the first particle size fitness is obtained, which quantifies the degree of closeness between the first control decision and the ideal state.

[0034] If the first particle size fitness does not reach the predetermined fitness target, it means that the current control decision still has room for optimization. Therefore, the iterative optimization is continued, the target value of the first indicator is continuously adjusted, and the production control decision is rebuilt. This process is repeated until the optimal control decision is found that makes the first particle size fitness meet the predetermined fitness target. This control decision can achieve the optimal control of the drug particle size under the premise of meeting the production constraints, thereby ensuring the high-quality production of the target nano drug.

[0035] Furthermore, step P33 of the embodiment of the present application also includes:

[0036] P33-1: Construct an initial simulation model according to the production control optimization constraints; P33-2: Perform granular simulation on the initial simulation model based on the first control decision to obtain a first control simulation record, which includes the first control simulation granularity value sequence.

[0037] In a possible embodiment of the present application, in order to more accurately evaluate the impact of production control decisions on the target nano-drug production process, the simulation process is further refined. First, an initial simulation model is constructed based on the previously determined production control optimization constraints. This model is a digital virtual system that can simulate the actual production environment. It integrates information on material properties, equipment performance, environmental conditions, and production processes, and can fully reflect the interactions and changes between various parameters in the production process.

[0038] The key to building the initial simulation model is to accurately capture the key variables in the production process and the mathematical relationships between them. Using advanced modeling techniques and algorithms, such as system dynamics and discrete event simulation, the production control optimization constraints are transformed into constraints in the simulation model to ensure that the simulation process can truly reflect the constraints of actual production.

[0039] After the initial simulation model is constructed, the system will perform granular simulation on the model based on the first control decision. This means that the system uses the simulation model to simulate the particle size change process of the target drug material at different stages. At each production stage, the change in particle size during the simulation process is recorded, and finally the first control simulation record is obtained. This record contains the sequence of first control simulation particle size values ​​for multiple production stages, that is, the particle size of nanomedicines at different stages. Particle size is one of the key quality indicators of nanomedicines. Particle size directly affects the release rate, stability and bioavailability of drugs. Therefore, the recording of particle size data during the simulation process is particularly important.

[0040] The first control simulation record is an important basis for evaluating the effectiveness of the first control decision. It not only contains the sequence of particle size values ​​changing over time or production stages (i.e., the first control simulation particle size value sequence), but may also contain other parameter information related to particle size, such as temperature, pressure, stirring speed, etc. This information provides rich data support for subsequent analysis and optimization.

[0041] Furthermore, step P34 of the embodiment of the present application also includes:

[0042] P34-1: Read the predetermined production weight ratio, and match the multiple weight coefficients of the multiple production stages in sequence according to the predetermined production weight ratio; P34-2: Extract the first production stage of the multiple production stages, and match the first weight coefficient of the first production stage among the multiple weight coefficients; P34-3: Match the first simulation granularity value and the first reference granularity value of the first production stage in the first control simulation granularity value sequence and the reference product granularity value sequence in sequence; P34-4: Obtain the first granularity difference between the first simulation granularity value and the first reference granularity value, and obtain the first stage granularity fitness in combination with the first weight coefficient; P34-5: Take the mean of the first stage granularity fitness as the first granularity fitness.

[0043] Specifically, in order to further refine the calculation of granularity fitness, according to the predetermined production weight ratio, the differences between the granularity values ​​and the benchmark values ​​of multiple production stages are gradually analyzed, and then the granularity fitness of each stage is calculated, and finally the overall first granularity fitness is obtained.

[0044] First, read the predetermined production weight ratio, which is formulated based on historical data, expert experience or production process requirements, and reflects the importance of each production stage in particle size control. Through the predetermined production weight ratio, a corresponding weight coefficient can be assigned to each production stage so that different attention can be given in subsequent calculations.

[0045] Next, the first production stage among the multiple production stages is extracted, and the first weight coefficient corresponding to the stage is matched among the multiple weight coefficients. This weight coefficient will be used for weighted processing when subsequently calculating the granularity fitness of the stage.

[0046] Then, the first simulation particle size value and the first reference particle size value of the first production stage are matched in the first control simulation particle size value sequence and the reference product particle size value sequence. These two values ​​represent the actual simulation result and the particle size standard under the ideal state, respectively, and are important bases for evaluating the particle size control effect of this stage.

[0047] Subsequently, the first granularity difference between the first simulation granularity value and the first reference granularity value is calculated, that is, the deviation of the simulation result of this stage from the ideal state, and the first stage granularity fitness is calculated in combination with the first weight coefficient, for example, the first granularity difference is multiplied by the first weight coefficient to obtain the first stage granularity fitness. This fitness value reflects the degree of deviation between the actual effect of granularity control in the first production stage and the ideal state, and is adjusted by the weight coefficient to reflect its importance.

[0048] Finally, the mean of the particle size fitness of all production stages (although only the first stage is described here, all stages will be involved in actual applications) is taken as the first particle size fitness. This mean combines the effect of particle size control in all production stages and can provide an overall evaluation index. By comparing the first particle size fitness under different control decisions, the optimal control decision can be selected to achieve optimal particle size control in the production process of the target nanomedicine.

[0049] Furthermore, step P34 of the embodiment of the present application also includes:

[0050] P34-6: Obtain the first reaction condition record in the first control simulation record; P34-7: Obtain the first reaction condition index based on the first air density and the first reaction pH value obtained by analyzing the first reaction condition record; P34-8: Adjust the first particle size adaptability based on the first reaction condition index.

[0051] Optionally, the environmental conditions during the production process also have an important impact on the particle size control effect. Therefore, the analysis and adjustment mechanism of the reaction conditions is introduced to ensure the comprehensiveness and accuracy of the evaluation results. This process can further optimize the production control of nanomedicines to make it more in line with actual production conditions.

[0052] First, the first reaction condition record related to the first production stage is extracted from the first control simulation record. These records include but are not limited to key parameters such as temperature, pressure, humidity, time and pH value, which together constitute the environmental conditions of the production stage. Slight changes in these environmental conditions may have a significant effect on the particle size of the drug. For example, temperature affects the activity of the reactants, pressure changes the equilibrium point of the reaction, humidity affects the evaporation rate of the solvent and the stability of the reaction environment, and pH value affects the acid-base balance of the chemical reaction.

[0053] Next, an in-depth analysis was performed based on the first reaction condition record. In particular, the first air density and the first reaction pH value under this condition were calculated using physical and chemical principles. These two parameters are important indicators that reflect the impact of environmental conditions on particle size control. Air density is related to the mass transfer rate of materials, while pH directly affects the balance and rate of chemical reactions, which in turn affects the particle size distribution of the product. Through these analyses, the system is able to identify the impact of specific reaction conditions on production control.

[0054] Then, the first reaction condition index is calculated according to the first air density and the first reaction pH value. Exemplarily, the optimal air density threshold and the optimal pH value threshold can be set according to the historical production environment data, and the difference between the first air density and the optimal air density threshold, and the difference between the first reaction pH value and the optimal pH value threshold are calculated respectively, and the two differences are weighted according to the importance of the two indicators to obtain the first reaction condition index. This index is a quantitative index that integrates multiple environmental parameters and is used to evaluate the overall impact of environmental conditions on the particle size control effect. By introducing this index, the difference in particle size control effect under different environmental conditions can be more accurately reflected.

[0055] Finally, the first granularity fitness is adjusted with the first reaction condition index. Specifically, the first reaction condition index is used as a weight factor and multiplied with the first granularity fitness to obtain an adjusted granularity fitness that takes into account the influence of environmental conditions. The adjusted granularity fitness not only reflects the effect of the control decision itself, but also takes into account the potential impact of environmental conditions on the granularity control effect, so it is more comprehensive and accurate.

[0056] P40: Activate the chromatographic probe to perform chromatographic detection on the pre-finished nano-drug produced according to the optimal control decision to obtain a chromatogram of the pre-finished product.

[0057] It should be understood that the chromatographic detection of the pre-finished nanomedicine produced according to the optimal control decision by the chromatographic probe to obtain the pre-finished chromatogram can accurately identify the distribution of chemical components in the pre-finished product to ensure that it meets the quality standards.

[0058] First, activate the chromatographic probe. Chromatographic probes usually refer to equipment used to analyze compound components, such as high performance liquid chromatography (HPLC) or gas chromatography (GC). These probes can accurately detect the presence and relative proportions of each component in a mixture. Chromatographic technology separates different components by their different retention times in the chromatographic column, and then captures the signal through the detector to generate a complete chromatogram.

[0059] Subsequently, the pre-made nanomedicine is injected into the chromatography system. In the chromatographic column, the components in the sample are separated according to the difference in their interaction with the stationary phase and the mobile phase. In this process, different components pass through the chromatographic column at different speeds, forming specific chromatographic peaks, thereby revealing the chemical composition and purity information of the sample.

[0060] Next, the chromatographic system detects the separated components and generates a pre-finished product chromatogram. This chromatogram is a two-dimensional graph, with the horizontal axis representing the retention time (the residence time of each component in the chromatographic column) and the vertical axis representing the signal intensity (reflecting the concentration of the component). The peak value and peak area of ​​the chromatogram represent the presence and content of different components. By analyzing this chromatogram, it can be determined whether the various components in the pre-finished product meet the design requirements, and the differences with the standard can be further compared. The pre-finished product chromatogram is not only an important basis for evaluating the quality of nanomedicines, but also the basic data for subsequent process optimization such as purification, modification or formulation adjustment.

[0061] P50: Compare the pre-finished product chromatogram with the reference product chromatogram to obtain a chromatogram comparison result, wherein the chromatogram comparison result includes a plurality of differential chromatogram segments.

[0062] Specifically, by comparing the chromatogram of the pre-finished product with the chromatogram of the reference product (i.e., the chromatogram of the standard pre-finished product), the difference between the component distribution of the pre-finished nanomedicine and the standard product is detected. This step is an important part of quality control to ensure that the produced nanomedicine is consistent with the design standard.

[0063] First, use chromatography analysis software or tools to load the pre-finished product chromatogram and the reference product chromatogram into the same interface for intuitive comparison. Chromatography analysis software usually has powerful data processing and image analysis functions, which can automatically or semi-automatically identify key features in the chromatogram, such as the position, height, width and peak shape of the chromatographic peak.

[0064] Subsequently, a comprehensive and detailed comparison of the two chromatograms is performed using specialized comparative analysis methods, such as the use of mathematical methods such as the Pearson correlation coefficient or the Euclidean distance to quantify the similarity between the two chromatograms. During this process, the software compares the signal intensities of the two spectra point by point and calculates the differences between them. These differences can be identified by analyzing the peak position, peak height, and peak area in the chromatogram. For example, if the peak position of a component in the pre-finished product in the chromatogram is different from that of the reference product, it indicates that the structure or properties of the component in the pre-finished product have changed; if the peak height or peak area is significantly different, it indicates that the concentration of the component deviates from the standard.

[0065] During the comparison process, the system will automatically identify multiple differential chromatographic segments. For example, by setting reasonable thresholds, areas with significant differences that may represent product quality problems are identified, namely differential chromatographic segments. Each differential segment represents the difference between the pre-finished product and the reference product in a certain component. For example, a certain segment in the chromatogram may show a different retention time (i.e., the retention time of the component in the chromatographic column), which may reflect that the chemical structure of the component in the pre-finished product has changed. By identifying these differential segments, the system can further analyze the problems that may arise in the production process of nano-drug pre-finished products.

[0066] P60: Retrieving the impurity prediction model to analyze the first segment feature set of the first differential chromatographic segment in the plurality of differential chromatographic segments to obtain a first impurity index, wherein the first segment feature set includes the first segment peak height and the first segment area.

[0067] Furthermore, step P60 of the embodiment of the present application also includes:

[0068] P61: traverse the first segment feature set in the target impurity database of the target nanomedicine to obtain the first traversed impurity; P62: when the first traversed impurity does not belong to the predetermined key abnormal impurity, calculate and obtain the first impurity index.

[0069] It should be understood that an in-depth impurity analysis is performed on the multiple differential chromatographic segments identified in the chromatogram comparison results, especially the first differential chromatographic segment. First, a pre-built and trained impurity prediction model is retrieved. This model is based on a large amount of experimental data and chemical knowledge, and can accurately identify and predict the impurity components and their contents in the drug. For the first differential chromatographic segment, its first segment feature set is extracted, which includes two key parameters: peak height and peak area of ​​the segment. The peak height reflects the concentration or content of the impurity component, while the peak area is related to the total amount or relative content of the impurity. These two parameters together constitute the basic data for evaluating the presence or absence of impurities and their importance.

[0070] Next, the first segment feature set is traversed and searched in the target impurity database of the target nanomedicine. This database contains various known impurity information that may exist in the target nanomedicine, including their chemical structure, physical properties, and toxicity assessment. Through traversal search, the first traversal impurity matching the first segment feature set is found, that is, the impurity component that may correspond to the difference chromatographic segment is preliminarily determined.

[0071] When the first-pass impurity is identified, the system will determine whether the impurity belongs to the predetermined key abnormal impurities. These key abnormal impurities usually refer to ingredients that are toxic and harmful or may have a significant impact on the quality of the drug. Although their content may not be high, their presence may cause serious quality problems. Therefore, even if the calculated first impurity index is low, as long as the impurity belongs to the predetermined key abnormal impurities, the emergency treatment process will be triggered.

[0072] If the first pass impurity is not a critical abnormal impurity, proceed to calculate the first impurity index. This step is to quantify its proportion in the pre-finished product by analyzing the content (peak height and area) of the impurity. If the impurity is a regular impurity and the content is low, it can be allowed to exist within the threshold range, but if the impurity content is too high, this problem will be marked for further processing.

[0073] P70: Determine whether the pre-finished product impurity index of the pre-finished product nanomedicine obtained by adding the first impurity index is within a predetermined impurity index threshold.

[0074] Specifically, by comprehensively evaluating the impurity index of each differential chromatographic segment in the pre-finished nano-drug, the pre-finished product impurity index of the batch of pre-finished products is finally obtained, ensuring that the impurity level of the pre-finished product is controlled within the pre-set standard range, thereby ensuring the safety and quality of the product.

[0075] First, the first impurity index obtained by analyzing each differential chromatographic segment is summed with other impurity indices. These impurity indices are obtained based on the feature sets of different segments in the chromatogram and analyzed by the impurity prediction model, representing the relative content and impact of different impurities in the pre-finished product. By summing up the various impurity indices, a comprehensive pre-finished product impurity index is generated, which is used to quantify the overall level of all potential impurities in the pre-finished product.

[0076] In order to determine whether the impurity index of the pre-finished product is within the acceptable range, it is necessary to compare it with the predetermined impurity index threshold. The predetermined impurity index threshold is determined based on factors such as the quality standards, safety requirements and actual production process of the drug. It represents the maximum allowable limit of impurity content in the drug. Exceeding this limit may have an adverse effect on the efficacy, stability or safety of the drug.

[0077] When making the comparison, a direct comparison method is used, that is, a one-to-one numerical comparison is made between the pre-finished product impurity index and the predetermined impurity index threshold. If the pre-finished product impurity index is lower than or equal to the threshold, it means that the impurity content is within an acceptable range, the product quality meets the requirements, and the next production or quality control process can be entered. On the contrary, if the pre-finished product impurity index exceeds the predetermined threshold, it means that the impurity content in the batch of pre-finished products is too high, and there may be quality problems. The system will trigger corresponding quality control measures, such as suspension of production, rework, or further impurity removal treatment.

[0078] P80: If so, a drug forming instruction is issued, and the pre-finished nano drug is dried and post-processed based on the drug forming instruction.

[0079] Optionally, if the system determines that the pre-finished product impurity index of the pre-finished nano drug is within a set predetermined threshold range, indicating that the product meets the quality standard, a drug forming instruction is issued. This instruction initiates the drug forming and subsequent drying process to ensure that the pre-finished product is converted into a final product.

[0080] Specifically, when the impurity index of the pre-finished product is confirmed to meet the quality requirements, the system will automatically issue a drug forming instruction. This instruction indicates that the quality assessment of the pre-finished nano drug has been completed and the product can enter the final processing stage. Drug forming is a crucial step that processes the nano drug pre-finished product into a finished drug that meets the specifications and is ready for packaging and distribution.

[0081] After the drug molding instruction is issued, the system will perform post-drying treatment on the nanodrug according to the preset molding process. Post-drying treatment is a key link to ensure the stability and quality of nanodrugs. Specific operations may include a variety of physical or chemical means, such as vacuum drying, freeze drying, spray drying, etc. The selection of these drying technologies depends on the physical and chemical properties of the drug and the requirements of the production process. The goal of the drying process is to remove excess solvent or moisture in the nanodrug, ensure that the drug has appropriate water content and stability, and avoid deterioration or degradation during storage or transportation. At the same time, the drying process can also improve the fluidity and processability of the drug, making the drug easier to carry out subsequent packaging processing.

[0082] Through drug molding instructions, only pre-finished nano drugs that meet quality standards can enter the molding process. Through strict impurity control and scientific post-drying treatment, a solid foundation is laid for the production of high-quality, safe and effective nano drug products.

[0083] In summary, the embodiments of the present application have at least the following technical effects:

[0084] The present application obtains the target raw materials and target solvents for producing the target nano-drug, optimizes the production control indicators, performs pre-production based on the optimal control decision, activates the chromatographic probe to perform chromatographic detection on the pre-finished nano-drug, compares the chromatogram of the benchmark product, analyzes the impurity index of the pre-finished product, and if it is within the predetermined impurity index threshold, performs drying and post-processing on the pre-finished nano-drug.

[0085] The technical effect of optimizing raw material particle size control, optimizing particle size and particle size distribution, and thus improving product quality has been achieved through big data analysis.

[0086] Embodiment 2

[0087] Based on the same inventive concept as the drug production quality identification method based on big data in the aforementioned embodiment, Figure 3 As shown, the present application provides a drug production quality identification platform based on big data, and the platform and method embodiments in the embodiments of the present application are based on the same inventive concept. Among them, the platform includes:

[0088] The target drug material acquisition module 11 is used to acquire target drug materials, and the target drug materials refer to target raw materials and target solvents for producing target nano-drugs.

[0089] The production control optimization constraint setting module 12 is used to use the target material characteristic information formed by the raw material characteristic information of the target raw material and the solvent characteristic information of the target solvent as the production control optimization constraint.

[0090] The production control index optimization module 13 is used to use a predetermined particle size evaluation function as a production control optimization evaluation, optimize the production control index of the target nano drug under the production control optimization constraint, and obtain the optimal control decision.

[0091] The pre-finished product chromatogram generation module 14 is used to activate the chromatographic probe, perform chromatographic detection on the pre-finished product nano-drug produced according to the optimal control decision, and obtain the pre-finished product chromatogram.

[0092] The chromatogram comparison module 15 is used to compare the pre-finished product chromatogram with the reference product chromatogram to obtain a chromatogram comparison result, wherein the chromatogram comparison result includes a plurality of difference chromatogram segments.

[0093] The impurity index prediction module 16 is used to call the impurity prediction model to analyze the first segment feature set of the first difference chromatographic segment in the multiple difference chromatographic segments to obtain a first impurity index.

[0094] The pre-finished product impurity index judging module 17 is used to judge whether the pre-finished product impurity index of the pre-finished product nano-drug obtained by adding the first impurity index is within a predetermined impurity index threshold.

[0095] The post-drying processing module 18 is used to issue a drug forming instruction if the pre-finished product impurity index is within a predetermined impurity index threshold, and perform post-drying processing on the pre-finished product nano drug based on the drug forming instruction.

[0096] Furthermore, the production control optimization constraint setting module 12 is also used to perform the following steps:

[0097] The raw material characteristic information includes raw material purity, raw material particle size and raw material crystallinity, and the solvent characteristic information includes solvent type, solvent polarity and solvent concentration.

[0098] Furthermore, the production control index optimization module 13 is also used to perform the following steps:

[0099] Extract the first indicator among the production control indicators, and match the first effective control threshold of the first indicator; form a first control decision according to the first indicator value in the first effective control threshold randomly extracted; obtain a first control simulation particle size value sequence, the first control simulation particle size value sequence refers to a particle size value sequence of multiple production stages obtained by simulating production control of the target drug material based on the first control decision; analyze the particle size value comparison result between the first control simulation particle size value sequence and the reference particle size value sequence according to the predetermined particle size evaluation function to obtain a first particle size fitness; continue to iterate until the first particle size fitness meets the predetermined fitness target, and use the first control decision at that time as the optimal control decision.

[0100] Furthermore, the production control index optimization module 13 is also used to perform the following steps:

[0101] An initial simulation model is constructed according to the production control optimization constraint; and a granular simulation is performed on the initial simulation model based on the first control decision to obtain a first control simulation record, wherein the first control simulation record includes the first control simulation granularity value sequence.

[0102] Furthermore, the production control index optimization module 13 is also used to perform the following steps:

[0103] Read a predetermined production weight ratio, and match multiple weight coefficients of the multiple production stages in sequence according to the predetermined production weight ratio; extract the first production stage of the multiple production stages, and match the first weight coefficient of the first production stage among the multiple weight coefficients; match the first simulation granularity value and the first reference granularity value of the first production stage in sequence in the first control simulation granularity value sequence and the reference product granularity value sequence; obtain the first granularity difference between the first simulation granularity value and the first reference granularity value, and obtain the first stage granularity fitness in combination with the first weight coefficient; take the mean of the first stage granularity fitness as the first granularity fitness.

[0104] Furthermore, the production control index optimization module 13 is also used to perform the following steps:

[0105] Obtain a first reaction condition record in the first control simulation record; obtain a first reaction condition index based on analyzing a first air density and a first reaction pH value obtained from the first reaction condition record; and adjust the first particle size adaptability according to the first reaction condition index.

[0106] Furthermore, the impurity index prediction module 16 is further configured to perform the following steps:

[0107] The first segment characteristic set includes a first segment peak height and a first segment area.

[0108] Furthermore, the impurity index prediction module 16 is further configured to perform the following steps:

[0109] The first segment feature set is traversed in the target impurity database of the target nanomedicine to obtain a first traversed impurity; when the first traversed impurity does not belong to a predetermined key abnormal impurity, the first impurity index is calculated.

[0110] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0112] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A drug production quality identification method based on big data, characterized in that: include: Obtaining target drug materials, wherein the target drug materials refer to target raw materials and target solvents for producing target nano-drugs; The target material characteristic information formed by the raw material characteristic information of the target raw material and the solvent characteristic information of the target solvent is used as a production control optimization constraint; Using a predetermined particle size evaluation function as a production control optimization evaluation, optimizing the production control index of the target nano-drug under the production control optimization constraint, and obtaining an optimal control decision; activating a chromatographic probe to perform chromatographic detection on the pre-finished nano-drug produced according to the optimal control decision to obtain a chromatogram of the pre-finished product; Comparing the pre-finished product chromatogram with the reference product chromatogram to obtain a chromatogram comparison result, wherein the chromatogram comparison result includes a plurality of differential chromatogram segments; Retrieving an impurity prediction model to analyze a first segment feature set of a first differential chromatographic segment among the plurality of differential chromatographic segments to obtain a first impurity index; Determining whether the pre-finished product impurity index of the pre-finished product nanodrug obtained by adding the first impurity index is within a predetermined impurity index threshold; If yes, a drug forming instruction is issued, and the pre-finished nano drug is dried and post-processed based on the drug forming instruction; The raw material characteristic information includes raw material purity, raw material particle size and raw material crystallinity, and the solvent characteristic information includes solvent type, solvent polarity and solvent concentration; The drug production quality identification method based on big data also includes: Extracting a first indicator from the production control indicators and matching a first effective control threshold of the first indicator; Forming a first control decision according to a first indicator value randomly extracted from the first effective control threshold; Acquire a first control simulation particle size value sequence, where the first control simulation particle size value sequence refers to a particle size value sequence of multiple production stages obtained by performing simulated production control on the target drug material based on the first control decision; Analyze the particle size value comparison result of the first control simulation particle size value sequence and the reference product particle size value sequence according to the predetermined particle size evaluation function to obtain a first particle size fitness; Continue iterating until the first granularity fitness meets a predetermined fitness target, and use the first control decision at that time as the optimal control decision; Wherein, the first segment feature set includes the first segment peak height and the first segment area; The drug production quality identification method based on big data also includes: Traversing the first segment feature set in the target impurity database of the target nanomedicine to obtain a first traversed impurity; When the first traversal impurity does not belong to the predetermined key abnormal impurity, the first impurity index is calculated, and as long as the impurity belongs to the predetermined key abnormal impurity, the emergency processing process will be triggered.

2. The drug production quality identification method based on big data according to claim 1 is characterized in that: include: Constructing an initial simulation model according to the production control optimization constraints; Based on the first control decision, a granular simulation is performed on the initial simulation model to obtain a first control simulation record, wherein the first control simulation record includes the first control simulation granularity value sequence.

3. The drug production quality identification method based on big data according to claim 1 is characterized in that: include: Reading a predetermined production weight ratio, and sequentially matching and obtaining a plurality of weight coefficients of the plurality of production stages according to the predetermined production weight ratio; extracting a first production stage of the plurality of production stages, and matching a first weight coefficient of the first production stage among the plurality of weight coefficients; sequentially matching the first simulation granularity value and the first reference granularity value of the first production stage in the first control simulation granularity value sequence and the reference product granularity value sequence; Obtaining a first granularity difference between the first simulation granularity value and the first reference granularity value, and obtaining a first-stage granularity adaptability in combination with the first weight coefficient; The mean value of the first stage granularity fitness is taken as the first granularity fitness.

4. The drug production quality identification method based on big data according to claim 2 is characterized in that: include: Obtaining a first reaction condition record in the first control simulation record; Obtaining a first reaction condition index based on analyzing a first air density and a first reaction pH value obtained from the first reaction condition record; The first particle size adaptability is adjusted using the first reaction condition index.

5. The drug production quality identification platform based on big data is characterized by: For executing the method according to any one of claims 1 to 4, the platform comprises: A target drug material acquisition module, which is used to acquire target drug materials, and the target drug materials refer to target raw materials and target solvents for producing target nano-drugs; A production control optimization constraint setting module, wherein the production control optimization constraint setting module is used to use the target material characteristic information formed by the raw material characteristic information of the target raw material and the solvent characteristic information of the target solvent as the production control optimization constraint; A production control index optimization module, wherein the production control index optimization module is used to optimize the production control index of the target nano-drug under the production control optimization constraint by using a predetermined granularity evaluation function as a production control optimization evaluation to obtain an optimal control decision; A pre-finished product chromatogram generation module, wherein the pre-finished product chromatogram generation module is used to activate a chromatographic probe to perform chromatographic detection on the pre-finished product nano-drug produced according to the optimal control decision to obtain a pre-finished product chromatogram; A chromatogram comparison module, the chromatogram comparison module is used to compare the pre-finished product chromatogram with the reference product chromatogram to obtain a chromatogram comparison result, the chromatogram comparison result including a plurality of differential chromatogram segments; An impurity index prediction module, the impurity index prediction module is used to call an impurity prediction model to analyze a first segment feature set of a first difference chromatographic segment among the multiple difference chromatographic segments to obtain a first impurity index; A pre-finished product impurity index determination module, the pre-finished product impurity index determination module is used to determine whether the pre-finished product impurity index of the pre-finished product nano-drug obtained by adding the first impurity index is within a predetermined impurity index threshold; A post-drying processing module is used to issue a drug forming instruction if the pre-finished product impurity index is within a predetermined impurity index threshold, and perform post-drying processing on the pre-finished product nano drug based on the drug forming instruction.

Images