Intelligent inventory management method and system for blood products based on medical big data

Abstract

The present invention relates to the technical field of medical blood inventory management, and in particular to a method and system for intelligent inventory management of blood products based on medical big data. The method comprises the following steps: by collecting and desensitizing historical medical record data of dermatology, generating desensitized historical medical record data, and further extracting the blood product treatment dosage data of skin diseases between different cases. Based on the desensitized historical data, a simulation evaluation of the difference in recovery capacity is performed to generate recovery capacity difference evaluation data, and thus predict future dosage demand data. Then, the current inventory is optimized in advance in combination with the predicted data to generate forward inventory allocation optimization data. Finally, an automated management architecture is designed for the optimized data, and the management architecture is imported into the medical management system. The present invention makes the medical blood inventory management technology more perfect by optimizing the medical blood inventory management technology.

Description

Technical Field

[0001] The present invention relates to the technical field of medical blood inventory management, and in particular to an intelligent inventory management method and system for blood products based on medical big data. Background Art

[0002] As an indispensable and important resource in hospital clinical treatment, blood products are widely used in many fields such as surgical operations, emergency rescue, anemia treatment, and tumor treatment. Especially in special cases such as hemodialysis, major trauma, cancer treatment, and skin burns, the demand for blood products is sudden and urgent, which poses a huge challenge to the inventory management of blood products. Traditional blood product management methods often rely on manual records and empirical judgments, which are difficult to accurately predict and adjust inventory levels in real time, and easily lead to waste, shortages, or expiration of blood products, which in turn affects the quality and efficiency of clinical treatment. Therefore, intelligent inventory management methods for blood products based on medical big data have emerged. Medical big data covers many aspects of information such as patients' medical records, diagnosis and treatment plans, blood use history, transfusion treatment effects, and blood product inventory. By analyzing and mining these data, accurate prediction and dynamic adjustment of blood product demand can be achieved, the inventory management process can be optimized, and resource waste and shortage can be reduced. Specifically, using big data technology, historical demand trends, disease types, surgical frequency, seasonal changes, and other factors can be analyzed to predict the demand for blood products, thereby achieving scientific and refined inventory management. However, the traditional intelligent inventory management method of blood products based on medical big data has the problem of being unable to accurately predict the patient's later blood product usage through historical dermatology records, and unable to perform accurate pre-scheduling optimization of usage. Summary of the invention

[0003] Based on this, it is necessary to provide an intelligent inventory management method and system for blood products based on medical big data to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a blood product intelligent inventory management method based on medical big data is provided, the method comprising the following steps:

[0005] Step S1: Obtain historical medical records of the dermatology department; desensitize the historical medical records of the dermatology department to obtain desensitized historical medical records; extract the amount of blood products used for the treatment of skin diseases between different types of cases from the desensitized historical medical records to obtain the amount of blood products used for the treatment of skin diseases between different cases;

[0006] Step S2: Performing a simulation evaluation of the recovery ability difference of the skin disease blood product treatment dosage data between different cases based on the historical medical desensitization record data to obtain recovery ability difference evaluation data; performing a subsequent dosage demand prediction based on the recovery ability difference evaluation data to obtain subsequent dosage demand prediction data;

[0007] Step S3: Obtain the inventory of dermatology blood products; optimize the pre-stock allocation of dermatology blood products according to the subsequent usage demand forecast data to obtain pre-stock allocation optimization data;

[0008] Step S4: Design an automated management architecture for the pre-inventory allocation optimization data to obtain a pre-inventory allocation management architecture; send the pre-inventory allocation management architecture to the medical management system to execute intelligent inventory management of blood products.

[0009] The present invention collects all the historical medical data of patients in the dermatology department, including basic information of patients, reasons for medical treatment, treatment process and treatment effect, etc. By comprehensively summarizing these data, basic data can be provided for subsequent treatment analysis, effect evaluation and dosage prediction. The purpose of desensitizing the historical medical records is to protect the privacy of patients, remove sensitive information (such as name, ID number, contact information, etc.), and ensure that personal privacy will not be disclosed during data analysis. Data desensitization not only meets the requirements of laws and regulations, but also provides a safer environment for subsequent data analysis. By analyzing the desensitized historical medical record data, the dosage data of skin blood products (such as plasma, immunoglobulin, blood product treatment, etc.) used in different case types (such as different skin diseases, treatment plans, etc.) are extracted. This process can help understand the differences in the use of blood products in different cases during treatment, and then provide a reference for formulating personalized treatment plans. This step extracts the corresponding skin disease blood product treatment dosage data for the case data of different skin diseases. By analyzing the use of blood products in the treatment of different skin diseases, the impact of different conditions on the demand for blood products can be evaluated, and the types of cases with high or low demand can be identified, providing data support for the optimization of treatment strategies. Based on historical medical records and skin blood product usage data, the differences in recovery ability of different cases during treatment are simulated. By establishing a recovery model (including clinical symptom recovery, immune response recovery, etc.), the effects of different treatment options and blood product usage on patient recovery are evaluated. This evaluation can reveal the differences in the effects of different treatment options and provide a basis for the optimization of future treatment options and personalized treatment. Based on the recovery ability difference evaluation data, predict the patient's future treatment needs, especially in terms of the dosage requirements of skin blood products. Through machine learning, regression analysis and other technologies, the patient's demand for blood products in subsequent treatment can be accurately predicted. This prediction can not only help hospitals rationally plan the inventory and supply of blood products, but also help to rationally allocate medical resources to ensure that patients' treatment needs are met in a timely manner.

[0010] By combining historical usage forecast data with the existing blood product inventory in the hospital, the forward inventory of blood products can be optimized and allocated. This optimization process takes into account future changes in demand for blood products. Through accurate demand forecasting, the inventory configuration is dynamically adjusted so that the inventory can meet actual demand fluctuations and avoid the problem of excess or shortage of inventory. The optimized forward inventory allocation plan can achieve more efficient resource utilization and reduce inventory management costs, while improving the response speed and service quality of medical institutions to sudden patient needs. According to the forward inventory allocation optimization data, the design and implementation of the automated management architecture can further improve the efficiency of blood product inventory management. The automated management architecture uses intelligent technology to realize real-time monitoring, automatic allocation and early warning mechanisms of inventory, reduce manual intervention, and improve management efficiency and accuracy. After connecting this management architecture with the medical management system, the inventory status can be tracked in real time and the optimization instructions can be automatically executed, thereby ensuring the dynamic adjustment and scientific management of blood product inventory, reducing operating costs while ensuring patient treatment needs, and improving the overall medical service level.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: Obtaining historical dermatology consultation record data;

[0013] Step S12: Desensitizing the historical medical records of the dermatology department to obtain desensitized historical medical records;

[0014] Step S13: classify the historical medical desensitization record data by case type to obtain dermatology medical case type classification data;

[0015] Step S14: extract the blood product treatment dosage of skin diseases between different case types from the historical desensitization record data according to the dermatology case classification data, and obtain the blood product treatment dosage data of skin diseases between different cases.

[0016] The present invention provides the basis of data source for the whole process by acquiring the historical medical record data of dermatology. These records include the patient's medical information, treatment process and medication, etc., which provide a comprehensive case background for subsequent analysis. The effective implementation of this link lays the foundation for the accuracy of subsequent steps and the reliability of data, and provides rich materials for further analysis of the characteristics of cases and medication trends. The data of historical medical records of dermatology are desensitized to protect the privacy of patients and ensure compliance with data protection regulations. After desensitization, all sensitive information involving personal privacy will be removed or encrypted, so that the data can be analyzed and used without revealing the identity of the patient. This step is a crucial link in the whole data processing process, which helps to ensure the legal and compliant use of data while maintaining the privacy and security of patients. By classifying the case types of the desensitized dermatology medical record data, different types of cases can be distinguished according to their characteristics. This step can help the system identify different categories of cases, which is convenient for subsequent analysis and processing of each type of case. This classification not only helps with the organization and processing of data, but also provides a structured information framework for understanding the usage patterns and demand for blood products in different case types. The resulting case type classification data can be used to further analyze the amount of skin blood products used in different case categories. This includes specific dosage data for blood products such as plasma, immunoglobulin, and albumin used by patients with skin diseases. This analysis helps understand the differences in the demand for blood products among patients with different diseases or courses of disease. For example, some critically ill patients may require higher doses of blood products, while mild patients may require less. By extracting these data, treatment plans and resource allocation strategies can be formulated more accurately, while providing medication guidance for hospitals or clinics to optimize the efficiency of blood product use.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: extracting physiological parameters from the historical medical consultation desensitization record data to obtain medical consultation physiological parameters;

[0019] Step S22: performing a simulation evaluation of the difference in recovery ability of the blood product treatment dosage data for skin diseases between different cases according to the physiological parameters of the consultation and the historical consultation desensitization record data, and obtaining recovery ability difference evaluation data;

[0020] Step S23: Based on the recovery ability difference evaluation data, blood volume demand pattern analysis is performed on the skin disease blood product treatment dosage data between different cases with different recovery ability differences to obtain blood volume recovery difference demand pattern data;

[0021] Step S24: performing subsequent usage demand prediction on the blood volume recovery difference demand pattern data to obtain subsequent usage demand prediction data.

[0022] The present invention extracts the patient's physiological parameters, such as basic vital signs data such as blood pressure, heart rate, body temperature, and blood oxygen saturation, from the existing historical medical desensitization data. These physiological parameters are crucial for assessing the patient's overall health status and response to treatment. By extracting and analyzing these data, the changing trend of the patient's physical state during the treatment process can be obtained. By combining the patient's physiological parameters with historical medical records, a simulation evaluation is performed on the recovery ability differences of different case types. A prediction model for recovery ability is constructed based on factors such as physical health status and immune response reflected by physiological parameters. For example, some patients may recover slowly due to weak immune systems or complications, and need more skin blood products. The simulation evaluation can reveal the recovery ability differences between different treatment plans and different cases, and predict the amount of blood products required during the treatment process based on these differences. This process not only helps doctors understand the patient's recovery trend, but also provides data support for subsequent treatment adjustments and the refined use of blood products. Pattern analysis is performed on the blood volume requirements of different cases. Different patients have different patterns of demand for blood products during treatment, which is closely related to multiple factors such as the severity of the patient's condition, recovery speed, and immune status. By analyzing these usage demand patterns, it is possible to identify the characteristics of blood product usage required by patient groups with different recovery abilities. For example, patients who recover more slowly require higher blood product usage, while patients who recover faster require less. Through this pattern analysis, hospitals can reasonably adjust treatment strategies according to different case types, optimize resource allocation, and improve treatment outcomes. After obtaining the blood usage demand pattern data for different recovery ability differences, the next step is to make subsequent usage demand forecasts for these data. Using forecasting techniques such as machine learning and time series analysis, accurate forecasts of future blood product demand can be made by combining historical treatment data, recovery ability differences, treatment plans, and other factors. This not only helps hospitals accurately assess the demand for blood products in future treatments, but also avoids problems such as over-purchasing or insufficient inventory.

[0023] Preferably, step S22 includes the following steps:

[0024] Step S221: Obtain the age, height and weight of the patient from the patient's physiological parameters; evaluate the degree of skin damage on the historical desensitization record data to obtain skin damage degree data;

[0025] Step S222: estimating the blood transfusion tolerance per unit time based on the patient's age, height, and weight to obtain the blood transfusion tolerance per unit time;

[0026] Step S223: estimating the skin tension damage gradient of the skin damage degree data to obtain skin tension damage gradient data;

[0027] Step S224: predicting the progressive cycle of injury recovery based on the skin tension injury gradient data according to the blood transfusion tolerance per unit time, and obtaining the progressive cycle data of injury recovery;

[0028] Step S225: Based on the injury recovery progressive cycle data, a recovery capacity difference simulation evaluation is performed on the skin disease blood product treatment dosage data between different cases to obtain recovery capacity difference evaluation data.

[0029] The present invention can provide individual physiological characteristic data for subsequent analysis by extracting the physiological parameters of the visit, such as the patient's age, height and weight. These parameters are crucial for calculating the individual's resource carrying capacity, tolerance and other aspects. At the same time, the evaluation of the degree of skin damage in the historical visit data can provide quantitative indicators for understanding the patient's skin damage. The purpose of this process is to provide the necessary basic data for the subsequent simulation of recovery ability differences to ensure the comprehensiveness and accuracy of the data. By estimating the transfusion tolerance per unit time based on the physiological parameters of the visit (such as age, height, weight), the amount of blood products that the patient can withstand within a specific time can be predicted. This estimation can take into account the influence of individual differences and physiological conditions, making the prediction of the blood product demand for each case more accurate. This step helps to provide a personalized, data-driven basis for the arrangement of blood usage for different cases, avoid excessive or insufficient usage allocation, and ensure the rational allocation of resources. By estimating the skin tension damage gradient of the skin damage degree data, the severity of skin damage and its impact on the recovery process can be more accurately measured. The skin damage gradient reflects the depth and breadth of the damage and can predict the difficulty of damage recovery. This step helps to establish a quantitative injury standard, provides a more scientific basis for subsequent recovery capacity assessment, and helps to optimize the use of blood products and adjust the changes in demand during the recovery process. Based on the estimated transfusion tolerance per unit time and skin tension injury gradient data, the progressive cycle of injury recovery can be predicted. This process predicts the key time nodes in the recovery process by comprehensively considering the patient's physiological conditions and degree of injury, and helps to deduce the changing trends in the duration and amount of blood product use. Based on the recovery progressive cycle data, a simulation assessment of the recovery capacity differences between different cases is performed, and the amount of blood products used is adjusted according to the patient's recovery capacity. This difference simulation assessment allows treatment plans to be tailored to each individual, thereby improving the accuracy of treatment and reducing unnecessary waste of resources.

[0030] Preferably, step S224 includes the following steps:

[0031] Performing fiber elastic loss gradient analysis on skin tension damage gradient data to obtain fiber elastic loss gradient data;

[0032] The cell self-healing rate is simulated and estimated based on the fiber elasticity loss gradient data and the skin tension damage gradient data to obtain the skin cell self-healing rate data;

[0033] Based on the skin cell self-healing rate data, the fiber elasticity loss gradient data is used to calculate the initial tension benchmark for damage recovery, and the initial tension benchmark data for damage recovery is obtained;

[0034] According to the blood transfusion tolerance per unit time, the initial tension baseline data of injury recovery is analyzed for gradual tension recovery increment, and the gradual tension recovery increment data is obtained;

[0035] The progressive cycle of damage recovery is predicted based on the progressive tension recovery increment data to obtain the progressive cycle data of damage recovery.

[0036] The present invention can effectively measure the elastic loss of the fiber structure during the skin damage process by performing fiber elastic loss gradient analysis on the skin tension damage gradient data. Fiber elasticity is an important component of the skin's recovery ability. Therefore, the analysis of the fiber elastic loss gradient provides a scientific basis for evaluating the skin's recovery potential after damage. This step can help to more accurately understand the depth and range of skin damage and the response to external forces during the recovery process, thereby providing data support for subsequent recovery strategies. Based on the fiber elastic loss gradient data and the skin tension damage gradient data, the cell self-healing rate simulation estimation is performed to predict the speed of skin self-repair after damage. The cell self-healing rate directly affects the time and quality of skin recovery. Estimating this process can help clarify the staged changes in skin recovery. This simulation can provide personalized recovery predictions for different damage situations, making the evaluation of the skin recovery process more accurate, thereby providing a basis for the use planning of blood products. Based on the estimated skin cell self-healing rate data, the initial tension benchmark calculation for damage recovery is performed to determine the tension state of the skin in the early stage of recovery. This calculation provides a preliminary reference for the mechanical properties during the recovery process, which can guide the subsequent tracking and analysis of the recovery process. Determining the initial tension baseline helps to evaluate how skin tension is gradually restored at different recovery stages and accurately predict the resource requirements (such as the use of blood products) during the recovery process, thereby reducing overtreatment or waste of resources. Based on the calculation of the initial tension baseline for recovery mentioned above, the incremental analysis of the gradual tension recovery is performed according to the transfusion tolerance per unit time. This analysis process mainly simulates the tension changes at each stage of the recovery process through the gradual evaluation of the initial tension baseline, thereby estimating the support resources required during the recovery period. This step ensures that during the skin tension recovery process, the recovery strategy can be flexibly adjusted according to the individual differences of the patients, making the use of resources more accurate and efficient, and avoiding unnecessary waste or insufficiency caused by too fast or too slow recovery progress. By analyzing the incremental data of gradual tension recovery, the time period of the entire injury recovery process can be predicted. This prediction provides a time frame for the recovery process and arranges resource allocation according to the expected recovery period. By understanding the time nodes of recovery in advance, the supply and use of blood products can be better planned, ensuring that the resource requirements during the recovery period are reasonably arranged, avoiding sudden resource shortages or surpluses, and improving the management efficiency of the recovery process.

[0037] Preferably, step S24 includes the following steps:

[0038] Step S241: using a preset staged recovery demand identification model to perform staged demand pattern hierarchical division on the blood volume recovery difference demand pattern data to obtain stage demand pattern hierarchical division data;

[0039] Step S242: performing a phased key usage change analysis on the blood volume recovery difference demand pattern data according to the phased demand pattern hierarchical division data to obtain phased key usage change data;

[0040] Step S243: performing phased demand fluctuation fitting on the phased key usage change data to obtain phased usage demand fluctuation fitting data;

[0041] Step S244: performing fluctuation inflection point trend analysis on the periodic consumption demand fluctuation fitting data to obtain demand fluctuation inflection point trend data;

[0042] Step S245: performing subsequent usage demand forecasting on the blood volume recovery difference demand pattern data according to the demand fluctuation inflection point trend data to obtain subsequent usage demand forecast data.

[0043] The present invention uses a preset staged recovery demand identification model to analyze the blood volume recovery difference demand pattern data, and performs staged demand pattern hierarchical division. Through this process, the demand pattern in the entire recovery process can be decomposed into multiple stages, and the demand characteristics of each stage are clearly identified. Hierarchical division not only helps to clearly identify the demand differences in each stage, but also provides more accurate staged data support for subsequent detailed analysis and resource planning, and improves the flexibility and accuracy of resource scheduling. Based on the hierarchical division data of the stage demand pattern, the blood volume recovery difference demand pattern data is further analyzed for staged key usage changes. The focus of this analysis is to find the key usage change points in each stage, and identify the change trend and key change amount of blood volume in each stage. By analyzing the key usage changes, the demand fluctuation characteristics of resources in different stages can be better understood, and data support can be provided for subsequent resource allocation and adjustment, ensuring the timeliness and effectiveness of resource supply. The staged key usage change data is subjected to a fitting analysis of demand fluctuations. This process fits the demand fluctuations in each stage through a mathematical model to obtain the trend and law of the usage demand fluctuations in each stage. By fitting the demand fluctuation, it can help predict the fluctuation range of blood volume at different stages, and then provide a scientific basis for more accurate demand forecasting and resource preparation, ensuring that there will be no shortage or surplus of resources due to excessive demand changes during the recovery process. The purpose of performing fluctuation inflection point trend analysis on the phased demand fluctuation fitting data is to identify the inflection point or turning point of demand fluctuation in each recovery stage. Fluctuation inflection points usually reflect sudden changes in demand or changes in phased demand, which is crucial for resource scheduling. By analyzing the trend of demand fluctuation inflection points, we can accurately grasp the key turning points in the recovery process, help predict the trend of future demand changes, and adjust resource allocation in advance to reduce resource shortages or waste. According to the data obtained through the fluctuation inflection point trend analysis, the subsequent demand demand is predicted. This prediction uses the aforementioned phased demand fluctuation data and combines the trend analysis results to make a forward-looking estimate of future resource demand. Through this prediction, it is possible to make advance plans for resource supply in the subsequent stages to ensure the continuous and stable supply of resources. The subsequent demand forecast provides a clear timetable and demand map for resource adjustment in the recovery process of each stage, helping to avoid excessive or insufficient resource consumption and making resource management more efficient and accurate.

[0044] Preferably, step S3 comprises the following steps:

[0045] Step S31: Obtaining the inventory of dermatology blood products;

[0046] Step S32: Estimating the usage fluctuation probability interval for the subsequent usage demand forecast data to obtain the subsequent usage fluctuation probability interval;

[0047] Step S33: Optimize the pre-stock allocation of dermatology blood products according to the subsequent usage fluctuation probability interval and the skin disease blood products treatment usage data between different cases to obtain pre-stock allocation optimization data.

[0048] The present invention obtains the existing inventory data of dermatology blood products in real time. This data provides a basis for subsequent inventory management and resource allocation, ensuring that the existing inventory status can be accurately grasped. Through a comprehensive understanding of the inventory situation, the shortage or surplus of inventory can be discovered in time, laying a data foundation for formulating a reasonable resource allocation plan. Effective inventory management helps to avoid waste or insufficient supply and ensure the efficiency of resource use. After obtaining the inventory data, the next step is to estimate the probability interval of the subsequent usage demand forecast data for the usage fluctuation. This step helps to foresee the changing trend of resource demand by performing a probability analysis on future usage fluctuations and evaluating the range of usage demand under different circumstances. Through the estimation of the fluctuation probability interval, it is possible to make full preparations for the demand fluctuations that occur, providing a more accurate basis for inventory allocation. This step helps to avoid resource shortages or backlogs caused by excessive demand fluctuations, thereby achieving more scientific inventory management. According to the subsequent usage fluctuation probability interval and the skin disease blood product treatment usage data between different cases, the inventory is optimized for pre-allocation. Through this optimization process, the inventory allocation method can be reasonably predicted and adjusted in advance to ensure that the resource requirements of each link are fully met. Forward inventory allocation optimization can dynamically adjust inventory based on the forecast of usage fluctuations to avoid waste caused by excessive inventory or tension caused by insufficient inventory. This method makes inventory management more flexible and efficient through scientific resource allocation, ensuring that demand at all stages can be met in a timely manner, and improving the overall efficiency and reliability of resource use.

[0049] Preferably, step S33 includes the following steps:

[0050] Step S331: Calculate the demand fluctuation extreme value interval for the subsequent usage fluctuation probability interval to obtain the demand fluctuation extreme value interval;

[0051] Step S332: performing dosage linear coupling according to the demand fluctuation extreme value interval and the dosage data of blood products for skin disease treatment between different cases to obtain the dosage linear coupling data before and after;

[0052] Step S333: Perform safety reserve usage analysis based on linear coupling data before and after usage and extreme value interval of demand fluctuation to obtain subsequent demand safety reserve usage data;

[0053] Step S334: Optimize the pre-inventory allocation of dermatology blood products according to the pre- and post-usage linear coupling data and the subsequent demand safety reserve usage data to obtain pre-inventory allocation optimization data.

[0054] The present invention analyzes the probability interval of subsequent usage fluctuations and calculates the extreme value interval of demand fluctuations. Through this calculation, the minimum and maximum ranges of demand in the future can be clearly defined. Understanding the extreme values ​​of demand fluctuations helps to determine the potential fluctuation range of inventory demand, thereby making full preparations for resource allocation. This step provides the necessary basis for subsequent inventory optimization and risk management, ensuring that there is sufficient response capability in extreme cases of demand fluctuations. Based on the extreme value interval of demand fluctuations and the dosage data of skin disease blood products for treatment between different cases, a linear coupling analysis of dosage is performed. This step linearly couples the demand fluctuation data at different time points with the actual dosage data to find out the regular relationship between dosage demand and fluctuation. Through linear coupling analysis, the mutual influence between dosage changes and demand fluctuations can be clearly revealed, providing a more accurate prediction basis for subsequent inventory allocation. This process helps to finely manage inventory and ensure that supply can respond to demand changes in real time. After obtaining the linear coupling data before and after the dosage, combined with the extreme value interval of demand fluctuations, a safe reserved dosage analysis is performed. The core purpose of this analysis is to retain a certain amount of safety stock in future demand fluctuations to cope with sudden or extreme demand fluctuations. By scientifically analyzing the extreme values ​​of demand fluctuations and usage trends, a reasonable "safety margin" can be provided for inventory allocation to ensure that there will be no resource shortages even when demand fluctuates greatly. This step ensures the robustness of inventory management and improves the ability to prevent supply risks. Based on the linear coupling data before and after usage and the subsequent demand safety reserve usage data, the forward inventory allocation of dermatology blood products is optimized. This step dynamically adjusts the allocation of forward inventory by comprehensively considering usage changes, demand fluctuations and safety reserve inventory to ensure that inventory can be adjusted in a timely manner according to actual demand and potential fluctuations. The implementation of forward inventory allocation optimization enables inventory resources to meet future needs more accurately, avoids waste or shortage of resources due to inaccurate demand forecasts, and improves the efficiency and flexibility of inventory management. This process effectively improves the adaptability to future demand changes and the accuracy of resource scheduling.

[0055] Preferably, step S4 comprises the following steps:

[0056] Step S41: normalizing the forward inventory allocation optimization data to obtain forward inventory allocation normalized data;

[0057] Step S42: Designing an automated management architecture for the normalized data of the front-end inventory allocation to obtain a front-end inventory allocation management architecture;

[0058] Step S43: Send the pre-inventory allocation management framework to the medical management system to perform intelligent inventory management of blood products.

[0059] The present invention normalizes the pre-inventory allocation optimization data, with the purpose of standardizing inventory data of different dimensions and dimensions into a unified numerical range. Through normalization, the dimensional differences of the original data can be eliminated, so that various indicators can be compared and analyzed at the same scale. This process not only improves the accuracy and consistency of data processing, but also provides a more concise and operable data format for subsequent intelligent decision-making and resource allocation, thereby providing a better basis for further optimization management. After obtaining the normalized pre-inventory allocation data, the design of the automated management architecture is carried out. The core of this step is to convert the optimization strategy of inventory allocation into an automated management framework through algorithms and intelligent systems. A reasonably designed automated architecture can make the inventory allocation decision process more efficient and accurate, reduce the error of manual intervention, and improve the response speed and execution of inventory management. Through the automated management architecture, it can be ensured that the inventory allocation and optimization plan can be monitored and adjusted in real time, improve the overall operational efficiency, and reduce resource waste or excessive backlog. After the automated management architecture is designed, the next step is to send the management architecture to the medical management system for execution. By integrating the pre-stock allocation management architecture into the medical management system, the intelligent management of blood product inventory can be achieved. This process monitors, adjusts and optimizes the inventory status in real time through system automation to ensure the accurate implementation of the inventory management strategy. The beneficial effect of this step is that it closely combines advanced inventory allocation optimization solutions with the medical management system, improves the transparency, accuracy and efficiency of inventory management, and thus improves resource utilization, ensuring that the system can respond intelligently to demand fluctuations, optimize inventory configuration, and reduce potential resource shortages or backlogs.

[0060] Preferably, the present invention also provides an intelligent inventory management system for blood products based on medical big data, which is used to execute the intelligent inventory management method for blood products based on medical big data as described above. The intelligent inventory management system for blood products based on medical big data comprises:

[0061] The product dosage extraction module is used to obtain the historical medical record data of the dermatology department; desensitize the historical medical record data of the dermatology department to obtain the desensitized historical medical record data; extract the treatment dosage of blood products for skin diseases between different case types from the desensitized historical medical record data to obtain the treatment dosage data of blood products for skin diseases between different cases;

[0062] The recovery capacity difference evaluation module is used to simulate and evaluate the recovery capacity difference of the skin disease blood product treatment dosage data between different cases based on the historical medical desensitization record data to obtain the recovery capacity difference evaluation data; and to predict the subsequent dosage demand based on the recovery capacity difference evaluation data to obtain the subsequent dosage demand prediction data;

[0063] The forward inventory allocation optimization module is used to obtain the inventory of dermatology blood products; the forward inventory allocation optimization of dermatology blood products is performed according to the subsequent usage demand forecast data to obtain the forward inventory allocation optimization data;

[0064] The management architecture design module is used to automatically design the management architecture for the forward inventory allocation optimization data to obtain the forward inventory allocation management architecture; the forward inventory allocation management architecture is sent to the medical management system to execute intelligent inventory management of blood products.

[0065] The beneficial effects of the present invention are as follows: the present invention optimizes a traditional intelligent inventory management method for blood products based on medical big data, solves the problem that the traditional intelligent inventory management method for blood products based on medical big data cannot accurately predict the amount of blood products required for the patient in the later period through the historical medical records of the dermatology department, and cannot perform accurate pre-scheduling optimization of the amount, improves the accuracy of predicting the amount of blood products required for the patient in the later period through the historical medical records of the dermatology department, and improves the accuracy of blood pre-scheduling optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A schematic diagram of the steps of an intelligent inventory management method for blood products based on medical big data;

[0067] Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0068] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0069] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0070] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0071] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0072] To achieve this, please refer to Figure 1 to Figure 2 , a blood product intelligent inventory management method based on medical big data, the method comprising the following steps:

[0073] Step S1: Obtain historical medical records of the dermatology department; desensitize the historical medical records of the dermatology department to obtain desensitized historical medical records; extract the amount of blood products used for the treatment of skin diseases between different types of cases from the desensitized historical medical records to obtain the amount of blood products used for the treatment of skin diseases between different cases;

[0074] Step S2: Performing a simulation evaluation of the recovery ability difference of the skin disease blood product treatment dosage data between different cases based on the historical medical desensitization record data to obtain recovery ability difference evaluation data; performing a subsequent dosage demand prediction based on the recovery ability difference evaluation data to obtain subsequent dosage demand prediction data;

[0075] Step S3: Obtain the inventory of dermatology blood products; optimize the pre-stock allocation of dermatology blood products according to the subsequent usage demand forecast data to obtain pre-stock allocation optimization data;

[0076] Step S4: Design an automated management architecture for the pre-inventory allocation optimization data to obtain a pre-inventory allocation management architecture; send the pre-inventory allocation management architecture to the medical management system to execute intelligent inventory management of blood products.

[0077] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a blood product intelligent inventory management method based on medical big data of the present invention. In this example, the blood product intelligent inventory management method based on medical big data includes the following steps:

[0078] Step S1: Obtain historical medical records of the dermatology department; desensitize the historical medical records of the dermatology department to obtain desensitized historical medical records; extract the amount of blood products used for the treatment of skin diseases between different types of cases from the desensitized historical medical records to obtain the amount of blood products used for the treatment of skin diseases between different cases;

[0079] In an embodiment of the present invention, the historical medical records of the dermatology department are obtained through medical big data. Static masking technology is required for data desensitization. The characters of the patient's personal information (including name, ID number, address, etc.) involved in each record are replaced to make it lose its identifiability, thereby protecting the patient's privacy. This operation uses regular expressions to identify and replace specific fields, and the generated result is "historical medical desensitization record data". Then, different case types are classified according to the data, and various types of skin diseases, such as eczema, psoriasis, dermatitis, etc., are identified for subsequent analysis. According to the type data of each case, the item-by-item retrieval method is used to extract the quantity, type and dosage of various blood products used in the corresponding case type in previous visits. After calculation, the blood product dosage data of various skin diseases in different medical instances are classified and summarized to generate "blood product treatment dosage data for skin diseases between different cases".

[0080] Step S2: Performing a simulation evaluation of the recovery ability difference of the skin disease blood product treatment dosage data between different cases based on the historical medical desensitization record data to obtain recovery ability difference evaluation data; performing a subsequent dosage demand prediction based on the recovery ability difference evaluation data to obtain subsequent dosage demand prediction data;

[0081] In an embodiment of the present invention, a virtual person is used to simulate and evaluate the difference in recovery ability. The virtual person setting is generated by a computer program, and the input includes basic attributes such as age, gender, and medical history, and is based on the patient's physiological parameters (such as age, weight, lesion area, degree of skin damage, etc.) and the actual difference in the amount of blood products used for skin disease treatment during the treatment process. First, the physiological parameters of each patient are extracted from the desensitization data, and a multivariate evaluation model for recovery ability is constructed in combination with their respective dosage conditions. The model uses a multivariate regression analysis method to evaluate the difference in recovery ability based on the relationship between the amount of blood products used for skin disease treatment and the recovery time between different cases. In this step, first, for different case categories, the specific amount of skin blood products is selected and the individualized attributes of the virtual person are comprehensively considered, and the recovery cycle of each skin disease and blood product type is repeatedly simulated by the Monte Carlo simulation method. In the specific implementation, based on the difference parameters of recovery speed, such as age, disease type, recovery cycle and other multivariate variables, the probability distribution of different recovery abilities is calculated using the Bayesian network algorithm to obtain the recovery ability difference evaluation data of different types of cases. The data can show the expected recovery time and degree of difference of different diseases under the same dose of blood products. Utilize the evaluation data to forecast subsequent usage demand. By combining the weighted moving average method with historical usage trends, estimate the future demand for blood products and generate forecast data for subsequent usage demand.

[0082] Step S3: Obtain the inventory of dermatology blood products; optimize the pre-stock allocation of dermatology blood products according to the subsequent usage demand forecast data to obtain pre-stock allocation optimization data;

[0083] In an embodiment of the present invention, detailed data on the existing inventory of blood products in the dermatology department is obtained, including the current inventory quantity, batch numbers of various blood products, expiration dates, etc. After the data is acquired, the inventory optimization algorithm is used to perform pre-deployment optimization by comparing the subsequent usage demand forecast data and inventory data. First, the optimal allocation plan is established using a linear programming model, using the expiration date of blood products, usage demand forecast value, and inventory quantity as parameters to gradually calculate and optimize the reasonable allocation of inventory to ensure that the blood products in stock can be supplied in a timely manner. During the allocation process, a dynamic constraint algorithm is used to allocate inventory cycles to form pre-inventory allocation optimization data. This data set includes the optimal inventory configuration, batch allocation, and future replenishment plans for various blood products.

[0084] Step S4: Design an automated management architecture for the pre-inventory allocation optimization data to obtain a pre-inventory allocation management architecture; send the pre-inventory allocation management architecture to the medical management system to execute intelligent inventory management of blood products.

[0085] In an embodiment of the present invention, an automated architecture design for intelligent inventory management is performed based on the forward inventory allocation optimization data. The management architecture design includes modules such as real-time monitoring of blood product inventory, inventory warning mechanism, automatic allocation and replenishment functions. In implementation, data processing and transmission are performed through a distributed database to build a forward inventory allocation management architecture. The inventory monitoring module uses database triggers and time event scheduling to perform regular or real-time detection of inventory data. The early warning module uses a threshold detection algorithm to automatically trigger an allocation instruction when the inventory level is lower than the safety value of the subsequent usage demand forecast data. The allocation function is implemented through the inventory automatic allocation module, which is transmitted to the medical management system through the API interface by the forward inventory allocation management architecture to ensure that the blood product inventory can be automatically replenished and optimally managed under intelligent conditions, and ultimately achieve the operational goal of comprehensive intelligent inventory management.

[0086] Preferably, step S1 comprises the following steps:

[0087] Step S11: Obtaining historical dermatology consultation record data;

[0088] Step S12: Desensitizing the historical medical records of the dermatology department to obtain desensitized historical medical records;

[0089] Step S13: classify the historical medical desensitization record data by case type to obtain dermatology medical case type classification data;

[0090] Step S14: extract the blood product treatment dosage of skin diseases between different case types from the historical desensitization record data according to the dermatology case classification data, and obtain the blood product treatment dosage data of skin diseases between different cases.

[0091] In an embodiment of the present invention, historical medical records of dermatology are extracted from medical big data or hospital information management systems or electronic health record systems. First, by establishing a connection interface, all dermatology-related medical records in the data source system are obtained. The data includes but is not limited to the patient's basic information (such as age, gender), visit time, diagnosis results, treatment plan, type and quantity of blood products used, clinical symptoms, etc. In order to ensure the integrity and accuracy of the data, data screening will be performed during the extraction process, and only valid records related to the use of blood products will be retained. The database is operated by SQL query language to filter out the blood product records used by patients in a specific time period to ensure the timeliness and representativeness of the data. These data will provide a basis for subsequent analysis. In order to protect the privacy of patients, the historical medical record data obtained is desensitized. Data desensitization uses information hiding and data mapping technology. First, the patient's name, ID number, address, contact information and other personal identity information are anonymized using data masking technology. These sensitive information will be replaced with non-identifiable characters, such as replacing the name with a randomly generated string, and the ID number is also displayed in a desensitized format. Secondly, pseudonymization is performed to replace the patient's unique identifier with a non-associated pseudo ID to avoid direct tracing back to the individual's identity. The desensitized data will be converted into historical medical records to ensure that the data meets privacy protection requirements and can be used for subsequent analysis. The desensitized historical medical records are classified into case types. This operation uses text classification technology based on natural language processing (NLP) to classify different cases according to the type of skin disease by analyzing the diagnosis description information in the medical records. The specific operation process is as follows: First, the diagnosis description of each medical record is segmented, and then the disease name, symptoms and related keywords are extracted based on the keyword matching algorithm. By building a rule base, common skin disease types such as eczema, psoriasis, acne, skin allergies, etc. are standardized and classified. If there is no clear disease type in the diagnosis description, a similarity matching algorithm (such as Jaccard similarity or cosine similarity) can also be used to match the text content with the pre-defined disease classification. The final generated "dermatology case type classification data" includes the category label and relevant diagnostic information of each case, which is convenient for the subsequent analysis of blood product usage. According to the "dermatology case type classification data" obtained in step S13, the skin blood product usage data between different case types is extracted. This step first summarizes the skin blood product usage of each case category, and counts the types, quantities and usage frequencies of blood products used in the treatment of different diseases. The specific method is: first, according to the case type classification data, the blood product usage records of the corresponding cases are screened out, and then the data is aggregated. The aggregation operation includes calculating the total amount of blood products required for each type of disease in a specific time period, as well as the differences in usage distribution among different patients.In this process, data grouping and aggregation functions (such as sum, average, etc.) are used, combined with statistical analysis methods, to conduct further descriptive statistical analysis of blood product usage, such as mean, standard deviation, distribution, etc. Ultimately, the output "blood product treatment usage data for skin diseases between different cases" will cover the blood product usage corresponding to each case category, providing data support for subsequent inventory management and demand forecasting.

[0092] Preferably, step S2 comprises the following steps:

[0093] Step S21: extracting physiological parameters from the historical medical consultation desensitization record data to obtain medical consultation physiological parameters;

[0094] Step S22: performing a simulation evaluation of the difference in recovery ability of the blood product treatment dosage data for skin diseases between different cases according to the physiological parameters of the consultation and the historical consultation desensitization record data, and obtaining recovery ability difference evaluation data;

[0095] Step S23: Based on the recovery ability difference evaluation data, blood volume demand pattern analysis is performed on the skin disease blood product treatment dosage data between different cases with different recovery ability differences to obtain blood volume recovery difference demand pattern data;

[0096] Step S24: performing subsequent usage demand prediction on the blood volume recovery difference demand pattern data to obtain subsequent usage demand prediction data.

[0097] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0098] Step S21: extracting physiological parameters from the historical medical consultation desensitization record data to obtain medical consultation physiological parameters;

[0099] In an embodiment of the present invention, the key physiological parameters of the patient are extracted from the historical medical desensitization record data. This step first determines the physiological parameter fields related to each case based on the desensitized data, including but not limited to the patient's body temperature, heart rate, blood pressure, blood oxygen saturation, blood indicators (such as white blood cell count, hemoglobin concentration, etc.), height, weight and age, and skin test results. Based on the extraction of these fields, a specific regular expression matching algorithm is used to identify numerical or textual physiological data item by item from the case text record. For missing data, data is completed through historical similar patient data and statistical methods, such as interpolation or mean filling, to make the physiological parameter information more complete and representative. The extracted and completed physiological data ultimately constitute the "physiological parameters of the medical visit", which provide a reference for the subsequent simulation evaluation of the recovery ability difference.

[0100] Step S22: performing a simulation evaluation of the difference in recovery ability of the blood product treatment dosage data for skin diseases between different cases according to the physiological parameters of the consultation and the historical consultation desensitization record data, and obtaining recovery ability difference evaluation data;

[0101] In the embodiment of the present invention, the recovery ability difference evaluation is simulated by the data of virtual patients, and the recovery ability difference between different case types during the use of blood products is calculated. First, the "physiological parameters for medical treatment" generated in step S21 is standardized, and each physiological parameter is converted into a standard score for comparison and weighting between different parameters. Then, according to the skin blood product dosage data and the physiological parameters for medical treatment of different case types, the fuzzy comprehensive evaluation method is applied to simulate the recovery ability difference evaluation. The specific operation process includes: setting the evaluation index weights of different case types, and based on the physiological parameters of the patient, the computer virtually simulates the patient's demand changes and recovery response for the treatment dosage of skin disease blood products. For patients with different conditions, the membership function in the fuzzy comprehensive evaluation is used to score the recovery effect, and the scoring basis includes dosage changes, recovery time, physiological parameter recovery curve, etc., and finally the recovery ability score corresponding to each type of case is obtained. The output result of step S22 is "recovery ability difference evaluation data", which can be used as an important basis for blood product demand forecasting and intelligent inventory management in subsequent steps.

[0102] Step S23: Based on the recovery ability difference evaluation data, blood volume demand pattern analysis is performed on the skin disease blood product treatment dosage data between different cases with different recovery ability differences to obtain blood volume recovery difference demand pattern data;

[0103] In the embodiment of the present invention, according to the "recovery capacity difference evaluation data" generated in step S22, the demand pattern of the amount of blood products for the treatment of skin diseases between different cases is analyzed. This step firstly classifies each case type by grouping statistics, and subdivides the cases into several categories with significant differences in recovery capacity based on factors such as disease category, recovery speed, and physiological parameters of patients as grouping criteria. Then, the demand pattern analysis is performed on the amount of blood products in each category to observe the blood product consumption pattern of patients in different categories. During the analysis, the relationship between the amount of blood used and time in each case is modeled by the time series analysis method, and the moving average method is used to smooth the data to remove the influence of random fluctuations on the pattern analysis. At the same time, the hierarchical clustering algorithm is applied to cluster the blood used in each case group, and the groups with similar usage change trends are merged into the same demand pattern. After completing the demand pattern analysis, the "blood volume recovery difference demand pattern data" is generated, and the data contains the demand pattern curves of each category and the expected blood volume distribution, which provides a detailed pattern reference for the subsequent dosage demand forecast.

[0104] Step S24: performing subsequent usage demand prediction on the blood volume recovery difference demand pattern data to obtain subsequent usage demand prediction data.

[0105] In an embodiment of the present invention, a usage demand forecast is performed based on the "blood volume recovery difference demand pattern data" to estimate in advance the demand for blood products for different case types in the future. This step first extends and extrapolates the usage curves of each category in the demand pattern data, and calculates the blood product demand of each type of patient in the forecast period based on the time series extrapolation method. The specific operation includes: taking the main trend in the demand pattern as a benchmark, combining the periodic characteristics of time variables such as seasonality and epidemiological factors, and applying the exponential smoothing method to dynamically predict the demand at future time points. Subsequently, regression analysis is used to calculate the demand fluctuations of the demand for each mode, and the abnormal peaks in the forecast results are corrected based on the recovery time differences and physiological parameters in the case characteristics. Through this forecasting method, the "subsequent usage demand forecast data" is finally generated, and the data is the expected blood product usage of different case types in a specific time period in the future. This data can be used for subsequent inventory management strategy optimization to provide support for the allocation of the blood product supply chain.

[0106] Preferably, step S22 includes the following steps:

[0107] Step S221: Obtain the age, height and weight of the patient from the patient's physiological parameters; evaluate the degree of skin damage on the historical desensitization record data to obtain skin damage degree data;

[0108] Step S222: estimating the blood transfusion tolerance per unit time based on the patient's age, height, and weight to obtain the blood transfusion tolerance per unit time;

[0109] Step S223: estimating the skin tension damage gradient of the skin damage degree data to obtain skin tension damage gradient data;

[0110] Step S224: predicting the progressive cycle of injury recovery based on the skin tension injury gradient data according to the blood transfusion tolerance per unit time, and obtaining the progressive cycle data of injury recovery;

[0111] Step S225: Based on the injury recovery progressive cycle data, a recovery capacity difference simulation evaluation is performed on the skin disease blood product treatment dosage data between different cases to obtain recovery capacity difference evaluation data.

[0112] In the embodiment of the present invention, a virtual patient is simulated in a computer, and all the following calculation processes and parameter inputs are input into the computer for simulation, and key physiological indicators are extracted from the physiological parameters of the visit, including the age, height and weight of the visit, as the basic variables for subsequent tolerance and recovery capacity evaluation. First, by reading the historical visit desensitization record data, the patient's age, height and weight recorded therein are selected, and stored in the corresponding fields respectively to ensure that the basic data of each patient are fully acquired. On this basis, the patient's skin injury is graded using a linear weighted method, and the case description text in the desensitization record is compared with different injury scores. The severity of skin injury is statistically analyzed, and skin injury degree data is constructed for subsequent analysis. A virtual patient is simulated in a computer, and the transfusion tolerance per unit time of each patient is calculated by the visit age, height and weight data. This step uses the tolerance evaluation formula of intravenous transfusion according to the patient's physiological parameters, wherein the upper limit threshold of the transfusion dose is regulated according to the BMI index of the patient's weight and height, and the tolerance range is determined in combination with the age factor. During the calculation process, the linear interpolation method is used to smoothly transition the thresholds between different physiological parameter groups. The final value of the transfusion tolerance is output through the interpolation function, matching the patient's physiological characteristics with the tolerance range, and generating the transfusion tolerance data per unit time. This data is output in the form of numerical values ​​and used as the input parameter of the next step to further calculate the recovery ability of skin tension damage. The skin tension damage gradient is estimated for the skin damage degree data of the virtual patient. First, the damage area, damage depth, and skin surface tension characteristics are analyzed. By establishing a damage gradient calculation model, the damage depth and tension index are graded and evaluated. In the specific implementation, the tension gradient change rate of the damaged part is calculated segmentally by the tension gradient equation. The damage diffusion coefficient is introduced in the calculation process, and the stretch and resilience characteristics of the skin under different tensions are considered to further derive the numerical weights of each segmented gradient. Subsequently, the multi-gradient weights are integrated to finally obtain the skin tension damage gradient data as the basic data for subsequent recovery cycle prediction. According to the transfusion tolerance per unit time, the skin tension damage gradient data is used to predict the progressive cycle of damage recovery. In this step, the recovery cycle of the damaged part is predicted by the numerical simulation algorithm through the transfusion tolerance parameters of the virtual patient. Specifically, the calculation method uses the simulation comparison method to recursively calculate the time node of recovery and the degree of injury healing based on the tolerance data per unit time. In the prediction process of each time node, the acceleration effect of blood transfusion on injury recovery is simulated, and the healing process is calculated using an increasing time step. During the simulation process, the gradual recovery is curve-fitted based on the injury gradient to generate the injury recovery progressive cycle data, providing the necessary cycle basis for simulating and evaluating the difference in recovery ability.Based on the progressive cycle data of injury recovery, the progressive cycle data of injury recovery is combined with the data of skin blood product usage between different cases, and a recovery capacity simulation model is established based on the hierarchical regression method. The difference in recovery speed between different cases is evaluated through stepwise regression analysis, and further correlation analysis is performed with the amount of blood products used for skin disease treatment, and the recovery capacity difference evaluation data is output. In the specific implementation, all input data are standardized to ensure the comparability of the calculation results. At the same time, the regression equation is optimized through multiple iterations to ensure the accuracy and stability of the evaluation.

[0113] Preferably, step S224 includes the following steps:

[0114] Performing fiber elastic loss gradient analysis on skin tension damage gradient data to obtain fiber elastic loss gradient data;

[0115] The cell self-healing rate is simulated and estimated based on the fiber elasticity loss gradient data and the skin tension damage gradient data to obtain the skin cell self-healing rate data;

[0116] Based on the skin cell self-healing rate data, the fiber elasticity loss gradient data is used to calculate the initial tension benchmark for damage recovery, and the initial tension benchmark data for damage recovery is obtained;

[0117] According to the blood transfusion tolerance per unit time, the initial tension baseline data of injury recovery is analyzed for gradual tension recovery increment, and the gradual tension recovery increment data is obtained;

[0118] The progressive cycle of damage recovery is predicted based on the progressive tension recovery increment data to obtain the progressive cycle data of damage recovery.

[0119] In an embodiment of the present invention, based on the skin tension injury gradient data, the elasticity loss of the skin fiber tissue of the virtual patient is analyzed to quantify the elasticity decline rate of the fiber. First, the key parameters in the skin tension injury gradient data, such as the tension value of the damaged part, the damage depth and the tension gradient distribution, are decomposed and numerically processed. On this basis, the segmented elastic determination method is introduced to differentiate the tension gradient of the fiber tissue, calculate the fiber elasticity decline rate of each small area, and perform integral calculation to form a fiber elasticity loss gradient map. The graphical data can further quantitatively describe the degree of attenuation of fiber elasticity with the damage process, and provide parameter basis for the self-healing rate estimation in the subsequent steps. Based on the fiber elasticity loss gradient data and the skin tension injury gradient data, the self-healing rate of skin cells is estimated. First, the elasticity loss coefficient in the above gradient data is combined with the damage tension gradient value to construct a multivariable differential equation of the cell self-healing rate, and define the initial rate and deceleration rate of cell recovery. Secondly, through the simulation analysis method of the self-healing rate function, the changes in the self-healing rate of various cells are simulated within a specific time, and the cell activity index of the damaged area is adjusted in real time. The differential quadrature method is introduced in the simulation calculation, and the self-healing rate is discretely calculated in multiple time periods to generate skin cell self-healing rate data to reflect the recovery ability of the skin at different time nodes. Based on the skin cell self-healing rate data, the fiber elasticity loss gradient data is calculated for the initial tension benchmark of damage recovery. In this step, the cell self-healing rate is first introduced into the initial tension evaluation of skin recovery through tension recovery simulation on the basis of the self-healing rate. During the calculation process, the initial tension value of each cell is calculated according to the recovery ability of the skin fiber layer through the tension benchmark estimation function. This method performs a multi-level analysis of regional tension through the gradual change of tension in the damaged area. Finally, the initial tension benchmark is output by integral calculation, and the initial tension benchmark data of damage recovery is generated as the input parameter for the subsequent tension recovery increment analysis. According to the initial tension benchmark data of damage recovery and the transfusion tolerance per unit time, the gradual tension recovery increment of damage recovery is analyzed. The specific calculation method includes taking each tension value in the initial tension benchmark data as the initial condition of tension recovery, and combining the tolerance data per unit time for multi-point incremental segmentation. The progressive tension recovery incremental analysis algorithm is used to generate gradual increments on the time axis. To ensure accuracy, slight adjustments are introduced at each incremental point, and the progressive tension recovery incremental data is constructed through the progressive tension values ​​to provide a dynamic incremental basis for the final recovery cycle prediction. Based on the progressive tension recovery incremental data, the progressive cycle of damage recovery is predicted. First, according to the time series and incremental values ​​in the progressive tension recovery data, the tension recovery process is segmented and a cycle prediction model is established. The model uses a periodic increment function and a time step to control the tension recovery speed. In the specific calculation process, a recursive algorithm is used to simulate the time axis evolution of tension recovery and generate recovery cycle data.After the data is output, the recovery time period of each stage is accurately located through the progressive cycle curve, and finally the damage recovery progressive cycle data is obtained.

[0120] Preferably, step S24 includes the following steps:

[0121] Step S241: using a preset staged recovery demand identification model to perform staged demand pattern hierarchical division on the blood volume recovery difference demand pattern data to obtain stage demand pattern hierarchical division data;

[0122] Step S242: performing a phased key usage change analysis on the blood volume recovery difference demand pattern data according to the phased demand pattern hierarchical division data to obtain phased key usage change data;

[0123] Step S243: performing phased demand fluctuation fitting on the phased key usage change data to obtain phased usage demand fluctuation fitting data;

[0124] Step S244: performing fluctuation inflection point trend analysis on the periodic consumption demand fluctuation fitting data to obtain demand fluctuation inflection point trend data;

[0125] Step S245: performing subsequent usage demand forecasting on the blood volume recovery difference demand pattern data according to the demand fluctuation inflection point trend data to obtain subsequent usage demand forecast data.

[0126] In an embodiment of the present invention, based on a preset staged recovery demand identification model, the blood volume recovery difference demand pattern data is hierarchically divided into staged demand patterns. First, a staged recovery demand identification model is defined, which generates a specific staged recovery demand distribution structure according to the staged change characteristics in the clinical data. In the specific operation process, the key variables of different demand stages are clustered by a hierarchical clustering algorithm, and the demand change trend in the blood volume recovery difference demand pattern data is hierarchically divided. In the hierarchical division, high-frequency demand fluctuations are separated from low-frequency demand fluctuations, so that the demand differences between stages can be more clearly distinguished. After hierarchical division analysis, the stage demand pattern hierarchical division data is formed as the basis for the key dosage change analysis in the subsequent steps. This step performs a staged key dosage change analysis on the blood volume recovery difference demand pattern data based on the stage demand pattern hierarchical division data. First, the hierarchical division data of the demand pattern in each stage is subjected to data decoupling processing, the key moments and peak points of the dosage increase and decrease in each level are extracted, and the significant change trend of the dosage in each stage is recorded. Next, through the time series analysis method, the key usage values ​​in each stage are regressed and fitted to calculate the rate of increase or decrease of blood volume demand, and these rate values ​​are marked as key change points. Subsequently, through the staged key usage change analysis, the staged key usage change data are generated to indicate the demand fluctuations and quantitative trends in different stages. According to the staged key usage change data, the demand pattern of each stage is subjected to fluctuation fitting. First, each key change point in the staged key usage change data is input into the polynomial fitting algorithm, and the fluctuation curve of each stage is generated through the mathematical operation of fluctuation fitting. During the fitting process, the least squares method is used to fit the fluctuation trend so that the fluctuation curve can more accurately reflect the actual demand changes. To ensure the accuracy of the fitting, the relative difference between the key point and other fluctuation points is quantified, and the sensitivity of the fitting curve is adjusted. Finally, the staged usage demand fluctuation fitting data is formed, which is used to present the fluctuation characteristics of demand in different stages, so as to facilitate further trend analysis of fluctuation inflection points. The staged usage demand fluctuation fitting data is analyzed to identify the inflection point trend in demand fluctuation. First, input the fluctuation fitting data, and analyze the trend change of the fitting curve through the inflection point detection algorithm. In the specific implementation, the inflection point identification method is used to calculate the derivative of the fluctuation curve, and the curvature change in each demand stage is detected to determine the inflection point position of the demand change. The inflection point trend analysis combines the duration, rate of change and direction of the trend to identify the characteristics and distribution of different inflection points in the fluctuation trend. Finally, the demand fluctuation inflection point trend data is generated as a basis for judging the subsequent demand trend. Based on the demand fluctuation inflection point trend data, the subsequent usage demand forecast is carried out on the blood volume recovery difference demand pattern data.First, the position and direction of each key inflection point in the inflection point trend data are input into the forecasting algorithm, and the subsequent changes in demand are estimated by trend extrapolation. The time series forecasting method is used to predict the increase or decrease in demand in the future time period based on the fluctuation direction and amplitude of the historical trend data, and generate subsequent demand forecast data.

[0127] Preferably, step S3 comprises the following steps:

[0128] Step S31: Obtaining the inventory of dermatology blood products;

[0129] Step S32: Estimating the usage fluctuation probability interval for the subsequent usage demand forecast data to obtain the subsequent usage fluctuation probability interval;

[0130] Step S33: Optimize the pre-stock allocation of dermatology blood products according to the subsequent usage fluctuation probability interval and the skin disease blood products treatment usage data between different cases to obtain pre-stock allocation optimization data.

[0131] In the embodiment of the present invention, the existing blood product inventory data of the dermatology department is fully obtained to provide accurate basic information for subsequent inventory allocation and demand forecasting. In the specific operation process, the different types of blood products in the dermatology department are first counted, and these inventories are registered in detail according to the product type, specification, storage temperature, batch number and other information. At the same time, the key parameters such as the validity period and storage conditions in the inventory data are confirmed to ensure that all types of blood products meet medical needs when used. This process needs to be systematically verified by registering the inventory data one by one to ensure the completeness and accuracy of the inventory information. To this end, the database query technology and the data acquisition device are used to synchronize the data of the dermatology blood product inventory to the inventory management database in the system to form a detailed inventory data list that can be called in real time. The subsequent usage demand forecast data is estimated for the usage fluctuation probability interval to obtain the subsequent usage fluctuation probability interval data. First, the subsequent usage demand forecast data is imported into the fluctuation probability calculation model. The model derives the probability of different usage levels occurring in a specific period in the future based on the frequency distribution and fluctuation characteristics of the historical usage. Using the Bayesian estimation method, the fluctuation amplitude of subsequent usage data is summarized and the future demand is divided into different probability intervals to predict the possibility of demand fluctuations under different conditions. In the operation, the time series of subsequent usage data is first segmented to determine the demand changes at specific time points. Subsequently, based on the variance calculation method in statistics, the upper and lower limits of the usage fluctuation are evaluated in the form of probability density, and the probability fluctuation interval of each demand value is generated and stored as the subsequent usage fluctuation probability interval data. According to the subsequent usage fluctuation probability interval and the skin disease blood product treatment usage data between different cases, the existing blood product inventory of the dermatology department is optimized for pre-stock allocation to obtain the pre-stock allocation optimization data. The specific operation includes: first, according to the blood product usage data of different cases, the priority demand categories of various blood products are set, and the usage risk levels of various inventories are graded in combination with the usage fluctuation probability interval data. Then, a stock allocation model based on demand priority is constructed through the linear programming method. Based on the priority demand for each category of blood products, the model calculates the optimal inventory allocation ratio of different types of blood products through a multi-objective optimization algorithm, and dynamically adjusts it according to the special usage requirements of dermatology. After the optimization is completed, the forward inventory allocation optimization data is saved in the database for future actual usage and rapid response.

[0132] Preferably, step S33 includes the following steps:

[0133] Step S331: Calculate the demand fluctuation extreme value interval for the subsequent usage fluctuation probability interval to obtain the demand fluctuation extreme value interval;

[0134] Step S332: performing dosage linear coupling according to the demand fluctuation extreme value interval and the dosage data of blood products for skin disease treatment between different cases to obtain the dosage linear coupling data before and after;

[0135] Step S333: Perform safety reserve usage analysis based on linear coupling data before and after usage and extreme value interval of demand fluctuation to obtain subsequent demand safety reserve usage data;

[0136] Step S334: Optimize the pre-inventory allocation of dermatology blood products according to the pre- and post-usage linear coupling data and the subsequent demand safety reserve usage data to obtain pre-inventory allocation optimization data.

[0137] In an embodiment of the present invention, the data of the subsequent usage fluctuation probability interval is deeply analyzed to obtain the extreme value interval of future usage demand. In the specific operation, the historical demand fluctuation frequency is firstly refined and analyzed according to the data of the fluctuation probability interval, and the upper and lower limits of the fluctuation range are identified by the maximum and minimum analysis method. The data is classified and layered according to the time dimension and the demand change dimension, and the slope of the usage change in different time periods is calculated by combining the calculus method, so as to obtain the extreme value interval of demand fluctuation for each specific time period. Through this method, an estimated boundary for the extreme usage change of subsequent demand is formed, providing a stable demand range framework for subsequent safety reservation and inventory allocation. The calculation results are stored as demand fluctuation extreme value interval data. Based on the demand fluctuation extreme value interval data and the skin disease blood product treatment dosage data between different cases, a linear coupling analysis of the demand dosage is performed to obtain the linear coupling data before and after the dosage. First, the skin blood product dosage data of different cases are grouped according to factors such as patient type, severity of the disease and demand cycle. The historical dosage and demand extreme value interval of similar cases are fitted and analyzed by the linear regression method. Then, with the help of multivariate linear equations, the continuity of the actual dosage before and after is obtained by coupling the correlation and demand change trend. This operation can effectively identify the dosage differences caused by factors such as symptoms and treatment stages, form the correlation function of different dosage patterns, and obtain a stable linear coupling relationship. The linear coupling data before and after the dosage output will be used as the basic data for the next step of safety reserve dosage analysis. In the safety reserve dosage analysis, the safety reserve dosage of subsequent demand is reasonably planned by combining the linear coupling data before and after the dosage and the extreme value interval of demand fluctuation. In the specific implementation process, the key risk points in each time period and demand change mode are first identified based on the intersection of the linear coupling data and the extreme value interval of demand fluctuation. Using the safety redundancy calculation method, a certain proportion of safety reserve is set to cover the demand peak in the extreme value interval. Then, the dynamic programming technology is used to gradually adjust the safety reserve dosage to ensure that the necessary safety redundancy is always maintained during the high demand fluctuation period. The safety reserve dosage data for subsequent demand obtained by this method provides a set of blood product reservation schemes that are adaptable to future uncertain demand changes. Combining the linear coupling data before and after usage and the subsequent demand safety reserve usage data, the existing dermatology blood product inventory is optimized for pre-inventory allocation to obtain the optimal inventory configuration after allocation. First, match the safety reserve usage data with the current inventory to determine which types of blood products need priority allocation and replenishment. Using the minimum cost flow algorithm, various types of blood product inventory are reallocated according to demand priority to ensure that critical usage is always within the safety reserve range. For inventory that needs to be replenished, an allocation priority strategy is adopted. By introducing blood product data from different sources, flexible configuration is performed while ensuring the safety of inventory circulation.Finally, this step optimizes the configuration and generates forward inventory allocation optimization data to ensure that subsequent usage fluctuations can be fully guaranteed.

[0138] Preferably, step S4 comprises the following steps:

[0139] Step S41: normalizing the forward inventory allocation optimization data to obtain forward inventory allocation normalized data;

[0140] Step S42: Designing an automated management architecture for the normalized data of the front-end inventory allocation to obtain a front-end inventory allocation management architecture;

[0141] Step S43: Send the pre-inventory allocation management framework to the medical management system to perform intelligent inventory management of blood products.

[0142] In an embodiment of the present invention, the pre-inventory allocation optimization data is first normalized to ensure that the inventory data of various blood products meet the unified standards and can be compared across dimensions. In the specific operation, the inventory of each blood product in the pre-inventory allocation optimization data is normalized according to its maximum and minimum values, and a linear normalization method is used. Through this method, the inventory data of each blood product is converted into a standardized value, which is convenient for unified comparison and further analysis between different types of blood products. After normalization, the pre-inventory allocation normalization data is obtained, which will serve as the basis for the subsequent management architecture design. Based on the pre-inventory allocation normalization data, an automated management architecture design is performed. First, according to the normalized data, a set of rule-based inventory management system is constructed to optimize the inventory management of blood products by determining multiple dimensions such as inventory thresholds, dynamic allocation mechanisms, and inventory turnover efficiency. In the specific operation process, a dynamic allocation model is set in combination with the dynamic change trend of the inventory, including adjusting the inventory allocation according to the prediction of the predetermined demand peak and fluctuation range. For example, when the inventory of a certain blood product is lower than the set safety threshold, the inventory replenishment mechanism is automatically enabled to ensure sufficient inventory through warehouse allocation or emergency procurement. For each blood product, an intelligent allocation module is designed to automatically adjust the inventory according to predetermined rules and normalized data. This design ensures real-time monitoring and allocation of blood product inventory to avoid excessive or insufficient inventory. Finally, an automated and dynamically responsive inventory management architecture is formed and stored as the front-end inventory allocation management architecture data. The front-end inventory allocation management architecture data is sent to the medical management system to perform intelligent inventory management of blood products. First, through a secure interface protocol, the designed front-end inventory allocation management architecture data is converted into a format that meets the processing standards of the medical management system. In this process, a data transmission protocol (such as an API interface protocol) is used to convert the data format and encrypt the transmission to ensure the security and integrity of the data transmission. Subsequently, the medical management system receives and parses the data, and uses the inventory management module in the system to intelligently schedule and manage blood products. According to the guiding principles of the front-end inventory allocation management architecture, the system monitors the inventory changes of blood products in real time and adjusts the allocation strategy of blood products according to demand fluctuations and inventory status. This operation can also be combined with patient demand data in the medical system to achieve accurate inventory allocation and efficient scheduling, ensuring that the usage needs of different treatment stages and different patient groups are met. Ultimately, through the automated operation of the system, the continuous optimization of intelligent inventory management can be achieved.

[0143] Preferably, the present invention also provides an intelligent inventory management system for blood products based on medical big data, which is used to execute the intelligent inventory management method for blood products based on medical big data as described above. The intelligent inventory management system for blood products based on medical big data comprises:

[0144] The product dosage extraction module is used to obtain the historical medical record data of the dermatology department; desensitize the historical medical record data of the dermatology department to obtain the desensitized historical medical record data; extract the treatment dosage of blood products for skin diseases between different case types from the desensitized historical medical record data to obtain the treatment dosage data of blood products for skin diseases between different cases;

[0145] The recovery capacity difference evaluation module is used to simulate and evaluate the recovery capacity difference of the skin disease blood product treatment dosage data between different cases based on the historical medical desensitization record data to obtain the recovery capacity difference evaluation data; and to predict the subsequent dosage demand based on the recovery capacity difference evaluation data to obtain the subsequent dosage demand prediction data;

[0146] The forward inventory allocation optimization module is used to obtain the inventory of dermatology blood products; the forward inventory allocation optimization of dermatology blood products is performed according to the subsequent usage demand forecast data to obtain the forward inventory allocation optimization data;

[0147] The management architecture design module is used to automatically design the management architecture for the forward inventory allocation optimization data to obtain the forward inventory allocation management architecture; the forward inventory allocation management architecture is sent to the medical management system to execute intelligent inventory management of blood products.

[0148] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0149] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A blood product intelligent inventory management method based on medical big data, characterized in that: The following steps are involved: Step S1: Obtaining historical dermatology consultation record data; Desensitize the historical medical records of the dermatology department to obtain the desensitized historical medical records; The historical medical desensitization record data were used to extract the amount of blood products used in the treatment of skin diseases between different case types, and the data on the amount of blood products used in the treatment of skin diseases between different cases were obtained; Step S2 includes: Step S21: extracting physiological parameters from the historical medical consultation desensitization record data to obtain medical consultation physiological parameters; Step S22: Performing a simulation evaluation of the difference in recovery ability of the blood product treatment dosage data for skin diseases between different cases according to the physiological parameters of the consultation and the historical consultation desensitization record data, and obtaining recovery ability difference evaluation data; wherein step S22 includes: Step S221: Obtain the age, height and weight of the patient from the patient's physiological parameters; evaluate the degree of skin damage on the historical desensitization record data to obtain skin damage degree data; Step S222: estimating the blood transfusion tolerance per unit time based on the patient's age, height, and weight to obtain the blood transfusion tolerance per unit time; Step S223: estimating the skin tension damage gradient of the skin damage degree data to obtain skin tension damage gradient data; Step S224: predicting the progressive cycle of injury recovery based on the skin tension injury gradient data according to the blood transfusion tolerance per unit time, and obtaining progressive cycle data of injury recovery; Step S225: Based on the injury recovery progressive cycle data, a recovery ability difference simulation evaluation is performed on the skin disease blood product treatment dosage data between different cases to obtain recovery ability difference evaluation data; Step S23: Based on the recovery ability difference evaluation data, blood volume demand pattern analysis is performed on the skin disease blood product treatment dosage data between different cases with different recovery ability differences to obtain blood volume recovery difference demand pattern data; Step S24: performing subsequent usage demand prediction on the blood volume recovery difference demand pattern data to obtain subsequent usage demand prediction data; wherein step S24 includes: Step S241: using a preset staged recovery demand identification model to perform staged demand pattern hierarchical division on the blood volume recovery difference demand pattern data to obtain staged demand pattern hierarchical division data; Step S242: performing a phased key usage change analysis on the blood volume recovery difference demand pattern data according to the phased demand pattern hierarchical division data to obtain phased key usage change data; Step S243: performing phased demand fluctuation fitting on the phased key usage change data to obtain phased usage demand fluctuation fitting data; Step S244: performing fluctuation inflection point trend analysis on the periodic consumption demand fluctuation fitting data to obtain demand fluctuation inflection point trend data; Step S245: performing subsequent usage demand forecasting on the blood volume recovery difference demand pattern data according to the demand fluctuation inflection point trend data to obtain subsequent usage demand forecasting data; Step S3: Obtain the inventory of dermatology blood products; optimize the pre-stock allocation of dermatology blood products according to the subsequent usage demand forecast data to obtain pre-stock allocation optimization data; Step S4: Design an automated management architecture for the pre-inventory allocation optimization data to obtain a pre-inventory allocation management architecture; send the pre-inventory allocation management architecture to the medical management system to execute intelligent inventory management of blood products.

2. The intelligent inventory management method for blood products based on medical big data according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Obtaining historical dermatology consultation record data; Step S12: Desensitizing the historical medical records of the dermatology department to obtain desensitized historical medical records; Step S13: classify the historical medical desensitization record data by case type to obtain dermatology medical case type classification data; Step S14: extract the blood product treatment dosage of skin diseases between different case types from the historical desensitization record data according to the dermatology case classification data, and obtain the blood product treatment dosage data of skin diseases between different cases.

3. The intelligent inventory management method for blood products based on medical big data according to claim 1 is characterized in that: Step S224 includes the following steps: Performing fiber elastic loss gradient analysis on skin tension damage gradient data to obtain fiber elastic loss gradient data; The cell self-healing rate is simulated and estimated based on the fiber elasticity loss gradient data and the skin tension damage gradient data to obtain the skin cell self-healing rate data; Based on the skin cell self-healing rate data, the fiber elasticity loss gradient data is used to calculate the initial tension benchmark for damage recovery, and the initial tension benchmark data for damage recovery is obtained; According to the blood transfusion tolerance per unit time, the initial tension baseline data of injury recovery is analyzed for gradual tension recovery increment, and the gradual tension recovery increment data is obtained; The progressive cycle of damage recovery is predicted based on the progressive tension recovery increment data to obtain the progressive cycle data of damage recovery.

4. The intelligent inventory management method for blood products based on medical big data according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: Obtaining the inventory of dermatology blood products; Step S32: Estimating the usage fluctuation probability interval for the subsequent usage demand forecast data to obtain the subsequent usage fluctuation probability interval; Step S33: Optimize the pre-stock allocation of dermatology blood products according to the subsequent usage fluctuation probability interval and the skin disease blood products treatment usage data between different cases to obtain pre-stock allocation optimization data.

5. The intelligent inventory management method for blood products based on medical big data according to claim 4 is characterized in that: Step S33 includes the following steps: Step S331: Calculate the demand fluctuation extreme value interval for the subsequent usage fluctuation probability interval to obtain the demand fluctuation extreme value interval; Step S332: performing dosage linear coupling according to the demand fluctuation extreme value interval and the dosage data of skin disease blood products for different cases to obtain the dosage linear coupling data before and after; Step S333: Perform safety reserve usage analysis based on linear coupling data before and after usage and extreme value interval of demand fluctuation to obtain subsequent demand safety reserve usage data; Step S334: Optimize the pre-inventory allocation of dermatology blood products according to the pre- and post-usage linear coupling data and the subsequent demand safety reserve usage data to obtain pre-inventory allocation optimization data.

6. The intelligent inventory management method for blood products based on medical big data according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: normalizing the forward inventory allocation optimization data to obtain forward inventory allocation normalized data; Step S42: Designing an automated management architecture for the normalized data of the front-end inventory allocation to obtain a front-end inventory allocation management architecture; Step S43: Send the pre-inventory allocation management framework to the medical management system to perform intelligent inventory management of blood products.

7. An intelligent inventory management system for blood products based on medical big data, characterized in that: Used to execute the intelligent inventory management method of blood products based on medical big data as claimed in claim 1, the intelligent inventory management system of blood products based on medical big data comprises: The product dosage extraction module is used to obtain the historical medical record data of the dermatology department; desensitize the historical medical record data of the dermatology department to obtain the desensitized historical medical record data; extract the treatment dosage of blood products for skin diseases between different case types from the desensitized historical medical record data to obtain the treatment dosage data of blood products for skin diseases between different cases; The recovery capacity difference evaluation module is used to simulate and evaluate the recovery capacity difference of the skin disease blood product treatment dosage data between different cases based on the historical medical desensitization record data to obtain the recovery capacity difference evaluation data; and to predict the subsequent dosage demand based on the recovery capacity difference evaluation data to obtain the subsequent dosage demand prediction data; The forward inventory allocation optimization module is used to obtain the inventory of dermatology blood products; the forward inventory allocation optimization of dermatology blood products is performed according to the subsequent usage demand forecast data to obtain the forward inventory allocation optimization data; The management architecture design module is used to automatically design the management architecture for the forward inventory allocation optimization data to obtain the forward inventory allocation management architecture; the forward inventory allocation management architecture is sent to the medical management system to execute intelligent inventory management of blood products.

Images