Parallelization task scheduling optimization method and system for blood detection process

Abstract

The present application relates to the technical field of blood chemical analysis, in particular to a parallel task scheduling optimization method and system for blood detection process, wherein the decay rate of temperature-sensitive components, protein components and light-sensitive components in blood is corrected according to real-time environmental parameters to obtain the corrected decay rate of blood under current environmental conditions; the concentration change rate of each detection item is calculated according to the corrected decay rate, a multi-dimensional priority evaluation matrix is established in combination with the clinical importance coefficient and the detection time requirement; the component pairs and reaction types with interaction are identified; a coupled reaction kinetics model is established to predict the concentration value of each component at a set time point; the detection items are sorted according to the multi-dimensional priority evaluation matrix; when the detection item combination contains mutually influencing components, the detection order is adjusted according to the coupled decay prediction result; the adjusted detection task is distributed to the corresponding parallel detection equipment according to equipment compatibility to generate a task scheduling scheme and execute it.

Description

Technical Field

[0001] This invention relates to the field of blood chemistry analysis technology, specifically to a parallel task scheduling optimization method and system for blood testing processes. Background Technology

[0002] In existing technologies, blood sample testing scheduling is primarily based on simple priority ranking algorithms, typically considering only single-dimensional factors such as clinical importance or testing time. Regarding blood sample stability, existing technologies generally recognize that various biochemical components in blood undergo varying degrees of degradation after being removed from the body. Environmental factors such as temperature, humidity, and light significantly affect the stability of components like proteins, enzymes, and hormones in blood samples. Traditional methods usually employ standardized storage conditions and fixed testing time limits to control sample quality; however, this one-size-fits-all approach cannot be dynamically adjusted according to actual environmental conditions, thus affecting the accuracy of test results.

[0003] While existing parallel detection technologies can improve throughput, they still have significant shortcomings in task scheduling. Traditional scheduling algorithms mainly rely on simple load balancing based on equipment availability and detection time, lacking in-depth consideration of the characteristics of blood samples. In particular, when multiple tests involve the same or related biochemical components, improper arrangement of the testing order may lead to interactions between components, thereby affecting the accuracy of the test results.

[0004] Furthermore, existing technologies typically employ static reference values ​​or empirical formulas when dealing with the chemical reaction relationships between blood components, lacking dynamic prediction models. This approach cannot accurately predict concentration changes of each component at different time points, nor can it effectively guide the optimization and adjustment of the testing sequence. Therefore, a parallel task scheduling optimization method for blood testing workflows that can comprehensively achieve multi-objective optimization and possess dynamic adjustment capabilities is needed to improve the overall efficiency of blood testing.

[0005] To address this, a parallel task scheduling optimization method and system for the blood testing process is proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a parallel task scheduling optimization method and system for blood testing processes. By establishing a database of biochemical component decay rates and environmental impacts, combining real-time parameter correction of concentration changes, constructing a multi-dimensional priority matrix and coupled reaction model, comprehensively sorting test items and dynamically adjusting the detection order of interacting components, a parallel task scheduling scheme is generated.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A parallel task scheduling optimization method for a blood testing process includes:

[0009] Based on real-time environmental parameters, the decay rates of temperature-sensitive components, protein components, and light-sensitive components in the blood are corrected to obtain the corrected decay rate of the blood under the current environmental conditions.

[0010] The concentration change rate of blood for each test item is calculated based on the corrected attenuation rate. A multi-dimensional priority evaluation matrix is ​​established by combining the clinical importance coefficient of blood for each test item with the test time requirements.

[0011] A database of chemical reaction relationships among various biochemical components in blood samples is pre-established, identifying component pairs with interactive effects and their reaction types; a coupled reaction kinetic model is established to predict the concentration values ​​of each component at a set time point;

[0012] All items to be detected are sorted according to a multi-dimensional priority evaluation matrix; when the combination of items to be detected contains mutually influential components, the detection order is adjusted according to the coupling attenuation prediction results; the adjusted detection tasks are assigned to the corresponding parallel detection devices according to device compatibility, and a task scheduling scheme containing detection order, device allocation and time arrangement is generated and executed.

[0013] Preferably, for temperature-sensitive components, the temperature correction factor is calculated using the Arrhenius equation, and the temperature-corrected decay rate is obtained by multiplying the base decay rate by the temperature correction factor.

[0014] For protein components, the combined influence factor of temperature and humidity is calculated based on the protein thermodynamic stability parameters. The protein corrected decay rate is obtained by multiplying the basic decay rate by the combined influence factor.

[0015] For photosensitive components, the light attenuation factor is calculated based on the light intensity and cumulative exposure time. The base attenuation rate is added to the light attenuation factor to obtain the corrected attenuation rate of the photosensitive component.

[0016] Preferably, the process of obtaining the corrected attenuation rate is as follows: real-time environmental data is collected, and environmental parameters are updated at preset time intervals;

[0017] Substitute real-time environmental parameters into the correction formula to obtain the environmental correction factor, which is then used to correct the temperature-corrected decay rate, protein-corrected decay rate, and photosensitized component-corrected decay rate. The corrected temperature-corrected decay rate, protein-corrected decay rate, and photosensitized component-corrected decay rate are then fused to obtain the corrected decay rate of each component at the current moment, and this information is stored and updated.

[0018] Preferably, the process of obtaining the concentration change rate includes:

[0019] Obtain the initial concentration values ​​of the corresponding components for each test item in the blood sample; multiply the corrected decay rate by the current concentration value to calculate the concentration change rate of each component at the current time; and calculate the concentration change rate of each component at future time points based on the first-order decay kinetic model.

[0020] Preferably, the process of obtaining the multi-dimensional priority evaluation matrix includes:

[0021] The attenuation urgency is calculated based on the concentration change rate, and the attenuation urgency coefficient is obtained by dividing the concentration change rate by the detection accuracy requirement. The clinical importance coefficient of each test item is obtained from the clinical database and determined according to the disease diagnosis weight and emergency priority. The time constraint coefficient is calculated based on the detection time requirement and the remaining available detection time. A three-dimensional evaluation matrix including the attenuation urgency coefficient, clinical importance coefficient, and time constraint coefficient is established, and the comprehensive priority score of each test item is calculated by weighted summation.

[0022] Preferably, the coupled reaction kinetic model includes:

[0023] The component identification layer extracts all biochemical components involved in the test item and their interactions from a chemical reaction relationship database;

[0024] The reaction parameter layer is used to obtain the reaction rate constant, equilibrium constant, and inhibition constant between each component pair.

[0025] The kinetic equation layer establishes a set of first-order ordinary differential equations describing the change of multi-component concentrations over time based on the law of mass action.

[0026] The numerical solution layer uses the Runge-Kutta algorithm to numerically integrate the differential equations, calculates the concentration change curves of each component within a set time interval, and outputs the predicted concentration values ​​at each time point.

[0027] Preferably, the process of sorting and adjusting the detection items according to the multi-dimensional priority evaluation matrix includes:

[0028] All items to be tested are initially sorted from high to low according to their comprehensive priority scores; component pairs that influence each other in the initial sort are identified, and the concentration prediction values ​​of these component pairs are obtained from the coupled reaction kinetic model; the degree of concentration influence between each component is calculated, and the test items whose influence exceeds the preset threshold are marked as high-impact items.

[0029] Adjust the testing order, prioritizing high-impact items before the components they affect, while maintaining the relative order of the overall priority ranking.

[0030] A parallel task scheduling optimization system for blood testing processes includes: an attenuation rate correction module, which corrects the attenuation rates of temperature-sensitive components, protein components, and light-sensitive components in blood based on real-time environmental parameters, to obtain the corrected attenuation rate of blood under the current environmental conditions.

[0031] The priority assessment module calculates the rate of change in blood concentration for each test item based on the corrected attenuation rate, and establishes a multi-dimensional priority assessment matrix by combining the clinical importance coefficient of the blood for each test item with the testing time requirements.

[0032] The coupling reaction module pre-establishes a database of chemical reaction relationships between various biochemical components in blood samples, identifies component pairs with interactive effects and their reaction types, and establishes a coupling reaction kinetic model to predict the concentration values ​​of each component at a set time point.

[0033] The parallel task scheduling module sorts all items to be detected according to a multi-dimensional priority evaluation matrix; when the combination of detection items contains mutually influential components, the detection order is adjusted according to the coupling attenuation prediction results; the adjusted detection tasks are allocated to the corresponding parallel detection devices according to device compatibility, and a task scheduling scheme containing detection order, device allocation and time arrangement is generated and executed.

[0034] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0035] 1. This invention establishes a pre-built database of basic decay rates for biochemical components and, by combining real-time environmental parameters to correct the decay rates of temperature-sensitive, protein-sensitive, and light-sensitive components, accurately predicts the concentration change trends of each component before detection. It calculates a decay urgency coefficient using information such as concentration change rate, detection accuracy requirements, and time constraints, constructing a multi-dimensional priority evaluation matrix to dynamically determine the detection order. This scheduling method, based on real-time environment and component characteristics, avoids sample distortion caused by detection delays and significantly reduces detection errors. This effectively improves the timeliness and stability of overall detection, ensuring more reliable blood test results.

[0036] 2. This invention introduces a coupled reaction kinetic model. For biochemical components in blood samples that interact, it constructs a set of ordinary differential equations based on the law of mass action and uses the Runge-Kutta algorithm for numerical solution to obtain predicted values ​​of the concentration of each component at different time points. This invention can identify the mutual influence relationships in the combination of detection items, calculate the degree of influence on the concentration of other components, and prioritize high-impact items before the detection of related components, while maintaining the relative stability of the overall priority ranking. It can effectively eliminate potential interference from chemical reactions and achieve scientific adjustment of the detection order.

[0037] 3. Based on determining the detection sequence, this invention further combines equipment compatibility and task scheduling requirements to automatically allocate detection tasks to different types of parallel detection equipment and generate a task scheduling scheme that includes detection sequence, equipment allocation, and time arrangement. By calculating the compatibility between tasks and equipment, detection tasks are rationally allocated to suitable equipment while meeting priority and concentration prediction requirements, achieving optimal resource scheduling. This improves the throughput of laboratory testing pipelines, reduces equipment idle time and waiting time, and avoids delays or invalid test results caused by resource conflicts. Simultaneously, the optimized equipment utilization and task scheduling strategy can also reduce overall testing costs and improve scalability and operational efficiency in large-scale sample testing scenarios. Attached Figure Description

[0038] Figure 1 A flowchart of a parallel task scheduling optimization method for a blood testing process provided by the present invention;

[0039] Figure 2 A flowchart of the task scheduling scheme provided by the present invention;

[0040] Figure 3 This invention provides a system architecture diagram for a parallelized task scheduling optimization system for a blood testing process. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0042] Example 1:

[0043] Please see Figure 1 This invention provides a parallel task scheduling optimization method for a blood testing process, the technical solution of which is as follows:

[0044] A database of basic decay rates and corresponding environmental impact coefficients for each biochemical component are pre-established. Based on real-time environmental parameters, the decay rates of temperature-sensitive components, protein components, and light-sensitive components are corrected to obtain the corrected decay rates under the current environmental conditions.

[0045] The establishment of the baseline decay rate database employed standardized experimental methods. First, under strictly controlled standard environmental conditions—temperature maintained at 20℃ ± 0.5℃, relative humidity controlled at 50% ± 5%, and in the dark—long-term stability tests were conducted on various biochemical components. During the tests, samples were periodically collected, and high-precision analytical instruments were used to measure the concentration changes of each component. Measurement time points included 0 hours, 1 hour, 2 hours, 4 hours, 8 hours, 12 hours, 24 hours, 48 ​​hours, and 72 hours. By fitting the concentration-time curves, the baseline decay rate constants of each component under standard conditions were obtained. The database stores information including component name, chemical structure identifier, molecular weight, baseline decay rate constant, measurement temperature, measurement humidity, measurement illumination conditions, data acquisition date, and data validity period.

[0046] The environmental impact coefficient database was obtained through systematic multi-factor experiments. For each biochemical component, different temperature, humidity, and light intensity gradients were designed. Temperature gradients were set at 4℃, 10℃, 15℃, 20℃, 25℃, 30℃, and 37℃; relative humidity gradients were set at 30%, 40%, 50%, 60%, 70%, and 80%; and light intensity gradients were set at 0 lx, 50 lx, 100 lx, 200 lx, 500 lx, and 1000 lx. Attenuation rate measurements were repeatedly performed under each environmental condition combination to obtain attenuation rate data for the component under different environmental conditions. Temperature, humidity, and light intensity coefficients were obtained through data fitting and regression analysis. The database structure includes component identification, environmental condition parameters, corresponding impact coefficient values, coefficient applicability range, measurement error range, and data update time.

[0047] The real-time environmental parameter acquisition system consists of a distributed sensor network. Temperature acquisition uses digital temperature sensors deployed in the blood sample storage area and around the testing equipment, with each sensor equipped with a wireless data transmission module. Humidity monitoring uses capacitive humidity sensors to measure relative humidity. Light intensity monitoring uses silicon photodiode sensors to measure light intensity in the visible light range.

[0048] Furthermore, for temperature-sensitive components, the Arrhenius equation is used to calculate the temperature correction factor, and the temperature-corrected decay rate is obtained by multiplying the basic decay rate by the temperature correction factor.

[0049] Specifically, for temperature-sensitive components, the activation energy parameter and standard reaction temperature of the component are first extracted from the component database. The activation energy data is obtained through a pre-conducted Arrhenius equation fitting experiment, which is carried out at at least five different temperature points, covering the actual application scenarios. During the real-time correction process, the current ambient temperature value is obtained, converted to Kelvin temperature units, and the temperature correction factor is calculated in combination with the component's activation energy parameter. This correction factor reflects the change factor of the component's decay rate under the current temperature condition relative to the standard temperature condition. Finally, the basic decay rate is multiplied by the temperature correction factor to obtain the corrected decay rate under the current temperature condition.

[0050] For protein components, the combined influence factor of temperature and humidity is calculated based on the protein thermodynamic stability parameters. The protein corrected decay rate is obtained by multiplying the basic decay rate by the combined influence factor.

[0051] Specifically: the correction of protein components takes into account the synergistic effect of temperature and humidity; during the real-time correction process, the current temperature and humidity values ​​are read, and the degree of joint influence of temperature and humidity on protein stability is calculated based on the hydration mechanism of protein molecules and the theory of thermal stability; the calculation of the joint influence factor comprehensively considers the influence of temperature on protein molecule motion and the influence of humidity on protein hydration layer, and the base decay rate is multiplied by the joint influence factor to obtain the protein correction decay rate considering the synergistic effect of temperature and humidity.

[0052] For photosensitive components, the light attenuation factor is calculated based on the light intensity and cumulative exposure time. The base attenuation rate is added to the light attenuation factor to obtain the corrected attenuation rate of the photosensitive component.

[0053] Specifically, during the real-time correction process, the ambient light intensity is continuously monitored, and the cumulative exposure time of each component is recorded. The calculation of the light attenuation factor is based on the first-order kinetics of the photochemical reaction, considering the linear relationship between light intensity and reaction rate, as well as the time integral effect of the cumulative exposure dose. The cumulative light dose is obtained by multiplying the real-time measured light intensity by the exposure time, and then the light attenuation factor is calculated based on the photosensitivity parameters of the component. This attenuation factor is combined with the basic attenuation rate through addition to obtain the corrected attenuation rate of the photosensitizing component.

[0054] In this embodiment, by constructing a basic decay rate database for biochemical components and an environmental impact coefficient database, and combining real-time collected environmental parameters such as temperature, humidity, and light, dynamic correction of the decay rates of various components during blood testing is achieved. The Arrhenius equation, a protein-temperature-humidity synergistic model, and a photodose kinetic formula are used to specifically correct temperature-sensitive, protein-based, and light-sensitive components, significantly improving the accuracy and adaptability of decay rate calculations. The system enables accurate assessment and rapid adjustment of sample stability under environmental changes, providing data support for the parallel scheduling of blood testing tasks and the reliability of results, thereby improving the overall efficiency and quality control level of the testing process.

[0055] Furthermore, the process of obtaining the corrected attenuation rate is as follows: real-time environmental data is collected, and environmental parameters are updated at preset time intervals;

[0056] Substitute real-time environmental parameters into the correction formula to obtain the environmental correction factor, which is then used to correct the temperature correction decay rate, protein correction decay rate, and photosensitizing component correction decay rate. The corrected temperature correction decay rate, protein correction decay rate, and photosensitizing component correction decay rate are then fused to obtain the correction decay rate of each component at the current moment, and stored and updated.

[0057] Specifically, within each update cycle, the latest environmental data is collected from each sensor node, and data validity is verified and outlier filtering is performed. The verified real-time environmental parameters are compared with standard environmental conditions in the environmental impact coefficient database. When the real-time parameters do not perfectly match the discrete data points in the database, a multidimensional linear interpolation method is used to calculate the temperature impact coefficient, humidity impact coefficient, and light impact coefficient that best reflect the current environmental conditions. Based on the degree of deviation between the current environmental conditions and the standard environmental conditions, an environmental correction factor is calculated. This correction factor reflects the adjustment range of the standard impact coefficient by the actual environmental conditions. For temperature-sensitive components, the environmental correction factor is multiplied by the temperature impact coefficient obtained from the database to obtain the adjusted temperature impact coefficient adapted to the current environment. This adjusted coefficient is then multiplied by the temperature correction decay rate to obtain the corrected decay rate of the temperature-sensitive component. For protein components, the environmental correction factor is applied to adjust the temperature impact coefficient and humidity impact coefficient respectively. Considering the combined effect of temperature and humidity on protein stability, the protein stability is adjusted by combining the correction factor. The decay rate is adjusted accordingly. For photosensitive components, the light influence coefficient is adjusted using an environmental correction factor based on changes in cumulative exposure time and light intensity. Combined with the time-dependent characteristics of photochemical reactions, the decay rate of photosensitive components is dynamically corrected. Finally, the corrected temperature-corrected decay rate, protein-corrected decay rate, and photosensitive component-corrected decay rate are numerically fused (weighted) to obtain the current corrected decay rate. In this embodiment, by periodically collecting and updating ambient temperature, humidity, and light intensity data, and combining them with an environmental influence coefficient database, dynamic correction of the decay rate of each biochemical component is achieved. In each update cycle, the sensor data is validated and anomaly filtered. Multidimensional linear interpolation is used to compensate for the lack of discrete points in the database to obtain the temperature, humidity, and light coefficients that are closest to reality. The environmental correction factor is calculated to adaptively adjust the standard coefficients. By weighting and fusing the basic decay rate with the corrected coefficients, the accurate corrected decay rate at the current moment is generated, ensuring the stability assessment of blood samples and the real-time reliability of detection data.

[0058] The concentration change rate of each test item is calculated based on the corrected attenuation rate, and a multi-dimensional priority evaluation matrix is ​​established by combining the clinical importance coefficient of the test item and the testing time requirements.

[0059] Furthermore, the process of obtaining the concentration change rate includes:

[0060] Obtain the initial concentration values ​​of the corresponding components for each test item in the blood sample; multiply the corrected decay rate by the current concentration value to calculate the concentration change rate of each component at the current time; calculate the concentration change rate of each component at future time points based on the first-order decay kinetic model.

[0061] Specifically, the initial concentration value acquisition and verification mechanism is as follows: A standardized pre-detection process is used to acquire the initial concentration values ​​of the corresponding components for each test item in the blood sample. First, newly collected blood samples are numbered, registered, and basic information is entered, including sampling time, sampling conditions, sample source, and a list of expected test items. For each test item, a small amount of serum or plasma is extracted from the sample for rapid concentration detection using mature quantitative detection technologies such as immunoturbidimetry, enzyme-linked immunosorbent assay (ELISA), or chemiluminescence immunoassay. Rapid detection uses specialized pre-detection equipment, with the detection time controlled within five minutes to ensure that the obtained initial concentration data accurately reflects the component concentration status at the time of sampling. The obtained initial concentration values ​​are compared and verified with the normal reference range for that component to identify abnormally high or low values ​​and mark any potential interfering factors. All initial concentration data, along with the detection timestamp, detection equipment number, and operator information, are stored in the sample database.

[0062] Calculation of the current concentration change rate: The concentration change rate of each component at the current moment is calculated based on the first-order decay kinetics principle. During the calculation, the latest corrected decay rate value for each component is first extracted from the real-time database, taking into account the influence of current environmental conditions. Then, the current concentration value of the component is read. For newly collected samples, the current concentration value is equal to the initial concentration value. For samples that have been stored for a period of time, the current concentration value needs to be updated based on the previous decay history. The corrected decay rate is multiplied by the current concentration value to obtain the absolute concentration change rate of the component at the current moment, expressed in concentration units per minute.

[0063] A method for predicting the rate of concentration change at future time points: Based on first-order decay kinetics, the method predicts the rate of concentration change of each component at any future time point. The prediction calculation uses a numerical integration method, dividing the time axis into several small time steps, typically one minute per step. Starting from the current moment, the change in component concentration within each time step is calculated step by step. Within each time step, assuming the decay rate remains constant, the current concentration value is multiplied by the corrected decay rate to obtain the concentration change for that time period, and then the component concentration value is updated for the next calculation. Considering that environmental conditions may change over time, the corrected decay rate is updated periodically during the prediction process. Through this step-by-step calculation method, the concentration values ​​and corresponding rates of change of each component at future time points are accurately predicted.

[0064] In this embodiment, a multi-dimensional priority evaluation matrix for testing items is constructed based on the coupled calculation of the corrected decay rate and the concentration change rate, enabling the scientific sorting and scheduling of blood sample testing tasks. The initial concentration of each component is quickly obtained and verified through a standardized pre-testing process. The current concentration change rate is calculated by combining the real-time corrected decay rate, and the concentration change trend at future time points is accurately predicted using first-order decay kinetics and step-integral methods. By comprehensively considering the concentration change rate, clinical importance coefficient, and testing timeliness requirements, the testing order and resource allocation can be dynamically adjusted to avoid data deviations caused by environmental factors or storage time for key components, significantly improving the stability and accuracy of test results and the overall processing efficiency of the laboratory.

[0065] Furthermore, the process of obtaining the multi-dimensional priority evaluation matrix includes:

[0066] The urgency of decay is calculated based on the rate of concentration change, and the urgency coefficient is obtained by dividing the rate of concentration change by the detection accuracy requirement. The clinical importance coefficient of each test item is obtained from the clinical database and determined according to the disease diagnosis weight and emergency priority. The time constraint coefficient is calculated based on the detection time requirement and the remaining available detection time. A three-dimensional evaluation matrix including the urgency coefficient, clinical importance coefficient and time constraint coefficient is established, and the comprehensive priority score of each test item is calculated by weighted summation.

[0067] Specifically, the calculation of the attenuation urgency coefficient is based on the relative relationship between the rate of concentration change and the required detection accuracy. A database of accuracy requirements for each detection item is pre-established, including technical indicators such as the minimum detection limit, linear range, precision requirements, and accuracy requirements for each item. These indicator data are derived from the technical specifications of the detection equipment, industry standard requirements, and clinical laboratory quality control specifications. When calculating the attenuation urgency coefficient, the key accuracy indicators for the detection item are first determined, typically using the coefficient of variation or relative standard deviation as the evaluation benchmark. The current rate of concentration change of the component is divided by the corresponding accuracy requirement value to obtain the dimensionless attenuation urgency coefficient. The larger the coefficient value, the more significant the impact of component concentration attenuation on the accuracy of the detection results, and the higher the urgency of the detection. The calculated attenuation urgency coefficient is standardized, mapping the value range to between zero and one, facilitating comparison and weight calculation with other evaluation indicators.

[0068] The establishment of the clinical importance coefficient database employs a multi-level hierarchical assessment method. Firstly, based on the role of each test item in disease diagnosis, they are categorized into four levels: key diagnostic indicators, important reference indicators, auxiliary diagnostic indicators, and routine monitoring indicators. Key diagnostic indicators include tests that can directly confirm the diagnosis, such as troponin in myocardial infarction, blood glucose in diabetes, and transaminase levels in liver function. Important reference indicators include tests with significant diagnostic value, such as white blood cell count in inflammatory responses and hemoglobin concentration in anemia. Auxiliary diagnostic indicators include tests that are helpful in diagnosis but not essential, such as various vitamins and trace elements. Routine monitoring indicators include routine biochemical indicators found in health checkups.

[0069] In the disease diagnosis weighting assessment, different weight scores are assigned based on the contribution of each test item to the diagnosis of a specific disease; the emergency priority assessment is based on the patient's clinical status and the urgency of their condition; referring to internationally accepted emergency triage standards, patients are divided into five levels: critical, severe, acute, secondary acute, and non-acute; the emergency priority coefficient is automatically determined according to the patient's triage level, with a priority coefficient of 1 for critical and severe patients, 0.8 for acute patients, 0.6 for secondary acute patients, and 0.4 for non-acute patients; the final value of the clinical importance coefficient is obtained by weighting the disease diagnosis weight and the emergency priority coefficient.

[0070] The time constraint coefficient reflects the time urgency and resource constraints of the testing task. It obtains the standard testing time requirements for each testing item, including sample preprocessing time, instrument testing time, result analysis time, and report generation time. It monitors the current equipment usage and task queue length in the laboratory in real time, calculating the remaining available testing time for various types of testing equipment. The calculation of remaining available time considers the equipment's operating status, maintenance plans, operator arrangements, and the time occupied by other scheduled tasks. It compares the standard testing time for each testing item with the remaining available time of the corresponding equipment to calculate the time pressure index. When the remaining available time is sufficient, the time constraint coefficient is close to 0.1; when the remaining available time is tight, the time constraint coefficient gradually increases; when the remaining available time is insufficient to complete the testing, the time constraint coefficient reaches 1.

[0071] The three-dimensional evaluation matrix uses a matrix data structure to store and manage multi-dimensional evaluation indicators. Rows correspond to different testing items, and columns correspond to three evaluation dimensions: attenuation urgency coefficient, clinical importance coefficient, and time constraint coefficient. Each evaluation dimension is assigned a corresponding weight, based on clinical expert opinions and historical data analysis results. In the standard configuration, the attenuation urgency coefficient is weighted at 0.4, reflecting the significant impact of sample quality on test results; the clinical importance coefficient is weighted at 0.4 to ensure priority for clinical needs; and the time constraint coefficient is weighted at 0.2 to balance resource utilization efficiency. The weight configuration can be adjusted according to different application scenarios. For example, the weight of the time constraint coefficient can be increased in emergency department applications, while the weight of the attenuation urgency coefficient can be increased in research applications, with the total weight always required to equal 1.

[0072] The overall priority score is calculated using a weighted summation method to ensure that the contributions of each evaluation dimension are reasonably reflected. The attenuation urgency coefficient, clinical importance coefficient, and time constraint coefficient for each test item are multiplied by their corresponding weights, and then the three products are summed to obtain the overall priority score for that item. During the calculation process, all input parameters are validated to ensure that the coefficient values ​​are within a reasonable range and the weights are correctly configured. To facilitate comparison between different batches of test tasks, the calculated overall priority score is standardized, with higher values ​​indicating higher priority. A confidence interval for the overall priority score of each test item is also calculated to account for the impact of input parameter uncertainties on the final result. All calculation results, along with the calculation timestamp, parameter version, and weight configuration information, are stored in the priority database. The task scheduling algorithm provides a reliable basis for decision-making and supports the tracing and analysis of historical data. In this embodiment, a priority evaluation matrix containing three-dimensional indicators—attenuation urgency, clinical importance, and time constraints—is constructed to achieve intelligent sorting and optimized resource allocation for blood testing items. The attenuation urgency is quantified based on the concentration change rate and detection accuracy requirements. A clinical importance coefficient is generated by combining disease diagnosis weights and emergency triage, and the remaining equipment time and task load are introduced to calculate the time constraint coefficient. The three-dimensional matrix is ​​weighted and summed to comprehensively reflect sample stability, clinical value, and testing resource status, generating a standardized comprehensive priority score and confidence interval. This mechanism can flexibly adjust the weights to adapt to different application scenarios, effectively ensuring the timeliness and accuracy of key item testing, while improving the overall scheduling efficiency and quality control level of the laboratory.

[0073] A database of chemical reaction relationships among various biochemical components in blood samples is pre-established, identifying component pairs with interactive effects and their reaction types; a coupled reaction kinetic model is established to predict the concentration values ​​of each component at a set time point;

[0074] Specifically, the establishment of the chemical reaction relationship database employs a combination of systematic literature review and experimental verification. Authoritative databases in biochemistry, clinical chemistry, and analytical chemistry, including the National Center for Biotechnology Information (NCBI) database, Chemical Abstracts, and peer-reviewed journal articles, are searched to collect information on interactions between common biochemical components in blood samples. The data collection scope covers various reaction types, including enzyme-catalyzed reactions, redox reactions, complexation reactions, competitive inhibition reactions, and non-competitive inhibition reactions. For each pair of interacting biochemical components, key information such as reaction mechanism, reaction conditions, reaction products, reaction rate, and equilibrium state is recorded. The database uses a relational data structure, with main fields including reaction number, reactant component identifier, product component identifier, reaction type classification, reaction mechanism description, applicable pH range, applicable temperature range, reaction activation energy, data source, and verification status.

[0075] Furthermore, the coupled reaction kinetic model includes:

[0076] The component identification layer extracts all biochemical components involved in the test item and their interactions from a chemical reaction relationship database;

[0077] Specifically, the component identification layer, as the first layer of the coupled reaction kinetics model, is responsible for accurately extracting all biochemical components and their interactions related to the current detection task from the chemical reaction relationship database; reading the current blood sample's test item list to establish a basic list of components to be detected; then, the system queries the chemical reaction relationship database to identify other components that have direct interactions with any component in the basic list and adds these components to the extended component list; to ensure the integrity of the model, a recursive search algorithm is used to continue querying the interaction relationships for each newly added component in the extended component list until no new related components are found; a component interaction relationship graph is established to graphically display the connections and reaction paths between components; each component node contains basic information such as component name, chemical identifier, molecular weight, and normal concentration range; each connection edge contains relationship information such as reaction type, reaction direction, and reaction intensity; the identified components are classified and labeled to distinguish between directly detected target components, indirectly affecting components, and intermediate product components;

[0078] The reaction parameter layer is used to obtain the reaction rate constant, equilibrium constant, and inhibition constant between each component pair.

[0079] Specifically: the reaction parameter layer is responsible for acquiring detailed reaction kinetic parameters between each component pair; a dedicated reaction parameter database has been established to store reaction kinetic parameters obtained from experimental measurements and literature reviews; for reversible chemical reactions, parameters such as the forward reaction rate constant, reverse reaction rate constant, equilibrium constant, and enthalpy change are recorded; for inhibitory reactions, parameters such as competitive inhibition constant, non-competitive inhibition constant, and mixed inhibition constant are recorded; during parameter acquisition, parameter values ​​measured under physiological conditions are prioritized, including physiological pH, physiological ionic strength, and body temperature; for reaction parameters lacking direct experimental data, estimation is performed using structure-activity relationship prediction methods or quantum chemical calculation methods; uncertainty analysis is conducted on all parameter data, and the measurement error range and confidence interval are recorded to provide a basis for error propagation analysis for numerical solutions;

[0080] The kinetic equation layer establishes a set of first-order ordinary differential equations describing the change of multi-component concentrations over time based on the law of mass action.

[0081] Specifically, the process involves: establishing a set of differential equations describing the concentration changes of multiple components over time based on the law of mass action and reaction kinetics; analyzing the component interaction diagram to identify all reaction processes involved by each component, including reactions consumed as reactants and reactions generated as products; for each biochemical component, establishing a mass balance equation, where the left side of the equation represents the rate of change of component concentration over time, and the right side represents the net formation rate of the component, i.e., the formation rate minus the consumption rate; the calculation of formation and consumption rates is based on the law of mass action, considering the product of reactant concentrations, the corresponding reaction rate constant, and the reaction order; handling complex multi-step reaction mechanisms by decomposing them into several elementary reaction steps, each following the law of mass action; for reversible reactions, considering both forward and reverse reaction rates; automatically checking the mathematical consistency of the equation set to ensure that atomic and charge conservation are satisfied; and finally establishing a set of differential equations containing the concentration change equations of all relevant components, forming a coupled first-order ordinary differential equation system.

[0082] The numerical solution layer uses the Runge-Kutta algorithm to numerically integrate the differential equation system, calculate the concentration change curves of each component within a set time interval, and output the predicted concentration values ​​at each time point.

[0083] The numerical solution layer employs a fourth-order Runge-Kutta algorithm to numerically integrate and solve the differential equation system. First, the time range and time step are set. The time range is typically set from the current time to the expected completion time of the detection, while the time step is determined based on the system's rigidity and accuracy requirements, generally on the order of seconds or minutes. The Runge-Kutta algorithm implementation process includes four calculation stages, each requiring the calculation of the concentration change rate of all components at a specific time point. At the beginning of each time step, the current component concentration value is read as the initial condition for the Runge-Kutta algorithm. The first stage calculates the change rate based on the current concentration value; the second and third stages calculate the change rate based on the predicted concentration at the midpoint of the time step; and the fourth stage calculates the change rate based on the predicted concentration at the end of the entire time step. The calculation results from the four stages are combined according to the weighted formula of the Runge-Kutta algorithm to obtain the change in concentration of each component within that time step, updating the component concentration values ​​for the calculation of the next time step. To ensure numerical stability, the accumulation of numerical errors during the calculation process is monitored, and when the error exceeds a preset threshold, the time step is automatically reduced and the calculation is recalculated. The output is time-series data containing the predicted concentration values ​​of each component at all time points.

[0084] In this embodiment, a database of chemical reaction relationships among biochemical components of blood samples is constructed, and the concentrations of multiple components at set time points are accurately predicted based on a coupled reaction kinetic model. The components to be detected and their interactions are extracted to establish a complete component relationship diagram. Key parameters such as rate constants, equilibrium constants, and inhibition constants are obtained from the reaction parameter database, and error and confidence level analyses are performed. Subsequently, a system of first-order differential equations describing the concentration changes of multiple components over time is constructed based on the law of mass action, comprehensively describing the generation and consumption processes. Finally, a fourth-order Runge-Kutta algorithm is used for numerical integration to obtain the predicted curve of concentration changes over time. This achieves quantitative modeling and dynamic simulation of chemical reactions among complex components in blood samples, providing a reliable basis for correcting test results and optimizing test timing, significantly improving the accuracy and timeliness of blood analysis.

[0085] All items to be detected are sorted according to a multi-dimensional priority evaluation matrix; when the combination of items to be detected contains mutually influential components, the detection order is adjusted according to the coupling attenuation prediction results; the adjusted detection tasks are assigned to the corresponding parallel detection devices according to device compatibility, and a task scheduling scheme containing detection order, device allocation and time arrangement is generated and executed.

[0086] Furthermore, the process of ranking and adjusting the detection items based on the multi-dimensional priority evaluation matrix includes:

[0087] All items to be tested are initially sorted from high to low according to their comprehensive priority scores; component pairs that influence each other in the initial sort are identified, and the concentration prediction values ​​of these component pairs are obtained from the coupled reaction kinetic model; the degree of concentration influence between each component is calculated, and the test items whose influence exceeds the preset threshold are marked as high-impact items.

[0088] Adjust the testing order, prioritizing high-impact items before the components they affect, while maintaining the relative order of the overall priority ranking.

[0089] Specifically, the initial ranking of the items to be tested is based on the comprehensive priority score calculated by a multi-dimensional priority evaluation matrix. First, the comprehensive priority scores of all items to be tested in the current batch are extracted from the priority database to verify the completeness and validity of the data. During the ranking process, detailed information of each item is recorded, including item number, item name, corresponding biochemical components, sample source, expected testing time, and equipment requirements. An initial ranking list is generated, which is arranged from high to low according to the comprehensive priority score, with the highest priority item at the top of the list.

[0090] Identification and Analysis of Interacting Component Pairs: A cross-comparison algorithm is used to identify interacting component pairs in the initial sorting. The identification process first traverses all detection items in the initial sorting list and extracts the biochemical component identifiers corresponding to each item. Then, the system queries the chemical reaction relationship database to retrieve information on component pairs that have direct or indirect interactions with these components. For each pair of interacting components, detailed information such as interaction type, intensity, and direction is recorded. The coupled reaction kinetic model is invoked, and the current component concentration value and expected detection time schedule are input to calculate the predicted concentration changes of these component pairs under different detection sequences. The prediction calculation considers the time delay effect during the detection process, i.e., the component detected first may continue to change concentration during the detection process, thus affecting the initial concentration value of the component detected later. An influence relationship matrix is ​​established, where the rows and columns of the matrix represent different detection items, and the matrix elements represent the degree of influence of the row item on the column item. The degree of influence is quantified by the relative percentage of concentration change. When the concentration change caused by the influence exceeds a specific multiple of the detection accuracy requirement of the component, a significant influence relationship is considered to exist.

[0091] High-impact item labeling and classification: Testing items are classified and labeled according to a preset impact threshold. The impact threshold is set based on the testing accuracy requirements and the clinically acceptable error range, typically two to five times the testing accuracy. When the maximum impact of a testing item on other items exceeds the preset threshold, the item is labeled as a high-impact item. During the labeling process, the impact characteristics of each high-impact item are recorded in detail, including the target component affected, the degree of impact, the duration of impact, and the type of impact mechanism. Further analysis of the interrelationships between high-impact items is conducted to identify combinations of items with cascading effects. For cascading effects, i.e., item A affects item B, and item B in turn affects item C, graph theory algorithms are used to analyze the impact propagation path and determine the optimal testing order. A priority sub-ranking of high-impact items is established, and within each high-impact item, they are still ranked according to their comprehensive priority score to ensure that clinically important items are prioritized.

[0092] The intelligent algorithm for adjusting the detection order employs a constraint optimization method to maximize overall detection efficiency while satisfying the constraints of influence relationships. The adjustment process first fixes the relative positions of high-impact items and then reorders them according to the time dependence of their influence relationships. Specifically, if item A has a significant impact on item B, and this impact accumulates and strengthens over time, item A is scheduled for detection before item B. If the impact weakens over time, the detection time of item A is delayed to reduce interference with item B. A dynamic programming algorithm is used to find the optimal adjustment scheme. The objective function comprehensively considers the impact of the detection order adjustment on the overall priority score ranking, the overall detection time after adjustment, and the expected loss of detection accuracy. During algorithm execution, a priority queue of candidate adjustment schemes is maintained. Each scheme in the queue contains a complete detection order arrangement and its corresponding objective function value. Infeasible or suboptimal adjustment schemes are pruned using a branch and bound method to improve the algorithm's execution efficiency. Finally, the adjustment scheme with the optimal objective function value is selected as the recommended detection order. After adjustment, it is verified whether the new detection order satisfies all influence relationship constraints, ensuring that there are no circular dependencies or time conflicts.

[0093] In this embodiment, intelligent sorting and optimized scheduling of blood testing tasks are achieved by combining a multi-dimensional priority evaluation matrix with a coupled reaction kinetic model. Based on the initial comprehensive priority sorting, the chemical interactions between testing items are identified and quantified, an influence relationship matrix is ​​constructed, and high-impact and cascading-impact items are screened. Using constraint optimization and dynamic programming algorithms, considering testing accuracy, time delay, and equipment compatibility, a testing sequence scheme that satisfies the influence constraints and has the best overall efficiency is generated. Finally, a scheduling plan including testing sequence, equipment allocation, and time arrangement is output, which effectively reduces interference between components and improves the accuracy of test results and the overall efficiency of the testing process.

[0094] This invention provides a parallel task scheduling optimization method for blood testing processes, as detailed in the following reference. Figure 2 A comprehensive data support and decision-making system was constructed, focusing on sample stability, clinical value, and resource constraints. First, a basic decay rate database and an environmental impact coefficient database were established. Combining real-time environmental parameters such as temperature, humidity, and light, the decay rates of temperature-sensitive, protein-sensitive, and photosensitive components were corrected using the Arrhenius equation, a temperature-humidity synergistic model, and photodose kinetics, respectively, to obtain high-precision decay rates. Dynamic updates of the decay rates were achieved by periodically collecting sensor data and utilizing multidimensional interpolation and correction factors. Based on this, the concentration change rate was calculated, and a multi-dimensional priority evaluation matrix was constructed based on clinical importance and time constraints, outputting a standardized comprehensive priority score. Furthermore, a chemical reaction relationship database and a coupled kinetic model were introduced to identify interactions between components and predict future concentration changes. This enabled intelligent sequencing of blood testing tasks and optimal allocation of equipment resources, effectively reducing environmental and chemical interference and improving the stability of test results and the overall processing efficiency of the laboratory.

[0095] Example 2:

[0096] This invention provides a parallel task scheduling optimization method for a blood testing process, applied to a parallel task scheduling optimization system for a blood testing process, with reference to... Figure 3The technical solution is as follows: A decay rate correction module pre-establishes a basic decay rate database and a corresponding environmental impact coefficient database for each biochemical component; based on real-time environmental parameters, it corrects the decay rates of temperature-sensitive components, protein components, and light-sensitive components to obtain the corrected decay rate under the current environmental conditions; a priority assessment module calculates the concentration change rate of each test item based on the corrected decay rate, and establishes a multi-dimensional priority assessment matrix by combining the clinical importance coefficient of the test item and the test time requirements; a coupling reaction module pre-establishes a database of chemical reaction relationships between various biochemical components in the blood sample, identifies component pairs with interactive effects and their reaction types; establishes a coupling reaction kinetic model to predict the concentration values ​​of each component at a set time point; a parallel task scheduling module sorts all test items according to the multi-dimensional priority assessment matrix; when the test item combination contains mutually influential components, it adjusts the test order based on the coupling decay prediction results; the adjusted test tasks are allocated to the corresponding parallel test devices according to device compatibility, generating a task scheduling scheme containing the test order, device allocation, and time arrangement, and then executing it.

[0097] This method also includes anomaly detection and real-time scheduling adjustment mechanisms:

[0098] Real-time monitoring of the operating status of each testing device, fluctuations in test results, and changes in sample status;

[0099] Set up anomaly detection rules, including equipment failure detection, detection of test results exceeding the expected range, and abnormal sample decay detection;

[0100] When an anomaly is detected, the emergency dispatch procedure is activated: the currently affected testing tasks are suspended, the priority of the remaining testing items is reassessed, and tasks are reallocated taking into account equipment availability and time constraints.

[0101] Record the anomaly handling process and its effects to continuously optimize anomaly detection rules and recovery strategies.

[0102] Specifically:

[0103] The anomaly judgment rules are constructed based on a multi-level threshold system and intelligent recognition algorithms. Equipment fault detection adopts a multi-parameter comprehensive judgment method, and a pre-established equipment fault feature database records typical parameter change patterns for different fault types. Common equipment faults include light source aging, detector sensitivity decline, mechanical component wear, temperature control system failure, and reagent supply abnormalities. Pattern recognition algorithms are used to analyze the changing trends of real-time monitoring data. When the change pattern of the monitoring parameters matches known fault features with a degree of over 80%, the equipment is judged to have the corresponding type of fault risk. For progressive faults, trend analysis methods are used to predict the time window of fault occurrence by fitting parameter change curves. For sudden faults, the system sets an emergency threshold, and triggers a fault alarm immediately when key parameters suddenly exceed the safe range.

[0104] The determination of test results exceeding the expected range adopts a multi-level screening mechanism; the first level of screening is based on the statistical reference interval, and when the test result exceeds the normal reference range of the item, it is marked as a suspicious result; the second level of screening is based on clinical rationality assessment, and the system evaluates the clinical rationality of the value based on the patient's basic information, medical history and other test results; the third level of screening is based on technical reproducibility verification, and repeat testing or verification using alternative testing methods are arranged.

[0105] The detection of abnormal sample decay is based on the prediction deviation analysis of the component decay kinetic model. The actual observed changes in component concentration are compared with the theoretically predicted decay curve, and the magnitude and duration of the deviation are calculated. When the deviation exceeds three times the standard deviation of the prediction error and the duration exceeds ten minutes, the sample is determined to have experienced abnormal decay. Possible causes of abnormal decay include sample contamination, unexpected chemical reactions, abnormal storage conditions, and sample processing errors.

[0106] The emergency dispatch procedure adopts a tiered response mechanism, which determines the response level based on the severity and scope of the anomaly. Level 1 response is for severe equipment failure or severe sample anomaly, requiring the immediate suspension of all affected testing tasks and the initiation of a comprehensive rescheduling. Level 2 response is for moderate anomalies, suspending directly affected testing tasks and making local scheduling adjustments. Level 3 response is for minor anomalies, requiring enhanced monitoring and preventative adjustments without suspending testing tasks.

[0107] The anomaly handling logging system employs a structured data recording format to meticulously record the entire process of each anomaly event. Records include the anomaly occurrence time, anomaly type, triggering conditions, detection environment, affected tasks, handling measures taken, handling process, recovery time, handling results, and subsequent impacts. The system evaluates handling effectiveness using a multi-indicator comprehensive evaluation method. Evaluation indicators include the accuracy of anomaly identification, the timeliness of handling response, the effectiveness of recovery strategies, the rationality of resource utilization, and the impact on overall efficiency. An effectiveness evaluation database is established to store performance data of different handling strategies under various conditions. Through comparative analysis, the most effective combination of handling strategies is identified, providing data support for strategy optimization.

[0108] The continuous optimization mechanism uses machine learning methods to analyze historical anomaly handling data, identify patterns in anomaly occurrences and areas for improvement in handling strategies; it periodically updates anomaly judgment thresholds, adjusting threshold settings based on accumulated monitoring data and handling experience to improve the accuracy and timeliness of anomaly detection; and it employs reinforcement learning algorithms to continuously improve the selection of handling strategies and parameter settings through trial-and-error learning and reward feedback.

[0109] This method achieves comprehensive risk control and dynamic optimization of the blood testing process. Through multi-dimensional monitoring of equipment operating status, test result fluctuations, and sample attenuation, it can quickly identify potential equipment problems such as light source aging, mechanical wear, and reagent abnormalities. It can accurately capture situations such as test values ​​exceeding reasonable ranges or abnormal sample attenuation, and respond according to severity, promptly pausing, rescheduling, or fine-tuning tasks. Structured recording and multi-index evaluation provide data support for strategy review and optimization, while machine learning and reinforcement learning continuously iterate the determination thresholds and recovery strategies. This mechanism effectively reduces the impact of faults and anomalies on testing quality and efficiency, ensuring process stability and efficient utilization of laboratory resources.

[0110] It also includes a multi-sample batch collaborative scheduling optimization step:

[0111] Record the testing requirements, sample characteristics, and time requirements for multiple blood samples within the same batch;

[0112] Construct a cross-sample resource optimization model to analyze the overlap of detection items and the complementarity of equipment requirements among different samples, and identify task combinations that can be shared and optimized for detection time;

[0113] Establish a batch-level multi-dimensional priority matrix to comprehensively consider the urgency of individual samples, the batch processing efficiency of testing items, and the overall utilization rate of equipment resources;

[0114] The testing order and equipment allocation of each sample within a batch are dynamically adjusted to achieve batch-level collaborative optimization scheduling.

[0115] Specifically, the cross-sample resource optimization model is constructed based on graph theory and combinatorial optimization theory. All detection tasks within a batch are represented as a graph structure, with nodes representing detection tasks and edges representing the possibility of resource sharing and optimization potential between tasks. Node attributes include the task's resource requirements, time requirements, priority, and constraints. Edge attributes include resource sharing type, optimization benefits, and implementation difficulty.

[0116] The analysis of overlapping test items identifies cases where different samples have the same test items; an overlap matrix is ​​established to statistically analyze the frequency and distribution of each test item within a batch; for frequently occurring test items, the feasibility and benefits of batch processing are systematically evaluated; factors to consider in batch processing include the batch processing capability of the test method, quality control requirements, the risk of cross-contamination between samples, and the complexity of time coordination; a batch processing benefit model is established to quantify the time and resource savings of batch processing compared to individual processing.

[0117] The equipment demand complementarity analysis identifies opportunities for collaborative optimization in equipment usage for different sample testing tasks; the equipment demand schedule for each sample is analyzed to identify idle and overlapping periods of equipment usage; for sample combinations with complementary equipment usage times, the system evaluates the feasibility of collaborative scheduling; collaborative scheduling considers practical constraints such as equipment changeover time, sample transmission time, operator coordination, and quality control requirements.

[0118] The batch-level multi-dimensional priority matrix extends the concept of single-sample priority assessment by adding batch-level assessment dimensions such as batch processing efficiency and system resource utilization. The matrix structure adopts a three-dimensional tensor form, with the first dimension corresponding to the samples within the batch, the second dimension corresponding to the test items within the samples, and the third dimension corresponding to multiple assessment dimensions. The urgency assessment of a single sample is based on factors such as the patient's clinical status, the diagnostic value of the test items, and time constraints. Batch-level collaborative optimization also includes cross-batch resource coordination. When multiple batches are processed in parallel, the system coordinates the resource allocation and time scheduling between different batches to avoid resource and time conflicts.

[0119] By constructing a graph theory-based cross-sample resource optimization model, the overlap of detection items and the complementarity of equipment requirements among different samples within a batch can be effectively identified, achieving optimal resource allocation and sharing. The three-dimensional batch-level priority matrix comprehensively considers the urgency of samples, batch processing efficiency, and overall equipment utilization, improving the scientific nature and comprehensiveness of scheduling decisions. By batch processing high-frequency detection items and optimizing equipment usage time arrangements, the overall detection time can be effectively reduced, and equipment utilization can be improved. The dynamic adjustment mechanism, based on real-time monitoring and feedback control, can respond promptly to changes during the execution process, ensuring the adaptability and robustness of the scheduling scheme.

[0120] Equipment compatibility analysis is based on a pre-established database of testing equipment capabilities and a database of testing item technical requirements. The equipment capability database records the technical specifications of all testing equipment in the laboratory, including testing principles, applicable testing item types, testing range, testing accuracy, sample volume, testing time, equipment status, and maintenance plans. The testing item technical requirements database records the technical requirements for each testing item, including the required testing methods, sample types, sample pretreatment requirements, testing environmental conditions, and result interpretation standards. A compatibility matrix is ​​generated by analyzing the compatibility of each testing item with each testing device through a matching algorithm. The compatibility assessment considers three dimensions: technical compatibility, performance compatibility, and time compatibility. Technical compatibility assesses whether the technical requirements of the item are within the technical capabilities of the equipment. Performance compatibility assesses whether the testing performance of the equipment meets the accuracy and precision requirements of the item. Time compatibility assesses whether the available time window of the equipment matches the time requirements of the item. The Hungarian algorithm is used to solve the optimal matching problem of equipment allocation, with the goal of maximizing equipment utilization efficiency while minimizing the overall testing time.

[0121] The task scheduling scheme is generated using a multi-objective optimization method, comprehensively considering multiple objectives such as detection efficiency, resource utilization, and result quality. First, a preliminary time arrangement scheme is generated based on the adjusted detection sequence and equipment allocation results. The time arrangement considers factors such as the estimated detection time for each detection item, equipment changeover time, sample transfer time, and quality control time. A Gantt chart model is established to graphically display the task scheduling timeline for each device, identifying time conflicts and resource bottlenecks. For device combinations with parallel detection capabilities, the system optimizes the time coordination of parallel tasks to ensure that sample transfer between different devices does not cause time waste or cross-contamination. Key performance indicators for each time arrangement scheme are calculated, including total detection time, average equipment utilization, task waiting time, and expected result accuracy. Based on these indicators, the system uses a genetic algorithm or simulated annealing algorithm to further optimize the time arrangement and find the Pareto optimal solution set. The final generated task scheduling scheme contains detailed execution instructions, specifying the start time, executing equipment, operator, quality control requirements, and contingency plan for each detection task. The scheme is output in a standardized format, supporting direct import into the laboratory information management system for execution, while retaining an interface for manual review and adjustment.

[0122] By establishing an intelligent matching system supported by dual databases, precise matching and optimal allocation of testing projects and equipment resources were achieved. A three-dimensional compatibility evaluation mechanism ensured comprehensive matching of technical requirements, performance indicators, and time constraints, improving the accuracy and reliability of equipment allocation. A multi-objective optimization method effectively balanced key indicators such as testing efficiency, resource utilization, and result quality, avoiding the limitations of single-objective optimization. The Gantt chart model provides intuitive timeline visualization, facilitating bottleneck identification and optimization of parallel processing strategies.

[0123] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A parallelization task scheduling optimization method for blood detection flow, characterized in that, Comprise: According to the real-time environmental parameters, the decay rates of temperature-sensitive components, protein components and light-sensitive components in blood are corrected to obtain the corrected decay rates of blood under the current environmental conditions; According to the corrected decay rates, the concentration change rates of each detection item blood are calculated, and a multi-dimensional priority evaluation matrix is established by combining the clinical importance coefficients of the detection item blood and the detection time requirements; A chemical reaction relationship database between each biochemical component in the blood sample is established in advance, and the component pairs and reaction types that have interactive effects are identified; A coupled reaction kinetics model is established to predict the concentration values of each component at a set time point; The coupled reaction kinetics model comprises: A component identification layer extracts all biochemical components involved in the detection item and their interaction relationships from the chemical reaction relationship database; A reaction parameter layer obtains the reaction rate constant, equilibrium constant and inhibition constant between each component pair; A kinetics equation layer establishes a first-order ordinary differential equation set describing the change of multi-component concentration with time according to the mass action law; A numerical solution layer uses the Runge-Kutta algorithm to numerically integrate the differential equation set, calculates the concentration change curve of each component within a set time interval, and outputs the concentration prediction value at each time point; According to the multi-dimensional priority evaluation matrix, all detection items are sorted; when the detection item combination contains components that interact with each other, the detection order is adjusted according to the coupled decay prediction results; the adjusted detection tasks are assigned to the corresponding parallel detection equipment according to the equipment compatibility, a task scheduling scheme including the detection order, equipment allocation and time arrangement is generated and executed.

2. The parallel task scheduling optimization method for blood detection process according to claim 1, characterized in that: For temperature-sensitive components in blood, a temperature correction factor is calculated using the Arrhenius equation, and the temperature correction decay rate is obtained by multiplying the basic decay rate and the temperature correction factor; For protein components in blood, a joint influence factor of temperature and humidity is calculated based on the protein thermodynamic stability parameters, and the protein correction decay rate is obtained by multiplying the basic decay rate and the joint influence factor; For light-sensitive components in blood, a light decay factor is calculated according to the light intensity and cumulative exposure time, and the light-sensitive component correction decay rate is obtained by adding the basic decay rate and the light decay factor.

3. The parallelization task scheduling optimization method of a blood test procedure according to claim 1, wherein: The correction decay rate acquisition process is: collecting real-time environmental data, updating environmental parameters every preset time interval; The real-time environmental parameters are substituted into the correction formula to obtain the environmental correction factor, which is used to correct the temperature correction decay rate, the protein correction decay rate and the light-sensitive component correction decay rate; The corrected temperature correction decay rate, protein correction decay rate and light-sensitive component correction decay rate are fused to obtain the corrected decay rate of each component in blood at the current time and stored and updated.

4. The parallel task scheduling optimization method for blood detection process according to claim 1, characterized in that: The concentration change rate acquisition process comprises: An initial concentration value of each component corresponding to each detection item in the blood sample is obtained; a concentration change rate of each component at the current time is calculated by multiplying the corrected decay rate by the current concentration value; and a concentration change rate of each component at a future time point is calculated according to a first-order decay kinetics model.

5. The parallelization task scheduling optimization method of a blood test procedure according to claim 1, wherein: The acquisition process of the multi-dimensional priority evaluation matrix comprises: According to the concentration change rate, the decay urgency is calculated, and the decay urgency coefficient is obtained by dividing the concentration change rate by the detection accuracy requirement; the clinical importance coefficient of each detection item is obtained from the clinical database according to the disease diagnosis weight and the emergency priority; the time constraint coefficient is calculated according to the detection time requirement and the remaining available detection time; a three-dimensional evaluation matrix containing the decay urgency coefficient, the clinical importance coefficient and the time constraint coefficient is established, and the comprehensive priority score of each detection item is calculated by weighted summation.

6. The parallelization task scheduling optimization method of a blood test procedure according to claim 1, wherein: The process of sorting and adjusting the detection items according to the multi-dimensional priority evaluation matrix comprises: All the detection items are initially sorted according to the comprehensive priority score from high to low; the component pairs with mutual influence in the initial sorting are identified, and the concentration prediction values of these component pairs are obtained from the coupled reaction kinetics model; the concentration influence degree between each component is calculated, and the detection items with an influence degree exceeding a preset threshold are marked as high-influence items; The detection order is adjusted, the high-influence items are arranged in priority before the detection of the components influenced by them, while the relative order of the overall priority sorting is kept unchanged.

7. A parallelization task scheduling optimization system for blood testing procedures, characterized in that, It comprises: A decay rate correction module corrects the decay rates of temperature-sensitive components, protein components and light-sensitive components in blood according to real-time environmental parameters to obtain the corrected decay rate of blood under the current environmental conditions; A priority evaluation module calculates the concentration change rate of each detection item blood according to the corrected decay rate, and establishes a multi-dimensional priority evaluation matrix combining the clinical importance coefficient of the detection item blood and the detection time requirement; A coupled reaction module pre-establishes a chemical reaction relationship database between each biochemical component in the blood sample, identifies the component pairs with mutual influence and the reaction types, establishes a coupled reaction kinetics model, and predicts the concentration values of each component at a set time point; A parallel task scheduling module sorts all the detection items according to the multi-dimensional priority evaluation matrix; when the detection item combination contains components with mutual influence, the detection order is adjusted according to the coupled decay prediction result; the adjusted detection tasks are distributed to the corresponding parallel detection equipment according to the equipment compatibility, a task scheduling scheme containing the detection order, equipment allocation and time arrangement is generated and executed.

Images