Temperature adaptive control system for organ perfusion

Abstract

The invention discloses a temperature self-adaptive control system for organ perfusion, and belongs to the technical field of temperature control, and the system comprises a sampling unit which is used for converting analog signals of temperature, flow velocity and pressure of one or more target nodes in perfusion blood circulation into digital signals to form an initial data set; the calculation unit is used for performing wavelet transformation and moving average processing on the initial data set to construct a feature matrix, and establishing a temperature change prediction model based on the feature matrix so as to determine a temperature compensation amount in combination with dynamic correction; and the temperature control unit is used for adjusting the temperature control power by adopting a progressive strategy based on the temperature compensation amount and synchronously adjusting the blood flow rate so as to dynamically optimize the control parameters according to the temperature control effect. Through the scheme of the invention, the real-time regulation and control precision can be improved, and the quick response capability to temperature mutation is enhanced.

Description

Technical Field

[0001] The present application relates to the field of temperature control, and in particular to a temperature adaptive control system for organ perfusion. Background Art

[0002] Temperature control of organ perfusion refers to a key technology that precisely controls the temperature of the perfusion fluid to maintain the physiological activity and function of the organ during the preservation and transportation of the organ in vitro. Its main purpose is to prolong the survival time of the organ after ex vivo, improve the success rate of organ transplantation, and provide a reliable experimental platform for organ repair, regenerative medicine research, and drug screening. After the organ is ex vivo, due to ischemia and hypoxia, cell metabolism disorder, irreversible damage will occur rapidly. Therefore, how to effectively prolong the ex vivo preservation time of the organ is a key issue that needs to be solved in the field of organ transplantation. The development of organ preservation technology has gone through a long process. From the early simple cold storage method, to the static cold storage method (SCS), to the currently widely studied mechanical perfusion technology (MP), the core goal is to delay the process of organ damage by reducing the organ metabolic rate, inhibiting cell apoptosis and necrosis. Among them, temperature control, as one of the key factors affecting the effect of organ preservation, runs through the entire development process of organ preservation technology. Low temperature can reduce the metabolic rate of cells, reduce oxygen consumption and the accumulation of metabolites, thereby delaying cell damage. However, low temperature itself can also cause certain damage to cells, such as cell membrane damage, intracellular ion balance disorder, reduced enzyme activity, etc. Therefore, how to choose the appropriate preservation temperature and accurately control temperature fluctuations is a key issue in organ preservation technology.

[0003] In the existing technology, static cold preservation (SCS) is the most commonly used organ preservation method. Its principle is to immerse the organ in a preservation solution at about 4°C to reduce the metabolic rate of the organ by low temperature. The SCS method is simple to operate and low-cost, but its preservation time is limited, usually only a few hours to tens of hours. For marginal donor organs or organs that need to be transported over long distances, the SCS method is difficult to meet clinical needs. Mechanical perfusion technology (MP) is a new organ preservation method developed in recent years. Its principle is to use a perfusion pump to continuously or pulse the perfusion fluid into the vascular system of the organ to simulate the blood circulation in the body, provide oxygen and nutrients to the organ, and remove metabolic waste. According to the perfusion temperature, MP can be divided into low-temperature mechanical perfusion (HMP), sub-hypothermic mechanical perfusion (SNMP) and normal temperature mechanical perfusion (NMP). HMP usually uses a perfusion temperature of 4-10°C, SNMP uses a perfusion temperature of 20-25°C, and NMP uses a perfusion temperature of 37°C, which is close to the physiological temperature. The MP method can significantly prolong the preservation time of organs, improve the function of organs, and provide the possibility for organ repair and regeneration. However, the existing MP technology also has some disadvantages. First, the complexity and high cost of perfusion equipment limit its wide clinical application. Second, there is no unified standard for the composition of the perfusion fluid and the perfusion parameters (such as pressure, flow, pulse frequency, etc.), which need to be optimized according to different organs and preservation conditions. Third, the existing temperature control methods mostly use water baths or semiconductor cooling, which have problems such as large temperature fluctuations, uneven temperature gradients, and difficulty in achieving precise temperature control. For example, the water bath temperature control method is greatly affected by the ambient temperature and it is difficult to achieve rapid temperature rise and fall; although the semiconductor cooling method can achieve precise temperature control, its cooling power is limited and it is difficult to meet the perfusion needs of large-volume organs. In addition, most of the existing temperature control systems lack real-time monitoring and feedback adjustment mechanisms, making it difficult to dynamically adjust according to the actual state of the organ. These shortcomings limit the further development and application of MP technology and also affect the effect of organ preservation.

[0004] Therefore, there is an urgent need for a technical solution that can improve the real-time control accuracy and enhance the ability to respond quickly to sudden temperature changes. Summary of the invention

[0005] In order to solve the deficiencies of the prior art, the embodiment of the present application provides a temperature adaptive control system for organ perfusion. The present application solves the technical problems of the prior art such as large temperature fluctuations and uneven temperature gradients.

[0006] An embodiment of the present application provides a temperature-adaptive control system for organ perfusion, comprising: a sampling unit, a computing unit and a temperature control unit; wherein the sampling unit is used to convert analog signals of temperature, flow rate and pressure of one or more target nodes in the perfused blood circulation into digital signals to form an initial data set; the computing unit is used to perform wavelet transform and sliding average processing on the initial data set to construct a feature matrix, and establish a temperature change prediction model based on the feature matrix to determine the temperature compensation amount in combination with dynamic correction; the temperature control unit is used to adjust the temperature control power using a progressive strategy based on the temperature compensation amount, and synchronously adjust the blood flow rate to dynamically optimize the control parameters according to the temperature control effect.

[0007] In a possible implementation, the initial data set is subjected to wavelet transform and sliding average processing to construct a feature matrix, and a temperature change prediction model is established based on the feature matrix to determine the temperature compensation amount in combination with dynamic correction, including: performing wavelet transform denoising on the temperature data in the initial data set, performing sliding average processing on the flow rate and pressure data, extracting characteristic parameters to construct a feature matrix; establishing a multivariate regression equation for temperature change prediction based on the feature matrix, introducing time attenuation weights and obtaining regression coefficients through iterative calculations to predict temperature change trends; calculating the temperature deviation based on the temperature change trend, and determining the temperature compensation amount using a piecewise linear compensation strategy in combination with a dynamic correction factor.

[0008] In a possible implementation, a progressive strategy is adopted to adjust the temperature control power based on the temperature compensation amount, and the blood flow rate is adjusted synchronously to dynamically optimize the control parameters according to the temperature control effect, including: adjusting the working parameters of the temperature control device based on the temperature compensation amount; adjusting the temperature control power using a progressive strategy, and the adjustment amplitude each time does not exceed a preset proportion of the current power; adjusting the blood flow rate through an airbag pressurization device, using a preset minimum step size; obtaining the temperature control effect in real time, and dynamically adjusting the control parameters according to the evaluation results; including: when the temperature control effect is lower than the preset standard, reducing the power adjustment step size.

[0009] In a possible implementation, the temperature data in the initial data set is subjected to wavelet transform denoising, the flow rate and pressure data are subjected to sliding average processing, and characteristic parameters are extracted to construct a characteristic matrix, including: performing wavelet transform denoising on the temperature data, selecting a preset wavelet and decomposition layer number; smoothing the flow rate and pressure data using a sliding average algorithm; extracting the statistical characteristics of the temperature at each measuring point; including: difference, arithmetic mean, weighted mean, standard deviation and / or coefficient of variation; calculating the temperature gradient between adjacent measuring points to obtain temperature distribution characteristics; and constructing a characteristic matrix from the processed characteristic data in chronological order.

[0010] In a possible implementation method, a multivariate regression equation for temperature change prediction is established based on the characteristic matrix, time decay weights are introduced, and regression coefficients are obtained through iterative calculation to predict temperature change trends, including: taking the temperature difference between adjacent sampling periods as the dependent variable, and selecting the target parameters in the characteristic matrix as the independent variable; determining the weight coefficient according to the stability index of the temperature sensor and the timeliness index of the data; obtaining the regression coefficient through iterative calculation, and establishing a multivariate regression equation for temperature change prediction; substituting the temperature difference, variance, change rate and temperature gradient in the characteristic matrix into the nonlinear polynomial regression equation; using a preset algorithm to calculate the Jacobian matrix and residual vector, and updating the regression coefficient; introducing time decay weights so that the farther the data is from the current moment, the smaller the weight; and obtaining the temperature change trend within a preset time through step-by-step iterative calculation.

[0011] In a possible implementation, the temperature deviation is calculated according to the temperature change trend, and the temperature compensation amount is determined by adopting a piecewise linear compensation strategy and combining with a dynamic correction factor, including: calculating the deviation value between the predicted temperature and the target temperature; determining the compensation amount according to the size of the deviation value by adopting a piecewise linear compensation strategy; introducing a dynamic correction factor to adjust the compensation amount according to the temperature change rate; when the temperature deviation is within a preset range, calculating the temperature compensation amount by a nonlinear function; when the temperature deviation exceeds a first preset range, calculating the compensation amount by adopting a piecewise linear compensation strategy; when the temperature deviation exceeds a second preset range, limiting the compensation amount by adopting a saturation control strategy; and multiplying the compensation amount by the dynamic correction factor to obtain the temperature compensation amount.

[0012] In a possible implementation, the temperature difference, variance, change rate and temperature gradient in the feature matrix are substituted into a nonlinear polynomial regression equation, including: extracting the arithmetic mean, weighted mean, standard deviation and coefficient of variation of the temperature of the current sampling period in the feature matrix as input parameters of the nonlinear polynomial regression equation; calculating the temperature difference between the current sampling period and the previous sampling period in the feature matrix, and dividing the difference by the sampling period to obtain the temperature change rate as the input parameter of the nonlinear polynomial regression equation; extracting the temperature gradient between adjacent measuring points in the feature matrix, and using the temperature gradient as the input parameter of the nonlinear polynomial regression equation; substituting the arithmetic mean, weighted mean, standard deviation, coefficient of variation, temperature change rate and temperature gradient of the extracted temperature into a pre-established nonlinear polynomial regression equation to calculate the temperature change prediction value.

[0013] In a possible implementation, a preset algorithm is used to calculate the Jacobian matrix and the residual vector, and the regression coefficient is updated, including: based on a nonlinear polynomial regression equation, calculating the difference between the predicted temperature change value and the actual temperature change value under the current regression coefficient, and forming all the differences into a residual vector; based on the nonlinear polynomial regression equation, calculating the partial derivatives of each regression coefficient in the equation, and forming all the partial derivatives into a Jacobian matrix; based on the obtained Jacobian matrix and residual vector, updating the regression coefficient according to an iterative formula including a damping factor; repeatedly calculating the residual vector, the Jacobian matrix and the updated regression coefficient until the preset convergence condition is met to obtain an updated regression coefficient.

[0014] In a temperature adaptive control system for organ perfusion provided above, the embodiment of the present application uses a temperature prediction model constructed by a characteristic matrix, uses temperature data information obtained during a sampling period, and combines a nonlinear polynomial regression method to predict the temperature change trend of each measuring point and accurately capture the rapid temperature change trend. Furthermore, in some embodiments, through a segmented compensation method, the adjustment sensitivity to small deviations can be specifically enhanced to avoid drastic temperature fluctuations, and the compensation amount can be adaptively adjusted according to the temperature change rate, which not only ensures the accuracy of temperature regulation, but also avoids temperature oscillations caused by over-compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A schematic block diagram of a temperature adaptive control system for organ perfusion provided in an embodiment of the present application; Figure 2 A schematic diagram of a flow chart of a method for determining a temperature compensation amount provided in an embodiment of the present application; Figure 3 A schematic diagram for comparing temperature control performance under multiple working conditions provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless otherwise specifically stated.

[0018] Those skilled in the art can understand that the terms "first", "second" and the like in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor represent the necessary logical order between them. It should also be understood that in the embodiments of the present disclosure, "multiple" can refer to two or more, and "at least one" can refer to one, two or more. It should also be understood that for any component, data or structure mentioned in the embodiments of the present disclosure, in the absence of explicit limitation or contrary revelation given in the context, it can generally be understood as one or more. In addition, the term "and / or" in the present disclosure is only a description of the association relationship of the associated objects, indicating that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present disclosure generally indicates that the associated objects before and after are an "or" relationship. It should also be understood that the description of each embodiment in the present disclosure emphasizes the differences between the embodiments, and the same or similar parts can refer to each other. For the sake of brevity, they will not be repeated one by one.

[0019] At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present disclosure and its application or use. The techniques, methods and devices known to ordinary technicians in the relevant fields may not be discussed in detail, but where appropriate, the techniques, methods and devices should be considered part of the specification. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] Figure 1A schematic block diagram of a temperature adaptive control system for organ perfusion provided in an embodiment of the present application. It should be understood that the system shown in the figure is exemplary and not restrictive. This means that the system architecture involved is not limited to a specific form or design, but is presented as an example. In other words, the architecture shown in the figure can be regarded as a way of expression to clearly describe related concepts and relationships, and does not exclude other forms of architecture. Therefore, when interpreting the architecture in the picture, it should be understood that the model is flexible and diverse, and its purpose is to provide an exemplary description rather than a restrictive provision on a specific form.

[0022] It should be noted that in the field of organ perfusion temperature control, the existing technology generally adopts a PID-based control scheme, which takes the deviation between the target temperature and the measured temperature as input, and calculates the adjustment amount of the cooling power through three links: proportional, integral, and differential. This method has the technical defect of insufficient temperature regulation accuracy. Specifically, due to the coupling of multiple physical processes such as blood flow, metabolism, and heat transfer during organ perfusion, the temperature change shows significant nonlinear characteristics, and there is a complex heat transfer relationship between different measuring points, which makes it difficult to adapt to the dynamic temperature change law by using fixed PID parameters. In practical applications, when the temperature is close to the target value, the regulation response is slow due to the small proportional coefficient; when the temperature deviation is large, regulation overshoot and oscillation are prone to occur. At the same time, since PID control does not take into account the spatial distribution characteristics of temperature, the temperature gradient between adjacent measuring points is difficult to be effectively controlled, affecting the temperature uniformity of organ perfusion. When using simple methods such as sliding average to predict the temperature change trend, it is impossible to accurately capture the temperature fluctuation characteristics, the prediction accuracy is limited, and it is not conducive to the precise control of temperature.

[0023] In view of the above technical problems, the embodiment of the present application adopts a nonlinear polynomial regression method to construct a temperature change prediction model, and uses multiple characteristic parameters such as temperature mean, variance, change rate and gradient as independent variables. The temperature change law is fitted by cubic terms, which can better describe the nonlinear characteristics of the system. At the same time, the time decay weight is introduced to highlight the importance of recent data and improve the adaptability of the prediction model to dynamic changes in temperature. In terms of determining the temperature compensation amount, a piecewise linear compensation strategy is adopted to dynamically adjust the compensation characteristics according to the deviation size. A nonlinear compensation function is used in the small deviation area, and the adjustment sensitivity to small deviations is enhanced by the introduction of quadratic terms; piecewise linear compensation is used in the medium deviation area to maintain the basic positive correlation between the compensation amount and the deviation; a saturation control strategy is used in the large deviation area to avoid temperature oscillation caused by excessive compensation. By introducing a dynamic correction factor that considers the influence of the temperature change rate, the compensation amount is appropriately reduced when the temperature changes violently, and the stability of temperature regulation is further improved. This adaptive control scheme based on temperature dynamic characteristics and spatial distribution solves the inherent defects of the existing PID control method in nonlinear system regulation and realizes precise temperature regulation during organ perfusion.

[0024] At the same time, simple algorithms such as conventional weighted average or linear regression cannot effectively extract high-order characteristic information from the temperature change law when processing temperature change data with significant nonlinear characteristics. The embodiment of the present application solves the nonlinear regression coefficient through the improved Levenberg-Marquardt algorithm, and adopts the residual convergence criterion to ensure the reliability of the iterative process, which can more accurately characterize the dynamic characteristics of temperature changes. In terms of compensation strategy, the traditional linear proportional compensation method has a single compensation characteristic and is difficult to adapt to the adjustment requirements under different deviation ranges. The three-segment adaptive compensation strategy designed in the embodiment of the present application, combined with a dynamic correction mechanism that considers the influence of the temperature change rate, not only ensures the adjustment sensitivity under small deviation conditions, but also can effectively suppress temperature oscillations under large deviation conditions, and has a significant improvement effect in technology.

[0025] Specifically, if Figure 1 As shown, the present application discloses a temperature adaptive control system for organ perfusion, including: a sampling unit 101, a calculation unit 102 and a temperature control unit 103; wherein the sampling unit 101 is used to convert the analog signals of the temperature, flow rate and pressure of one or more target nodes in the perfused blood circulation into digital signals to form an initial data set. It includes: establishing a blood circulation system, setting an air bag pressurizing device outside the blood transfusion bag, and applying pressure to circulate the blood through the air pressure control system; setting temperature sensors at the blood transfusion bag, the oxygenator input end, the oxygenator output end, the organ container and other positions; setting a pressure sensor and a flow sensor in the pipeline to collect blood circulation parameters; and converting the acquired temperature, flow rate and pressure analog signals into digital signals through a data acquisition module.

[0026] In one embodiment, first, temperature data and blood circulation parameters are obtained. The blood is circulated through a blood transfusion bag stored in a 4°C constant temperature box, and a pressure of 80-120 mmHg is applied by an air bag pressurizing device arranged outside the blood transfusion bag, so that the blood flows at a rate of 150-180 mL / min (close to the normal blood flow rate of the hepatic artery). In this process, temperature data is collected in real time at key positions such as the blood transfusion bag, the oxygenator input end, the oxygenator output end, and the organ container, and the sampling period is 0.5 seconds. At the same time, parameters such as blood flow rate and pressure are recorded, and the collection interval is 0.5 seconds. The acquired temperature, flow rate and pressure data are converted into digital signals through a data acquisition module to form an initial data set.

[0027] In another embodiment of the present application, a blood transfusion bag stored in a 4°C incubator is connected to a circulation pipeline, wherein an airbag pressurizing device is arranged outside the blood transfusion bag. The airbag pressurizing device applies pressure within the range of 80-120 mmHg through an adjustable air pressure control system, so that the blood is circulated and transported at a flow rate of 150-180 mL / min. In practical applications, the pressure range can be appropriately adjusted according to the organ type. For example, for more fragile tissues, the pressure can be controlled within the range of 85-115 mmHg. A high-precision temperature sensor network is arranged at the key nodes of the circulatory system. Specifically, it includes a temperature sensor at the blood transfusion bag, a temperature sensor at the input end of the oxygenator, a temperature sensor at the output end of the oxygenator, and a temperature sensor for the organ container. The sampling period of these sensors is 0.5 seconds, and the sampling accuracy reaches ±0.1°C. At the same time, a pressure sensor and a flow sensor are arranged in the pipeline to monitor blood circulation parameters in real time, and the sampling interval is also 0.5 seconds. The analog signals collected by all sensors are converted into digital signals by a 16-bit A / D converter and transmitted to the data acquisition module via the RS485 bus. The data acquisition module packages these data in a preset format to form a comprehensive data package containing parameters such as temperature, pressure, and flow.

[0028] The calculation unit 102 is used to perform wavelet transform and sliding average processing on the initial data set to construct a feature matrix, and establish a temperature change prediction model based on the feature matrix to determine the temperature compensation amount in combination with dynamic correction. Figure 2 As shown): the temperature data in the initial data set is subjected to wavelet transform denoising, the flow rate and pressure data are subjected to sliding average processing, and characteristic parameters are extracted to construct a characteristic matrix; a multivariate regression equation for temperature change prediction is established based on the characteristic matrix, and a time attenuation weight is introduced and the regression coefficient is obtained through iterative calculation to predict the temperature change trend; the temperature deviation is calculated according to the temperature change trend, and the temperature compensation amount is determined by adopting a piecewise linear compensation strategy and combining a dynamic correction factor.

[0029] Specifically, first, the initial data is preprocessed and features are extracted. The temperature data is denoised using wavelet transform, and the db4 wavelet is selected. The decomposition layer is 3 layers, and the coefficients greater than 0.2 are retained. The flow rate and pressure data are smoothed using a 60-point sliding average. The characteristic parameters such as the mean, variance, and change rate of the temperature at each measuring point are extracted, and the temperature gradient between adjacent measuring points is calculated. The processed feature data are arranged in chronological order to establish an 80×15 feature matrix.

[0030] Then, a temperature change prediction model is constructed. A regression analysis method based on weighted least squares can be used. Specifically, the temperature difference between two adjacent sampling periods (with an interval of 0.5 seconds) is used as the dependent variable, and the key parameters in the aforementioned feature matrix are selected as independent variables to establish a multiple regression equation. The weight coefficient is determined by an adaptive weighting strategy, that is, the weight is dynamically adjusted according to the reliability of the data. In practice, the reliability evaluation criteria include: the stability index of the temperature sensor (the weight is 1.0 when the standard deviation is less than 0.15℃, the weight is 0.85 when the standard deviation is in the range of 0.15-0.25℃, and the weight is 0.7 when the standard deviation is greater than 0.25℃), the timeliness index of the data (the weight of the data within the last 2 seconds is 1.0, the weight of the data within 2-5 seconds is 0.9, and the weight of the data within 5-10 seconds is 0.8). The regression coefficient is obtained by iterative calculation, and the termination condition of the iteration is that the difference between the coefficients of two adjacent calculations is less than 0.001 or the number of iterations reaches the preset value (usually 50-100 times). Based on the obtained regression equation, the system can predict the temperature change trend of each measuring point in the next 30 seconds.

[0031] Next, a hierarchical control strategy can be used for temperature deviation analysis and compensation. For the temperature inside the organ container, the system compares the real-time temperature with the target temperature of 4°C; for the oxygenator outlet temperature, it compares it with the target value of 37°C.

[0032] The temperature control unit 103 is used to adjust the temperature control power using a progressive strategy based on the temperature compensation amount, and to adjust the blood flow rate synchronously, so as to dynamically optimize the control parameters according to the temperature control effect. This includes: adjusting the working parameters of the temperature control device based on the temperature compensation amount; adjusting the temperature control power using a progressive strategy, and the adjustment amplitude each time does not exceed the preset proportion of the current power; adjusting the blood flow rate through the airbag pressurization device, using a preset minimum step size; obtaining the temperature control effect in real time, and dynamically adjusting the control parameters according to the evaluation results; including: when the temperature control effect is lower than the preset standard, reducing the power adjustment step size.

[0033] Preferably, based on the aforementioned temperature compensation amount, the operating parameters of the refrigeration system are adjusted. When cooling is required, the refrigeration power is increased, and the adjustment range is 85%-115% of the reference power; when heating is required, the refrigeration power is reduced, and the adjustment range is 70%-95% of the reference power. A progressive adjustment strategy is adopted, and the adjustment amplitude does not exceed 10% of the current power each time. At the same time, the blood flow rate is appropriately adjusted through the airbag pressurization device to achieve temperature control in a coordinated manner. The flow rate adjustment step is 5mL / min, and the adjusted flow rate is always maintained in the range of 140-190mL / min to ensure that the blood supply to the organs is sufficient but not excessive.

[0034] Specifically, precise temperature regulation is performed based on the aforementioned compensation amount. The power regulation of the refrigeration system adopts a progressive strategy, and the adjustment amplitude does not exceed 10% of the current power each time. When cooling is required, the system adjusts the cooling power to 85%-115% of the reference power; when heating is required, the power is adjusted to 70%-95% of the reference power. At the same time, the system coordinates the blood flow rate through the airbag pressurization device, and the flow rate regulation adopts a minimum step of 5mL / min to ensure the stability of the regulation. The adjusted flow rate is always maintained in the range of 140-190mL / min, which is close to the normal physiological blood flow rate of the hepatic artery, which can ensure adequate blood supply to the organ and avoid excessive perfusion. In actual operation, the system evaluates the temperature control effect every 0.5 seconds and dynamically adjusts the control parameters according to the evaluation results. When the temperature control effect is not ideal (such as the temperature deviation exceeds ±0.3℃ for 5 consecutive cycles), the power adjustment step is reduced to 7.5% of the current value, and enhanced control is entered to improve the control accuracy.

[0035] Figure 2 A flow chart of a method for determining a temperature compensation amount provided in an embodiment of the present application. Figure 2 As shown, at step S201, the temperature data in the initial data set is subjected to wavelet transform denoising, the velocity and pressure data are subjected to sliding average processing, and characteristic parameters are extracted to construct a characteristic matrix. This includes: performing wavelet transform denoising on the temperature data, selecting a preset wavelet and decomposition layer number; smoothing the velocity and pressure data using a sliding average algorithm; extracting the statistical characteristics of the temperature at each measuring point; including: difference, arithmetic mean, weighted mean, standard deviation and / or coefficient of variation; calculating the temperature gradient between adjacent measuring points to obtain temperature distribution characteristics; constructing a characteristic matrix with the processed characteristic data in chronological order.

[0036] Preferably, the data preprocessing and feature extraction process of the embodiment of the present application adopts a multi-level signal processing method. First, the temperature data is subjected to wavelet denoising, the db4 wavelet function is selected, and a 3-layer decomposition depth is set. In practical applications, the wavelet coefficient threshold is set to 0.2, and coefficients below this threshold will be filtered, thereby effectively removing the measurement noise while retaining the valid information of the temperature signal.

[0037] For flow rate and pressure data, a 60-point sliding average algorithm is used for smoothing. In specific implementation, a sliding window range of 52-68 points can be selected according to the fluctuation characteristics of the data. For data segments with more drastic changes, the sliding window can be appropriately reduced to 52 points to improve the system response speed; for relatively stable data segments, the sliding window can be increased to 68 points to obtain a better smoothing effect.

[0038] In the feature extraction phase, the system calculates the statistical characteristics of the temperature at each measuring point, including but not limited to: the arithmetic mean, weighted mean, standard deviation, coefficient of variation, etc. within each sampling period. At the same time, the temperature gradient between adjacent measuring points is calculated to obtain the temperature distribution characteristics of each part of the system. These feature data are arranged in time series to construct an 80×15 feature matrix, in which 80 rows represent time series and 15 columns contain various feature parameters.

[0039] At step S202, a multivariate regression equation for temperature change prediction is established based on the feature matrix, and a time decay weight is introduced and the regression coefficient is obtained through iterative calculation to predict the temperature change trend. This includes: taking the temperature difference between adjacent sampling periods as the dependent variable, selecting the target parameter in the feature matrix as the independent variable; determining the weight coefficient according to the stability index of the temperature sensor and the timeliness index of the data; obtaining the regression coefficient through iterative calculation, and establishing a multivariate regression equation for temperature change prediction; substituting the temperature difference, variance, change rate and temperature gradient in the feature matrix into the nonlinear polynomial regression equation; using a preset algorithm to calculate the Jacobian matrix and the residual vector, and updating the regression coefficient; introducing the time decay weight so that the farther the data is from the current moment, the smaller the weight; and obtaining the temperature change trend within a preset time through step-by-step iterative calculation.

[0040] Among them, the temperature difference, variance, change rate and temperature gradient in the feature matrix are substituted into the nonlinear polynomial regression equation, including: extracting the arithmetic mean, weighted mean, standard deviation and coefficient of variation of the temperature of the current sampling period in the feature matrix as input parameters of the nonlinear polynomial regression equation; calculating the temperature difference between the current sampling period and the previous sampling period in the feature matrix, and dividing the difference by the sampling period to obtain the temperature change rate as the input parameter of the nonlinear polynomial regression equation; extracting the temperature gradient between adjacent measuring points in the feature matrix, and using the temperature gradient as the input parameter of the nonlinear polynomial regression equation; substituting the arithmetic mean, weighted mean, standard deviation, coefficient of variation, temperature change rate and temperature gradient of the extracted temperature into the pre-established nonlinear polynomial regression equation to calculate and obtain the temperature change prediction value.

[0041] Among them, a preset algorithm is used to calculate the Jacobian matrix and the residual vector, and to update the regression coefficient, including: based on the nonlinear polynomial regression equation, calculating the difference between the predicted temperature change value and the actual temperature change value under the current regression coefficient, and forming all the differences into a residual vector; based on the nonlinear polynomial regression equation, calculating the partial derivatives of each regression coefficient in the equation, and forming all the partial derivatives into a Jacobian matrix; based on the obtained Jacobian matrix and residual vector, updating the regression coefficient according to an iterative formula including a damping factor; repeatedly calculating the residual vector, the Jacobian matrix and the updated regression coefficient until the preset convergence condition is met to obtain the updated regression coefficient.

[0042] Furthermore, in one embodiment, a temperature change prediction model can be constructed based on the established 80×15 feature matrix. First, a nonlinear polynomial regression method is used to construct a basic temperature change prediction model, and the temperature mean, variance, change rate and temperature gradient in the feature matrix are used as independent variables, and the temperature difference between adjacent sampling periods is used as the dependent variable. Specifically, for each measurement point, the temperature change prediction value ΔT is calculated by the following formula: ΔT = α1(T-Ttarget) 3 + α2(T-Ttarget) 2 + α3(dT / dt) + α4 T + α5σT 2 , where T is the current temperature, Ttarget is the target temperature, dT / dt is the temperature change rate, T is the temperature gradient, σT 2 is the temperature variance, α i (i=1, 2, 3, 4, 5) are regression coefficients. The regression coefficients are determined by the improved Levenberg-Marquardt algorithm, and the optimal solution is obtained through iterative calculation. In each round of iteration, the Jacobian matrix J and the residual vector r are first calculated, and then according to the formula Δα = -(JT J + λdiag(J T J)) -1 J T r updates the regression coefficient, where λ is the damping factor and its initial value is set to 0.01. When the difference between the residual sum of squares of two iterations is less than 10 -6 When , the iteration is considered to have converged. In order to improve the prediction accuracy, the time decay weight is introduced when determining the regression coefficient. The farther the data is from the current moment, the smaller the weight is. The weight coefficient is calculated by e (-βΔt) Calculation, where Δt is the time difference and β is the attenuation coefficient, which is taken as 0.15. Based on the obtained regression coefficient, the temperature change trend in the next 30 seconds is discretized into 60 prediction points, and the temperature change value of each prediction point is obtained by step-by-step iterative calculation, where the initial value of each iteration adopts the calculation result of the previous prediction point.

[0043] At step S203, the temperature deviation is calculated according to the temperature change trend, and the temperature compensation amount is determined by adopting a piecewise linear compensation strategy and combining a dynamic correction factor. This includes: calculating the deviation value between the predicted temperature and the target temperature; determining the compensation amount by adopting a piecewise linear compensation strategy according to the deviation value; introducing a dynamic correction factor to adjust the compensation amount according to the temperature change rate; when the temperature deviation is within a preset range, using a nonlinear function to calculate the temperature compensation amount; when the temperature deviation exceeds the first preset range, using a piecewise linear compensation strategy to calculate the compensation amount; when the temperature deviation exceeds the second preset range, using a saturation control strategy to limit the compensation amount; multiplying the compensation amount by the dynamic correction factor to obtain the temperature compensation amount.

[0044] Specifically, the temperature compensation amount can be determined by an adaptive algorithm. First, the deviation value εt = Tpredict - Ttarget between the predicted temperature and the target temperature is calculated, where Tpredict is the predicted temperature and Ttarget is the target temperature. Then, the compensation amount is determined by a piecewise linear compensation strategy based on the size of the deviation value. Specifically, when |εt| ≤ 0.3℃, the temperature compensation amount ΔTc is calculated by the nonlinear function ΔTc = k1εt + k2εt 2sign(εt), where k1 is the proportional coefficient, with a value of 1.23, k2 is the nonlinear adjustment coefficient, with a value of 0.87, and sign(εt) is the sign function of the deviation; when 0.3℃ < |εt|≤ 0.8℃, a piecewise linear compensation strategy is adopted, and the compensation amount is calculated by ΔTc = ±[0.3k1 + k2(|εt| - 0.3)], where the positive and negative signs are the same as the signs of the deviation value; when |εt| > 0.8℃, in order to prevent over-compensation, a saturation control strategy is adopted to limit the compensation amount to the maximum value range, that is, |ΔTc| ≤ 0.95℃. When determining the compensation amount, the influence of the temperature change rate is also considered, and a dynamic correction factor μ = 1 - γ|dT / dt| is introduced, where γ is the correction coefficient, with a value of 0.32, and dT / dt is the temperature change rate. The final temperature compensation amount needs to be multiplied by the dynamic correction factor, that is, ΔTc_final = μΔTc. In this way, the compensation amount can be appropriately reduced when the temperature changes drastically to avoid overshoot.

[0045] In addition, the temperature change trend can also be considered. When the temperature shows a continuous upward or downward trend, the compensation amount will be appropriately increased to suppress the temperature drift in advance. Specifically, if the temperature changes in the same direction for three consecutive sampling periods, the compensation amount will be increased by 8.5%; if the temperature changes in the same direction for five consecutive sampling periods, the compensation amount will be increased by 12%.

[0046] Figure 3 A schematic diagram for comparing temperature control performance under multiple working conditions provided in an embodiment of the present application. Figure 3 Four sub-graphs comprehensively demonstrate the control performance of this method on key parameters such as temperature, flow rate and pressure, and compare and analyze it with various conventional control methods.

[0047] Figure 3 (a) shows the temperature control effect of the organ container, and the target temperature is set to 4°C. Under the same working conditions, the adaptive control method of the embodiment of the present application, traditional PID control and fuzzy control were used for testing. Experimental data show that the adaptive control method controls the temperature fluctuation within the range of ±0.08°C, while the temperature fluctuations of PID control and fuzzy control reach ±0.15°C and ±0.2°C respectively. This is due to the fact that the embodiment of the present application uses db4 wavelet three-layer decomposition to preprocess the temperature data, and smoothes the flow rate and pressure data through 60-point sliding average, which effectively improves the control accuracy.

[0048] Figure 3(b) reflects the temperature control effect at the outlet of the oxygenator, with a target temperature of 37°C. The adaptive control method of the embodiment of the present application is compared with model predictive control (MPC) and traditional cascade control. The results show that the adaptive control method stabilizes the temperature within 900 seconds in the range of 37±0.3°C, while MPC control and cascade control require 1200 seconds and 1500 seconds, respectively, to achieve similar stabilization effects. This is mainly due to the temperature compensation strategy adopted in the embodiment of the present application: when the temperature deviation |εt|≤0.3°C, the compensation amount is calculated by the nonlinear function ΔTc = 1.23εt + 0.87εt²sign(εt); when 0.3°C<|εt|≤0.8°C, piecewise linear compensation is used; when |εt|>0.8°C, a saturation control strategy is used to limit the compensation amount to ±0.95°C.

[0049] Figure 3 (c) Demonstrates the control effect of blood flow rate. The adaptive control method of the embodiment of the present application stabilizes the flow rate within the range of 165 ± 8 mL / min, while the flow rate fluctuations of switch control and PI control reach ± ​​15 mL / min and ± 12 mL / min, respectively. This is due to the minimum adjustment step of 5 mL / min adopted in the embodiment of the present application, and the precise adjustment of the flow rate is achieved by the airbag pressurization device. During the experiment, the flow rate was always maintained within the range of 140-190 mL / min, which not only ensured adequate blood supply to the organs, but also avoided overperfusion.

[0050] Figure 3 (d) reflects the system pressure control performance. Under the working condition of target pressure 100mmHg, the adaptive control method of the embodiment of the present application is compared with sliding mode control and traditional PID control. The results show that adaptive control controls the pressure fluctuation within the range of ±5mmHg, while the pressure fluctuations of sliding mode control and PID control reach ±10mmHg and ±8mmHg respectively. This reflects the advantage of the embodiment of the present application adopting a progressive strategy in the temperature regulation control link: when cooling is required, the refrigeration power is adjusted to 85%-115% of the reference power; when heating is required, the power is adjusted to 70%-95% of the reference power. The adjustment amplitude each time does not exceed 10% of the current power, which ensures the stability of the control process.

[0051] By analyzing the experimental data, it can be seen that the temperature adaptive control method of the embodiment of the present application is superior to the conventional control method in all control indicators. This is mainly based on the following technical principles: First, the data preprocessing is carried out by combining wavelet transform and sliding average to improve the signal quality; second, by establishing an 80×15 feature matrix, the accurate prediction of the temperature change trend is achieved; third, the dynamic correction factor μ = 1 - 0.32|dT / dt| is introduced to enable the compensation amount to be adaptively adjusted according to the temperature change rate; finally, a hierarchical control strategy is adopted, and different compensation algorithms are used within different deviation ranges to achieve precise temperature control.

[0052] When the system detects that the temperature deviation exceeds ±0.3°C for 5 consecutive cycles, it will automatically enter the enhanced control mode and reduce the power adjustment step to 7.5% of the current value, further improving the control accuracy. This multi-level adaptive control strategy ensures that the system can maintain the optimal control effect under various working conditions. The experimental results show that this method can not only achieve precise control of the dual temperature zones of 4°C for the organ container and 37°C for the oxygenator outlet, but also ensure the stability of blood flow rate and pressure, providing reliable temperature control technical support for the organ perfusion process.

[0053] In summary, in the process of organ perfusion temperature control, the temperature regulation accuracy directly affects the preservation quality of the organ. The temperature distribution of key locations such as the blood transfusion bag, oxygenator input end, oxygenator output end and organ container in the organ perfusion system shows complex spatial heterogeneity. The blood in the blood transfusion bag needs to be stored at 4°C and heated to about 37°C by the oxygenator to be supplied to the organ. During this process, there are significant differences in the dynamic change characteristics of the temperature at each measuring point. The simple temperature control method used in the prior art has the problem of large temperature fluctuations. The spatial unevenness and temporal fluctuation of this temperature make it difficult to simultaneously meet the temperature control requirements of each measuring point using a fixed parameter control scheme, especially in the organ container.

[0054] In response to the technical problems in the above-mentioned organ perfusion process, the temperature prediction model constructed based on the 80×15 feature matrix in the embodiment of the present application makes full use of the temperature data information obtained in the 0.5 second sampling period. Through the nonlinear polynomial regression method, characteristic parameters such as temperature mean, variance, rate of change and gradient are integrated into the prediction model to achieve accurate prediction of the temperature change trend of each measuring point. Compared with the prior art, this model shows better prediction performance when dealing with rapid temperature changes in the oxygenator: by fitting the temperature change law through the cubic term, it can better describe the rapid temperature rise process of blood in the oxygenator; the introduced time attenuation weight e (-0.15Δt) The improved Levenberg-Marquardt algorithm uses a residual square sum less than 10 when solving the regression coefficient.-6 The convergence criterion ensures the reliability of temperature prediction. This prediction model is particularly suitable for the differences in the dynamic change characteristics of temperature at each measuring point in the organ perfusion system: for blood transfusion bags and organ containers with relatively slow temperature changes, the model can provide stable prediction results; for oxygenator areas with drastic temperature changes, the model can also accurately capture the rapid temperature change trend.

[0055] In terms of temperature compensation strategy, the segmented compensation method designed in this extended scheme specifically solves the difficulty of temperature control during organ perfusion. When the temperature in the organ container is close to the target range of 4±0.2℃, a nonlinear compensation function with k1=1.23 and k2=0.87 is used. The introduction of quadratic terms enhances the adjustment sensitivity to small deviations, which helps to control the temperature stably within the allowable range. When the oxygenator outlet temperature deviates from the target value of 37±0.5℃ by a large margin, a segmented linear compensation and saturation control strategy is adopted to limit the compensation amount to within 0.95℃, avoiding drastic temperature fluctuations. At the same time, a dynamic correction factor of γ=0.32 is introduced to adaptively adjust the compensation amount according to the temperature change rate. This mechanism shows good anti-disturbance performance during the rapid temperature rise stage of blood flowing through the oxygenator. This compensation strategy ensures the accuracy of temperature regulation and avoids temperature oscillation caused by over-compensation by reasonably setting segmentation points and saturation limits.

[0056] It can be known from the description of the above implementation mode that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., including several instructions for enabling a computer device (which can be a personal computer, a server, or a network communication device such as a media gateway, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0057] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0058] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0059] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. 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 application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A temperature adaptive control system for organ perfusion, characterized in that: include: Sampling unit, calculation unit and temperature control unit; wherein, The sampling unit is used to convert analog signals of temperature, flow rate and pressure of one or more target nodes in the perfused blood circulation into digital signals to form an initial data set; The calculation unit is used to perform wavelet transform and sliding average processing on the initial data set to construct a characteristic matrix, and establish a temperature change prediction model based on the characteristic matrix to determine the temperature compensation amount in combination with dynamic correction; The temperature control unit is used to adjust the temperature control power using a progressive strategy based on the temperature compensation amount, and synchronously adjust the blood flow rate to dynamically optimize the control parameters according to the temperature control effect.

2. The control system according to claim 1, characterized in that: in, The initial data set is subjected to wavelet transformation and sliding average processing to construct a characteristic matrix, and a temperature change prediction model is established based on the characteristic matrix to determine the temperature compensation amount in combination with dynamic correction, including: Performing wavelet transform denoising on the temperature data in the initial data set, performing sliding average processing on the flow velocity and pressure data, extracting characteristic parameters and constructing a characteristic matrix; A multivariate regression equation for temperature change prediction is established based on the characteristic matrix, a time decay weight is introduced, and a regression coefficient is obtained through iterative calculation to predict the temperature change trend; The temperature deviation is calculated according to the temperature change trend, and the temperature compensation amount is determined by adopting a piecewise linear compensation strategy in combination with a dynamic correction factor.

3. The control system according to claim 1, characterized in that: in, Based on the temperature compensation amount, a progressive strategy is adopted to adjust the temperature control power, and the blood flow rate is adjusted synchronously to dynamically optimize the control parameters according to the temperature control effect, including: Adjusting the working parameters of the temperature control device based on the temperature compensation amount; Use a progressive strategy to adjust the temperature control power, and each adjustment range does not exceed the preset proportion of the current power; Blood flow rate is regulated by a balloon pressurization device, using a preset minimum step size; Obtain the temperature control effect in real time and dynamically adjust the control parameters according to the evaluation results; this includes: when the temperature control effect is lower than the preset standard, reduce the power adjustment step.

4. The control system according to claim 2, characterized in that: in, The temperature data in the initial data set is subjected to wavelet transform denoising, the velocity and pressure data are subjected to sliding average processing, and characteristic parameters are extracted to construct a characteristic matrix, including: Perform wavelet transform denoising on the temperature data, and select the preset wavelet and decomposition level; The flow rate and pressure data are smoothed using a sliding average algorithm; Extract statistical characteristics of the temperature at each measuring point; including: difference, arithmetic mean, weighted mean, standard deviation and / or coefficient of variation; Calculate the temperature gradient between adjacent measuring points to obtain temperature distribution characteristics; The processed feature data is used to construct a feature matrix in chronological order.

5. The control system according to claim 2, characterized in that: in, A multivariate regression equation for temperature change prediction is established based on the characteristic matrix, a time decay weight is introduced, and the regression coefficient is obtained through iterative calculation to predict the temperature change trend, including: The temperature difference between adjacent sampling periods is used as the dependent variable, and the target parameters in the feature matrix are selected as the independent variables; Determine the weight coefficient according to the stability index of the temperature sensor and the timeliness index of the data; The regression coefficients are obtained through iterative calculation, and a multivariate regression equation for temperature change prediction is established; Substitute the temperature difference, variance, change rate and temperature gradient in the characteristic matrix into the nonlinear polynomial regression equation; Use the preset algorithm to calculate the Jacobian matrix and residual vector and update the regression coefficients; Introduce time decay weights so that the weight of data farther from the current moment is smaller; The temperature change trend within the preset time is obtained through step-by-step iterative calculation.

6. The control system according to claim 2, characterized in that: in, The temperature deviation is calculated according to the temperature change trend, and the temperature compensation amount is determined by adopting a piecewise linear compensation strategy and combining a dynamic correction factor, including: Calculate the deviation between the predicted temperature and the target temperature; According to the deviation value, the compensation amount is determined by adopting piecewise linear compensation strategy; Introducing a dynamic correction factor to adjust the compensation amount according to the temperature change rate; When the temperature deviation is within the preset range, a nonlinear function is used to calculate the temperature compensation amount; When the temperature deviation exceeds a first preset range, a piecewise linear compensation strategy is used to calculate the compensation amount; When the temperature deviation exceeds a second preset range, a saturation control strategy is used to limit the compensation amount; The temperature compensation is obtained by multiplying the compensation amount by the dynamic correction factor.

7. The control system according to claim 5, characterized in that: in, Substitute the temperature difference, variance, rate of change and temperature gradient in the feature matrix into the nonlinear polynomial regression equation, including: Extracting the arithmetic mean, weighted mean, standard deviation and coefficient of variation of the temperature in the current sampling period in the feature matrix as input parameters of the nonlinear polynomial regression equation; Calculate the temperature difference between the current sampling period and the previous sampling period in the characteristic matrix, and divide the difference by the sampling period to obtain the temperature change rate as an input parameter of the nonlinear polynomial regression equation; Extract the temperature gradient between adjacent measuring points in the characteristic matrix, and use the temperature gradient as the input parameter of the nonlinear polynomial regression equation; The arithmetic mean, weighted mean, standard deviation, coefficient of variation, temperature change rate and temperature gradient of the extracted temperatures were substituted into the pre-established nonlinear polynomial regression equation to calculate the predicted value of temperature change.

8. The control system according to claim 5, characterized in that: in, The preset algorithm is used to calculate the Jacobian matrix and residual vector and update the regression coefficients, including: Based on the nonlinear polynomial regression equation, the difference between the predicted temperature change value under the current regression coefficient and the actual temperature change value is calculated, and all the differences are combined into a residual vector; Based on the nonlinear polynomial regression equation, calculate the partial derivatives of each regression coefficient in the equation, and form the Jacobian matrix with all partial derivatives; Based on the obtained Jacobian matrix and residual vector, the regression coefficients are updated according to an iterative formula including a damping factor; The residual vector, Jacobian matrix and updated regression coefficients are calculated repeatedly until the preset convergence conditions are met to obtain updated regression coefficients.

Images