Negative pressure drainage device control method and system for wound nursing

Abstract

The invention provides a negative pressure drainage device control method and system for wound nursing, and relates to the technical field of negative pressure drainage control. Dynamic wound data are transmitted back to a negative pressure host of a negative pressure drainage device through an IPX interface; according to static patient characteristics and dynamic wound data, negative pressure drainage double decisions are carried out, initial negative pressure dynamic parameters are output, and a negative pressure host intermittently adopts the initial negative pressure dynamic parameters to drive a negative pressure drainage device to fill a wound with dressing to execute closed-loop negative pressure suction with a real-time drainage period as a constraint. And performing dynamic adjustment of the initial negative pressure kinetic parameters according to the time sequence percolate characteristics. The technical problems that in the prior art, parameter setting of a negative pressure drainage device depends on operation experience of medical staff, and the real-time response parameter adjusting capacity for characteristic changes of percolate in the negative pressure adsorption process is lacked are solved. And the effects of performing negative pressure dynamic regulation and control based on the real-time evolution of the wound microenvironment, improving the adaptability of negative pressure drainage and the patient state and reducing the drainage blockage rate are achieved.

Description

Technical Field

[0001] This invention relates to the field of negative pressure drainage control technology, and in particular to a control method and system for a negative pressure drainage device used in wound care. Background Technology

[0002] The key deficiency of existing negative pressure drainage technology lies in the lack of mechanical and clinical adaptability of its control logic.

[0003] Traditional equipment relies on medical staff to preset static negative pressure parameters based on experience. These parameters remain fixed throughout the treatment process and cannot respond to the dynamic biological needs of wound healing.

[0004] A more prominent problem is that existing technologies lack a real-time sensing and feedback mechanism for the physical properties of drainage fluid. When high-viscosity exudate causes a sharp increase in flow resistance or a decrease in drainage efficiency, the negative pressure drainage device still mechanically executes the original parameters, failing to provide early warning of blockage risks or trigger adaptive adjustments. This rigid control mode, detached from the actual state of the wound, severely limits the safety of negative pressure drainage technology.

[0005] In summary, the parameter settings of existing negative pressure drainage devices rely on the operational experience of medical staff and lack the technical capability to adjust parameters in real time to respond to changes in the characteristics of exudate during the negative pressure adsorption process. Summary of the Invention

[0006] In view of the above-mentioned defects or improvement needs of the prior art, the present invention provides a control method and system for a negative pressure drainage device for wound care, which addresses the technical problem that the parameter setting of the existing negative pressure drainage device depends on the operating experience of medical staff and lacks the ability to adjust parameters in real time to respond to changes in the characteristics of exudate during the negative pressure adsorption process.

[0007] To achieve the above objectives, the present invention provides a negative pressure drainage device control method and system for wound care.

[0008] A first aspect of the present invention provides a method for controlling a negative pressure drainage device for wound care, the method comprising:

[0009] After the wound-filling dressing has been sealed over the target patient's wound for a duration sufficient for steady-state operation, the micro-thermocouples, impedance electrodes, and flexible piezoresistive films integrated into the wound-filling dressing are driven to collect dynamic wound data. This dynamic wound data is then transmitted back to the negative pressure unit of the negative pressure drainage device via a medical-grade IPX interface. The dynamic wound data includes the metabolic heat distribution of the wound tissue, the wound stress vector field, and the exudate transport flux. Static patient characteristics are retrieved from the HIS system using the target patient's unique identification code. A dual decision-making process for negative pressure drainage is performed based on both the static patient characteristics and the dynamic wound data, outputting initial negative pressure kinetic parameters. These initial negative pressure kinetic parameters include a real-time drainage cycle. During the process where the negative pressure unit, constrained by the real-time drainage cycle, intermittently uses the initial negative pressure kinetic parameters to drive the negative pressure drainage device to perform closed-loop negative pressure suction on the wound-filling dressing, the initial negative pressure kinetic parameters are dynamically adjusted according to the time-series exudate characteristics.

[0010] In one implementation, the following processing is also performed:

[0011] The initial negative pressure kinetic parameters are used to drive the negative pressure drainage device. During the closed-loop negative pressure suction process of the wound filling dressing, the exudate characteristics are collected at the output end of the negative pressure drainage tube through a microfluidic sensing unit to obtain real-time exudate characteristics, including real-time viscosity and real-time flow rate. Based on the real-time exudate characteristics and the initial negative pressure kinetic parameters, the probability of congestion transition is calculated, and a real-time congestion risk coefficient is output. If the real-time congestion risk coefficient meets a preset congestion threshold, incremental PID control is performed on the initial negative pressure kinetic parameters based on the real-time exudate characteristics, and updated negative pressure kinetic parameters are output.

[0012] In one implementation, the congestion state transition probability is calculated based on the real-time exudate characteristics and initial negative pressure kinetic parameters, and a real-time congestion risk coefficient is output. The following processing is also performed:

[0013] Based on the real-time drainage cycle and a set time matching mechanism, historical equipment congestion logs are retrieved to obtain multiple sample exudate characteristics, multiple sample negative pressure kinetic parameters, and multiple sample congestion risk coefficients. The multiple sample congestion risk coefficients are then classified into states according to a preset congestion state classification, resulting in W state transition paths. Based on the association of the multiple sample congestion risk coefficients with the W state transition paths, cross-state transition frequency statistics are performed on the multiple sample exudate characteristics and multiple sample negative pressure kinetic parameters to construct a congestion risk transition matrix. The real-time viscosity and real-time flow velocity are used as observational evidence and loaded into the congestion risk transition matrix to update the output state transition probability. The state transition probability is corrected using the pressure target value in the initial negative pressure kinetic parameters, and the real-time congestion risk coefficient is output.

[0014] In one implementation, if the real-time congestion risk coefficient meets a preset congestion threshold, then incremental PID control is performed on the initial negative pressure dynamic parameters based on the real-time exudate characteristics, and updated negative pressure dynamic parameters are output. The following processing is also performed:

[0015] The initial negative pressure kinetic parameters are used to adapt to the target viscosity and target flow rate; the viscosity deviation value is calculated based on the real-time viscosity and target viscosity, and the flow rate deviation value is calculated based on the real-time flow rate and target flow rate; a pre-constructed PID parallel analysis model is built, wherein the PID parallel analysis model includes parallel viscosity error correction channels and flow rate error correction channels; the viscosity deviation value and flow rate deviation value are mapped and input into the viscosity error correction channels and flow rate error correction channels of the PID parallel analysis model to perform proportional-derivative operations, and output a double-correction positive term; the double-correction positive term is used to perform incremental PID control on the pressure target value in the initial negative pressure kinetic parameters, and output the updated negative pressure kinetic parameters.

[0016] In one implementation, the following processing is also performed:

[0017] By scanning the wound of the target patient, the geometric and edge features of the wound are quantified and output. Based on the measured exudate characteristics input by medical staff, the geometric and edge features of the wound, a dressing requirement feature vector is constructed. Using the dressing requirement feature vector, material mechanics matching is performed in the dressing configuration database to obtain the wound filling dressing. The micro thermocouple is integrated into the base contact layer of the wound filling dressing, the impedance electrodes are distributed in an interdigitated array on the pore inner wall of the wound filling dressing, and the flexible piezoresistive film is integrated into the closed membrane junction area of ​​the wound filling dressing.

[0018] In one implementation, a dual decision-making process for negative pressure drainage is performed based on static patient characteristics and dynamic wound data, outputting initial negative pressure kinetic parameters, and the following processing is also performed:

[0019] The static patient characteristics are decomposed to obtain patient age, surgical type code, and serum protein level; wound stage matching is performed based on patient age, surgical type code, and measured exudate characteristics, and stage labels are output; the negative pressure drainage decision space is invoked based on the stage labels; control decision particles are constructed based on the wound tissue metabolic heat distribution, wound stress vector field, exudate transport flux, patient age, surgical type code, and serum protein level; the control decision particles are loaded into the negative pressure drainage decision space to perform multidimensional control parameter weighted quantization, and the initial negative pressure dynamic parameters are output.

[0020] In one implementation, the control decision particle points are loaded into the negative pressure drainage decision space to perform multidimensional control parameter weighted quantization, and the initial negative pressure dynamic parameters are output. The following processing is also performed:

[0021] After projecting the control decision particle point onto the negative pressure drainage decision space, using the control decision particle point as the starting point and the surgical type code as the particle point screening constraint, P related particle points are selected based on a preset association radius. P decision confidence weights are set according to the P spatial distances between the P related particle points and the control decision particle point. The P related negative pressure dynamic parameters and P related drainage cycles stored in the P related particle points are extracted. The initial negative pressure dynamic parameters and real-time drainage cycle are output by weighting the P related negative pressure dynamic parameters and P related drainage cycles using the P decision confidence weights.

[0022] In one implementation, if the real-time congestion risk coefficient does not meet the preset congestion threshold, the real-time diversion cycle is reset after triggering a reverse pulse to clear the congestion.

[0023] A second aspect of the present invention provides a negative pressure drainage device control system for wound care, the system comprising:

[0024] The wound data acquisition unit is used to drive the miniature thermocouple, impedance electrode, and flexible piezoresistive film integrated in the wound filling dressing to acquire dynamic wound data after the wound filling dressing has been sealed and covered with the target patient's wound for a steady-state duration. The wound data transmission unit is used to transmit the dynamic wound data back to the negative pressure unit of the negative pressure drainage device via a medical-grade IPX interface. The dynamic wound data includes the metabolic heat distribution of wound tissue, the wound stress vector field, and the exudate transport flux. The static data retrieval unit is used to retrieve data from the target patient's unique identification code using H... The IS system retrieves static patient characteristics; the negative pressure drainage decision unit is used to make dual decisions on negative pressure drainage based on static patient characteristics and dynamic wound data, and outputs initial negative pressure dynamic parameters, wherein the initial negative pressure dynamic parameters include a real-time drainage cycle; the negative pressure drainage parameter adjustment and update unit is used to dynamically adjust the initial negative pressure dynamic parameters based on the time-series exudate characteristics during the process in which the negative pressure host intermittently uses the initial negative pressure dynamic parameters to drive the negative pressure drainage device to perform closed-loop negative pressure suction on the wound filling dressing, constrained by the real-time drainage cycle.

[0025] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0026] The method provided in this invention involves, after sealing the wound with a wound packing dressing for a target patient's wound for a steady-state duration, driving a miniature thermocouple, impedance electrode, and flexible piezoresistive film integrated within the wound packing dressing to collect dynamic wound data; transmitting the dynamic wound data back to the negative pressure unit of a negative pressure drainage device via a medical-grade IPX interface, wherein the dynamic wound data includes the metabolic heat distribution of wound tissue, the wound stress vector field, and the exudate transport flux; retrieving static patient characteristics from the HIS system using the target patient's unique identification code; performing a dual decision-making process for negative pressure drainage based on the static patient characteristics and dynamic wound data, and outputting initial negative pressure kinetic parameters, wherein the initial negative pressure kinetic parameters include a real-time drainage cycle; and dynamically adjusting the initial negative pressure kinetic parameters based on the time-series exudate characteristics during the process in which the negative pressure unit intermittently uses the initial negative pressure kinetic parameters to drive the negative pressure drainage device to perform closed-loop negative pressure suction on the wound packing dressing, constrained by the real-time drainage cycle. It achieves the technical effect of dynamically adjusting negative pressure based on the real-time evolution of the wound microenvironment, improving the adaptability of negative pressure drainage to the patient's condition, and reducing the drainage blockage rate. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 A schematic diagram of the control method for a negative pressure drainage device for wound care provided by the present invention is shown.

[0029] Figure 2 A schematic diagram of the overall structure of a negative pressure drainage device for wound care provided by the present invention is shown.

[0030] Figure 3 The diagram illustrates a flow chart of the matching output of initial negative pressure dynamic parameters in a negative pressure drainage device control method for wound care provided by the present invention.

[0031] Figure 4 The diagram illustrates the process of calculating and outputting a real-time congestion risk coefficient in a negative pressure drainage device control method for wound care provided by the present invention.

[0032] Figure 5 A schematic diagram of the structure of a negative pressure drainage device control system for wound care provided by the present invention is shown.

[0033] Figure labeling: 1. Negative pressure unit, 2. Empty chamber, 3. Fluid collection bag, 4. Drainage tube, 5. Catheter, 6. Negative pressure pump, 7. Wound filling dressing, 11. Wound data acquisition unit, 12. Wound data transmission unit, 13. Static data retrieval unit, 14. Negative pressure drainage decision unit, 15. Negative pressure drainage parameter adjustment and update unit. Detailed Implementation

[0034] This invention provides a control method and system for a negative pressure drainage device for wound care, which addresses the technical problem that the parameter settings of existing negative pressure drainage devices rely on the operational experience of medical staff and lack the ability to adjust parameters in real time to respond to changes in the characteristics of exudate during the negative pressure adsorption process.

[0035] 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 merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0036] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0037] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0038] Example 1: A flowchart of a negative pressure drainage device control method for wound care provided by this embodiment of the invention, see below. Figure 1 The method includes:

[0039] Step A100: After the wound filling dressing has been sealed over the target patient's wound for a period of time that meets the steady-state duration, drive the micro thermocouple, impedance electrode and flexible piezoresistive film integrated in the wound filling dressing to collect dynamic wound data.

[0040] In one implementation, the method provided by the present invention further includes:

[0041] Step A110: By scanning the target patient's wound, quantify and output the wound's geometric features and edge features.

[0042] Step A120: Construct a dressing requirement feature vector based on the measured exudate characteristics, wound geometry, and wound edge features input by medical staff.

[0043] Step A130: Using the dressing requirement feature vector, perform material mechanics matching in the dressing configuration database to obtain the wound filling dressing.

[0044] The micro thermocouple is integrated into the base contact layer of the wound filling dressing, the impedance electrodes are distributed in an interdigitated array on the inner wall of the pores of the wound filling dressing, and the flexible piezoresistive film is integrated into the sealing membrane junction area of ​​the wound filling dressing.

[0045] It should be understood that the selection of wound filling dressings is limited by the wound type and geometric characteristics. Therefore, in this embodiment, the physical morphology of the wound is first obtained by laser scanning or three-dimensional imaging technology to obtain the output wound geometric features and wound edge features. The wound geometric features include parameters such as wound surface area, maximum depth and volume, while the wound edge features involve indicators such as wound edge sharpness and epithelial migration distance, providing a spatial structural basis for subsequent dressing selection.

[0046] Meanwhile, this embodiment involves human-machine collaboration. Specifically, it combines key parameters of exudate input by medical staff based on clinical testing (such as viscosity grade, pH value, and 24-hour exudate volume) with the aforementioned scanning features to form a multi-dimensional vector. This vector integrates biochemical characteristics and physical morphology to guide the selection of dressing functions. For example, high-absorbency foam materials are matched for high-exudate wounds, and cuttable dressing bases are selected for irregular wounds.

[0047] The dressing configuration database stores multiple sample dressing selections and multiple sample demand vector features, and the index composition of the sample demand vector features and the dressing demand vector features is consistent.

[0048] After calculating the similarity between the dressing requirement feature vector and multiple sample requirement feature vectors based on Pearson correlation coefficient or Euclidean distance, the wound filling dressing is selected from the multiple sample dressings according to the decreasing sorting result of the multiple sample similarity. The wound filling dressing is adapted to the mechanical requirements of the current wound. While ensuring effective drainage of exudate from the target patient's wound, its internal sensor integration can also effectively monitor and collect physiological data of the patient's wound.

[0049] At the same time, it should be understood that, such as Figure 2 As shown, the specific structure of the negative pressure drainage device including the wound filling dressing is as follows:

[0050] The suction hole of the wound filling dressing 7 is preferably located in the center. The suction hole of the wound filling dressing 7 is connected to the drainage tube 4. An annular adhesive strip is provided on the edge of the wound filling dressing 7 that is in contact with the patient's body. The adhesive strip is used to ensure that the dressing is sealed and covered on the patient's wound.

[0051] The wound filling dressing 7 has a sensor component embedded on the side that adheres to the patient's wound to sense the condition of the wound. The sensor component integrated in the wound filling dressing 7 is arranged as follows: a miniature thermocouple temperature sensor is embedded in the dressing substrate with a contact layer 0.5-1mm away from the wound surface to directly sense the metabolic heat of granulation tissue; an impedance electrode humidity sensor is arranged in an interdigital array on the inner wall of the dressing pores with an electrode spacing ≤200μm, and the exudate flux is inferred by measuring the change in dielectric constant within the pores; a flexible piezoresistive film pressure sensor is placed at the physical interface between the dressing and the external sealing membrane, completely covering the negative pressure application area, and is used to monitor the uniformity of negative pressure distribution and tissue deformation stress.

[0052] This includes a sensor assembly consisting of a miniature thermocouple, an impedance electrode, and a flexible piezoresistive film, which is electrically connected to the negative pressure host 1. In this embodiment and thereafter, the parameter adjustment and control of negative pressure drainage are all implemented by the negative pressure host 1.

[0053] The negative pressure unit 1 has an internal chamber 2, in which a fluid collection bag 3 and a negative pressure pump 6 are detachably assembled. The fluid collection bag 3 and the negative pressure pump 6 are connected by a conduit 5. The negative pressure pump 6 is connected to the wound filling dressing 7 by the drainage tube 4 mentioned above. The contents of the chamber 2 can be protected by covering it with a front cover.

[0054] Meanwhile, the output end of the drainage tube 4 is equipped with a microfluidic sensing unit to continuously capture the time-series exudate characteristics of the output end of the drainage tube 4.

[0055] Step A200: The dynamic wound data is transmitted back to the negative pressure host of the negative pressure drainage device via the medical-grade IPX interface. The dynamic wound data includes the metabolic heat distribution of wound tissue, the stress vector field of the wound, and the exudate transport flux.

[0056] Specifically, in this embodiment, the miniature thermocouple, impedance electrode, and flexible piezoresistive film are all connected to the negative pressure host through a medical-grade IPX interface, and the subsequent parameter adjustment and analysis process of the negative pressure drainage device is completed within the negative pressure host.

[0057] In the current scenario, the medical-grade IPX interface serves as a data bridge, and its waterproof and dustproof characteristics ensure reliable transmission of raw data collected by dressing sensors in complex clinical environments.

[0058] The specific dynamic wound data transmitted back includes wound tissue metabolic heat distribution, wound stress vector field, and exudate transport flux. Specifically, the wound tissue metabolic heat distribution maps cellular metabolic activity at a depth of 0.5 mm from the wound base (temperature measurement accuracy ±0.2℃); the wound stress vector field analyzes the vacuum pressure component and tissue deformation stress component during negative pressure suction (three-dimensional vector accuracy ±1.5%FS); and the exudate transport flux quantifies the mass of exudate passing through a unit dressing area per unit time and its diffusion rate (flux resolution 0.01 mg·mm²). -2 ·h-1), together the three constitute a dynamic digital twin of the wound healing microenvironment.

[0059] Step A300: Retrieve static patient characteristics from the HIS system using the target patient's unique identification code.

[0060] Specifically, static clinical features are extracted from the secure data pool of the Hospital Information System (HIS) using unique identifiers such as the patient's medical insurance card number or electronic medical record ID. These static patient features encompass baseline parameters strongly correlated with wound healing, including patient age (affecting tissue regeneration rate), standardized coding of surgical type (e.g., ICD-10-PCS coding defining surgical trauma level), serum albumin level (indicating nutritional status), and markers of chronic disease history (e.g., diabetes markers). This process strictly adheres to the HIPAA medical privacy protocol, ensuring anonymized data transmission to the negative pressure host decision-making center.

[0061] Step A400: Make a dual decision on negative pressure drainage based on static patient characteristics and dynamic wound data, and output initial negative pressure dynamic parameters, wherein the initial negative pressure dynamic parameters include the real-time drainage cycle.

[0062] In one implementation method, Figure 3 As shown, the method A400 of this invention includes the following steps: Based on both static patient characteristics and dynamic wound data, a dual decision-making process for negative pressure drainage is performed, outputting initial negative pressure kinetic parameters.

[0063] Step A410: Decompose the static patient characteristics to obtain the patient's age, surgery type code, and serum protein level.

[0064] Step A420: Match the wound stage based on the patient's age, surgical type code, and measured exudate characteristics, and output the stage label.

[0065] Step A430: Match and call the negative pressure drainage decision space according to the stage label.

[0066] Step A440: Construct control decision particles based on the wound tissue metabolic heat distribution, wound stress vector field, exudate transport flux, patient age, surgical type coding, and serum protein level.

[0067] Step A450: Load the control decision particle points into the negative pressure drainage decision space, perform multi-dimensional control parameter weighted quantization, and output the initial negative pressure dynamic parameters.

[0068] In one implementation, the control decision particle points are loaded into the negative pressure drainage decision space to perform multidimensional control parameter weighted quantization, and the initial negative pressure dynamic parameters are output. The method step A450 provided by this invention includes:

[0069] Step A451: After projecting the control decision particle point onto the negative pressure drainage decision space, using the control decision particle point as the starting point and the surgical type code as the particle point screening constraint, P related particle points are selected based on the preset association radius.

[0070] Step A452: Based on the P spatial distances between the P associated particle points and the P control decision particle points, set P decision confidence weights.

[0071] Step A453: Extract the P associated negative pressure dynamic parameters and P associated drainage cycles stored in the P associated particle points.

[0072] Step A454: Use the P decision confidence weights to weight the P associated negative pressure kinetic parameters and P associated drainage cycles to output the initial negative pressure kinetic parameters and real-time drainage cycle.

[0073] Specifically, in this embodiment, based on step A300, the specific data composition of the static patient characteristics is known. Therefore, this embodiment directly performs data decomposition of the static patient characteristics to obtain the patient's age, surgery type code, and serum protein level.

[0074] The wound stage matching engine employs a multi-level decision tree: when the patient is ≥65 years old and the exudate pH is >7.4, or the surgical code is infection-related (e.g., code with purulent discharge) and the 24-hour exudate volume is >50ml, the "inflammatory stage" label is output; if the serum protein is in the critical range of 30-35g / L and the exudate viscosity is 30-50cP, and the wound metabolic heat distribution temperature difference is ≤1.5℃ / cm², the system is optimized. 2 If the viscosity of the exudate is less than 10 cP and the stress vector field fluctuation amplitude is less than 5%, it is marked as "proliferative phase"; when the viscosity of the exudate is less than 10 cP and the stress vector field fluctuation amplitude is less than 5%, it is determined to enter the "mature phase". In special scenarios such as diabetic patients with high viscosity exudate (>80 cP), even if the proliferative phase conditions are met, the "delayed proliferative phase" label will still be output, pointing to the unique negative pressure control logic.

[0075] Based on the obtained stage labels, the negative pressure drainage decision space is matched and invoked according to the stage labels. It should be understood that each stage has a pre-built corresponding negative pressure drainage decision space.

[0076] The construction method of the negative pressure drainage decision space in each postoperative stage is consistent. Based on this, this embodiment takes the construction of the negative pressure drainage decision space in the current stage as an example to elaborate on the technical solution in detail.

[0077] The five indicators—the metabolic heat distribution of the wound tissue, the stress vector field of the wound, the exudate transport flux, the patient's age, and the serum protein level—are used as coordinates in a multidimensional space to construct an initial space (a five-dimensional coordinate system).

[0078] The system retrieves multiple sample decision-related data and multiple sample negative pressure dynamic parameters. The sample decision-related data is consistent with the index composition of the control decision particle point, and each sample negative pressure dynamic parameter is coded with the sample drainage cycle and sample surgery type.

[0079] After locating multiple sample decision particle points corresponding to the multiple sample decision association data in the initial space, the multiple sample negative pressure dynamic parameters, multiple sample drainage cycles, and multiple sample surgical type codes are stored in the corresponding particle points to complete the construction of the negative pressure drainage decision space at the current stage.

[0080] Control decision particles are constructed based on the wound tissue metabolic heat distribution, wound stress vector field, exudate transport flux, patient age, surgical type coding, and serum protein level.

[0081] After projecting the control decision particle points onto the negative pressure drainage decision space, a spatial sphere is constructed based on a preset correlation radius, with the control decision particle points as the starting point. Several sample decision particle points within this spatial sphere constitute the multiple correlation particle points.

[0082] Based on obtaining multiple associated particle points, the surgical type code is used as the particle point filtering constraint to filter out P associated particle points that are consistent with the current wound surgical type.

[0083] Based on the P spatial distances between the P associated particle points and the P control decision particle points, P decision confidence weights are set.

[0084] Extract the P associated negative pressure dynamic parameters (sample negative pressure dynamic parameters) and P associated drainage cycles (sample drainage cycles) stored in the P associated particle points.

[0085] The initial negative pressure dynamic parameters and the real-time drainage cycle are output by weighting the P associated negative pressure dynamic parameters and the P associated drainage cycles using the P decision confidence weights.

[0086] This embodiment integrates the patient's inherent physiological characteristics (static features) with the real-time dynamics of the wound microenvironment to generate a dynamic parameter package containing target pressure values, ramp rate, fluctuation amplitude, and key drainage cycles, as well as a dynamically set drainage cycle. This achieves the technical effect of matching negative pressure drainage parameter settings with individual wound conditions while laying the foundation for subsequent timing adjustments.

[0087] Step A500: During the process in which the negative pressure host intermittently drives the negative pressure drainage device to perform closed-loop negative pressure suction on the wound filling dressing using the initial negative pressure dynamic parameters, constrained by the real-time drainage cycle, the initial negative pressure dynamic parameters are dynamically adjusted according to the time-series exudate characteristics.

[0088] Under the strict time-series framework of the negative pressure host following the real-time drainage cycle setting, during the intermittent initiation of the negative pressure suction cycle using the initial negative pressure kinetic parameters, the time-series exudate characteristics (real-time viscosity and flow rate) at the drainage tube output end are continuously captured by the microfluidic sensing unit, and the dynamic congestion risk coefficient is calculated based on the Markov chain model. When the risk coefficient exceeds the 0.85 threshold, the system triggers a dual-channel incremental PID control mechanism: the viscosity channel responds to the continuous high viscosity state through proportional-integral operation, and the flow rate channel captures the flow rate change through proportional-derivative operation. The two work together to generate the pressure target value correction amount, and refresh the kinetic parameters in real time within the constraint range of ±40% of the absolute value of the initial pressure, thereby realizing closed-loop dynamic optimization based on the evolution of exudate biological characteristics.

[0089] This embodiment achieves the technical effect of personalized negative pressure dynamic control based on the real-time evolution of the wound microenvironment, improving the adaptability of negative pressure drainage to the patient's condition and reducing the drainage blockage rate.

[0090] In one implementation, the method further includes:

[0091] Step A610: Using the initial negative pressure dynamic parameters to drive the negative pressure drainage device, during the closed-loop negative pressure suction process of the wound filling dressing, the exudate characteristics are collected at the output end of the negative pressure drainage tube through the microfluidic sensing unit to obtain real-time exudate characteristics, wherein the real-time exudate characteristics include real-time viscosity and real-time flow rate.

[0092] Step A620: Calculate the probability of congestion transition based on the real-time exudate characteristics and initial negative pressure dynamic parameters, and output the real-time congestion risk coefficient.

[0093] Step A630: If the real-time congestion risk coefficient meets the preset congestion threshold, then incremental PID control is performed on the initial negative pressure dynamic parameters according to the real-time exudate characteristics, and the updated negative pressure dynamic parameters are output.

[0094] In one implementation, if the real-time congestion risk coefficient does not meet the preset congestion threshold, the real-time diversion cycle is reset after triggering a reverse pulse to clear the congestion.

[0095] In one implementation, Figure 4 The method A620 provided by this invention, which calculates the probability of congestion transition based on the real-time exudate characteristics and initial negative pressure kinetic parameters, and outputs a real-time congestion risk coefficient, includes the following steps:

[0096] Step A621: Based on the real-time drainage cycle and the set time matching mechanism, the historical equipment congestion log is matched and retrieved to obtain multiple sample exudate characteristics, multiple sample negative pressure dynamic parameters, and multiple sample congestion risk coefficients.

[0097] Step A622: Divide the congestion risk coefficients of the multiple samples into states according to the preset congestion state classification to obtain W state transition paths.

[0098] Step A623: Based on the association relationship of the congestion risk coefficients of the multiple samples with the W state transition paths, perform cross-state transition frequency statistics on the exudate characteristics and negative pressure dynamic parameters of the multiple samples to construct a congestion risk transition matrix.

[0099] Step A624: Use the real-time viscosity and real-time flow velocity as observational evidence, load them into the congestion risk transition matrix, and update the output state transition probability.

[0100] Step A625: Correct the state transition probability using the pressure target value in the initial negative pressure dynamic parameters, and output the real-time congestion risk coefficient.

[0101] In one implementation, if the real-time congestion risk coefficient meets a preset congestion threshold, then incremental PID control is performed on the initial negative pressure dynamic parameters based on the real-time exudate characteristics, and updated negative pressure dynamic parameters are output. Step A630 of the method provided by this invention includes:

[0102] Step A631: Invoke the target viscosity and target flow rate adapted by the initial negative pressure kinetic parameters.

[0103] Step A632: Calculate the viscosity deviation value based on the real-time viscosity and the target viscosity, and calculate the flow rate deviation value based on the real-time flow rate and the target flow rate.

[0104] Step A633: Pre-construct a PID parallel analysis model, wherein the PID parallel analysis model includes parallel viscosity error correction channels and flow rate error correction channels.

[0105] Step A634: Map the viscosity deviation value and flow rate deviation value into the viscosity error correction channel and flow rate error correction channel of the PID parallel analysis model, perform proportional-derivative operations, and output double-correction positive terms.

[0106] Step A635: Use the dual maintenance positive term to perform incremental PID control on the pressure target value in the initial negative pressure dynamic parameters, and output the updated negative pressure dynamic parameters.

[0107] Specifically, in this embodiment, the initial negative pressure dynamic parameters are used to drive the negative pressure drainage device to perform closed-loop negative pressure suction on the wound filling dressing. At the output end of the negative pressure drainage tube, the microfluidic sensing unit performs exudate characteristic acquisition to obtain real-time exudate characteristics, wherein the real-time exudate characteristics include real-time viscosity and real-time flow rate.

[0108] Step A620: Calculate the probability of congestion transition based on the real-time exudate characteristics and initial negative pressure dynamic parameters, predict the possibility of the pipeline migrating from unobstructed to partially or completely blocked under the current state, and output the real-time congestion risk coefficient.

[0109] The specific method for obtaining the real-time congestion risk coefficient is as follows:

[0110] Using the current real-time drainage cycle duration as the key index (e.g., a 15-minute cycle only matches historical 15±2 minute cycle data), similar cases with the same time-series characteristics are retrieved from encrypted historical equipment congestion logs. The exudate viscosity / flow rate records, negative pressure parameter configurations, and final congestion risk levels of these historical cases under similar cycles are extracted to construct a reference dataset strongly correlated with the current operating environment. Specifically, the reference dataset includes multiple sample exudate characteristics, multiple sample negative pressure kinetic parameters, and multiple sample congestion risk coefficients.

[0111] Based on the preset congestion status classification (e.g., risk coefficient <0.3 is defined as smooth state, 0.3-0.7 is partially congested state, and ≥0.7 is completely congested state), the congestion risk coefficient of each sample in the historical log is classified into the corresponding state category. Based on the state change relationship between adjacent samples in the time series, all possible state transition direction combinations are extracted (e.g., smooth state → partially congested state, partially congested state → smooth state, etc.), and finally W independent state transition paths are formed.

[0112] For each defined state transition path (e.g., unobstructed state → partially blocked state), all historical sample data belonging to that path are statistically analyzed: the frequency of occurrence of exudate viscosity value, flow velocity value, and associated negative pressure dynamic parameters (pressure value / cycle, etc.) under that path are accumulated, and the congestion risk transition matrix (two-dimensional transition matrix) is constructed in this way: the matrix row index represents the state before migration, the column index represents the state after migration, and each matrix cell stores the cumulative occurrence of each parameter under that migration direction.

[0113] The current measured real-time viscosity and velocity values ​​are input into the matrix update mechanism: Based on the new observations, the frequency distribution of parameters under each transition path is corrected by Bayes (for example, a real-time viscosity of 50 cP will increase the weight of high viscosity-related paths), the trigger probability of the migration direction between all states is recalculated, and the updated state transition probability is output (which is essentially a vector, including the probability of maintaining the unobstructed state, the probability of migrating to a partially blocked state, etc.).

[0114] Using the initial negative pressure target value (e.g., -130 mmHg) as an adjustment factor, the updated state transition probability is weighted and corrected: the high-pressure target value proportionally increases the probability of a worsening path (e.g., unobstructed state → completely blocked state), while the low-pressure target value suppresses the probability of worsening. Finally, based on the corrected probability distribution, the real-time congestion risk coefficient in the range of 0-1 is generated by weighted summation (unobstructed state probability × 0.2 + partially blocked state probability × 0.5 + completely blocked state probability × 1.0).

[0115] Step A630: If the real-time congestion risk coefficient meets the preset congestion threshold (for example, if the real-time risk coefficient exceeds the 0.85 threshold, it indicates that the probability of blockage is >85%), then incremental PID control is performed on the initial negative pressure dynamic parameters according to the real-time exudate characteristics, and the updated negative pressure dynamic parameters are output.

[0116] Specifically, the preset physiological target values ​​are analyzed from the initial negative pressure dynamic parameters: target viscosity 30 cP (normal tissue exudate viscosity benchmark) and target flow rate 2 ml / min (ideal drainage rate in the middle stage of wound healing), which serve as the benchmark reference values ​​for PID control.

[0117] The viscosity deviation (absolute difference) is calculated based on the real-time viscosity and the target viscosity, and the velocity deviation (instantaneous acceleration) is calculated based on the real-time flow velocity and the target flow velocity.

[0118] A pre-built PID parallel analysis model is provided, wherein the PID parallel analysis model has a built-in decoupled dual-channel control architecture: the viscosity channel independently operates proportional (P)-integral (I) calculation to deal with the problem of continuous high viscosity, and the flow velocity channel independently performs proportional (P)-derivative (D) calculation to capture flow velocity change events. The two channels are physically isolated to avoid mutual interference.

[0119] The viscosity deviation value is input into the PI calculator of the viscosity error correction channel (proportional coefficient amplifies the deviation, integral term accumulates the continuous deviation), and the flow velocity acceleration is input into the PD calculator of the flow velocity error correction channel (proportional response instantaneous change, differential processing of abrupt change rate). The two channels independently output correction components and synthesize them into a two-dimensional vector, outputting a double-correction positive term (viscosity correction amount + flow velocity correction amount).

[0120] The dual maintenance positive items are converted into pressure adjustment values ​​according to preset gains. For example, the viscosity correction component is converted at 0.8 mmHg / cP, and the flow rate correction component is converted at 1.2 mmHg·min. 2 The system converts / ml and superimposes to generate a total pressure correction value (e.g., -12mmHg). Based on the initial pressure target value, it incrementally adjusts the pressure (e.g., -130→-142mmHg), while maintaining the initial values ​​of parameters such as the ramp rate. The updated negative pressure dynamic parameters are then output.

[0121] This embodiment achieves the technical effect of early and accurate warning of the risk of drainage tube blockage and dynamic closed-loop control of negative pressure parameters to maintain the homeostasis of the wound healing microenvironment.

[0122] In one implementation, if the real-time congestion risk coefficient does not meet the preset congestion threshold, the real-time diversion cycle is reset after triggering a reverse pulse to clear the congestion.

[0123] Specifically, when the real-time congestion risk coefficient exceeds the preset threshold, a preventive dredging mechanism is automatically triggered: the negative pressure pump is controlled to execute a reverse pulse sequence at a specific frequency (alternating between brief positive and negative pressure oscillations) to physically remove potential deposits from the inner wall of the pipe; at the same time, the real-time drainage cycle is shortened and reset to a fixed ratio of the original value, thereby proactively preventing low-risk sludge accumulation trends by increasing the suction frequency, thus forming a synergistic guarantee mechanism for blockage prevention and cycle adaptation.

[0124] Meanwhile, it should be noted that a strict human-machine collaborative decision-making mechanism is implemented in the control process of the negative pressure drainage device: the initial negative pressure dynamic parameters, real-time congestion risk coefficient and dynamic adjustment suggestions automatically generated by the negative pressure host are pushed to the clinical decision support dashboard of the human-machine interface in real time, and the historical comparison curve of the pressure target value, the real-time monitoring waveform of the exudate viscosity and flow rate and the time evolution path of the congestion risk coefficient are presented synchronously through visual charts.

[0125] After reading the data on the control panel, medical staff can perform operations on the recommended parameters of the negative pressure unit via touch or voice commands: manually correcting the parameters within the safety boundaries (pressure target value is allowed to fluctuate by ±20%, drainage cycle is allowed to be adjusted by ±30%), or directly confirming execution and emergency termination of negative pressure drainage; ultimately, the negative pressure unit will only drive the negative pressure drainage device to perform the corresponding operation after obtaining the medical staff's biometric-bound electronic signature authorization. In other words, the control / adjustment data of the negative pressure drainage device generated in this embodiment is not directly used for equipment control, and the actual parameter adjustment and control settings of the equipment are still based on the decision of the medical staff.

[0126] Example 2: Based on the same inventive concept as the negative pressure drainage device control method for wound care described in the foregoing examples, this invention provides a negative pressure drainage device control system for wound care. See [link to example]. Figure 5 As shown, the system includes:

[0127] The wound data acquisition unit 11 is used to drive the micro thermocouple, impedance electrode and flexible piezoresistive film integrated in the wound filling dressing to acquire dynamic wound data after the wound filling dressing has been sealed and covered in the target patient's wound for a period of time that meets the steady state duration.

[0128] The wound data feedback unit 12 is used to transmit the dynamic wound data back to the negative pressure host of the negative pressure drainage device through a medical-grade IPX interface. The dynamic wound data includes the metabolic heat distribution of wound tissue, the stress vector field of the wound, and the exudate transport flux.

[0129] The static data retrieval unit 13 is used to retrieve static patient characteristics from the HIS system using the unique identification code of the target patient.

[0130] The negative pressure drainage decision unit 14 is used to make dual decisions on negative pressure drainage based on static patient characteristics and dynamic wound data, and output initial negative pressure dynamic parameters, wherein the initial negative pressure dynamic parameters include a real-time drainage cycle.

[0131] The negative pressure drainage parameter adjustment and update unit 15 is used to dynamically adjust the initial negative pressure dynamic parameters according to the time-series exudate characteristics during the process in which the negative pressure host intermittently drives the negative pressure drainage device to perform closed-loop negative pressure suction on the wound filling dressing with the initial negative pressure dynamic parameters as a constraint of the real-time drainage cycle.

[0132] In one implementation, the system is further used for:

[0133] The initial negative pressure kinetic parameters are used to drive the negative pressure drainage device. During the closed-loop negative pressure suction process of the wound filling dressing, the exudate characteristics are collected at the output end of the negative pressure drainage tube through a microfluidic sensing unit to obtain real-time exudate characteristics, including real-time viscosity and real-time flow rate. Based on the real-time exudate characteristics and the initial negative pressure kinetic parameters, the probability of congestion transition is calculated, and a real-time congestion risk coefficient is output. If the real-time congestion risk coefficient meets a preset congestion threshold, incremental PID control is performed on the initial negative pressure kinetic parameters based on the real-time exudate characteristics, and updated negative pressure kinetic parameters are output.

[0134] In one implementation, the system is further used for:

[0135] Based on the real-time drainage cycle and a set time matching mechanism, historical equipment congestion logs are retrieved to obtain multiple sample exudate characteristics, multiple sample negative pressure kinetic parameters, and multiple sample congestion risk coefficients. According to a preset congestion state classification, the multiple sample congestion risk coefficients are divided into states, resulting in W state transition paths. Based on the association of the multiple sample congestion risk coefficients among the W state transition paths, the state transition frequency of the multiple sample exudate characteristics and multiple sample negative pressure kinetic parameters is statistically analyzed to construct a congestion risk transition matrix. The real-time viscosity and real-time flow velocity are used as observational evidence and loaded into the congestion risk transition matrix to update the output state transition probability. The pressure target value in the initial negative pressure kinetic parameters is used to correct the state transition probability, and the real-time congestion risk coefficient is output.

[0136] In one implementation, the system is further used for:

[0137] The initial negative pressure kinetic parameters are used to adapt to the target viscosity and target flow rate; the viscosity deviation value is calculated based on the real-time viscosity and target viscosity, and the flow rate deviation value is calculated based on the real-time flow rate and target flow rate; a pre-constructed PID parallel analysis model is built, wherein the PID parallel analysis model includes parallel viscosity error correction channels and flow rate error correction channels; the viscosity deviation value and flow rate deviation value are mapped and input into the viscosity error correction channels and flow rate error correction channels of the PID parallel analysis model to perform proportional-derivative operations, and output a double-correction positive term; the double-correction positive term is used to perform incremental PID control on the pressure target value in the initial negative pressure kinetic parameters, and output the updated negative pressure kinetic parameters.

[0138] In one implementation, the wound data acquisition unit 11 is further configured to:

[0139] By scanning the wound of the target patient, the geometric and edge features of the wound are quantified and output. Based on the measured exudate characteristics input by medical staff, the geometric and edge features of the wound, a dressing requirement feature vector is constructed. Using the dressing requirement feature vector, material mechanics matching is performed in the dressing configuration database to obtain the wound filling dressing. The micro thermocouple is integrated into the base contact layer of the wound filling dressing, the impedance electrodes are distributed in an interdigitated array on the pore inner wall of the wound filling dressing, and the flexible piezoresistive film is integrated into the closed membrane junction area of ​​the wound filling dressing.

[0140] In one implementation, the negative pressure drainage decision unit 14 is further configured to:

[0141] The static patient characteristics are decomposed to obtain patient age, surgical type code, and serum protein level; wound stage matching is performed based on patient age, surgical type code, and measured exudate characteristics, and stage labels are output; the negative pressure drainage decision space is invoked based on the stage labels; control decision particles are constructed based on the wound tissue metabolic heat distribution, wound stress vector field, exudate transport flux, patient age, surgical type code, and serum protein level; the control decision particles are loaded into the negative pressure drainage decision space to perform multidimensional control parameter weighted quantization, and the initial negative pressure dynamic parameters are output.

[0142] In one implementation, the negative pressure drainage decision unit 14 is further configured to:

[0143] After projecting the control decision particle point onto the negative pressure drainage decision space, using the control decision particle point as the starting point and the surgical type code as the particle point screening constraint, P related particle points are selected based on a preset association radius. P decision confidence weights are set according to the P spatial distances between the P related particle points and the control decision particle point. The P related negative pressure dynamic parameters and P related drainage cycles stored in the P related particle points are extracted. The initial negative pressure dynamic parameters and real-time drainage cycle are output by weighting the P related negative pressure dynamic parameters and P related drainage cycles using the P decision confidence weights.

[0144] In one implementation, the system is further used for:

[0145] If the real-time congestion risk coefficient does not meet the preset congestion threshold, the real-time diversion cycle is reset after triggering the reverse pulse to clear the congestion.

[0146] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for controlling a negative pressure drainage device for wound care, characterized in that, include: After the wound filling dressing is sealed and covered on the target patient's wound for a period of time that meets the steady-state duration, the micro thermocouple, impedance electrode and flexible piezoresistive film integrated in the wound filling dressing are driven to collect dynamic wound data. The dynamic wound data is transmitted back to the negative pressure host of the negative pressure drainage device via a medical-grade IPX interface. The dynamic wound data includes the metabolic heat distribution of wound tissue, the stress vector field of the wound, and the exudate transport flux. Static patient characteristics are retrieved from the HIS system using the target patient's unique identification code; The system makes a dual decision on negative pressure drainage based on static patient characteristics and dynamic wound data, and outputs initial negative pressure dynamic parameters, wherein the initial negative pressure dynamic parameters include the real-time drainage cycle. During the process in which the negative pressure host intermittently drives the negative pressure drainage device to perform closed-loop negative pressure suction on the wound filling dressing, constrained by the real-time drainage cycle, the initial negative pressure dynamic parameters are dynamically adjusted according to the time-series exudate characteristics.

2. The control method for a negative pressure drainage device for wound care as described in claim 1, characterized in that, Also includes: The initial negative pressure dynamic parameters are used to drive the negative pressure drainage device to perform closed-loop negative pressure suction on the wound filling dressing. At the output end of the negative pressure drainage tube, the microfluidic sensing unit performs exudate feature acquisition to obtain real-time exudate characteristics, wherein the real-time exudate characteristics include real-time viscosity and real-time flow rate. Based on the real-time exudate characteristics and initial negative pressure dynamic parameters, the probability of congestion state transition is calculated, and the real-time congestion risk coefficient is output. If the real-time congestion risk coefficient meets the preset congestion threshold, then incremental PID control is performed on the initial negative pressure dynamic parameters based on the real-time exudate characteristics, and the updated negative pressure dynamic parameters are output.

3. The control method for a negative pressure drainage device for wound care as described in claim 2, characterized in that, Based on the real-time exudate characteristics and initial negative pressure kinetic parameters, the probability of congestion transition is calculated, and a real-time congestion risk coefficient is output, including: Based on the real-time drainage cycle, a time matching mechanism is set, and historical equipment congestion logs are matched and retrieved to obtain multiple sample exudate characteristics, multiple sample negative pressure dynamic parameters, and multiple sample congestion risk coefficients. Based on a preset congestion state classification, the congestion risk coefficients of the multiple samples are divided into states to obtain W state transition paths; Based on the association of the congestion risk coefficients of the multiple samples with the W state transition paths, cross-state transition frequency statistics are performed on the exudate characteristics and negative pressure dynamic parameters of the multiple samples to construct a congestion risk transition matrix. The real-time viscosity and real-time flow velocity are used as observational evidence and loaded into the congestion risk transition matrix to update the output state transition probability. The state transition probability is corrected using the target pressure value in the initial negative pressure dynamic parameters, and the real-time congestion risk coefficient is output.

4. The control method for a negative pressure drainage device for wound care as described in claim 3, characterized in that, If the real-time congestion risk coefficient meets the preset congestion threshold, then incremental PID control is performed on the initial negative pressure dynamic parameters based on the real-time exudate characteristics, and updated negative pressure dynamic parameters are output, including: The target viscosity and target flow velocity are adapted by invoking the initial negative pressure kinetic parameters; The viscosity deviation value is calculated based on the real-time viscosity and the target viscosity, and the flow velocity deviation value is calculated based on the real-time flow velocity and the target flow velocity; A pre-constructed PID parallel analysis model is provided, wherein the PID parallel analysis model includes parallel viscosity error correction channels and flow rate error correction channels; The viscosity deviation value and flow rate deviation value are mapped and input into the viscosity error correction channel and flow rate error correction channel of the PID parallel analysis model to perform proportional-derivative operations and output double-correction positive terms; The dual maintenance positive term is used to perform incremental PID control on the pressure target value in the initial negative pressure dynamic parameters, and the updated negative pressure dynamic parameters are output.

5. A control method for a negative pressure drainage device for wound care as described in claim 1, characterized in that, Also includes: By scanning the wound of the target patient, the geometric features and edge features of the wound are quantified and output. A dressing requirement feature vector is constructed based on the measured exudate characteristics, wound geometry, and wound edge features input by medical staff. Using the aforementioned dressing requirement feature vector, material mechanics matching is performed in the dressing configuration database to obtain wound filling dressings; The micro thermocouple is integrated into the base contact layer of the wound filling dressing, the impedance electrodes are distributed in an interdigitated array on the inner wall of the pores of the wound filling dressing, and the flexible piezoresistive film is integrated into the sealing membrane junction area of ​​the wound filling dressing.

6. The control method for a negative pressure drainage device for wound care as described in claim 5, characterized in that, Based on both static patient characteristics and dynamic wound data, a dual decision-making process for negative pressure drainage is performed, outputting initial negative pressure kinetic parameters, including: The static patient characteristics were decomposed to obtain patient age, surgical type code, and serum protein level; Based on the patient's age, surgical type code, and measured exudate characteristics, wound stage matching is performed, and stage labels are output. The negative pressure drainage decision space is invoked based on the stage labels mentioned above; Control decision particles are constructed based on the wound tissue metabolic heat distribution, wound stress vector field, exudate transport flux, patient age, surgical type coding, and serum protein level. The control decision particle points are loaded into the negative pressure drainage decision space to perform multidimensional control parameter weighted quantization, and the initial negative pressure dynamic parameters are output.

7. A control method for a negative pressure drainage device for wound care as described in claim 6, characterized in that, The control decision particle points are loaded into the negative pressure drainage decision space to perform multi-dimensional control parameter weighted quantization, and the initial negative pressure dynamic parameters are output, including: After projecting the control decision particle points onto the negative pressure drainage decision space, P related particle points are selected based on the preset association radius, starting from the control decision particle points and using the surgical type code as the particle point screening constraint. Based on the P spatial distances between the P associated particle points and the P control decision particle points, set P decision confidence values; Extract the P associated negative pressure dynamic parameters and P associated drainage cycles stored in the P associated particle points; The initial negative pressure dynamic parameters and the real-time drainage cycle are output by weighting the P associated negative pressure dynamic parameters and the P associated drainage cycles using the P decision confidence weights.

8. A control method for a negative pressure drainage device for wound care as described in claim 2, characterized in that, If the real-time congestion risk coefficient does not meet the preset congestion threshold, the real-time diversion cycle is reset after triggering the reverse pulse to clear the congestion.

9. A control system for a negative pressure drainage device for wound care, characterized in that, The steps for implementing the method according to any one of claims 1 to 8 include: The wound data acquisition unit is used to drive the micro thermocouple, impedance electrode and flexible piezoresistive film integrated in the wound filling dressing to acquire dynamic wound data after the wound filling dressing has been sealed and covered in the target patient's wound for a period of time that meets the steady state time. The wound data feedback unit is used to transmit the dynamic wound data back to the negative pressure host of the negative pressure drainage device through a medical-grade IPX interface. The dynamic wound data includes the metabolic heat distribution of wound tissue, the stress vector field of the wound, and the exudate transport flux. The static data retrieval unit is used to retrieve static patient characteristics from the HIS system using the target patient's unique identification code. The negative pressure drainage decision unit is used to make dual decisions on negative pressure drainage based on static patient characteristics and dynamic wound data, and output initial negative pressure dynamic parameters, wherein the initial negative pressure dynamic parameters include the real-time drainage cycle; The negative pressure drainage parameter adjustment and update unit is used to dynamically adjust the initial negative pressure dynamic parameters according to the time-series exudate characteristics during the process in which the negative pressure host intermittently drives the negative pressure drainage device to perform closed-loop negative pressure suction on the wound filling dressing with the initial negative pressure dynamic parameters, constrained by the real-time drainage cycle.

Images