A burn care method integrating pressure dynamic monitoring and intelligent liquid infiltration management

Abstract

This invention relates to the field of intelligent monitoring technology, specifically to a burn care method integrating dynamic pressure monitoring and intelligent exudate management. The method includes the following steps: collecting pressure data through an embedded pressure sensor array, generating a local pressure distribution map, analyzing pressure concentration areas, acquiring exudate data and generating an exudate diffusion map, combining the two to analyze the interaction area, assessing healing risk and generating a nursing instruction set, converting it into a time series and combining it with the data, adjusting the order of nursing instructions through a chi-square test, and generating a nursing adjustment instruction set. In this invention, pressure data is collected through a pressure sensor array to generate a local pressure distribution map, identify pressure concentration areas, and convert exudate impedance signals into quantitative data to enhance monitoring accuracy. By superimposing pressure and exudate distribution, potential risk areas are analyzed, providing a basis for nursing decisions. Healing risk is assessed through pressure gradient and exudate diffusion rate, supporting personalized nursing instructions and improving nursing quality and timeliness.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, and in particular to a burn care method that integrates dynamic pressure monitoring and intelligent exudate management. Background Technology

[0002] The field of intelligent monitoring technology encompasses methods for real-time acquisition and quantitative analysis of physiological parameters. Its core lies in the continuous acquisition of multi-dimensional information such as pressure, temperature, humidity, and exudate to achieve dynamic monitoring of the body's condition. The system comprehensively covers various medical and nursing monitoring methods, including load change detection based on pressure sensing elements, exudate identification based on fluid conductivity characteristics, and status assessment based on continuous signals.

[0003] One of the burn care methods integrating dynamic pressure monitoring and intelligent exudate management refers to a specific scheme for the joint monitoring and management of pressure changes and exudate in burn wound care. It encompasses real-time acquisition of wound contact pressure through the installation of a pressure-sensitive layer, quantitative collection and pathway control of exudate through an exudate adsorption structure, and coordinated recording of pressure signals and fluid flow status through an integrated monitoring approach.

[0004] While existing technologies in burn care encompass basic methods for pressure monitoring and exudate detection, their processing is relatively simplistic and lacks in-depth integration and analysis of multi-dimensional data. Pressure monitoring typically relies solely on pressure sensors to detect changes in load. While this reflects changes in wound pressure, it fails to accurately reflect the details of pressure distribution and does not provide clear identification of pressure concentration zones. For exudate monitoring, current technologies depend on traditional fluid collection and conduction methods, lacking real-time dynamic tracking and quantification of exudate status. This makes it difficult to precisely adjust exudate management and care measures according to actual conditions. This approach may result in the failure to promptly identify localized situations of excessive wound pressure or excessive exudate, thus affecting wound healing. For example, unmonitored excessive pressure may lead to wound ischemia and necrosis; while uncontrolled excessive exudate may lead to infection or other complications. The simplistic nature of existing technologies in pressure and exudate management results in a lack of precision and personalization in burn care, failing to fully utilize data mining techniques to improve monitoring and care effectiveness. Summary of the Invention

[0005] To address the technical problems existing in the prior art, this invention provides a burn care method integrating dynamic pressure monitoring and intelligent exudate management, comprising the following steps: To achieve the above objectives, the present invention adopts the following technical solution: a burn care method integrating dynamic pressure monitoring and intelligent exudate management, comprising the following steps: S1: Embed a miniature pressure sensor array in the burn dressing to collect pressure data of the wound and surrounding tissues, perform noise reduction and smoothing processing, construct a pressure gradient, input it into the least squares method for gradient fitting, and generate a local pressure distribution map. S2: Based on the local pressure distribution map, analyze the pressure concentration area of ​​the wound, obtain the exudate impedance change signal through the bioelectrical impedance sensor embedded in the dressing, and convert it into exudate electrical data after signal amplification and filtering. Then, combine the pressure concentration area for clustering to generate an exudate diffusion distribution map. S3: Based on the spatial superposition of the local pressure distribution map and the seepage diffusion distribution map, analyze the directional relationship and cross-region characteristics of the two, and divide them into cooperative interaction zone and antagonistic interaction zone, generating interaction zone division results; S4: Based on the results of the interaction zone division, calculate the average pressure gradient and exudate diffusion rate in multiple interaction zones and analyze the ratio, assess the wound healing risk level of the interaction zone, match the corresponding nursing instructions, and generate a nursing instruction set. S5: The nursing instruction set is converted into a nursing time series, and the pressure monitoring data and exudation monitoring data of the corresponding interaction area are extracted. The offset is determined by establishing the data residual distribution. When the offset exceeds the nursing instruction offset threshold, the execution order of the nursing instructions is adjusted to generate a burn nursing adjustment instruction set.

[0006] As a further embodiment of the present invention, the local pressure distribution map includes a pressure gradient field, a pressure gradient direction, and local abnormal pressure points; the exudate diffusion distribution map includes exudate concentration distribution, diffusion radius parameters, and exudate electrical data; the interaction zone division result includes a collaborative interaction zone and an antagonistic interaction zone; the nursing instruction set includes wound decompression nursing instructions, exudate removal nursing instructions, and dressing change nursing instructions; and the burn nursing adjustment instruction set includes the adjusted execution order, nursing time sequence, and offset determination.

[0007] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Embed a miniature pressure sensor array in the burn dressing to collect pressure data of the wound and surrounding tissues and arrange them according to the time series. Remove and smooth the noise and abnormal points in the series to generate a pressure data series. S102: Based on the pressure data sequence, calculate the pressure difference between adjacent sensing nodes in the pressure data sequence, and vectorize the difference result on a two-dimensional plane to obtain the pressure gradient field. S103: Call the pressure gradient field, perform least squares fitting on the gradient value and position in the spatial coordinate system, and map the local pressure intensity and distribution range based on the fitting result to obtain a local pressure distribution map.

[0008] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Based on the local pressure distribution map, retrieve the regions in the distribution map where the pressure intensity value exceeds the preset pressure threshold, and aggregate them at the boundary points within the two-dimensional coordinate range. Mark the coordinate set within the aggregated range as the pressure concentration area to obtain the coordinate set of the pressure concentration area. S202: Based on the coordinate set of the pressure concentration area, obtain the exudate impedance change signal collected by the bioelectrical impedance sensor embedded in the dressing, amplify it, remove high-frequency noise components through a filter, and map the filtered impedance change value in time order to obtain the exudate electrical data sequence. S203: Call the coordinate set of the pressure concentration area and the seepage electrical data sequence, perform correlation calculation between the seepage electrical data value and the corresponding pressure area under a unified index space, and perform clustering of the seepage data in the same area, and perform spatial diffusion mapping of the clustering results to generate a seepage diffusion distribution map. The pressure threshold is set by statistically analyzing the pressure intensity values ​​of multiple nodes in the local pressure distribution map.

[0009] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Based on the local pressure distribution map and the seepage diffusion distribution map, the spatial position indices of the two distribution maps are superimposed point by point under a unified spatial coordinate system, and the coordinate values ​​of the overlapping points are recorded to generate a spatial superimposed coordinate set. S302: Call the spatial superimposed coordinate set, compare the direction vectors of adjacent points one by one. If the vector angle does not reach the direction similarity threshold, mark them as the same direction. If the vector angle exceeds the direction similarity threshold, mark them as opposite directions, and obtain the direction relationship marking sequence. S303: Based on the directional relationship marker sequence, retrieve the same-direction marker region and the opposite-direction marker region, aggregate the same-direction region into a cooperative interaction region, aggregate the opposite-direction region into an adversarial interaction region, and generate the interaction region division result within the overall coordinate range.

[0010] As a further aspect of the present invention, the direction similarity threshold is set by calculating the angle between all adjacent direction vectors, arranging them in order of magnitude, and then taking the median of the arrangement results as the representative value of the direction distribution.

[0011] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Based on the interaction zone division results, extract the pressure monitoring data and area parameters in the multiple interaction zones and normalize them. Then, perform differential calculation on the normalized pressure monitoring data in each interaction zone, and then perform interval averaging on the differential calculation results to generate an average pressure gradient sequence. S402: Call the average pressure gradient sequence, extract the seepage volume data and diffusion radius parameters in the same area, calculate the seepage diffusion rate distribution value and perform weighting to generate a seepage diffusion rate sequence; S403: Based on the average pressure gradient sequence and the exudate diffusion rate sequence, calculate the ratio of their values ​​in the corresponding interaction zone, classify the risk level of the ratio result, and then match nursing instruction items under multiple risk level results to generate a nursing instruction set.

[0012] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Based on the nursing instruction set, extract the execution time and action sequence of multiple nursing instructions, convert the time parameters into an equally spaced index sequence, and then rearrange the action sequence labels according to the index sequence to generate a nursing time sequence. S502: Call the nursing time series, retrieve the pressure monitoring data and exudation monitoring data in the interaction area and aggregate them according to time points, establish the data residual distribution based on the difference results between the monitoring data and the index time points, and perform offset determination to obtain the offset determination result; S503: Based on the offset determination result, the determined offset is compared with the nursing instruction offset threshold. When the offset exceeds the nursing instruction offset threshold, the execution order of the nursing instructions is rearranged to generate a burn nursing adjustment instruction set.

[0013] As a further aspect of the present invention, the nursing instruction offset threshold is set by statistically retrieving the joint distribution of the pressure monitoring data and the effusion monitoring data in the nursing instruction execution data, extracting the maximum fluctuation range, and then weighting the mean and standard deviation of the maximum fluctuation range.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: This invention effectively achieves dynamic monitoring and assessment in burn care through refined wound pressure and exudate management. The application of a miniature pressure sensor array allows for real-time acquisition of pressure data from the wound and surrounding tissues, generating detailed local pressure distribution maps. This process identifies areas of concentrated wound pressure, providing precise location for subsequent exudate analysis. Unlike traditional single-mode pressure monitoring, combining the acquisition and processing of exudate impedance signals, through signal amplification and filtering, transforms exudate data into quantifiable electrical data, further enhancing the accuracy and detail of monitoring. By spatially superimposing pressure and exudate distribution maps and analyzing their directional relationship and intersection characteristics, potential risk areas in wound healing can be revealed, providing a scientific basis for nursing decisions. Furthermore, by analyzing the ratio of pressure gradient to exudate diffusion rate, the wound healing risk level can be assessed, supporting the generation of personalized nursing instructions and ensuring the accuracy and timeliness of nursing plans. Overall, the innovative solution improves the accuracy, real-time performance, and adaptability of monitoring through in-depth integration and analysis of multi-dimensional information, providing more comprehensive support for risk prediction and intervention in burn care and significantly enhancing the quality of care. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the invention, terms such as "exemplarily," "for example," etc., are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as an example in the invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the term "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the invention, the meaning expressed and / or may be both, or either one may be preferred.

[0019] In this embodiment of the invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing their distinction, their intended meanings are consistent. Similarly, the terms "of," "corresponding," and "relevant" may sometimes be used interchangeably. It should be noted that, without emphasizing their distinction, their intended meanings are consistent.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] Please see Figure 1 This invention provides a burn care method integrating dynamic pressure monitoring and intelligent exudate management, comprising the following steps: S1: Embed a miniature pressure sensor array in the burn dressing to collect pressure data of the wound and surrounding tissues, perform noise reduction and smoothing processing, construct a pressure gradient, input it into the least squares method for gradient fitting, and generate a local pressure distribution map. S2: Based on the analysis of local pressure distribution map, the pressure concentration area of ​​the wound is analyzed. The exudate impedance change signal is obtained by the bioelectrical impedance sensor embedded in the dressing. After signal amplification and filtering, it is converted into exudate electrical data. Then, it is clustered with the pressure concentration area to generate an exudate diffusion distribution map. S3: Based on the spatial overlay of the local pressure distribution map and the seepage diffusion distribution map, analyze the directional relationship and cross-region characteristics of the two, and divide them into cooperative interaction zone and antagonistic interaction zone, generating the interaction zone division results; S4: Based on the results of the interaction zone division, calculate the average pressure gradient and exudate diffusion rate in multiple interaction zones and analyze the ratio, assess the wound healing risk level of the interaction zone, match the corresponding nursing instructions, and generate a nursing instruction set. S5: Convert the nursing instruction set into a nursing time series and extract the pressure monitoring data and effusion monitoring data of the corresponding interaction area. Determine the offset by establishing the data residual distribution. When the offset exceeds the nursing instruction offset threshold, adjust the execution order of the nursing instructions and generate a burn nursing adjustment instruction set.

[0023] The local pressure distribution map includes the pressure gradient field, pressure gradient direction, and local abnormal pressure points; the exudate diffusion distribution map includes the exudate concentration distribution, diffusion radius parameter, and exudate electrical data; the interaction zone division results include the collaborative interaction zone and the antagonistic interaction zone; the nursing instruction set includes wound decompression nursing instructions, exudate removal nursing instructions, and dressing change nursing instructions; and the burn nursing adjustment instruction set includes the adjusted execution order, nursing time sequence, and offset determination.

[0024] Please see Figure 2 The specific steps of S1 are as follows: S101: Embed a miniature pressure sensor array in the burn dressing to collect pressure data of the wound and surrounding tissues and arrange them according to the time series. Remove and smooth the noise and abnormal points in the series to generate a pressure data series. A 3x3 miniature pressure sensor array was embedded in the burn dressing, with a sensor spacing of 10 mm. Each sensor was sampled at a frequency of 1 Hz. The collected pressure values, corresponding timestamps, and sensor location codes were correlated and arranged chronologically. The raw pressure data sequence was processed to remove noise and outliers. The noise threshold was set by applying multiple constant pressures ranging from 2.0 kPa to 5.0 kPa to a silicone model similar to human tissue, collecting data, calculating the standard deviation at each pressure level, and taking three times the average as the threshold, set at 0.12 kPa. The outlier threshold was set by collecting data under two conditions: normal limb activity and simulated external impact. The rate of pressure change was analyzed, and a critical value of 1.8 kPa / s was used to distinguish between the two conditions. During data processing, each data point was compared with the arithmetic mean of the readings at two consecutive time points. If the absolute value of the difference was greater than 0.12 kPa, it was identified as a noise point and replaced with the arithmetic mean. Subsequently, the pressure change rate between each data point and the previous time point is calculated. If the change rate is greater than 1.8 kPa / s, it is identified as an outlier and replaced with the valid pressure data from the previous time point. After traversing all data points from all sensors and performing the above corrections, a pressure data sequence that reflects the true pressure change trend is generated.

[0025] S102: Based on the pressure data sequence, calculate the pressure difference between adjacent sensing nodes in the pressure data sequence, and vectorize the difference result on a two-dimensional plane to obtain the pressure gradient field. Based on the generated pressure data sequence, the pressure data of each sensor node at a specific time point (T=5s) are selected, and the pressure difference between adjacent sensor nodes is calculated. The sensor array is arranged in a 3*3 configuration, and the coordinates of each sensor are assigned according to its physical location. As shown in Table 1, the coordinates and pressure values ​​of each sensor node at T=5s are listed.

[0026] Table 1: Pressure data of each sensor node at time T=5s f1 (0，20) 3.15 f2 (10，20) 3.20 f3 (20，20) 3.22 f4 (0，10) 3.85 f5 (10，10) 4.10 f6 (20，10) 3.90 f7 (0，0) 3.40 f8 (10，0) 3.55 f9 (20，0) 3.48 As shown in Table 1, taking sensor f5 (coordinates (10, 10)) located at the center of the array as an example, the pressure difference between it and its surrounding adjacent nodes is calculated. In the horizontal direction, the difference between the pressure value of f5 (4.10 kPa) and the pressure value of f4 (3.85 kPa) is 0.25 kPa; the difference between the pressure value of f6 (3.90 kPa) and the pressure value of f5 (4.10 kPa) is -0.20 kPa. In the vertical direction, the difference between the pressure value of f5 (4.10 kPa) and the pressure value of f2 (3.20 kPa) is 0.90 kPa; the difference between the pressure value of f8 (3.55 kPa) and the pressure value of f5 (4.10 kPa) is -0.55 kPa. These pressure difference results are then vectorized by dividing the pressure difference by the distance between the nodes (10 mm) to obtain the pressure gradient components in each direction. For example, at the midpoint of the line connecting f4 and f5, the horizontal pressure gradient is 0.25 kPa divided by 10 mm, which is 0.025 kPa / mm. Repeat this calculation for all adjacent node pairs to obtain a discrete dataset consisting of position coordinates and corresponding pressure gradient vectors. This dataset is the pressure gradient field.

[0027] S103: Call the pressure gradient field, perform least squares fitting on the gradient value and position in the spatial coordinate system, and map the local pressure intensity and distribution range based on the fitting result to obtain the local pressure distribution map. The generated discrete pressure gradient field data is retrieved. In a two-dimensional spatial coordinate system, least-squares fitting is performed on the gradient values ​​and location coordinates to construct a continuous bivariate function that describes the pressure distribution across the entire region. This process aims to determine a set of function coefficients that minimizes the difference between the calculated gradient at each discrete point derived from the function and the actual gradient vector obtained in S102. Based on the fitting results, the local pressure intensity is mapped to the distribution range. First, a high-pressure region judgment benchmark is established, referencing the pressure values ​​in clinical data that significantly affect capillary reperfusion in burn wounds; this benchmark is set to 4.0 kPa. Subsequently, a high-resolution (e.g., 1 mm * 1 mm) virtual grid is created within a 20 mm * 20 mm area covered by the sensor array. For each point in the grid, the specific pressure value at that point is calculated using the fitted continuous pressure function. The calculated pressure value is compared with the 4.0 kPa benchmark. If the calculated pressure value at a point is greater than 4.0 kPa, it is marked as a high-pressure point. For example, the pressure value at coordinate point (8, 12) is calculated to be 4.022 kPa, which is greater than 4.0 kPa, so this point is marked. After calculating and comparing all grid points, all marked high-pressure points are aggregated to form one or more continuous regions. A local pressure distribution map is generated using the boundary coordinate set of these regions.

[0028] Please see Figure 3 The specific steps of S2 are as follows: S201: Based on the local pressure distribution map, retrieve the regions in the distribution map where the pressure intensity value exceeds the preset pressure threshold, and aggregate them at the boundary points within the two-dimensional coordinate range. Mark the coordinate set within the aggregated range as the pressure concentration area to obtain the coordinate set of the pressure concentration area. Based on the generated local pressure distribution map, the pressure intensity values ​​of each node in the map are retrieved to determine the pressure concentration areas. First, a preset pressure threshold is set. The threshold setting process is as follows: All pressure intensity values ​​of a 21*21 grid (441 nodes) under the dressing are collected at the current time. These 441 values ​​are sorted in ascending order, and the value at the 85th percentile is selected as the preset pressure threshold. For example, if the pressure values ​​of the 441 nodes range from 3.10 kPa to 4.18 kPa, after sorting by value, the pressure value of the 375th point is 4.05 kPa, so the preset pressure threshold is set to 4.05 kPa. Then, all 441 nodes in the local pressure distribution map are traversed, and the pressure intensity value of each node is compared with the 4.05 kPa threshold. The coordinates of all nodes with pressure intensity values ​​exceeding 4.05 kPa are filtered out, forming a list of high-pressure point coordinates. Next, boundary aggregation is performed on the coordinates in the list. Starting from the first coordinate point in the list, all other high-pressure points with a spatial distance of less than or equal to 1.5 mm are retrieved and grouped into the same set. This retrieval process is repeated, starting from new points within the set, until no new points can be added. After this process, an aggregation range is determined. If there are still unaggregated coordinate points in the list, another unaggregated point is selected as a new starting point, and the aggregation process is repeated. All high-pressure points are divided into one or more aggregation ranges. The coordinate set within each independent aggregation range is marked as a pressure concentration zone, thus obtaining the pressure concentration zone coordinate set.

[0029] S202: Based on the coordinate set of the pressure concentration area, the exudate impedance change signal collected by the bioelectrical impedance sensor embedded in the dressing is obtained, amplified and processed, and high-frequency noise components are removed by a filter. The filtered impedance change value is numerically mapped in time order to obtain the exudate electrical data sequence. Based on the obtained coordinate set of the pressure concentration area, the raw exudate impedance change signal collected by the bioelectrical impedance sensor embedded in the dressing, corresponding to the position of the pressure sensor, is acquired. This is a weak voltage signal, the amplitude of which is proportional to the change in inter-electrode impedance caused by the amount of wound exudate. First, this raw signal is input to a preamplifier circuit with a gain coefficient set to 100, linearly amplifying the amplitude of the raw signal by a factor of 100. For example, a signal with an original amplitude of 5 millivolts becomes 500 millivolts after amplification. Subsequently, the amplified signal passes through a low-pass filter. The cutoff frequency of this filter is set to 0.2 Hz based on experimental data. The rationale is that spectral analysis of the wound exudate signal monitored continuously for 24 hours revealed that the effective signal frequency components related to changes in exudate volume are all below 0.1 Hz, while high-frequency noise components generated by body activity or the circuit itself are mainly distributed above 0.5 Hz. Therefore, a cutoff frequency of 0.2 Hz can filter out noise. The filtering process attenuates the amplitude of components with frequencies above 0.2 Hz by more than 95%. A composite signal containing both valid signal and noise is filtered, effectively removing the high-frequency noise component. Finally, the filtered continuous voltage signal is sampled at the same time sampling rate (1Hz) as the pressure data, and the voltage value of each sampling point is converted into a specific impedance value (unit: ohms) through an analog-to-digital converter. The samples are then arranged in chronological order to obtain the seepage electrical data sequence.

[0030] S203: Call the coordinate set of the pressure concentration area and the seepage electrical data sequence, perform correlation calculation between the seepage electrical data value and the corresponding pressure area under a unified index space, and perform clustering of the seepage data in the same area. Then, perform spatial diffusion mapping on the clustering results to generate a seepage diffusion distribution map. The pressure threshold is set statistically based on the pressure intensity values ​​of multiple nodes in the local pressure distribution map; The system retrieves the coordinate set of the determined pressure concentration zone and the generated exudate electrical data sequence. Under a unified two-dimensional coordinate index space, the exudate electrical data value of each bioelectrical impedance sensor at a specific time is associated with the pressure zone where the sensor is located. For example, if the pressure concentration zone covers the coordinates (10, 10) of sensor f5, then the impedance reading of the bioelectrical impedance sensor corresponding to f5 at that time is associated with that pressure concentration zone. Next, the exudate data within the same pressure concentration zone are clustered. The criterion for this clustering is set through previous experiments: in a dry state, the baseline impedance value of the dressing is 1600 ohms; an impedance value below 1300 ohms is defined as saturated (level 3); an impedance value between 1300 and 1500 ohms is defined as wet (level 2); and an impedance value above 1500 ohms is defined as dry (level 1). Taking T=5s as an example, the S5 sensor at the core of the pressure concentration zone is associated with an impedance value of 1280 ohms. Since 1280 is less than 1300, the core region is classified as saturated. The adjacent sensors f4 and f6 have impedance readings of 1350 ohms and 1360 ohms respectively, both falling within the 1300-1500 ohm range; therefore, these two locations are classified as wet. The clustering results are then spatially mapped outwards from the saturated region to represent wet regions, until a dry region is identified, generating a permeation diffusion distribution map.

[0031] Please see Figure 4 The specific steps of S3 are as follows: S301: Based on the local pressure distribution map and the seepage diffusion distribution map, the spatial location indexes of the two distribution maps are superimposed point by point under a unified spatial coordinate system, and the coordinate values ​​of the overlapping points are recorded to generate a spatial superimposed coordinate set. Based on the generated local pressure distribution map and the generated seepage diffusion distribution map, the spatial location indices of the two distribution maps are overlaid point by point in a unified 21*21 spatial coordinate system. This coordinate system covers a physical area of ​​20mm*20mm with a resolution of 1mm. First, the coordinate set of the pressure concentration area determined by S201 is extracted from the local pressure distribution map. The coordinate set contains all coordinate points with pressure values ​​exceeding 4.05kPa. At the same time, all coordinate points marked as wet or saturated states are extracted from the seepage diffusion distribution map to form the coordinate set of the seepage influence area. Subsequently, the intersection operation of these two coordinate sets is performed. Specifically, each coordinate point in the pressure concentration area coordinate set is traversed, and it is determined whether it exists in the seepage influence area coordinate set. If a coordinate point satisfies both conditions, it is determined to be a coincident point, and its coordinate value is recorded. For example, the pressure value of coordinate point (9, 11) in the local pressure distribution map is 4.12kPa, which exceeds the threshold of 4.05kPa, and belongs to the pressure concentration area; at the same time, in the seepage diffusion distribution map, the state of this point is saturated. Therefore, coordinate point (9, 11) is recorded. Conversely, coordinate point (8, 11) has a pressure value of 4.08 kPa, which belongs to the pressure concentration zone, but its state is dry in the seepage diffusion distribution map, so this point is not recorded. After traversing and judging all coordinate points in the pressure concentration zone, the set of coordinates of all recorded coincident points constitutes the spatial superimposed coordinate set.

[0032] S302: Call the spatial overlay coordinate set, compare the direction vectors of adjacent points one by one. If the vector angle does not reach the direction similarity threshold, mark them as the same direction. If the vector angle exceeds the direction similarity threshold, mark them as opposite directions, and obtain the direction relationship marking sequence. The generated spatial overlay coordinate set is invoked. For each pair of adjacent points in the coordinate set, the direction vector is calculated and compared. The direction vector originates from two physical fields: the pressure gradient field and the exudate concentration gradient field. The pressure gradient vector, calculated in S102, points in the direction of the fastest increase in pressure. The calculation process for the exudate concentration gradient vector is similar to that of the pressure gradient. It is obtained by calculating the impedance difference between adjacent bioelectrical impedance sensors and dividing it by the sensor spacing (10 mm). The direction points in the direction of the fastest increase in impedance (i.e., the fastest decrease in exudate concentration). The direction similarity threshold is set by applying pressure and fluid sources at different angles to the in vitro wound model. When the angle between the pressure application direction and the main direction of fluid diffusion is less than 45 degrees, the two show a synergistic effect; when the angle is greater than 45 degrees, there is a significant inhibitory or reversing effect. Based on this experimental data, the direction similarity threshold is set to 45 degrees. For each point in the spatial overlay coordinate set, the corresponding pressure gradient vector and exudate concentration gradient vector are obtained, and the angle between the two vectors is calculated. Table 2 shows the vector and angle calculation results for some overlapping points.

[0033] Table 2: Calculation Table of Direction Vector Relationship of Spatial Coincident Points coordinate Pressure gradient vector Permeation gradient vector Vector Angle Directional relationship labeling (9，11) (0.02，0.08) (0.01，0.05) 3.6 same direction (10，11) (0.01，0.09) (-0.02，0.04) 71.6 relative direction (9，12) (0.02，0.07) (0.01，0.06) 2.9 same direction As shown in Table 2, taking coordinate point (9, 11) as an example, its pressure gradient vector is (0.02, 0.08), and its seepage gradient vector is (0.01, 0.05). By calculating the dot product of the two vectors and dividing by the product of their respective moduli, the cosine of the angle between them is obtained. Then, the inverse cosine function is calculated, and the angle between the two vectors is found to be 3.6 degrees. This does not reach the 45-degree directional similarity threshold, so the directional relationship of this point is marked as being in the same direction. For coordinate point (10, 11), the calculated angle between its pressure and seepage gradient vectors is 71.6 degrees, exceeding the 45-degree threshold, so it is marked as being in a relative direction. After completing this calculation and labeling for all points in the spatially superimposed coordinate set, a directional relationship labeling sequence is obtained.

[0034] S303: Based on the directional relationship marker sequence, retrieve the same-direction marker regions and the opposite-direction marker regions, aggregate the same-direction regions into cooperative interaction regions, aggregate the opposite-direction regions into adversarial interaction regions, and generate the interaction region division results within the overall coordinate range; Based on the generated directional relationship marker sequence, the set of coordinate points marked as being in the same direction and the set of coordinate points marked as being in opposite directions are retrieved from the sequence. First, spatial aggregation is performed on the set of coordinate points marked in the same direction. A coordinate point is randomly selected from the set as the starting point, and all other coordinate points marked in the same direction within 1.5 mm of this point are retrieved and grouped into a cluster. Then, starting from newly added points in this cluster, this retrieval process is repeated until no new coordinate points can be added to the current cluster. After constructing a cluster, if there are still unassigned points in the set of coordinate points marked in the same direction, another unassigned point is selected as the new starting point, and the above aggregation process is repeated until all coordinate points in the same direction are assigned to a cluster. Each such cluster of coordinate points is defined as a cooperative interaction zone. Subsequently, the exact same spatial aggregation method is used to process all sets of coordinate points marked in opposite directions. Each cluster formed by aggregating points in opposite directions is defined as an adversarial interaction zone. For example, coordinate points (9, 11) and (9, 12) are both marked as being in the same direction, and their spatial distance is 1mm, which is less than the aggregation distance of 1.5mm. Therefore, they are aggregated into the same cooperative interaction zone. Coordinate point (10, 11), however, is marked as being in a relative direction, and it will aggregate with other nearby points in opposite directions to form an adversarial interaction zone. Within the overall 21*21 coordinate range, the specific coordinate ranges of all cooperative and adversarial interaction zones are output, generating the interaction zone division results.

[0035] Please see Figure 5 The specific steps of S4 are as follows: S401: Based on the results of the interaction zone division, extract the pressure monitoring data and area parameters in multiple interaction zones and normalize them. Then, perform differential calculation on the normalized pressure monitoring data in each interaction zone, and then perform interval averaging on the differential calculation results to generate an average pressure gradient sequence. Based on the generated interaction zone division results, pressure monitoring data and area parameters are extracted from the collaborative interaction zone (Zone A) and the adversarial interaction zone (Zone B). Taking T=5s as an example, Zone A contains 15 coordinate points with an area of ​​15 square millimeters; Zone B contains 12 coordinate points with an area of ​​12 square millimeters. First, the extracted pressure data and area parameters are normalized. The normalization interval is set as follows: Pressure data normalization is based on the pressure value range of the entire monitoring area (441 nodes), with a minimum value of 3.10 kPa and a maximum value of 4.18 kPa. The normalization calculation for any pressure value P is (P−3.10)÷(4.18−3.10). Area parameter normalization is based on the total area of ​​the entire monitoring area of ​​400 square millimeters. The normalization calculation for the area S of any interaction zone is S÷400. For example, the pressure at a point in zone A is 4.12 kPa, and its normalized value is (4.12 − 3.10) ÷ 1.08 = 0.9444. The normalized area of ​​zone A is 15 ÷ 400 = 0.0375. Subsequently, within each interaction zone, differential calculation is performed on the normalized pressure monitoring data. This calculation is performed between adjacent coordinate points, calculating the difference in normalized pressure values ​​and dividing by the physical distance (1 mm) between the two points. For example, the normalized pressure values ​​of two adjacent points (9,11) and (9,12) in zone A are 0.944 and 0.926, respectively, and the differential calculation result is |0.944 − 0.926| ÷ 1 = 0.018. This differential calculation is performed on all adjacent point pairs within the interaction zone. Then, an interval average is calculated on all differential calculation results within an interaction zone, i.e., the arithmetic mean of all differential results is obtained. If 20 difference results are calculated within region A, with a total sum of 0.620, then the interval average for region A is 0.620 ÷ 20 = 0.031. The same calculation process is performed on region B to obtain the interval average, for example, 0.055. The calculation results for each interaction region at continuous time points (e.g., per second) are arranged in chronological order to generate the average pressure gradient sequence for each region.

[0036] S402: Call the average pressure gradient sequence, extract the seepage volume data and diffusion radius parameters in the same region, calculate the seepage diffusion rate distribution value and perform weighting to generate the seepage diffusion rate sequence. The generated average pressure gradient sequence was used, and the exudate volume data and diffusion radius parameters of the same interaction zone (taking the collaborative interaction zone A as an example) at the corresponding time point (T=5s) were extracted. The exudate volume data was obtained by converting the impedance value acquired by S202. The conversion relationship was established based on a calibration experiment: a known volume (0 to 5 μL) of physiological saline was added to a dressing equipped with a bioelectrical impedance sensor, and the corresponding stable impedance readings were recorded. The experimental data showed that the impedance value and the exudate volume were approximately linearly negatively correlated, with the relationship being: exudate volume (μL) = (1600 - impedance value) ÷ 200. At T=5s, the average impedance value corresponding to the 15 coordinate points in zone A was 1250 ohms. Substituting this into the relationship, the average exudate volume of zone A was calculated to be (1600−1250)÷200 = 1.75 μL. The diffusion radius parameter is calculated based on the area of ​​region A, which is 15 square millimeters. Approximating region A as a circle, its radius is approximately 15 ÷ π ≈ 2.18. Subsequently, the exudate diffusion rate distribution is calculated. The calculation requires the diffusion radius at the previous time point (T=4s). If the diffusion radius of region A is 2.10 mm at T=4s, then the exudate diffusion rate at T=5s is 2.18 − 2.10 = 0.08. This rate value is then weighted. The weighting coefficient is set in relation to the current exudate volume of the region, specifically: weight = 1 + (current volume ÷ threshold volume). The threshold volume is set to 5 microliters based on the typical exudate volume requiring dressing replacement in clinical data. Therefore, the weighting coefficient for region A is 1 + (1.75 ÷ 5) = 1.35. The weighted exudate diffusion rate is 0.08 × 1.35 = 0.108 mm / s. This calculation is repeated for all interactive regions at consecutive time points to generate an exudate diffusion rate sequence.

[0037] S403: Based on the average pressure gradient sequence and the exudate diffusion rate sequence, the ratio of the two values ​​in the corresponding interaction zone is calculated, and the ratio results are classified into risk levels. Then, nursing instruction items are matched under multiple risk level results to generate a nursing instruction set. Based on the generated average pressure gradient sequence and the generated exudate diffusion rate sequence, the ratio of their values ​​in the corresponding interaction zone (taking the synergistic interaction zone A at T=5s as an example) is calculated. The average pressure gradient value is divided by the exudate diffusion rate value. Substituting the calculation result from the previous step, the ratio in zone A is 0.031÷0.108≈0.287. The ratio result is then input into the preset wound healing risk level classification rules for risk level classification. The classification rules are based on statistical analysis of historical pressure and exudate monitoring data from 500 burn patients, as shown in Table 3.

[0038] Table 3: Classification of Wound Healing Risk Levels Level 1 (Low Risk) Less than 0.20 Exudate diffuses smoothly with minimal pressure impact. Level 2 (Medium Risk) 0.20 to 0.80 Permeation diffusion is moderately affected by pressure. Level 3 (High Risk) Greater than 0.80 Permeation diffusion is severely hampered by pressure. As shown in Table 3, the calculated ratio for area A is 0.287, falling within the range of 0.20 to 0.80. This result indicates that the healing risk in area A is classified as level 2 (medium risk). Subsequently, pre-defined nursing instructions were matched based on the multi-risk level results. This matching relationship was established according to clinical nursing guidelines. For example: level 1 risk matches continued standard monitoring; level 2 risk matches notification to the patient to make minor adjustments to their position and adjust the monitoring frequency to once every 2 hours; level 3 risk matches issuing an advanced alert, indicating the need for immediate positional intervention and assessing whether dressing changes are necessary. Based on the level 2 risk classification result for area A, the system matches the corresponding nursing instructions. After completing the risk classification and instruction matching for all interactive areas, all instructions and their corresponding area information are combined to generate the final nursing instruction set: Instruction Item 1: Target Area: Collaborative Interaction Area A, Risk Level: Level 2 (Medium Risk), Corresponding Data: Pressure Gradient / Exudate Rate Ratio = 0.287, Nursing Instruction: Notify the patient to make minor adjustments to their position and adjust the monitoring frequency to once every 2 hours. Instruction Item 2: Target Area: Anti-interaction Zone B Risk Level: Level 3 (High Risk) Related Data: Pressure Gradient / Exudate Rate Ratio = 2.619, Nursing Instruction: Issue an advanced alert indicating the need for immediate postural intervention and assess whether dressing change is necessary.

[0039] Please see Figure 6 The specific steps of S5 are as follows: S501: Based on the nursing instruction set, extract the execution time and action sequence of multiple nursing instructions, convert the time parameters into an equally spaced index sequence, and then rearrange the action sequence labels according to the index sequence to generate a nursing time sequence; Based on the generated nursing instruction set, high-level alerts are extracted to indicate the need for immediate postural intervention, assess the need for dressing change, and notify the patient to perform postural fine-tuning. The monitoring frequency is adjusted to once every 2 hours. The execution time and sequence of these two instructions are then determined. First, the actions in the instructions are digitized and mapped. This mapping is defined in a pre-defined nursing action coding table. For example, immediate postural intervention is mapped to label 101, assessing the need for dressing change is mapped to label 102, performing postural fine-tuning is mapped to label 201, and performing routine monitoring is mapped to label 202. The execution sequence of the actions is determined based on the risk level of the instruction. High-risk (level 3) instructions have higher priority than medium-risk (level 2) instructions. Therefore, at the initial time (T=0), the execution sequence is determined to include labels 101 and 102. Subsequently, the time parameters are converted into an equally spaced index sequence. A 30-minute time interval is set to divide the nursing cycle for the next 12 hours, generating an index sequence from 0 to 23, where index 0 represents the current moment, index 1 represents 30 minutes later, and so on. Next, the action sequence labels are rearranged according to this index sequence. The instruction generated at the initial moment (T=0) is placed at index 0. The instruction adjusting the monitoring frequency to once every 2 hours is a periodic action, labeled 202; therefore, this label is placed at the corresponding time index: 2 hours (index 4), 4 hours (index 8), 6 hours (index 12), 8 hours (index 16), and 10 hours (index 20). For index positions with no assigned nursing actions, the action list is empty. After completing the above steps, a nursing time sequence spanning 12 hours in 30-minute units is generated. This sequence contains action labels 101 and 102 at index 0, action label 202 at indices 4, 8, 12, 16, and 20, and the action lists at the remaining indices are empty.

[0040] S502: Call the nursing time series, retrieve the pressure monitoring data and effusion monitoring data in the interaction area and aggregate them by time point, establish the data residual distribution based on the difference between the monitoring data and the index time point, and perform offset determination to obtain the offset determination result; The generated nursing time series was retrieved, and for the high-risk area—the confrontational interaction zone B—pressure and effusion monitoring data were retrieved after the execution of nursing action 101 (immediate postural intervention). The data retrieval time range was 2 hours after the instruction was executed, with a sampling frequency of once every 30 minutes, corresponding to indices 0 to 4 of the nursing time series. The pressure monitoring data (unit: kPa) and effusion monitoring data (volume converted from electrical impedance, unit: μL) at each sampling time point were aggregated. Simultaneously, an expected response model based on historical successful intervention data was established internally. This model predicts that after successful postural intervention, the pressure in the high-pressure area should decrease linearly by 15% within 2 hours, and the effusion volume should stop increasing or decrease slightly. Based on this model, a set of expected monitoring data was generated. Subsequently, a data residual distribution was established based on the difference between the actual monitoring data and the expected monitoring data. Table 4 shows the monitoring data and residual calculation results of the confrontational interaction zone B within 2 hours after intervention.

[0041] Table 4: Post-intervention monitoring data and residual distribution Time Index Actual pressure Expected pressure Pressure residual Actual seepage Expected seepage Permeation Residual 0 4.15 4.15 0.00 2.10 2.10 0.00 1 4.10 3.99 0.11 2.15 2.09 0.06 2 4.05 3.84 0.21 2.22 2.08 0.14 3 4.02 3.68 0.34 2.28 2.07 0.21 4 3.98 3.53 0.45 2.35 2.06 0.29 As shown in Table 4, the pressure residual and seepage residual data were input into a pre-defined statistical test procedure for offset determination. This procedure first categorizes each set of residual data into three classes based on their sign and magnitude: significantly negative deviation (less than -0.1), normal range (-0.1 to 0.1), and significantly positive deviation (greater than 0.1). Then, the frequency of observations falling into each class is counted. For the pressure residual, all five data points fall into the significantly positive deviation class. This observed frequency is compared with the expected frequency (which is expected to be within the normal range) to calculate an offset statistic. This statistic is calculated as follows: for each class, the square of (observed frequency - expected frequency) is calculated, then divided by the expected frequency, and finally the results for all classes are summed. If the expected frequency is [0.5, 4, 0.5] and the observed frequency is [0, 0, 5], then the calculated offset statistical value is (0-0.5)^2 / 0.5+(0-4)^2 / 4+(5-0.5)^2 / 5=0.5+4+4.05=8.55. This value is the final offset determination result.

[0042] S503: Based on the offset determination result, compare the determined offset with the nursing instruction offset threshold. When the offset exceeds the nursing instruction offset threshold, rearrange the execution order of the nursing instructions to generate a burn nursing adjustment instruction set. Based on the generated offset judgment result of 8.55, the determined offset is compared with a preset nursing instruction offset threshold. The threshold setting process is as follows: By retrospectively analyzing the residual data of 1000 historical successful nursing intervention cases, and calculating the offset statistics for each case, a distribution containing 1000 statistical values ​​is formed. The maximum fluctuation range of the offset statistics is extracted, and the mean and standard deviation are selected for weighted calculation. After calculation, the threshold is determined to be 5.99. The threshold represents a relatively high level of offset caused by random fluctuations when the nursing intervention is effective. The current offset of 8.55 calculated by S502 is compared with the threshold of 5.99. Since 8.55 is greater than 5.99, it is determined that the execution effect of the current nursing instruction has significantly deviated, that is, the initial positional intervention measures have failed to achieve the expected effect. When the determined offset exceeds the nursing instruction offset threshold, the system rearranges the execution order of the nursing instructions. The specific rearrangement logic is: the nursing priority of the area where the current offset occurs (adversarial interaction area B) is raised to the highest, and a higher-level intervention instruction is generated. The priority of the routine monitoring instruction (tag 202) originally scheduled for later execution in collaborative interaction zone A was reduced. The system generated a new action tag 103, representing a request for manual assessment and intervention by healthcare professionals, and inserted it at the very beginning of the nursing timeline, requiring immediate execution. The original periodic monitoring task was suspended. Finally, a burn care adjustment instruction set was generated, containing the following: Emergency Adjustment Instruction: Nursing measures in counter-interaction zone B are ineffective, and the risk level remains high. Current offset 8.55, exceeding the threshold of 5.99. Immediate Execution Instruction 103: Request healthcare professionals to perform manual assessment and intervention in zone B. Suspend all other routine nursing tasks until this instruction is completed.

[0043] The above description is merely a specific 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 technical scope 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 burn care method integrating dynamic pressure monitoring and intelligent exudate management, characterized in that, Includes the following steps: S1: Embed a miniature pressure sensor array in the burn dressing to collect pressure data of the wound and surrounding tissues, perform noise reduction and smoothing processing, construct a pressure gradient, input it into the least squares method for gradient fitting, and generate a local pressure distribution map. The specific steps of S1 are as follows: S101: Embed a miniature pressure sensor array in the burn dressing to collect pressure data of the wound and surrounding tissues and arrange them according to the time series. Remove and smooth the noise and abnormal points in the series to generate a pressure data series. S102: Based on the pressure data sequence, calculate the pressure difference between adjacent sensing nodes in the pressure data sequence, and vectorize the difference result on a two-dimensional plane to obtain the pressure gradient field. S103: Call the pressure gradient field, perform least squares fitting on the gradient value and position in the spatial coordinate system, and map the local pressure intensity and distribution range according to the fitting result to obtain a local pressure distribution map. S2: Based on the local pressure distribution map, analyze the pressure concentration area of ​​the wound, obtain the exudate impedance change signal through the bioelectrical impedance sensor embedded in the dressing, and convert it into exudate electrical data after signal amplification and filtering. Then, combine the pressure concentration area for clustering to generate an exudate diffusion distribution map. The specific steps of S2 are as follows: S201: Based on the local pressure distribution map, retrieve the regions in the distribution map where the pressure intensity value exceeds the preset pressure threshold, and aggregate them at the boundary points within the two-dimensional coordinate range. Mark the coordinate set within the aggregated range as the pressure concentration area to obtain the coordinate set of the pressure concentration area. S202: Based on the coordinate set of the pressure concentration area, obtain the exudate impedance change signal collected by the bioelectrical impedance sensor embedded in the dressing, amplify it, remove high-frequency noise components through a filter, and map the filtered impedance change value in time order to obtain the exudate electrical data sequence. S203: Call the coordinate set of the pressure concentration area and the seepage electrical data sequence, perform correlation calculation between the seepage electrical data value and the corresponding pressure area under a unified index space, and perform clustering of the seepage data in the same area, and perform spatial diffusion mapping of the clustering results to generate a seepage diffusion distribution map. The pressure threshold is set statistically by analyzing the pressure intensity values ​​of multiple nodes in the local pressure distribution map. S3: Based on the spatial superposition of the local pressure distribution map and the seepage diffusion distribution map, analyze the directional relationship and cross-region characteristics of the two, and divide them into cooperative interaction zone and antagonistic interaction zone, generating interaction zone division results; The specific steps for S3 are as follows: S301: Based on the local pressure distribution map and the seepage diffusion distribution map, the spatial position indices of the two distribution maps are superimposed point by point under a unified spatial coordinate system, and the coordinate values ​​of the overlapping points are recorded to generate a spatial superimposed coordinate set. S302: Call the spatial superimposed coordinate set, compare the direction vectors of adjacent points one by one. If the vector angle does not reach the direction similarity threshold, mark them as the same direction. If the vector angle exceeds the direction similarity threshold, mark them as opposite directions, and obtain the direction relationship marking sequence. S303: Based on the directional relationship marking sequence, retrieve the same-direction marking regions and the opposite-direction marking regions, aggregate the same-direction regions into cooperative interaction regions, aggregate the opposite-direction regions into adversarial interaction regions, and generate the interaction region division results within the overall coordinate range; S4: Based on the results of the interaction zone division, calculate the average pressure gradient and exudate diffusion rate in multiple interaction zones and analyze the ratio, assess the wound healing risk level of the interaction zone, match the corresponding nursing instructions, and generate a nursing instruction set.

2. The burn care method integrating dynamic pressure monitoring and intelligent exudate management according to claim 1, characterized in that, The local pressure distribution map includes a pressure gradient field, pressure gradient direction, and local abnormal pressure points; the exudate diffusion distribution map includes exudate concentration distribution, diffusion radius parameters, and exudate electrical data; the interaction zone division results include collaborative interaction zones and antagonistic interaction zones; and the nursing instruction set includes wound decompression nursing instructions, exudate removal nursing instructions, and dressing change nursing instructions.

3. The burn care method integrating dynamic pressure monitoring and intelligent exudate management according to claim 1, characterized in that, The directional similarity threshold is set by calculating the angle between all adjacent directional vectors, arranging them in order of magnitude, and then taking the median of the arrangement results as the representative value of the directional distribution.

4. The burn care method integrating dynamic pressure monitoring and intelligent exudate management according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Based on the interaction zone division results, extract the pressure monitoring data and area parameters in the multiple interaction zones and normalize them. Then, perform differential calculation on the normalized pressure monitoring data in each interaction zone, and then perform interval averaging on the differential calculation results to generate an average pressure gradient sequence. S402: Call the average pressure gradient sequence, extract the seepage volume data and diffusion radius parameters in the same area, calculate the seepage diffusion rate distribution value and perform weighting to generate a seepage diffusion rate sequence; S403: Based on the average pressure gradient sequence and the exudate diffusion rate sequence, calculate the ratio of their values ​​in the corresponding interaction zone, classify the risk level of the ratio result, and then match nursing instruction items under multiple risk level results to generate a nursing instruction set.

5. The burn care method integrating dynamic pressure monitoring and intelligent exudate management according to claim 1, characterized in that, The method further includes: S5: The nursing instruction set is converted into a nursing time series, and the pressure monitoring data and exudation monitoring data of the corresponding interaction area are extracted. The offset is determined by establishing the data residual distribution. When the offset exceeds the nursing instruction offset threshold, the execution order of the nursing instructions is adjusted to generate a burn nursing adjustment instruction set. The burn care adjustment instruction set includes the adjusted execution order, care time sequence, and offset determination.

6. The burn care method integrating dynamic pressure monitoring and intelligent exudate management according to claim 5, characterized in that, The specific steps of S5 are as follows: S501: Based on the nursing instruction set, extract the execution time and action sequence of multiple nursing instructions, convert the time parameters into an equally spaced index sequence, and then rearrange the action sequence labels according to the index sequence to generate a nursing time sequence. S502: Call the nursing time series, retrieve the pressure monitoring data and exudation monitoring data in the interaction area and aggregate them according to time points, establish the data residual distribution based on the difference results between the monitoring data and the index time points, and perform offset determination to obtain the offset determination result; S503: Based on the offset determination result, the determined offset is compared with the nursing instruction offset threshold. When the offset exceeds the nursing instruction offset threshold, the execution order of the nursing instructions is rearranged to generate a burn nursing adjustment instruction set.

7. The burn care method integrating dynamic pressure monitoring and intelligent exudate management according to claim 6, characterized in that, The nursing instruction offset threshold is set by statistically retrieving the joint distribution of pressure monitoring data and effusion monitoring data in the nursing instruction execution data, extracting the maximum fluctuation range, and then weighting the mean and standard deviation of the maximum fluctuation range.

Images