Multi-dimensional data processing method and system for intelligent scoring of emergency patients

Abstract

The embodiment of the invention provides a multi-dimensional data processing method and system for intelligent scoring of emergency patients. The method comprises the following steps: acquiring dynamic temperature data and humidity fluctuation data of a burn wound surface through a flexible wearable monitoring device; inputting the time sequence change of the dynamic temperature data into a temperature-drug release model to generate drug concentration distribution; constructing an infection diffusion map taking the anatomical structure as a topological reference, wherein the infection diffusion map comprises an edge attribute representing the flow direction of tissue fluid and a node attribute representing a necrotic region; combining the metabolic rate parameter in the node attribute and the fluid resistance parameter in the edge attribute, iteratively updating the infection propagation intensity, and generating a weight distribution diagram; matching burn depth grading standards according to geometric deformation characteristics and gradient attenuation trends of the weight distribution map, and generating infection risk dynamic scores integrated with drug concentration distribution feedback. According to the technical scheme provided by the embodiment of the invention, the multi-dimensional data processing efficiency and precision can be improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of multi-dimensional data processing, and in particular, to a multi-dimensional data processing method and system for intelligent scoring of emergency patients. Background Art

[0002] During the burn treatment process, accurately monitoring the dynamic changes of the wound surface is crucial for preventing infection and promoting healing. Especially in the case of severe burns, parameters such as the temperature and humidity of the wound surface environment will directly affect the wound recovery process and the effectiveness of drugs. Therefore, a technical solution that can monitor these parameters in real time and accurately, and adjust the drug release strategy according to the monitoring results is needed to adapt to the specific conditions of different patients and their wound surfaces, and achieve personalized treatment.

[0003] Currently, there are already some technical means for monitoring the state of burn wound surfaces on the market, including using traditional dressing materials combined with regular manual inspections, and some intelligent dressings that initially attempt to integrate sensors. These methods can provide basic information about the wound surface to a certain extent, such as temperature and humidity. There are also some studies that have explored intelligent dressings made of thermosensitive materials, aiming to automatically adjust the drug release rate according to the temperature change of the wound surface. Although this progress provides a new idea for personalized treatment, the existing technologies still have limitations in how to accurately control the drug release amount and distribution, and at the same time lack a systematic framework to integrate multiple monitoring parameters to comprehensively evaluate the infection risk and guide subsequent treatment strategies. These existing technologies and methods form the basis of the current burn wound surface monitoring and treatment, but there is still room for further development.

[0004] The existing solutions mainly have the following deficiencies: First, the method of combining traditional dressings with manual inspections is simple and easy to implement, but it mainly relies on the experience judgment of medical staff and lacks the ability to continuously monitor the wound surface environment parameters, resulting in the inability to detect early infection signs or evaluate the healing progress in a timely manner; Second, although the intelligent dressings that initially integrate sensor technology can provide basic physiological parameters in real time, they are weak in data analysis and utilization. These devices are often limited to data collection and do not fully utilize the acquired information for in-depth processing, such as constructing an infection diffusion model or predicting the drug release effect, which limits their application potential in personalized medicine; Finally, although the technology of using thermosensitive materials to adjust the drug release rate shows a new treatment direction, it still faces challenges in accurately controlling the drug release amount and distribution, which limits the maximization of the treatment effect. Therefore, it is particularly necessary to develop a new type of flexible wearable monitoring device and its supporting analysis system. Summary of the Invention

[0005] The embodiments of the present application provide a multi-dimensional data processing method and system for intelligent scoring of emergency patients, aiming to solve the problems of low efficiency and poor accuracy in multi-dimensional data processing in the prior art.

[0006] In a first aspect, the embodiments of the present application provide a multi-dimensional data processing method for intelligent scoring of emergency patients, including:

[0007] Obtain the dynamic temperature data and humidity fluctuation data of the burn wound through a flexible wearable monitoring device, and the flexible wearable monitoring device is integrated into an intelligent dressing layer with a thermal response spraying structure;

[0008] Generate a temperature-drug release model according to the thermosensitive parameters of the intelligent dressing layer, input the time series change of the dynamic temperature data into the temperature-drug release model, and generate a drug concentration distribution adapted to the morphology of the burn wound;

[0009] Based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, construct an infection diffusion map with the anatomical structure as the topological reference, and the infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic areas;

[0010] Input the infection diffusion map into a multi-layer topological structure, combine the metabolic rate parameters in the node attributes with the fluid resistance parameters in the edge attributes, and iteratively update the infection propagation intensity to generate a weight distribution map including the infection core area and the diffusion boundary;

[0011] Match the geometric deformation characteristics and gradient attenuation trend of the weight distribution map with the burn depth grading standard to generate an infection risk dynamic score integrating the feedback of the drug concentration distribution.

[0012] Optionally, the based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, construct an infection diffusion map with the anatomical structure as the topological reference, and the infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic areas, including:

[0013] Couple the gradient change direction between adjacent regions within the anatomical unit of the spatial gradient of the humidity fluctuation data with the drug penetration range of the corresponding anatomical unit in the drug concentration distribution to generate a superposition effect parameter;

[0014] Define the nodes and edges in the topological reference according to the spatial segmentation relationship of the anatomical structure, bind the nodes to the spatial coordinates of the anatomical unit, and bind the edges to the adjacent connection paths between the anatomical units to generate reference topological parameters;

[0015] Based on the superposition effect parameter and the reference topology parameter, correct the interstitial fluid flow direction attribute of the edge through the vector direction of the spatial gradient of the humidity fluctuation data, and at the same time define the necrosis region attribute of the node according to the coincidence degree between the drug-uncovered region in the drug concentration distribution and the region with abnormal decrease in the spatial gradient of the humidity fluctuation data, and generate the edge attribute parameter and the node attribute parameter;

[0016] Fuse the reference topology parameter, the edge attribute parameter and the node attribute parameter into the same topology mapping framework to generate an infection diffusion map including the dynamic path of the complete interstitial fluid and the evolution relationship of the necrosis region.

[0017] Optionally, the step of, based on the superposition effect parameter and the reference topology parameter, correcting the interstitial fluid flow direction attribute of the edge through the vector direction of the spatial gradient of the humidity fluctuation data, and at the same time defining the necrosis region attribute of the node according to the coincidence degree between the drug-uncovered region in the drug concentration distribution and the region with abnormal decrease in the spatial gradient of the humidity fluctuation data, and generating the edge attribute parameter and the node attribute parameter, includes:

[0018] Based on the superposition effect parameter and the reference topology parameter, decompose the spatial gradient of the humidity fluctuation data into a parallel component and an orthogonal component, and dynamically couple the orthogonal component with the reference topology parameter to generate the correction weight factor of the edge;

[0019] Perform a direction field superposition on the correction weight factor and the vector direction of the spatial gradient of the humidity fluctuation data to perform step-by-step direction calibration on the interstitial fluid flow direction attribute of the edge, and generate a corrected interstitial fluid flow direction attribute sequence;

[0020] Segment the low-density region of the drug concentration distribution, extract the geometric boundary of the drug-uncovered region, and at the same time, based on the rate of change of the spatial gradient amplitude of the humidity fluctuation data, detect the geometric boundary of the region with abnormal decrease in the gradient, and perform a spatial topology superposition on the geometric boundaries of the drug-uncovered and gradient-abnormally decreasing regions, and define the overlapping region as the necrosis region attribute of the node;

[0021] Jointly encode the corrected interstitial fluid flow direction attribute sequence and the edge connection weight in the reference topology parameter to generate the edge attribute parameter, and at the same time perform a dynamic attenuation operation on the necrosis region attribute and the initial node health in the reference topology parameter to generate the node attribute parameter.

[0022] Optionally, perform low-density region segmentation on the drug concentration distribution, extract the geometric boundary of the drug-uncovered region, and at the same time, based on the rate of change of the spatial gradient amplitude of the humidity fluctuation data, detect the geometric boundary of the region with abnormal gradient decline. Perform spatial topological superposition on the geometric boundaries of the drug-uncovered and gradient-abnormally declining regions, and define the overlapping region as the necrosis region attribute of the node, including:

[0023] Perform low-density region segmentation on the drug concentration distribution, extract the geometric boundary of the drug-uncovered region, and store the geometric boundary as the first boundary set;

[0024] Perform temporal cumulative calculation on the rate of change of the spatial gradient amplitude of the humidity fluctuation data to generate the geometric boundary of the region with abnormal gradient decline, and store the geometric boundary as the second boundary set;

[0025] Input the first boundary set and the second boundary set into the spatial topological superposition module, perform a Boolean intersection operation on the drug-uncovered region and the gradient-abnormally declining region, output the polygon vertex coordinate set of the overlapping region, and construct a spatial distribution vector map of the overlapping region based on the polygon vertex coordinate set;

[0026] Calculate the real-time ratio of the area of the overlapping region to the area of the drug-uncovered region according to the spatial distribution vector map, and extract the number of contour curvature extreme points as the morphological complexity index;

[0027] When the morphological complexity index exceeds the preset complexity threshold, mark the overlapping region as the necrosis region attribute of the node, and associate the necrosis region attribute with the node spatial topology database.

[0028] Optionally, input the infection spread map into a multi-layer topological structure, combine the metabolic rate parameter in the node attribute and the fluid resistance parameter in the edge attribute, and iteratively update the infection propagation intensity to generate a weight distribution map including the infection core area and the diffusion boundary, including:

[0029] Based on the multi-layer topological structure, initialize the node attributes and edge attributes of the infection spread map by level, associate the metabolic rate parameter with the activation state of each level of nodes, and embed the fluid resistance parameter into the edge connection strength between levels to generate the initial weight distribution of the multi-layer topological structure;

[0030] Based on the edge connection strength, perform heterologous interference compensation on the metabolic rate parameters of adjacent nodes to generate the metabolic interference value of each node, and calculate the incremental component of the infection propagation intensity to iteratively update the infection propagation intensity;

[0031] Extract the drug residue attenuation factors of each node according to the diffusion path of the residue amount in the multi-layer topology according to the drug concentration distribution, non-linearly superimpose the drug residue attenuation factors with the gradient change rate of the updated infection transmission intensity to generate a dynamic correction coefficient, and adaptively scale the weight ratio of the metabolic interference value by using the dynamic correction coefficient;

[0032] Perform multi-level cascading on the path integral result of the scaled metabolic interference value and the fluid resistance parameter, and re-divide the boundary permeability of the infection core area to generate a dynamic probability field of the diffusion boundary;

[0033] Based on the initial weight distribution, alternately iterate the incremental component of the infection transmission intensity and the backpropagation compensation of the dynamic probability field to generate a weight distribution map including the infection core area and the diffusion boundary.

[0034] Optionally, the extracting the drug residue attenuation factors of each node according to the diffusion path of the residue amount in the multi-layer topology according to the drug concentration distribution, non-linearly superimposing the drug residue attenuation factors with the gradient change rate of the updated infection transmission intensity to generate a dynamic correction coefficient, and adaptively scaling the weight ratio of the metabolic interference value by using the dynamic correction coefficient includes:

[0035] Traverse the node sequence of the drug diffusion path in the multi-layer topology, perform a convolution operation on the number of path branches of each node and the concentration gradient difference between adjacent layers, and output the numerical sequence of the drug residue attenuation factors of each node;

[0036] Perform a sliding window difference calculation on the spatial gradient amplitude of the updated infection transmission intensity in the time series dimension, perform two-way propagation compensation on the gradient change rate between adjacent nodes, and generate a gradient change rate parameter matrix;

[0037] Input the numerical sequence of the drug residue attenuation factors and the gradient change rate parameter matrix into a non-linear coupling unit, perform a piecewise power-law transformation operation, and generate a dynamic correction coefficient;

[0038] Divide the weight levels of the metabolic interference value according to the numerical distribution range of the dynamic correction coefficient, and adaptively scale the weight ratio of the metabolic interference value according to the grade boundary.

[0039] Optionally, the matching the burn depth grading standard according to the geometric deformation characteristics and gradient attenuation trend of the weight distribution map to generate a dynamic infection risk score integrating the drug concentration distribution feedback includes:

[0040] Perform multi-scale curvature extreme point detection on the geometric deformation characteristics of the weight distribution map, extract the density distribution of curvature extreme points and the change rate of connected domain area at different scales, and construct a weight deformation feature vector;

[0041] Perform a spatial gradient direction field decomposition on the gradient decay trend, calculate the gradient mean change rate and direction consistency coefficient of the anisotropic region, and generate a gradient decay trend parameter matrix by combining the gradient decay acceleration accumulated over time series;

[0042] Input the weight deformation eigenvector and the gradient decay trend parameter matrix into a hierarchical fusion module, and output a deformation-decay coupling coefficient sequence according to the preset skin layer thickness parameters in the burn depth grading standard;

[0043] Construct a three-dimensional probability distribution surface based on the deformation-decay coupling coefficient sequence, and use the local inhibition rate feedback by the drug concentration distribution as a constraint condition to calculate the ratio of the integral value within the truncated region to the standard burn depth volume as the dynamic infection risk scoring base;

[0044] Perform layer-by-layer interpolation calibration according to the dynamic infection risk scoring base and the depth thresholds of the epidermis, dermis, and subcutaneous tissue in the burn depth grading standard, and generate a dynamic infection risk score with the feedback of the drug concentration distribution.

[0045] In a second aspect, an embodiment of the present application provides a multi-dimensional data processing system for intelligent scoring of emergency patients, including:

[0046] An acquisition module that acquires dynamic temperature data and humidity fluctuation data of the burn wound through a flexible wearable monitoring device, and the flexible wearable monitoring device is integrated into an intelligent dressing layer with a thermal response spraying structure;

[0047] An input module that generates a temperature-drug release model according to the thermosensitive parameters of the intelligent dressing layer, and inputs the time series change of the dynamic temperature data into the temperature-drug release model to generate a drug concentration distribution adapted to the morphology of the burn wound;

[0048] A construction module that constructs an infection diffusion map with the anatomical structure as the topological reference based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, and the infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic regions;

[0049] An update module that inputs the infection diffusion map into a multi-layer topological structure, and combines the metabolic rate parameters in the node attributes and the fluid resistance parameters in the edge attributes to iteratively update the infection propagation intensity and generate a weight distribution map including an infection core area and a diffusion boundary;

[0050] An integration module that matches the geometric deformation characteristics and gradient decay trend of the weight distribution map with the burn depth grading standard to generate a dynamic infection risk score integrating the feedback of the drug concentration distribution.

[0051] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a multi-dimensional data processing method for intelligent scoring of emergency patients as described in the first aspect above.

[0052] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which when executed by a computer, implements a multi-dimensional data processing method for intelligent scoring of emergency patients as described in the first aspect.

[0053] In an embodiment of the present application, dynamic temperature data and humidity fluctuation data of a burn wound are obtained through a flexible wearable monitoring device, and the flexible wearable monitoring device is integrated into an intelligent dressing layer with a thermally responsive spraying structure; a temperature-drug release model is generated according to the thermosensitive parameters of the intelligent dressing layer, and the time series change of the dynamic temperature data is input into the temperature-drug release model to generate a drug concentration distribution adapted to the morphology of the burn wound; based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, an infection diffusion map with anatomical structure as the topological reference is constructed, and the infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic areas; the infection diffusion map is input into a multi-layer topological structure, and in combination with the metabolic rate parameter in the node attribute and the fluid resistance parameter in the edge attribute, the infection propagation intensity is iteratively updated to generate a weight distribution map including an infection core area and a diffusion boundary; according to the geometric deformation characteristics and gradient attenuation trend of the weight distribution map, the burn depth grading standard is matched to generate an infection risk dynamic score integrating the feedback of the drug concentration distribution.

[0054] The technical solution of the present application has the following beneficial effects:

[0055] In the present application, the dynamic temperature and humidity data of the burn wound are obtained in real time through a flexible wearable monitoring device, and the drug release is adjusted by using an intelligent dressing layer with a thermally responsive spraying structure. It can not only accurately monitor the wound state, but also adjust the drug concentration distribution according to actual needs to adapt to different wound morphologies. Based on the superposition effect of the humidity fluctuation data and the drug concentration distribution, an infection diffusion map is constructed, and combined with the anatomical structure characteristics, it ensures the visual management of the infection diffusion process. Further, the infection propagation intensity is iteratively updated through a multi-layer topological structure to generate a weight distribution map including an infection core area and a diffusion boundary, providing a scientific basis for personalized treatment. Finally, by matching the geometric deformation characteristics with the burn depth grading standard, an infection risk dynamic score integrating the drug concentration feedback is generated to ensure the efficiency and pertinence of the treatment and promote the rehabilitation process of the patient.

[0056] Furthermore, by coupling the spatial gradient of humidity fluctuation data with the drug concentration distribution, a superposition effect parameter is generated, and the node and edge attributes in the topological benchmark are defined in combination with the spatial segmentation relationship of the anatomical structure, making the identification of tissue fluid flow direction and necrosis area more accurate. This method realizes the accurate tracking and prediction of the infection spread path by correcting the tissue fluid flow direction attribute of the edge and defining the necrosis area attribute of the node. Integrating this information into a unified topological mapping framework not only enhances the integrity and accuracy of the infection spread map, but also provides strong support for the subsequent formulation of treatment strategies. In addition, this method helps to achieve a comprehensive assessment of complex burn wounds, thereby optimizing the treatment plan, reducing the occurrence of complications, and improving the cure rate.

[0057] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0059] Figure 1 The flowchart of a multi-dimensional data processing method for intelligent scoring of emergency patients provided by the present application is shown;

[0060] Figure 2 The structural schematic diagram of a multi-dimensional data processing system for intelligent scoring of emergency patients provided by the present application is shown;

[0061] Figure 3 The structural schematic diagram of a computing device provided by the present application is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] In order to enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application.

[0063] In some of the processes described in the specification, claims, and the above-mentioned drawings of this application, multiple operations appear in a specific order. However, it should be clearly understood that these operations can be executed not in the order in which they appear herein or in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. Additionally, these processes can include more or fewer operations, and these operations can be executed sequentially or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0064] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0065] This project aims to develop a system that can dynamically monitor the status of burn wounds and intelligently adjust drug release, which is achieved by integrating a flexible wearable device with an intelligent dressing layer with thermoresponsive characteristics. This system can not only obtain the temperature and humidity data of burn wounds in real time, but also generate personalized drug treatment plans based on these data and predict the trend of infection spread, providing scientific basis and technical support for clinical practice.

[0066] Figure 1 A flowchart of a multi-dimensional data processing method for intelligent scoring of emergency patients is provided for the embodiments of the present application, as Figure 1 shown, the method includes:

[0067] 101. Obtain the dynamic temperature data and humidity fluctuation data of the burn wound through a flexible wearable monitoring device, and the flexible wearable monitoring device is integrated into an intelligent dressing layer with a thermoresponsive spraying structure;

[0068] In this step, the flexible wearable monitoring device is a device that fits the human skin, is soft and comfortable, and is used to continuously collect physiological parameters such as the temperature and humidity of the burn wound.

[0069] The dynamic temperature data includes the recorded surface temperature of the wound that changes over time, and the humidity fluctuation data reflects the change in the humidity of the environment around the wound.

[0070] The intelligent dressing layer with a thermoresponsive spraying structure is a special material that can adjust its physical or chemical properties according to temperature changes, so as to achieve intelligent regulation of the local environment.

[0071] The intelligent dressing layer with a thermoresponsive spraying structure is a covering made of special materials that can adjust their physical properties (such as expansion or contraction) according to temperature changes, thus achieving intelligent regulation of the local environment, such as adjusting humidity by changing breathability.

[0072] In the embodiments of the present application, first, the intelligent dressing layer is applied to the burn wound surface to ensure close contact with the skin and activate its thermoresponsive function. Then, the flexible wearable monitoring device is used to start collecting temperature and humidity data in real time. These data are transmitted wirelessly to the central processing system for preliminary analysis. During the data analysis process, any abnormal fluctuations need to be identified. For example, an increase in body temperature at night may indicate an exacerbation of the inflammatory response. Finally, based on these analysis results, the function settings of the intelligent dressing can be further optimized, such as adjusting the drug release rate to cope with potential infection risks. This process emphasizes the importance of data collection, transmission, analysis, and feedback regulation.

[0073] In a practical case, a doctor used the above technologies and methods for a severely burned patient. During the treatment process, the monitoring system found that the increase in the patient's body temperature at night led to an increase in the humidity of the wound surface, which might be a signal of an exacerbation of the inflammatory response. To address this situation, the doctor adjusted the drug release rate using the thermoresponsive characteristics of the intelligent dressing and strengthened the night care measures. In addition, through in-depth analysis of the humidity fluctuation data, the team also predicted possible infection risks and took preventive treatments in advance, thus effectively controlling the development of the condition.

[0074] 102. Generate a temperature-drug release model according to the thermosensitive parameters of the intelligent dressing layer, input the temporal variation of the dynamic temperature data into the temperature-drug release model, and generate a drug concentration distribution adapted to the morphology of the burn wound surface;

[0075] In this step, the thermosensitive parameters of the intelligent dressing layer refer to the sensitivity of the material to temperature changes, such as the expansion coefficient or color change, etc.

[0076] The temperature-drug release model is a mathematical model established based on these parameters, which is used to describe how temperature affects the release rate of drugs from the dressing.

[0077] The drug concentration distribution adapted to the morphology of the burn wound surface is a drug release pattern customized according to the specific shape, size, and depth of each patient's wound surface, aiming to ensure that the drug can achieve the best effect where it is most needed.

[0078] In the embodiments of the present application, first, researchers construct a temperature-drug release model based on the thermosensitive parameters of the intelligent dressing layer, which usually involves a series of experiments to determine the optimal drug release rate at different temperatures. Then, the dynamic temperature data obtained in step 101 is input into the model to calculate the drug concentration distribution most suitable for the current situation. During this process, the model parameters need to be continuously adjusted and verified to ensure that the drug release can accurately match the actual situation of the patient. At the same time, individual differences of the patient, such as age, weight and other factors, also need to be considered to develop a personalized treatment plan.

[0079] Based on the data of the previous step, the doctor adjusted the treatment plan, increased the drug release amount at night, and successfully controlled the inflammation. In the following days, as the patient's condition improved, the doctor continued to fine-tune the drug release strategy and gradually reduced the night-time dose. At the same time, the patient's recovery was closely monitored, and the treatment effect was evaluated and corresponding adjustments were made by comparing the temperature and humidity data at different time periods. This method not only improves the treatment efficiency, but also reduces unnecessary drug use and the risk of side effects. For example, in a certain treatment, when it was observed that the patient's body temperature tended to be stable, the doctor promptly reduced the drug dose to avoid the adverse effects caused by over-medication.

[0080] 103. Based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, construct an infection diffusion map with the anatomical structure as the topological reference. The infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic areas.

[0081] In this step, the spatial gradient of the humidity fluctuation data shows the distribution difference of moisture on the wound surface, reflecting the change trend of the humidity of the surrounding environment of the wound surface.

[0082] The superposition effect of the drug concentration distribution considers how the drug release affects these humidity changes.

[0083] The infection diffusion map is a chart that combines anatomical features to depict potential infection paths, including edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic areas.

[0084] The edge attributes describe the flow of tissue fluid between different regions, while the node attributes mark the possible necrotic tissue or high-infection-risk areas.

[0085] In the embodiments of the present application, first, the spatial gradient of the humidity fluctuation data is analyzed to determine the high-humidity regions and their potential causes. Then, it is combined with the drug concentration distribution to construct an infection spread map. In this process, various factors need to be comprehensively considered, such as anatomical features, the direction and speed of tissue fluid flow, etc., to accurately predict the possibility of infection spread. Next, the inter-tissue fluid flow and metabolic activities are simulated to predict the direction and speed of possible infection spread. Finally, based on the generated infection spread map, the medical team can take targeted measures to prevent further spread of the infection.

[0086] During the treatment process, the doctors noticed that the humidity in certain areas was abnormally elevated, indicating a possible infection risk. So, they used the infection spread map generated by the system to identify the areas most likely to be affected and adjusted the local drug concentration accordingly. At the same time, the team also adopted a series of adjuvant treatment methods, such as local debridement and the application of antibacterial agents, to further reduce the infection risk. After some time of effort, the patient's infection was effectively controlled, and the wound healing speed was significantly accelerated. For example, when a relatively high humidity was detected in a specific area, the doctor immediately increased the drug release amount in that area and performed local cleaning, effectively curbing the spread of the infection.

[0087] 104. Input the infection spread map into a multi-layer topological structure, combine the metabolic rate parameter in the node attributes with the fluid resistance parameter in the edge attributes, and iteratively update the infection propagation intensity to generate a weight distribution map including the infection core area and the diffusion boundary;

[0088] In this step, the multi-layer topological structure is a complex network that simulates the human tissue hierarchy and their interactions, including different layers such as the skin, fat, and muscle.

[0089] The metabolic rate parameter measures the speed of cell activity and reflects the tissue health status; the fluid resistance parameter describes the difficulty of liquid flow between tissues and affects the transfer efficiency of nutrients and drugs.

[0090] The weight distribution map shows different regions of the infection propagation intensity, helping medical staff understand the trend of infection spread and formulating corresponding intervention measures accordingly.

[0091] In the embodiments of the present application, first, a multi-layer topological structure is defined, including anatomical features and physiological functions. Then, based on the metabolic rate and fluid resistance parameters, the infection transmission intensity is iteratively updated to form a weight distribution map. This process requires comprehensive consideration of various factors to ensure that the map can accurately reflect the actual situation. During the construction of the weight distribution map, the accuracy of the model needs to be repeatedly verified to ensure that it can truly reflect the actual transmission path of the infection. Finally, based on the generated weight distribution map, the medical team can take targeted measures to inhibit the expansion of the infection core area and protect more healthy tissues.

[0092] Through continuous monitoring and adjustment, the doctor successfully inhibited the expansion of the infection core area and protected more healthy tissues. For example, during a treatment, when a high fluid resistance was detected in a specific area, it indicated that there might be a poor local blood circulation situation here. In response to this problem, the medical team immediately took action and enhanced the blood circulation promotion measures in this area, such as local massage and hot compress, effectively alleviating the symptoms and promoting the healing of the wound surface. In this way, not only was the infection controlled, but also the patient's recovery process was accelerated.

[0093] 105. Match the geometric deformation characteristics and gradient decay trend of the weight distribution map with the burn depth grading standard to generate a dynamic infection risk score integrating the feedback of the drug concentration distribution.

[0094] In this step, the geometric deformation characteristics refer to the changing trend of the burn wound surface shape over time, and the gradient decay trend reflects the reduction speed of the infection severity, that is, whether the infection range shrinks over time.

[0095] The dynamic infection risk score integrates multiple factors such as the geometric deformation of the wound surface, humidity change, and drug concentration distribution, providing a comprehensive risk assessment index to help doctors make more accurate treatment decisions.

[0096] In the embodiments of the present application, first, the geometric deformation characteristics are compared with the standard grading to determine the recovery stage of the current wound surface. Then, the dynamic infection risk score is generated in combination with the gradient decay trend. The whole process emphasizes personalization and dynamic adjustment to ensure that each patient can obtain the most suitable treatment plan for themselves. During the implementation process, the score needs to be updated regularly to adjust the treatment strategy in a timely manner. In addition, it is also necessary to combine with other clinical indicators, such as blood test results, to comprehensively evaluate the patient's health status.

[0097] After a series of precise treatments, the patient's wound healed well, the infection risk decreased significantly, and the rehabilitation process accelerated. For example, in a certain patient, the doctor used the system to continuously monitor the geometric deformation and humidity changes of the wound over several weeks and found that as the treatment progressed, the wound area gradually shrank and the humidity level tended to stabilize. Based on these positive changes, the medical team adjusted the treatment strategy in a timely manner and finally achieved the ideal treatment effect. In the later stage of treatment, as the infection risk score continued to decline, the doctor gradually reduced the drug dosage and instead focused on the natural healing ability of the wound, promoting faster recovery.

[0098] In summary, steps 101 to 105 achieve the precise monitoring of the burn wound status and the formulation of personalized treatment plans. It can not only identify and respond to possible complications in a timely manner, but also effectively promote wound healing, reduce hospital stay and medical costs, and improve the patient's quality of life. The entire process emphasizes the closed-loop management from data collection, analysis to feedback adjustment to ensure that each patient can obtain the best treatment experience. Through real-time monitoring and dynamic adjustment, the effectiveness and safety of the treatment are greatly improved.

[0099] To further improve the precise prediction of the infection spread path during burn treatment and optimize the personalized treatment plan, in some embodiments, in step 103, based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, an infection spread map is constructed with the anatomical structure as the topological reference. The infection spread map includes edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic areas, including:

[0100] 1031. Couple the gradient change direction between adjacent regions within the anatomical unit of the spatial gradient of the humidity fluctuation data with the drug penetration range of the corresponding anatomical unit in the drug concentration distribution to generate a superposition effect parameter;

[0101] In step 1031, the gradient change direction between adjacent regions within the anatomical unit of the spatial gradient of the humidity fluctuation data is coupled with the drug penetration range of the corresponding anatomical unit in the drug concentration distribution to generate a superposition effect parameter. Here, the superposition effect parameter refers to evaluating the changes in the local environment by analyzing the interaction between humidity fluctuation and drug concentration, so as to identify potential infection risk areas.

[0102] In the embodiments of the present application, first, humidity fluctuation data is obtained from a flexible wearable monitoring device and divided into different anatomical units. Each anatomical unit represents a specific local area. Secondly, the spatial gradient of the humidity fluctuation data within these areas is calculated to determine the direction and intensity of the change in humidity value with position. Then, according to the drug concentration distribution map, the drug penetration range within each anatomical unit is identified. Finally, the spatial gradient of the humidity fluctuation data is coupled with the drug concentration distribution to generate a superposition effect parameter. This step requires a complex mathematical model to quantify the interaction between the two, such as using the weighted average method to comprehensively consider the effects of humidity and drugs.

[0103] 1032. Define the nodes and edges in the topological reference according to the spatial segmentation relationship of the anatomical structure. The nodes are bound to the spatial coordinates of the anatomical units, and the edges are bound to the adjacent connection paths between the anatomical units to generate reference topological parameters.

[0104] In step 1032, the nodes and edges in the topological reference are defined according to the spatial segmentation relationship of the anatomical structure, where the nodes are bound to the spatial coordinates of the anatomical units and the edges are bound to the adjacent connection paths between the anatomical units to generate reference topological parameters. The reference topological parameters are used to describe the three-dimensional spatial relationship of the anatomical structure, helping to understand the flow direction of tissue fluid and the distribution of necrotic areas.

[0105] In the embodiments of the present application, first, a personalized three-dimensional anatomical model is established using the scan results of the patient's computed tomography. Each anatomical unit is assigned a unique spatial coordinate as a node. Secondly, based on anatomical knowledge, the connection paths between adjacent anatomical units are defined as edges. These edges not only represent physical connections but also reflect the possibility of tissue fluid flow. Then, through image processing techniques, these nodes and edges are mapped into three-dimensional space to form reference topological parameters. This step requires high-precision imaging techniques and algorithms to ensure the accuracy of the model.

[0106] 1033. Based on the superposition effect parameter and the reference topological parameter, correct the tissue fluid flow direction attribute of the edge through the vector direction of the spatial gradient of the humidity fluctuation data, and at the same time define the necrotic area attribute of the node according to the coincidence degree between the drug-uncovered area in the drug concentration distribution and the abnormally decreasing area of the spatial gradient of the humidity fluctuation data to generate edge attribute parameters and node attribute parameters.

[0107] In step 1033, based on the superposition effect parameters and the reference topology parameters, the tissue fluid flow direction attribute of the edge is corrected through the vector direction of the spatial gradient of the humidity fluctuation data. Meanwhile, according to the coincidence degree between the drug-uncovered area in the drug concentration distribution and the abnormal decline area of the spatial gradient of the humidity fluctuation data, the necrosis area attribute of the node is defined, and the edge attribute parameters and the node attribute parameters are generated. This step aims to combine the humidity fluctuation and drug concentration information to accurately label the potential infection paths and necrosis areas.

[0108] In the embodiment of the present application, first, the superposition effect parameters generated in step 1031 are used to adjust the edge attribute in the reference topology parameter, that is, the tissue fluid flow direction. Specifically, according to the vector direction of the spatial gradient of the humidity fluctuation data, the fluid flow direction of each edge is corrected. Secondly, the uncovered area in the drug concentration distribution is analyzed and compared with the abnormal decline area of the spatial gradient of the humidity fluctuation data. If the two coincide, it is marked as a potential necrosis area. Then, this information is integrated into the three-dimensional anatomical model to generate the edge attribute parameters and the node attribute parameters. Finally, these attributes are displayed through a visualization tool to help doctors better understand the trend of infection spread.

[0109] 1034. Integrate the reference topology parameters, the edge attribute parameters and the node attribute parameters into the same topological mapping framework to generate an infection spread map including the complete dynamic path of tissue fluid and the evolution relationship of necrosis areas.

[0110] In step 1034, the reference topology parameters, the edge attribute parameters and the node attribute parameters are integrated into the same topological mapping framework to generate an infection spread map including the complete dynamic path of tissue fluid and the evolution relationship of necrosis areas. This map provides a comprehensive prediction of the infection spread path to help formulate personalized treatment strategies.

[0111] In the embodiment of the present application, first, all the parameters generated in steps 1032 and 1033 are imported into a unified topological mapping framework. Secondly, a specially developed software tool is used to integrate these parameters together to form a complete three-dimensional infection spread map. This step requires powerful computing capabilities and efficient algorithm support. Then, through a detailed analysis of the map, the key infection spread paths and the evolution trends of necrosis areas are extracted. Finally, based on these analysis results, personalized treatment plans are formulated, such as adjusting the drug release rate, arranging debridement surgery, etc.

[0112] The following is a specific example:

[0113] In a case of a severely burned patient, the doctor first applied the method of step 1031 to obtain the humidity fluctuation data of the patient's wound surface and its surrounding area from the flexible wearable monitoring device. By calculating the spatial gradient, the direction and intensity of the change of humidity value with position were determined, and were coupled with the drug concentration distribution to generate the superposition effect parameters. Next, in step 1032, a personalized three-dimensional anatomical model was established using the patient's CT scan results, the connection paths between anatomical units were defined, and the reference topological parameters were generated. In step 1033, the tissue fluid flow direction was corrected according to the superposition effect parameters, and the potential necrotic areas were marked. Finally, in step 1034, all parameters were integrated into a unified topological mapping framework to generate a detailed infection spread map. Based on this map, the doctor found that in some areas, due to the ineffective coverage of the drug, the humidity remained high, indicating a high risk of infection. Therefore, they timely adjusted the treatment plan, increased the local drug release amount, and performed necessary surgical operations, effectively controlling the development of the infection. As the treatment progressed, the patient's condition was significantly improved, the wound healing speed was accelerated, and the infection risk was greatly reduced.

[0114] In summary, the accurate monitoring of the burn wound state and the formulation of personalized treatment plans were achieved through steps 1031 to 1034. It can not only timely identify and respond to possible complications, but also effectively promote wound healing, reduce the hospital stay and medical costs, and improve the quality of life of the patient. The entire process emphasizes the closed-loop management from data collection, analysis to feedback adjustment to ensure that each patient can obtain the best treatment experience. Through real-time monitoring and dynamic adjustment, the effectiveness and safety of the treatment are greatly improved. This method not only improves the treatment efficiency, but also reduces the unnecessary use of drugs and the risk of side effects.

[0115] In order to further improve the accurate prediction of the infection spread path during the burn treatment process and optimize the personalized treatment plan, in some embodiments, based on the superposition effect parameters and the reference topological parameters in step 1033, the tissue fluid flow direction attribute of the edge is corrected through the vector direction of the spatial gradient of the humidity fluctuation data, and at the same time, the necrotic area attribute of the node is defined according to the coincidence degree between the drug-uncovered area in the drug concentration distribution and the abnormally decreasing area of the spatial gradient of the humidity fluctuation data, generating the edge attribute parameters and the node attribute parameters, including:

[0116] Based on the superposition effect parameter and the reference topology parameter, decompose the spatial gradient of the humidity fluctuation data into a parallel component and an orthogonal component, and dynamically couple the orthogonal component with the reference topology parameter to generate the corrected weight factor of the edge; perform a direction field superposition on the vector direction of the corrected weight factor and the spatial gradient of the humidity fluctuation data to perform step-by-step direction calibration on the interstitial fluid flow direction attribute of the edge, and generate a corrected interstitial fluid flow direction attribute sequence; perform low-density region segmentation on the drug concentration distribution, extract the geometric boundary of the drug-uncovered region, and at the same time, based on the change rate of the spatial gradient amplitude of the humidity fluctuation data, detect the geometric boundary of the gradient abnormally decreasing region, perform a spatial topology superposition on the geometric boundaries of the drug-uncovered and gradient abnormally decreasing regions, and define the overlapping region as the necrosis region attribute of the node; jointly encode the corrected interstitial fluid flow direction attribute sequence and the edge connection weight in the reference topology parameter to generate the edge attribute parameter, and at the same time perform a dynamic attenuation operation on the necrosis region attribute and the initial node health in the reference topology parameter to generate the node attribute parameter.

[0117] In this embodiment, based on the superposition effect parameter and the reference topology parameter, the spatial gradient of the humidity fluctuation data is decomposed into a parallel component and an orthogonal component, and the orthogonal component is dynamically coupled with the reference topology parameter to generate the corrected weight factor of the edge. The corrected weight factor here is used to adjust the direction and intensity of the interstitial fluid flow. The vector direction of the spatial gradient of the humidity fluctuation data is combined with the corrected weight factor through direction field superposition to perform step-by-step calibration on the interstitial fluid flow direction attribute of the edge, thereby generating a corrected interstitial fluid flow direction attribute sequence. For the drug concentration distribution, a low-density region segmentation method is used to extract the geometric boundary of the uncovered region, and at the same time, the geometric boundary of the gradient abnormally decreasing region is detected by using the change rate of the spatial gradient amplitude of the humidity fluctuation data. The geometric boundaries of these two types of regions are topologically superposed in space, and the overlapping part is defined as the necrosis region attribute of the node. Finally, the edge attribute parameter is generated by jointly encoding the corrected interstitial fluid flow direction attribute sequence and the edge connection weight in the reference topology parameter; and the node attribute parameter is generated by combining the necrosis region attribute and the initial node health in the reference topology parameter through dynamic attenuation operation.

[0118] In the embodiments of the present application, first, the parallel component and the orthogonal component of the spatial gradient are calculated from the humidity fluctuation data. Then, these components are dynamically coupled with the anatomical unit adjacency relationship in the reference topological parameters to generate a corrected weight factor. Next, the corrected weight factor is subjected to a direction field superposition with the spatial gradient vector direction of the humidity fluctuation data to gradually calibrate the interstitial fluid flow direction attribute of each edge, forming a corrected interstitial fluid flow direction attribute sequence. For the drug concentration distribution, a low-density region segmentation technique is used to identify the regions not covered by the drug, and the regions with abnormal gradient decline are detected based on the change rate of the spatial gradient amplitude of the humidity fluctuation data. The geometric boundaries of these two types of regions are spatially superimposed, and the overlapping part is marked as the potential necrotic region. Finally, the edge attribute parameters are generated by jointly encoding the corrected interstitial fluid flow direction attribute sequence and the edge connection weight in the reference topological parameters; and the node attribute parameters are generated by using dynamic attenuation operation in combination with the necrotic region attribute and the initial node health in the reference topological parameters. The whole process relies on high-precision data processing techniques and complex mathematical models to ensure the accuracy of the results.

[0119] The following is a specific example:

[0120] In a case of a severely burned patient, the doctor first applied the methods of steps 1031 and 1032 to obtain detailed humidity fluctuation data and a three-dimensional anatomical model. Next, in step 1033, they decomposed the spatial gradient of the humidity fluctuation data into parallel and orthogonal components, and dynamically coupled them with the anatomical unit adjacency relationship in the reference topological parameters to generate a corrected weight factor. Using these factors, they gradually calibrated the interstitial fluid flow direction to form a corrected interstitial fluid flow direction attribute sequence. At the same time, by performing low-density region segmentation on the drug concentration distribution, the regions not covered by the drug were identified, and the regions with abnormal gradient decline were detected based on the change rate of the spatial gradient amplitude of the humidity fluctuation data. After spatially superimposing the geometric boundaries of these two types of regions, it was found that some overlapping parts might indicate the existence of necrotic regions. Based on this information, the doctor adjusted the treatment plan, increased the local drug release amount, and performed necessary surgical operations.

[0121] In order to further improve the accurate prediction of the infection spread path during the burn treatment process and optimize the personalized treatment plan, in some embodiments, the low-density region segmentation of the drug concentration distribution is performed to extract the geometric boundary of the region not covered by the drug, and at the same time, based on the change rate of the spatial gradient amplitude of the humidity fluctuation data, the geometric boundary of the region with abnormal gradient decline is detected, and the geometric boundaries of the region not covered by the drug and the region with abnormal gradient decline are subjected to spatial topological superposition, and the overlapping region is defined as the necrotic region attribute of the node, including:

[0122] Segment the low-density regions of the drug concentration distribution, extract the geometric boundaries of the regions not covered by the drug, and store the geometric boundaries as a first boundary set; perform a temporal cumulative calculation on the rate of change of the spatial gradient amplitude of the humidity fluctuation data to generate the geometric boundaries of the regions with abnormal gradient decline, and store the geometric boundaries as a second boundary set; input the first boundary set and the second boundary set into a spatial topology overlay module, perform a Boolean intersection operation on the regions not covered by the drug and the regions with abnormal gradient decline, output the polygon vertex coordinate set of the overlapping region, and construct a spatial distribution vector map of the overlapping region based on the polygon vertex coordinate set; calculate the real-time proportion of the area of the overlapping region to the area of the region not covered by the drug according to the spatial distribution vector map, and extract the number of extreme points of the contour curvature as a morphological complexity index; when the morphological complexity index exceeds a preset complexity threshold, mark the overlapping region as the necrotic region attribute of the node, and associate the necrotic region attribute with the node spatial topology database.

[0123] In this embodiment, the segmentation of the low-density regions of the drug concentration distribution refers to identifying the anatomical units not covered or insufficiently covered by the drug, extracting the geometric boundaries of these regions, and storing them as a first boundary set. The rate of change of the spatial gradient amplitude of the humidity fluctuation data generates the geometric boundaries of the regions with abnormal gradient decline through temporal cumulative calculation, and stores them as a second boundary set. The spatial topology overlay module performs a Boolean intersection operation on these two types of boundaries and outputs the polygon vertex coordinate set of the overlapping region for constructing a spatial distribution vector map. This vector map not only shows the position and shape of the overlapping region but also provides the area proportion and morphological complexity index. When the morphological complexity index exceeds the preset threshold, the overlapping region is marked as the necrotic region attribute and associated with the node spatial topology database to help doctors evaluate the infection risk.

[0124] In the embodiment of the present application, first, use a mathematical model to segment the low-density regions of the drug concentration distribution, extract the geometric boundaries of the uncovered regions, and form a first boundary set. Then, based on the rate of change of the spatial gradient amplitude of the humidity fluctuation data, perform a temporal cumulative calculation to generate the geometric boundaries of the regions with abnormal gradient decline, and form a second boundary set. Then, input the two boundary sets into the spatial topology overlay module, perform a Boolean intersection operation, and output the polygon vertex coordinate set of the overlapping region. Construct a spatial distribution vector map according to these coordinates, calculate the real-time proportion of the area of the overlapping region to the area of the region not covered by the drug, and extract the number of extreme points of the contour curvature as a morphological complexity index. Finally, if the morphological complexity index exceeds the preset threshold, mark the overlapping region as a necrotic region and update the node spatial topology database. The whole process relies on accurate data processing and complex mathematical models to ensure the accuracy of the results.

[0125] The following is a specific example:

[0126] In a treatment case of a severely burned patient, the doctor first used the above method to obtain detailed drug concentration distribution and humidity fluctuation data. By segmenting the low-density areas of the drug concentration distribution, the geometric boundaries of the areas not covered by the drug were identified, forming the first boundary set. At the same time, based on the rate of change of the spatial gradient amplitude of the humidity fluctuation data, the geometric boundaries of the areas with abnormal gradient decline were detected, forming the second boundary set. These two boundary sets were input into the spatial topology superposition module. After performing the Boolean intersection operation, it was found that some overlapping areas might indicate the presence of necrotic tissue. Based on this information, the doctor adjusted the treatment plan, increased the local drug release amount, and performed necessary surgical operations. As the treatment progressed, the wound healing of the patient improved significantly, and the risk of infection was greatly reduced. By this method, not only was the treatment efficiency improved, but also unnecessary drug use was reduced, and the risk of side effects was lowered, ultimately achieving the best rehabilitation effect. This series of operations demonstrated how to improve the effect of burn treatment, shorten the rehabilitation time, and improve the quality of life of patients through scientific data analysis means.

[0127] To further improve the efficiency and accuracy of infection control, in some embodiments, in step 104, inputting the infection spread map into a multi-layer topology structure, combining the metabolic rate parameter in the node attribute with the fluid resistance parameter in the edge attribute, and iteratively updating the infection propagation intensity to generate a weight distribution map including the infection core area and the diffusion boundary, includes:

[0128] 1041. Initialize the node attributes and edge attributes of the infection spread map based on the multi-layer topology structure, associate the metabolic rate parameter with the node activation state of each layer, and embed the fluid resistance parameter into the edge connection strength between layers to generate the initial weight distribution of the multi-layer topology structure;

[0129] In step 1041, the node attributes and edge attributes of the infection spread map are initialized based on the multi-layer topology structure. Here, the metabolic rate parameter is associated with the node activation state of each layer, which is used to describe the ability of different tissue regions to resist infection; the fluid resistance parameter is embedded into the edge connection strength between layers, representing the ease of transmission of pathogens or drugs between different tissues. These parameters are obtained through a combination of clinical data collection, experimental measurement, and mathematical modeling.

[0130] In the embodiments of the present application, first, a basic topology structure is constructed based on the medical image data of the patient, then the initial activation state of the nodes is determined according to the metabolic rate of different tissue types, and then the edge connection strength between layers is set according to anatomical knowledge to simulate the actual fluid resistance situation, and finally an initial weight distribution map reflecting the specific physiological conditions of the patient is generated as the basis for subsequent iteration.

[0131] 1042. Perform heterologous interference compensation on the metabolic rate parameters of adjacent nodes based on the edge connection strength, generate the metabolic interference value of each node, and calculate the incremental component of the infection transmission strength to iteratively update the infection transmission strength;

[0132] In step 1042, the metabolic interference value refers to the change in the metabolic rate caused by heterologous interference between adjacent nodes. In this step, the incremental component of the infection transmission strength is calculated to iteratively update the infection transmission strength in the entire network. This process takes into account the influence of the local environment on the infection diffusion speed.

[0133] In the embodiments of the present application, first, the key nodes most affected by the infection are identified. Second, the heterologous interference compensation received by each node from its neighbor nodes is calculated. Then, the infection transmission strength of each node is adjusted according to the obtained metabolic interference value. Finally, the infection transmission mode of the entire network is updated to more accurately simulate the actual infection diffusion process.

[0134] 1043. Extract the drug residue attenuation factor of each node according to the diffusion path of the residue amount of the drug concentration distribution in the multi-layer topology structure, non-linearly superimpose the drug residue attenuation factor and the gradient change rate of the updated infection transmission strength to generate a dynamic correction coefficient, and adaptively scale the weight ratio of the metabolic interference value by using the dynamic correction coefficient;

[0135] In step 1043, the drug residue attenuation factor refers to the degree of weakening of the drug effect exhibited by each node on the diffusion path in the multi-layer topology structure according to the residue amount of the drug concentration distribution. This factor is used to quantify the effect of the drug action over time and helps to more accurately predict the change trend of infection transmission during the treatment process. The dynamic correction coefficient is a regulatory parameter generated by non-linearly superimposing the drug residue attenuation factor and the gradient change rate of the updated infection transmission strength, and is used to adaptively adjust the weight ratio of the metabolic interference value.

[0136] In the embodiments of the present application, first, the drug residue attenuation factor at each node in the patient's body is extracted. Second, the non-linear relationship between these factors and the gradient change rate of the current infection transmission strength is calculated. Then, a dynamic correction coefficient for each node is generated based on the above results. Finally, the metabolic interference value is adaptively scaled by using this coefficient to ensure that the model can reflect the treatment progress and infection status in real time.

[0137] 1044. Perform multi-level cascading on the scaled metabolic interference value and the path integral result of the fluid resistance parameter, re-divide the boundary permeability of the infection core area, and generate a dynamic probability field of the diffusion boundary;

[0138] In step 1044, the scaled metabolic interference value is combined with the path integral result of the fluid resistance parameter, redefining the concept of the boundary permeability of the infection core area. Here, "path integral" refers to considering the cumulative effect of all possible paths from the source point to the target point. This method helps to determine how pathogens or drugs move between different tissue levels and accordingly updates the boundary of the core area of infection spread.

[0139] In the embodiments of the present application, first, the adaptively scaled metabolic interference value is integrated with the fluid resistance parameter between levels. Second, the new boundary permeability is calculated through the path integral method. Then, based on this information, the scope of the infection core area is re-divided. Finally, a dynamic field map showing the probability of infection spread is generated to support clinical decision-making.

[0140] 1045. Based on the initial weight distribution, by alternately iterating the incremental component of the infection propagation intensity and the backpropagation compensation of the dynamic probability field, a weight distribution map including the infection core area and the diffusion boundary is generated.

[0141] In step 1045, based on the initial weight distribution, by alternately iterating the incremental component of the infection propagation intensity and the backpropagation compensation mechanism of the dynamic probability field, a complete weight distribution map including the infection core area and the diffusion boundary is finally generated. This process aims to continuously optimize the model parameters to make them closer to the actual situation, thereby improving the prediction accuracy.

[0142] In the embodiments of the present application, first, the weight distribution of the entire system is initialized. Second, the update of the infection propagation intensity and the adjustment of the dynamic probability field are alternately executed. Then, the feedback mechanism is used to gradually improve the model performance. Finally, a detailed weight distribution map is output, clearly marking the key areas and the edge boundaries of the infection spread, providing a scientific basis for formulating personalized treatment plans.

[0143] The following is a specific example:

[0144] When treating a severely burned patient, the doctor first constructed a multi-layer topological structure model based on the patient's scanned image data according to step 1041. Subsequently, the infection transmission intensity was dynamically adjusted according to step 1042, taking into account the heterologous interference compensation of antibiotics to the surrounding healthy tissues. Then, according to the drug concentration distribution and its attenuation factor in step 1043, a dynamic correction coefficient was calculated to adaptively adjust the metabolic interference value. According to step 1044, the dynamic probability field of the diffusion boundary was optimized. Finally, according to step 1045, by alternately iteratively updating the infection transmission intensity and the backpropagation compensation mechanism, a weight distribution map accurately showing the infection diffusion was generated, guiding the doctor to implement a precise treatment plan, effectively controlling the infection and promoting the patient's recovery. This method not only improves the accuracy of treatment, but also significantly shortens the patient's recovery time, demonstrating the great potential of personalized medicine.

[0145] In summary, by implementing the solutions of steps 1041 to 1045, not only can the precise control of infection diffusion be achieved, but also personalized treatment planning can be carried out according to the specific conditions of the patient. This comprehensive analysis method significantly improves the medical efficiency and curative effect, reduces unnecessary drug use, and improves the patient's recovery speed and quality of life.

[0146] In order to further improve the efficiency and accuracy of infection control, in some embodiments, according to the diffusion path of the residual amount of the drug concentration distribution in the multi-layer topological structure in step 1043, the drug residue attenuation factor of each node is extracted, and the drug residue attenuation factor is non-linearly superimposed with the gradient change rate of the updated infection transmission intensity to generate a dynamic correction coefficient, and the weight ratio of the metabolic interference value is adaptively scaled by using the dynamic correction coefficient, including:

[0147] Traverse the node sequence of the drug diffusion path in the multi-layer topological structure, perform a convolution operation on the number of path branches of each node and the concentration gradient difference between adjacent layers, and output the numerical sequence of the drug residue attenuation factor of each node; perform a sliding window difference calculation on the spatial gradient amplitude of the updated infection transmission intensity in the time series dimension, and perform two-way propagation compensation on the gradient change rate of adjacent nodes to generate a gradient change rate parameter matrix; input the numerical sequence of the drug residue attenuation factor and the gradient change rate parameter matrix into a non-linear coupling unit, perform a piecewise power-law transformation operation to generate a dynamic correction coefficient; divide the weight level of the metabolic interference value according to the numerical distribution interval of the dynamic correction coefficient, and adaptively scale the weight ratio of the metabolic interference value according to the level boundary.

[0148] In this embodiment, the drug residue attenuation factor refers to a data sequence obtained by analyzing the node sequence on the diffusion path of the drug in the multi-layer topological structure and performing a convolution operation by combining the number of path branches of each node and the concentration gradient difference between adjacent layers. These data are used to quantify the change of drug effect over time at each node, helping to precisely adjust the treatment strategy. The gradient change rate parameter matrix is obtained through a sliding window difference calculation based on the spatial gradient amplitude of the infection propagation intensity, reflecting the changing trend of the infection diffusion speed between different nodes. The non-linear coupling unit combines the above two parameters by performing a piecewise power-law transformation operation to generate a dynamic correction coefficient, which can adaptively adjust the weight ratio of the metabolic interference value according to the actual situation.

[0149] In the embodiment of the present application, first, all nodes on the drug diffusion path are traversed in the multi-layer topological structure, and a convolution operation is performed on the number of path branches of each node and the concentration gradient difference between adjacent layers to output a numerical sequence of the drug residue attenuation factor. Then, a sliding window difference calculation is used for the spatial gradient amplitude of the updated infection propagation intensity in the time series dimension to generate a gradient change rate parameter matrix. Then, these two results are input into the non-linear coupling unit to generate a dynamic correction coefficient through a piecewise power-law transformation. Finally, the weight levels of the metabolic interference value are divided according to the distribution interval of the dynamic correction coefficient, and the weight ratio of the metabolic interference value is adaptively scaled accordingly to optimize the model prediction accuracy.

[0150] The following is a specific example:

[0151] In the treatment case of a severely burned patient, the doctor constructed a multi-layer topological structure using the computer tomography data of different tissue levels in the patient's body and monitored the diffusion of antibiotics in the body. By accurately calculating the drug residue attenuation factor and the gradient change rate of the infection propagation intensity, a dynamic correction coefficient was generated. This enables the doctor to adjust the treatment plan according to real-time data, such as increasing or decreasing the antibiotic dose in a specific area. At the same time, the weight ratio of the metabolic interference value was adjusted based on this coefficient, so as to more accurately predict and control the spread of infection, ultimately achieving personalized and precise treatment, effectively improving the patient's recovery speed and quality of life. This method not only improves the treatment efficiency but also reduces unnecessary drug use, demonstrating its great potential in clinical applications.

[0152] In order to further improve the dynamic monitoring ability of the infection risk, in some embodiments, the step of matching the burn depth grading standard according to the geometric deformation characteristics and gradient attenuation trend of the weight distribution map in step 105 to generate a dynamic infection risk score integrating the drug concentration distribution feedback includes:

[0153] 1051. Detect multi-scale curvature extreme points of the geometric deformation features of the weight distribution map, extract the density distribution of curvature extreme points and the change rate of the connected domain area at different scales, and construct a weight deformation feature vector.

[0154] In step 1051, the multi-scale curvature extreme point detection refers to analyzing the geometric deformation features in the weight distribution map at different scales, extracting the density distribution of curvature extreme points and the change rate of the connected domain area, and constructing a weight deformation feature vector. These data are used to describe the spatial structure features of the infection diffusion pattern and its changes over time. The density distribution of curvature extreme points reflects the density of deformation in a specific area, while the change rate of the connected domain area measures the speed of the expansion of the infected area.

[0155] In the embodiment of the present application, first, a series of different scale parameters are selected. Second, the curvature extreme point detection algorithm is applied at each scale. Then, the density of curvature extreme points and the change rate of the connected domain area at each scale are calculated. Finally, these features are combined into a vector as the basis for describing the geometric deformation of the weight distribution map.

[0156] 1052. Decompose the gradient decay trend into a spatial gradient direction field, calculate the change rate of the gradient mean value and the direction consistency coefficient of the anisotropic region, and generate a gradient decay trend parameter matrix in combination with the gradient decay acceleration accumulated in time series.

[0157] In step 1052, the spatial gradient direction field decomposition is a process of detailed analysis of the gradient decay trend. By calculating the change rate of the gradient mean value and the direction consistency coefficient of the anisotropic region, and combining the gradient decay acceleration accumulated in time series, a gradient decay trend parameter matrix is generated. This process helps to understand the directional and speed change laws of infection diffusion.

[0158] In the embodiment of the present application, first, the weight distribution map is decomposed into a spatial gradient direction field. Second, the change rate of the gradient mean value and the direction consistency coefficient of the anisotropic region are calculated. Then, the gradient decay acceleration information in the time series is combined. Finally, a parameter matrix reflecting the gradient decay trend is generated to guide the subsequent analysis.

[0159] 1053. Input the weight deformation feature vector and the gradient decay trend parameter matrix into the hierarchical fusion module, and output a deformation-decay coupling coefficient sequence according to the preset skin layer thickness parameters in the burn depth grading standard.

[0160] In step 1053, the deformation-attenuation coupling coefficient sequence is a series of numerical values generated by inputting the weighted deformation eigenvector and the gradient attenuation trend parameter matrix into the hierarchical fusion module. These coefficients represent the degree of interaction between geometric deformation and gradient change during the infection diffusion process, which helps to more accurately match the burn depth grading standard. The skin layer thickness parameter is used to calibrate the infection risk assessment at different levels, ensuring that the scoring system can accurately reflect the specific conditions of each level.

[0161] In the embodiment of the present application, first, the extracted weighted deformation eigenvector and the calculated gradient attenuation trend parameter matrix are used as inputs. Secondly, the hierarchical fusion module is used to process these data, adjust the model output according to the preset skin layer thickness parameter. Then, a deformation-attenuation coupling coefficient sequence is generated. Finally, these coefficients are used to quantify the infection risk at different burn depths.

[0162] 1054. Construct a three-dimensional probability distribution surface based on the deformation-attenuation coupling coefficient sequence, and use the local inhibition rate feedback by the drug concentration distribution as a constraint condition. Calculate the ratio of the integral value within the truncated region to the standard burn depth volume as the dynamic scoring base of the infection risk.

[0163] In step 1054, the three-dimensional probability distribution surface is a mathematical model constructed based on the deformation-attenuation coupling coefficient sequence, which reflects the spatial distribution characteristics of the infection risk. The local inhibition rate is an index in the drug concentration distribution feedback, used to measure the effective control ability of the drug on the infection in a specific region. The ratio of the integral value within the truncated region to the standard burn depth volume is used as the dynamic scoring base of the infection risk. This process aims to adjust the scoring standard in combination with the actual treatment effect.

[0164] In the embodiment of the present application, first, construct a three-dimensional probability distribution surface according to the deformation-attenuation coupling coefficient sequence. Secondly, consider the influence of the local inhibition rate. Then, calculate the integral value within the truncated region and compare it with the standard burn depth volume. Finally, obtain a basic score to guide the subsequent risk assessment and treatment strategy adjustment.

[0165] 1055. Perform layer-by-layer interpolation calibration according to the dynamic scoring base of the infection risk and the depth thresholds of the epidermis, dermis, and subcutaneous tissue in the burn depth grading standard, and generate the dynamic infection risk score with the drug concentration distribution feedback.

[0166] In step 1055, layer-by-layer interpolation calibration is a method used to finely adjust according to the dynamic scoring base of the infection risk and the depth thresholds of the epidermis, dermis, and subcutaneous tissue in the burn depth grading standard, so as to generate the final dynamic infection risk score with the drug concentration distribution feedback. This method can improve the accuracy and applicability of the scoring system, making it better serve clinical decision-making.

[0167] In the embodiments of the present application, first, the dynamic scoring base for infection risk is determined. Secondly, layer-by-layer interpolation calibration is performed according to different depth thresholds of the epidermis, dermis, and subcutaneous tissue. Then, the scoring is adjusted by combining the feedback of the drug concentration distribution. Finally, a comprehensive dynamic scoring for infection risk is generated. This scoring not only considers the current state of infection but also incorporates possible future development trends, providing strong support for doctors to help them make more scientific and reasonable treatment decisions. This method significantly improves the accuracy and efficiency of treatment and promotes the rapid recovery of patients.

[0168] The following is a specific example:

[0169] In the case of a severely burned patient, the doctor used the method of steps 1051-1055 to conduct a detailed analysis of the wound. First, geometric deformation features were extracted through multi-scale curvature extreme point detection, and then spatial gradient direction field decomposition was performed to obtain the gradient attenuation trend. Next, a sequence of deformation-attenuation coupling coefficients was output in combination with the skin layer thickness parameter, and a three-dimensional probability distribution surface was constructed. Finally, the dynamic scoring of infection risk was adjusted based on the feedback of the drug concentration distribution, thereby realizing a personalized and precise treatment plan, and significantly improving the treatment effect.

[0170] In summary, adopting the solution of the above steps 1051 to 1055 greatly improves the prediction accuracy of the infection risk of severely burned patients. Through in-depth analysis of geometric deformation features and gradient attenuation trends, combined with the feedback of drug concentration distribution, high-risk areas can be more accurately identified and treatment strategies can be adjusted in a timely manner. This not only improves the effectiveness of treatment but also reduces the occurrence of complications and improves the prognosis of patients. In addition, this method can also help doctors better understand the development pattern of infection, optimize resource allocation, and improve the overall medical quality.

[0171] Figure 2 The following is a schematic structural diagram of a multi-dimensional data processing system for intelligent scoring of emergency patients provided by the embodiments of the present application, as Figure 2 shown, the device includes:

[0172] An acquisition module 21, which acquires dynamic temperature data and humidity fluctuation data of the burn wound through a flexible wearable monitoring device, and the flexible wearable monitoring device is integrated in an intelligent dressing layer with a thermally responsive spraying structure;

[0173] An input module 22, which generates a temperature-drug release model according to the thermosensitive parameters of the intelligent dressing layer, inputs the time series change of the dynamic temperature data into the temperature-drug release model, and generates a drug concentration distribution adapted to the morphology of the burn wound;

[0174] The construction module 23 constructs an infection diffusion map based on the anatomical structure as the topological reference, with the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution. The infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing necrotic regions.

[0175] The update module 24 inputs the infection diffusion map into a multi-layer topological structure, combines the metabolic rate parameter in the node attributes with the fluid resistance parameter in the edge attributes, and iteratively updates the infection propagation intensity to generate a weight distribution map including the infection core area and the diffusion boundary.

[0176] The integration module 25 matches the burn depth grading standard according to the geometric deformation characteristics and gradient attenuation trend of the weight distribution map, and generates a dynamic infection risk score integrating the feedback of the drug concentration distribution.

[0177] Figure 2 The multi-dimensional data processing system for intelligent scoring of emergency patients described above can execute Figure 1 The multi-dimensional data processing method for intelligent scoring of emergency patients described in the embodiments shown. Its implementation principle and technical effects will not be elaborated. For the multi-dimensional data processing system for intelligent scoring of emergency patients in the above embodiments, the specific ways for each module and unit to execute operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0178] In a possible design, Figure 2 The multi-dimensional data processing system for intelligent scoring of emergency patients in the embodiments shown can be implemented as a computing device, such as Figure 3 shown, and this computing device can include a storage component 31 and a processing component 32;

[0179] The storage component 31 stores one or more computer instructions, and among them, the one or more computer instructions are called and executed by the processing component 32.

[0180] The processing component 32 is used for the Figure 1 multi-dimensional data processing method for intelligent scoring of emergency patients in the above

[0181] embodiments. Among them, the processing component 32 can include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component can also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method.

[0182] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0183] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0184] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above-mentioned peripheral interface module may be an output device, an input device, etc.

[0185] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0186] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above-mentioned processing component, storage component, etc. may be basic server resources leased or purchased from a cloud computing platform.

[0187] The embodiment of the present application also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above Figure 1 multi-dimensional data processing method for intelligent scoring of emergency patients shown in the embodiment.

[0188] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0189] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0190] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0191] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-dimensional data processing method for intelligent scoring of emergency patients, characterized in that: include: The dynamic temperature data and humidity fluctuation data of the burn wound surface are obtained by a flexible wearable monitoring device, wherein the flexible wearable monitoring device is integrated into a smart dressing layer having a thermal responsive spray structure; generating a temperature-drug release model according to the thermosensitive parameters of the smart dressing layer, inputting the temporal changes of the dynamic temperature data into the temperature-drug release model, and generating a drug concentration distribution adapted to the morphology of the burn wound; Based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, an infection diffusion map based on the anatomical structure as the topological reference is constructed, wherein the infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing the necrotic area; Inputting the infection diffusion map into a multi-layer topological structure, combining the metabolic rate parameter in the node attribute and the fluid resistance parameter in the edge attribute, iteratively updating the infection propagation intensity, and generating a weight distribution map including the infection core area and the diffusion boundary; According to the geometric deformation characteristics of the weight distribution map and the gradient attenuation trend, the burn depth classification standard is matched to generate a dynamic infection risk score integrating the drug concentration distribution feedback.

2. The method according to claim 1, characterized in that The spatial gradient based on the humidity fluctuation data and the superposition effect of the drug concentration distribution are used to construct an infection diffusion map based on the anatomical structure as a topological reference. The infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing the necrotic area, including: The gradient change direction between adjacent regions in the anatomical unit of the spatial gradient of the humidity fluctuation data is coupled with the drug penetration range of the corresponding anatomical unit in the drug concentration distribution to generate a superposition effect parameter; Define nodes and edges in the topological benchmark according to the spatial segmentation relationship of the anatomical structure, wherein the nodes are bound to the spatial coordinates of the anatomical units, and the edges are bound to the adjacent connected paths between the anatomical units, to generate benchmark topological parameters; Based on the superposition effect parameter and the reference topology parameter, the tissue fluid flow direction attribute of the edge is corrected by the vector direction of the spatial gradient of the humidity fluctuation data, and the necrotic area attribute of the node is defined according to the overlap between the drug uncovered area in the drug concentration distribution and the area where the spatial gradient of the humidity fluctuation data drops abnormally, to generate edge attribute parameters and node attribute parameters; The benchmark topological parameters, the edge attribute parameters and the node attribute parameters are integrated into the same topological mapping framework to generate an infection diffusion map including the complete tissue fluid dynamic path and the evolution relationship of the necrotic area.

3. The method according to claim 2, characterized in that Based on the superposition effect parameter and the reference topology parameter, the tissue fluid flow direction attribute of the edge is corrected by the vector direction of the spatial gradient of the humidity fluctuation data, and the necrotic area attribute of the node is defined according to the overlap between the drug uncovered area in the drug concentration distribution and the area where the spatial gradient of the humidity fluctuation data drops abnormally, to generate edge attribute parameters and node attribute parameters, including: Based on the superposition effect parameter and the reference topology parameter, the spatial gradient of the humidity fluctuation data is decomposed into a parallel component and an orthogonal component, and the orthogonal component is dynamically coupled with the reference topology parameter to generate a modified weight factor of the edge; Superimposing the correction weight factor with the vector direction of the spatial gradient of the humidity fluctuation data in the direction field, calibrating the tissue fluid flow direction attribute of the edge step by step, and generating a corrected tissue fluid flow direction attribute sequence; The drug concentration distribution is segmented into low-density areas, and the geometric boundaries of the drug-uncovered area are extracted. At the same time, based on the spatial gradient amplitude change rate of the humidity fluctuation data, the geometric boundaries of the gradient abnormally decreased area are detected, and the geometric boundaries of the drug-uncovered area and the gradient abnormally decreased area are spatially topologically superimposed, and the overlapping area is defined as the necrotic area attribute of the node; The modified tissue fluid flow attribute sequence is jointly encoded with the edge connection weight in the benchmark topology parameter to generate the edge attribute parameter, and the necrotic area attribute and the node initial health degree in the benchmark topology parameter are dynamically attenuated to generate the node attribute parameter.

4. The method according to claim 3, characterized in that The method of performing low-density area segmentation on the drug concentration distribution, extracting the geometric boundary of the drug-uncovered area, and detecting the geometric boundary of the gradient abnormally decreasing area based on the spatial gradient amplitude change rate of the humidity fluctuation data, and spatially topologically superimposing the geometric boundaries of the drug-uncovered area and the gradient abnormally decreasing area, and defining the overlapping area as the necrotic area attribute of the node, includes: Segmenting the drug concentration distribution into low-density areas, extracting geometric boundaries of areas not covered by the drug, and storing the geometric boundaries as a first boundary set; Performing time series accumulation calculation on the spatial gradient amplitude change rate of the humidity fluctuation data to generate a geometric boundary of the gradient abnormal drop area, and storing the geometric boundary as a second boundary set; Input the first boundary set and the second boundary set into a spatial topology superposition module, perform a Boolean intersection operation on the drug uncovered area and the gradient abnormally decreased area, output a polygon vertex coordinate set of the overlapping area, and construct a spatial distribution vector diagram of the overlapping area based on the polygon vertex coordinate set; Calculating the real-time ratio of the area of ​​the overlapping region to the area of ​​the drug uncovered region according to the spatial distribution vector diagram, and extracting the number of extreme points of contour curvature as a morphological complexity index; When the morphological complexity index exceeds a preset complexity threshold, the overlapping area is marked as a necrotic area attribute of the node, and the necrotic area attribute is associated with a node space topology database.

5. The method according to claim 1, characterized in that: The infection diffusion map is input into a multi-layer topological structure, and the metabolic rate parameter in the node attribute and the fluid resistance parameter in the edge attribute are combined to iteratively update the infection propagation intensity, and generate a weight distribution map including the infection core area and the diffusion boundary, including: Initializing the node attributes and edge attributes of the infection diffusion map by level based on a multi-layer topological structure, associating metabolic rate parameters with the node activation states of each level, and embedding fluid resistance parameters into the edge connection strength between levels, thereby generating an initial weight distribution of the multi-layer topological structure; Based on the edge connection strength, the metabolic rate parameters of the adjacent nodes are compensated for heterogeneous interference, a metabolic interference value of each node is generated, and an incremental component of the infection propagation intensity is calculated to iteratively update the infection propagation intensity; According to the diffusion path of the residual amount of the drug concentration distribution in the multi-layer topological structure, the drug residual attenuation factor of each node is extracted, the drug residual attenuation factor is nonlinearly superimposed with the gradient change rate of the updated infection transmission intensity to generate a dynamic correction coefficient, and the weight ratio of the metabolic interference value is adaptively scaled using the dynamic correction coefficient; Performing multi-layer cascades on the scaled metabolic interference value and the path integral result of the fluid resistance parameter, re-dividing the boundary permeability of the infection core area, and generating a dynamic probability field of the diffusion boundary; Based on the initial weight distribution, a weight distribution map including an infection core area and a diffusion boundary is generated by alternately iterating the incremental component of the infection propagation intensity and the back propagation compensation of the dynamic probability field.

6. The method according to claim 5, characterized in that The method extracts the drug residue attenuation factor of each node according to the diffusion path of the residual amount of the drug concentration distribution in the multi-layer topological structure, nonlinearly superimposes the drug residue attenuation factor with the updated gradient change rate of the infection transmission intensity to generate a dynamic correction coefficient, and uses the dynamic correction coefficient to adaptively scale the weight ratio of the metabolic interference value, including: Traversing the node sequence of the drug diffusion path in the multi-layer topological structure, performing convolution operation on the number of path branches of each node and the concentration gradient difference of adjacent layers, and outputting a numerical sequence of drug residue attenuation factors of each node; Perform sliding window difference calculation on the spatial gradient amplitude of the updated infection propagation intensity in the time series dimension, perform bidirectional propagation compensation on the gradient change rate of adjacent nodes, and generate a gradient change rate parameter matrix; Inputting the drug residue attenuation factor numerical sequence and the gradient change rate parameter matrix into a nonlinear coupling unit, performing a piecewise power law transformation operation, and generating a dynamic correction coefficient; The weight levels of the metabolic interference value are divided according to the numerical distribution interval of the dynamic correction coefficient, and the weight ratio of the metabolic interference value is adaptively scaled according to the level boundary.

7. The method according to claim 1, characterized in that The method of matching the burn depth classification standard according to the geometric deformation characteristics of the weight distribution graph and the gradient attenuation trend to generate a dynamic infection risk score integrating the drug concentration distribution feedback includes: Perform multi-scale curvature extreme point detection on the geometric deformation features of the weight distribution graph, extract the density distribution of curvature extreme point and the area change rate of the connected domain at different scales, and construct a weight deformation feature vector; Decomposing the gradient attenuation trend by spatial gradient direction field, calculating the gradient mean change rate and direction consistency coefficient of the anisotropic region, and generating a gradient attenuation trend parameter matrix by combining the time-series accumulated gradient attenuation acceleration; Input the weighted deformation feature vector and the gradient attenuation trend parameter matrix into a hierarchical fusion module, and output a deformation-attenuation coupling coefficient sequence according to a skin layer thickness parameter preset in a burn depth grading standard; A three-dimensional probability distribution surface is constructed based on the deformation-attenuation coupling coefficient sequence, and the local inhibition rate of drug concentration distribution feedback is used as a constraint condition to calculate the ratio of the integral value in the truncated area to the standard burn depth volume as the infection risk dynamic scoring base; A layer-by-layer interpolation calibration is performed based on the infection risk dynamic score base and the depth thresholds of the epidermis, dermis, and subcutaneous tissue in the burn depth grading standard to generate an infection risk dynamic score with feedback on the drug concentration distribution.

8. A multi-dimensional data processing method for intelligent scoring of emergency patients, characterized in that: include: An acquisition module, which acquires dynamic temperature data and humidity fluctuation data of a burn wound surface through a flexible wearable monitoring device, wherein the flexible wearable monitoring device is integrated into a smart dressing layer having a thermally responsive spray structure; An input module generates a temperature-drug release model according to the thermosensitive parameters of the smart dressing layer, inputs the temporal changes of the dynamic temperature data into the temperature-drug release model, and generates a drug concentration distribution adapted to the burn wound morphology; A construction module, based on the superposition effect of the spatial gradient of the humidity fluctuation data and the drug concentration distribution, constructs an infection diffusion map with the anatomical structure as the topological reference, wherein the infection diffusion map includes edge attributes representing the flow direction of tissue fluid and node attributes representing the necrotic area; An updating module, inputting the infection diffusion map into a multi-layer topological structure, combining the metabolic rate parameter in the node attribute and the fluid resistance parameter in the edge attribute, iteratively updating the infection propagation intensity, and generating a weight distribution map including the infection core area and the diffusion boundary; The integration module matches the burn depth classification standard according to the geometric deformation characteristics and gradient attenuation trend of the weight distribution map, and generates a dynamic infection risk score that integrates the drug concentration distribution feedback.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a multi-dimensional data processing method for intelligent scoring of emergency patients as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a multi-dimensional data processing method for intelligent scoring of emergency patients as described in any one of claims 1 to 7 is implemented.

Images