Burn ward infection prevention and control intelligent early warning system based on big data

Abstract

The invention discloses a burn ward infection prevention and control intelligent early warning system based on big data, and relates to the field of big data analysis and monitoring. The system comprises a multi-modal data acquisition module for acquiring multi-modal data; the data preprocessing module is used for preprocessing the collected data; the feature construction module is used for extracting infection risk features according to the preprocessed data; the dynamic infection risk assessment module is used for constructing a patient infection risk assessment model by using a multi-modal feature learning model according to the infection risk features, and taking the infection risk features as input and the patient infection risk indexes as output; and the intelligent early warning module sets a multi-stage early warning mechanism according to the infection risk index. Through multi-modal data fusion, an intelligent early warning mechanism and dynamic infection risk assessment based on multi-modal feature learning, the infection risk is accurately predicted, early warning hierarchical management is optimized, the adaptability of the system to risk changes is improved, and efficient and prospective infection prevention and control decision support is provided for clinic.

Description

Technical Field

[0001] The present invention relates to the field of big data analysis and monitoring, and specifically to an intelligent early warning system for infection prevention and control in a burn ward based on big data. Background Art

[0002] Patients in burn wards are extremely vulnerable to infection due to damaged skin barriers. Infection not only delays wound healing but may also lead to systemic inflammatory responses and even endanger life. Therefore, infection prevention and control in burn wards is an important part of clinical nursing management. Traditional infection prevention and control mainly rely on manual monitoring and empirical judgment, but this method has problems such as strong subjectivity, poor real-time performance, and difficulty in accurately predicting infection risks, and cannot meet the refined management requirements of the modern medical environment.

[0003] With the development of big data, artificial intelligence, and multi-modal data fusion technologies, it has become possible to develop an intelligent early warning system for infection prevention and control based on data-driven. By collecting real-time patient physiological indicators, ward environment monitoring data, and medical staff operation behavior data, and combining with a multi-modal feature learning model, it is possible to accurately evaluate the infection risk of patients and provide intelligent early warnings to assist medical staff in taking timely intervention measures, thereby reducing the infection rate and improving the patient recovery efficiency.

[0004] The present invention provides an intelligent early warning system for infection prevention and control in a burn ward based on big data. This system constructs an accurate, real-time, and efficient infection risk prediction model through modules such as multi-modal data collection, data preprocessing, feature construction, dynamic risk assessment, intelligent early warning, and decision support. Compared with traditional methods, the present invention has higher accuracy and adaptability in data fusion, intelligent analysis, risk prediction, and decision optimization, providing strong support for the intelligent management of burn wards. Summary of the Invention

[0005] Based on the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide an intelligent early warning system for infection prevention and control in a burn ward based on big data to solve the above technical problems.

[0006] To achieve the above purpose, the present invention provides the following technical solution: An intelligent early warning system for infection prevention and control in a burn ward based on big data, comprising:

[0007] A multi-modal data collection module that collects patient physiological indicators, ward environment monitoring data, and medical staff operation behavior data;

[0008] A data preprocessing module that performs outlier detection, denoising, and normalization processing on the collected data;

[0009] A feature construction module that extracts infection risk features based on the preprocessed data;

[0010] The dynamic infection risk assessment module constructs a patient infection risk assessment model using a multi-modal feature learning model according to the infection risk characteristics, taking the infection risk characteristics as the input and the patient infection risk index as the output;

[0011] The intelligent early warning module sets up a multi-level early warning mechanism according to the infection risk index.

[0012] The present invention is further configured such that the multi-modal data acquisition module includes:

[0013] The patient physiological index acquisition unit acquires the patient's physiological indexes;

[0014] The environmental monitoring unit acquires the ward environmental monitoring data;

[0015] The medical staff operation behavior monitoring unit is used to record the operation behavior data of the medical staff.

[0016] The present invention is further configured such that the data preprocessing module includes:

[0017] Data outlier detection: construct a dynamic time series change function to evaluate whether the data at the current moment exceeds the normal fluctuation range according to the change trend of the data itself. The formula is: where, A t is whether it is an outlier at the t-th moment, 1 means outlier, 0 means normal, is the change factor at the t-th moment, γ t is the dynamic sensitivity threshold at the t-th moment, and its calculation formula is: α is the adaptive adjustment coefficient;

[0018] Data denoising: use an adaptive denoising method to construct a composite denoising model, and its calculation logic is: is the denoised signal, is the signal smoothing operation, is the feature function, and its calculation logic is: τ i is the adjustment parameter;

[0019] Data normalization: use an adaptive non-linear mapping normalization method to dynamically perform normalization according to the data distribution characteristics. The formula is: where, is the normalized data, α i is the adjustment coefficient for each dimension, β is the non-linear mapping exponent, X t-i is the data at the past moment.

[0020] The present invention is further configured such that the feature construction module extracts infection risk features according to the preprocessed data, including:

[0021] Construct a patient physiological abnormality degree according to the patient's physiological indicators, where the patient's physiological indicators include: body temperature, heart rate, respiratory rate, blood oxygen saturation, white blood cell count, C-reactive protein, and lactic acid concentration;

[0022] Construct an environmental pollution index according to the ward environment monitoring data, where the ward environment monitoring data includes: air microorganism concentration, PM2.5 / PM10 particulate matter concentration, carbon dioxide concentration, relative humidity, and room temperature;

[0023] Construct a medical staff behavior deviation degree according to the medical staff operation behavior data, where the medical staff operation behavior data includes: hand hygiene compliance rate, nursing operation time, alcohol disinfection amount, and glove replacement frequency.

[0024] The present invention is further configured such that the calculation formula for constructing the patient physiological abnormality degree according to the patient's physiological indicators is: Λ i 、Δ j and Ω k are adjustment coefficients for the indicators, is the abnormality expression of indicator i, represents the dynamic change amount of the physiological indicator, represents the clinical relevant risk degree of each indicator, β i 、γ j and δ k are weight indices, and N, M, and L are the number of indicators.

[0025] The present invention is further configured such that the calculation logic for constructing the environmental pollution index according to the ward environment monitoring data is: Among them, E env (t) is the environmental pollution index, A(t) is the air microorganism concentration at time t, A ref is the air microorganism concentration under normal conditions, H(t) is the relative humidity at time t, H thr is the relative humidity under normal conditions, λ1, α1, and β1 are adjustment parameters, PM(t) is the PM2.5 / PM10 particulate matter concentration, T(t) is the room temperature at time t, T norm is the room temperature under normal conditions, δ1, μ1, and η1 are adjustment parameters, CO2(t) is the carbon dioxide concentration at time t, is the carbon dioxide concentration within the safe range, ρ1, ξ1, and θ1 are adjustment parameters, and γ1 and κ1 are weight coefficients.

[0026] The present invention is further configured such that the deviation degree of medical staff behavior is constructed based on the operation behavior data of medical staff, and its calculation logic is as follows: where D care (t) is the deviation degree of medical staff behavior, W(t) is the hand hygiene compliance rate, C(t) is the nursing operation time, S(t) is the amount of alcohol disinfection, G(t) is the glove replacement frequency, and G norm , C thr and W ref are operation indicators under normal conditions, α, β, δ, μ, λ, and η are adjustment parameters, and γ is a weight coefficient.

[0027] The present invention is further configured such that the construction of the patient infection risk assessment model includes:

[0028] Obtain the multi-modal signals after historical synchronization and the corresponding patient infection risks, construct infection risk features based on the multi-modal signals after historical synchronization, and set the infection risk features and the corresponding patient infection risk indices as a data set;

[0029] Divide the data set into a training set and a validation set;

[0030] Use the training set to train the long short-term memory network. When the loss function converges or reaches the maximum number of iterations, the training is completed. Use the validation set to verify the trained long short-term memory network. The input of the long short-term memory network is the infection risk feature, and the output is the real-time patient infection risk index;

[0031] Set the long short-term memory network that passes the verification as the patient infection risk assessment model.

[0032] The present invention is further configured such that the infection risk index has the following calculation formula where R risk (t) is the infection risk index, D care (t) is the deviation degree of medical staff behavior, E env (t) is the environmental pollution index, F(t) is the degree of patient physiological abnormality, σ(·) is the activation function, and α3, β3, γ3, λ3, δ3, and μ3 are adjustment parameters.

[0033] The present invention is further configured such that the multi-level warning mechanism is set according to the infection risk index, including:

[0034] Set the multi-level warning mechanism according to the infection risk index, and the warning levels are divided into: Among them, τ1 and τ2 are preset risk thresholds. When the early warning level is low risk, normal monitoring is carried out without special intervention; when the early warning level is medium risk, the monitoring frequency is increased to remind medical staff to optimize operations; when the early warning level is high risk, an immediate warning is issued and interventions are taken.

[0035] The present invention provides an intelligent early warning system for preventing and controlling infections in a burn ward based on big data, and the beneficial effects generated include:

[0036] 1. Multi-modal data fusion to improve the accuracy of infection risk assessment: Through the multi-modal fusion of patients' physiological indicators, ward environment monitoring data, and medical staff's operation behavior data, a more comprehensive infection risk assessment system is constructed, avoiding the one-sidedness that may be caused by a single data source, and improving the accuracy and reliability of prediction;

[0037] 2. Intelligent early warning mechanism to optimize risk classification management: According to the infection risk index, a multi-level early warning mechanism is set, and corresponding intervention measures are triggered according to different early warning levels, which not only avoids waste of resources caused by over-early warning, but also ensures that high-risk patients receive timely attention, and improves the refined level of hospital infection management;

[0038] 3. Infection risk assessment based on multi-modal feature learning to improve the dynamic adaptability of risk prediction: Combining the deviation degree of medical staff's behavior, environmental pollution index, and physiological abnormality degree of patients, a dynamic infection risk assessment model is established using a long short-term memory network, which can capture the changing trend of infection risk in real time. Compared with traditional static assessment methods, it can identify potential infection risks in advance and provide more forward-looking decision-making basis for clinical practice.

[0039] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the specific implementation manners of the present application are hereinafter specifically exemplified. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. In the drawings:

[0041] Figure 1 is a flowchart of an intelligent early warning system for preventing and controlling infections in a burn ward based on big data shown in an exemplary embodiment of the present invention;

[0042] Figure 2Schematic diagram of a smart early warning system for infection prevention and control in a burn ward based on big data, shown for an exemplary embodiment of the present invention. Detailed implementation manners

[0043] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, rather than for limiting the protection scope of the present invention.

[0044] It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present invention schematically. Therefore, only the components related to the present invention are shown in the drawings, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0045] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0046] Embodiment 1

[0047] A smart early warning system for infection prevention and control in a burn ward based on big data, as Figure 1 shown, includes:

[0048] A multi-modal data acquisition module that acquires patients' physiological indicators, ward environment monitoring data, and medical staff's operation behavior data;

[0049] A data preprocessing module that performs outlier detection, denoising, and normalization processing on the acquired data;

[0050] A feature construction module that extracts infection risk features according to the preprocessed data;

[0051] A dynamic infection risk assessment module that constructs a patient infection risk assessment model using a multi-modal feature learning model according to the infection risk features, takes the infection risk features as input, and the patient infection risk index as output;

[0052] A smart early warning module that sets a multi-level early warning mechanism according to the infection risk index.

[0053] The present invention is further configured such that the multimodal data acquisition module includes:

[0054] a patient physiological index acquisition unit for acquiring patient physiological indexes;

[0055] an environmental monitoring unit for acquiring ward environmental monitoring data;

[0056] a medical staff operation behavior monitoring unit for recording medical staff operation behavior data.

[0057] The present invention is further configured such that the data preprocessing module includes:

[0058] Data outlier detection: constructing a dynamic time series change function to evaluate whether the data at the current moment exceeds the normal fluctuation range according to the change trend of the data itself. The formula is: where A t is whether it is an outlier at the t-th moment, 1 means abnormal, 0 means normal, is the change factor at the t-th moment, and γ t is the dynamic sensitivity threshold at the t-th moment, and its calculation formula is: α is an adaptive adjustment coefficient. Specifically, it is calculated through the difference between the maximum and minimum values of the data to ensure that the threshold can adapt to the fluctuation range of the data, adaptively adjust the detection threshold according to the change trend of the data, avoid misjudgment caused by a fixed threshold, improve the sensitivity and accuracy of outlier detection, and calculate the sensitivity threshold through the maximum and minimum change ranges, which can adapt to different data patterns, enabling the system to have stronger adaptability when processing different patient physiological parameters, ward environmental data, and medical staff operation data;

[0059] Data denoising: using an adaptive denoising method to construct a composite denoising model, and its calculation logic is: is the denoised signal, is the signal smoothing operation, is the characteristic function, and its calculation logic is: τ i is a regulation parameter. Specifically, a composite denoising model is adopted. The core idea is to combine signal smoothing processing and noise suppression factor calculation to perform dynamic denoising on the original data X t to improve the stability of the signal, while retaining key feature information, being able to accurately remove noise while retaining key information, improving the signal quality and analysis accuracy of the burn ward infection prevention and control intelligent early warning system, and providing reliable data support for subsequent infection risk assessment;

[0060] Data normalization uses an adaptive non - linear mapping normalization method to adaptively normalize according to the data distribution characteristics dynamically. Its formula is: Among them, is the normalized data, α i is the adjustment coefficient for each dimension, β is the non - linear mapping exponent, and X t-i is the data at the past moment.

[0061] The present invention is further configured such that the feature construction module extracts infection risk features according to the pre - processed data, including:

[0062] Construct the patient physiological abnormality degree according to the patient's physiological indicators. The patient's physiological indicators include: body temperature, heart rate, respiratory rate, blood oxygen saturation, white blood cell count, C - reactive protein, and lactate concentration. The patient physiological abnormality degree reflects the individual's health status. Abnormal physiological parameters may indicate the presence of infection or other inflammatory reactions;

[0063] Construct the environmental pollution index according to the ward environmental monitoring data. The ward environmental monitoring data includes: air microbial concentration, PM2.5 / PM10 particulate matter concentration, carbon dioxide concentration, relative humidity, and room temperature. The environmental pollution index quantifies the potential infection risk in the ward environment, such as microbial concentration, air quality, etc.;

[0064] Construct the medical staff behavior deviation degree according to the medical staff operation behavior data. The medical staff operation behavior data includes: hand hygiene compliance rate, nursing operation time, alcohol disinfection amount, and glove replacement frequency. The medical staff behavior deviation degree evaluates whether the medical staff strictly implements standard protection measures to reduce the infection risk caused by human factors;

[0065] Integrate the patient's physiology, ward environment, and medical staff behavior data to construct a more complete infection risk feature, improve the accuracy of infection prediction. The constructed features can be used as the input of the infection risk assessment model, providing high - quality feature data for subsequent intelligent early warning and decision - making support.

[0066] The present invention is further configured such that the construction of the patient physiological abnormality degree according to the patient's physiological indicators has the following calculation formula: Λ i 、Δ j and Ω k are the adjustment coefficients of the indicators, is the abnormal expression of indicator i, represents the dynamic change amount of the physiological indicator, represents the clinical - related risk degree of each indicator, β i 、γ j and δ k are the weight exponents, N, M, and L are the number of indicators. Specifically, Represents the degree of physiological index abnormality. Calculate the abnormality degree of each physiological index relative to the normal range. A larger value indicates that the index may be abnormal. Its expression is X i (t) is the current physiological index value of the patient, and X norm,i is the normal reference value of the index, Represents the dynamic change of physiological index, which is used to evaluate the fluctuation of physiological index. If the index fluctuates violently, there may be potential risks. Its expression is Calculate the time change rate of physiological index, Represents the clinical relevant risk degree. According to medical experience, different risk weights are set for different physiological indexes to ensure that high-risk indexes have a greater impact on the calculation results. Its expression is X thr,k is the danger threshold of this index. If is too small, it means that this index has exceeded the safe range and may lead to a higher infection risk. The patient's physiological abnormality degree combines the index abnormality degree, dynamic change and clinical risk, which is more comprehensive than a single feature. It can accurately identify the patient's physiological abnormality, adapt to the fluctuation of physiological indexes, and can accurately reflect the patient's real-time health status. It is applicable to high-infection-risk scenarios such as burn wards, and provides accurate physiological data analysis capabilities for early infection risk prediction and personalized intervention.

[0067] The present invention is further set that the environmental pollution index is constructed according to the ward environmental monitoring data, and its calculation logic is as follows: Among them, E env (t) is the environmental pollution index, A(t) is the air microbial concentration at time t, and A ref is the air microbial concentration under normal conditions, H(t) is the relative humidity at time t, and H thr is the relative humidity under normal conditions, λ1, α1 and β1 are adjustment parameters, PM(t) is the PM2.5 / PM10 particulate matter concentration, T(t) is the room temperature at time t, and T norm is the room temperature under normal conditions, δ1, μ1 and η1 are adjustment parameters, and CO2(t) is the carbon dioxide concentration at time t, For the carbon dioxide concentration within the safe range, ρ1, ξ1, and θ1 are adjustment parameters, γ1 and κ1 are weight coefficients. Specifically, the environmental pollution index quantifies the impact of ward air pollution on the infection risk through comprehensive analysis of various environmental monitoring parameters. The core logic includes the air microorganism concentration impact factor, which measures the abnormality of the microorganism concentration in the ward air; the particulate matter pollution factor, as PM2.5 / PM10 particulate matter may carry pathogens and affect air quality; the carbon dioxide concentration change factor, as CO2 reflects the ventilation situation in the ward, and poor ventilation may lead to the accumulation of pathogens. The air microorganism concentration A(t) monitors the concentration of bacteria and viruses in the air, and exceeding the standard may increase the infection risk. Calculate the degree of air pollution. Calculate the impact of particulate matter pollution. Calculate the CO2 impact, which reflects the ventilation situation in the ward. High CO2 may mean poor air circulation and easy accumulation of pathogens. The environmental pollution index comprehensively considers air microorganisms, particulate matter, and ventilation conditions, which is more comprehensive than a single pollution index and improves the accuracy of risk assessment. By introducing humidity, temperature, and CO2 adjustment factors, the false alarm rate is reduced and the stability of the environmental pollution index is improved.

[0068] The present invention is further configured such that the deviation degree of medical staff behavior is constructed based on the operation behavior data of medical staff, and its calculation logic is: Wherein, D care (t) is the deviation degree of medical staff behavior, W(t) is the hand hygiene compliance rate, C(t) is the nursing operation time, S(t) is the amount of alcohol disinfection, G(t) is the glove replacement frequency, G norm , C thr and W ref are operation indicators under normal conditions, α, β, δ, μ, λ, and η are adjustment parameters, and γ is a weight coefficient. Specifically, the deviation degree of medical staff behavior is used to measure whether the operation behavior of medical staff conforms to the infection prevention and control standards and quantifies the impact of human factors on the infection risk. It is used to measure the impact of hand hygiene compliance and nursing duration. Insufficient hand hygiene compliance rate or too long nursing operation time may increase the risk of cross-infection. Therefore, the relationship between the two is calculated, and the impact is strengthened through non-linear transformation. It is used to measure the impact of the amount of disinfectant used and the glove replacement frequency. Insufficient alcohol disinfection or untimely glove replacement may lead to hand contamination, thus increasing the probability of infection transmission. Therefore, the interactive impact of the two is calculated. The deviation degree of medical staff behavior accurately quantifies the deviation degree of medical staff behavior through multi-dimensional factors such as hand hygiene, nursing duration, disinfection, and glove replacement, and uses an exponential function to amplify the impact of abnormal behaviors, making it more sensitive to severely deviated behaviors and improving the detection accuracy.

[0069] The present invention is further configured such that the patient infection risk assessment model is constructed, including:

[0070] Obtain the multi-modal signals after historical synchronization and the corresponding patient infection risks, construct infection risk features based on the multi-modal signals after historical synchronization, and set the infection risk features and the corresponding patient infection risk indices as a data set;

[0071] Divide the said data set into a training set and a validation set;

[0072] Use the training set to train a long short-term memory network. When the loss function converges or reaches the maximum number of iterations, the training is completed. Use the validation set to verify the trained long short-term memory network. The input of the long short-term memory network is the infection risk feature, and the output is the real-time patient infection risk index;

[0073] Set the long short-term memory network that passes the verification as the patient infection risk assessment model.

[0074] The present invention is further configured that the said infection risk index has a calculation formula of wherein, R risk (t) is the infection risk index, D care (t) is the deviation degree of medical staff behavior, E env (t) is the environmental pollution index, F(t) is the physiological abnormality degree of the patient, σ(·) is the activation function, and α3, β3, γ3, λ3, δ3, and μ3 are adjustment parameters. Specifically, the numerator part of the infection risk index formula is the risk accumulation term, which calculates the individual impacts of medical staff behavior, environmental pollution, and patient physiological abnormality, and uses exponential amplification to enhance the sensitivity to high-risk values and avoid the possible underestimation of risks caused by simple linear weighting. The denominator part is the risk interaction term, which calculates the interactions between different factors. represents the impact of medical staff behavior on environmental pollution, represents the impact of environmental pollution on patient physiological abnormality. The activation function (·) is used to perform non-linear normalization on the final risk index to ensure that the output value is between 0 and 1, making it more interpretable and applicable to the risk warning system. Combining medical staff behavior, environmental pollution, and patient physiological status, it comprehensively evaluates the infection risk. Using a non-linear amplification strategy, it ensures that high-risk factors will not be masked by the average effect and improves the abnormal recognition ability. Calculating the interactions considering different factors prevents a single factor from overly dominating the evaluation result.

[0075] The present invention is further configured that a multi-level warning mechanism is set according to the infection risk index, including:

[0076] Set a multi-level warning mechanism according to the infection risk index, and the warning levels are divided into: Among them, τ1 and τ2 are preset risk thresholds. When the early warning level is low risk, normal monitoring is carried out without special intervention; when the early warning level is medium risk, the monitoring frequency is increased to remind medical staff to optimize operations; when the early warning level is high risk, an immediate warning is issued and interventions are taken.

[0077] Embodiment 2

[0078] Please refer to Figure 2 , the exemplary intelligent early warning system for preventing and controlling infections in a burn ward based on big data includes:

[0079] A multi-modal data collection module that collects patients' physiological indicators, ward environment monitoring data, and medical staff's operation behavior data;

[0080] A data preprocessing module that performs outlier detection, denoising, and normalization on the collected data;

[0081] A feature construction module that extracts infection risk features based on the preprocessed data;

[0082] A dynamic infection risk assessment module that uses a multi-modal feature learning model to construct a patient infection risk assessment model based on the infection risk features, with the infection risk features as the input and the patient infection risk index as the output;

[0083] An intelligent early warning module that sets a multi-level early warning mechanism based on the infection risk index.

[0084] It should be noted that the intelligent early warning system for preventing and controlling infections in a burn ward based on big data provided in the above embodiment belongs to the same concept as the intelligent early warning system for preventing and controlling infections in a burn ward based on big data provided in the above embodiment. The specific ways in which each module and unit perform operations have been described in detail in the method embodiment and will not be repeated here. In practical applications, the intelligent early warning system for preventing and controlling infections in a burn ward based on big data provided in the above embodiment can, according to needs, allocate the above functions to different functional modules, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above. This will not be limited here either.

[0085] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0086] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood with reference to the context.

[0087] In this application, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0088] It should be understood that in various embodiments of the present application, the magnitude of the sequence numbers of the above processes does not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0089] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0090] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0091] In several embodiments provided in this application, it should be understood that the disclosed systems can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0092] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0093] In addition, the functional units in each embodiment of this application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0094] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0095] As described above, the above are only specific implementation manners of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. An intelligent early warning system for infection prevention and control in a burn ward based on big data, characterized in that, It includes: A multi-modal data acquisition module that acquires patients' physiological indicators, ward environmental monitoring data, and medical staff's operation behavior data; A data preprocessing module that performs outlier detection, denoising, and normalization on the acquired data; A feature construction module that extracts infection risk features based on the preprocessed data; A dynamic infection risk assessment module that constructs a patient infection risk assessment model using a multi-modal feature learning model based on the infection risk features, with the infection risk features as the input and the patient infection risk index as the output; An intelligent early warning module that sets up a multi-level early warning mechanism based on the infection risk index.

2. The intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 1, wherein The multi-modal data acquisition module includes: A patient physiological indicator acquisition unit that acquires patients' physiological indicators; An environmental monitoring unit that acquires ward environmental monitoring data; A medical staff operation behavior monitoring unit for recording medical staff's operation behavior data.

3. An intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 1, characterized in that, The data preprocessing module includes: Data outlier detection, construct a dynamic time series change function, and evaluate whether the data at the current moment exceeds the normal fluctuation range according to the change trend of the data itself. The formula is: Among them, A t Indicates whether it is an outlier at the t-th moment, 1 means outlier, and 0 means normal. Is the change factor at the t-th moment, γ t Is the dynamic sensitivity threshold at the t-th moment, and its calculation formula is: α is the adaptive adjustment coefficient; Data denoising, using an adaptive denoising method, constructs a composite denoising model, and its calculation logic is as follows: is the denoised signal, is the signal smoothing operation, is the feature function, and its calculation logic is as follows: τ i is the adjustment parameter; Data normalization, using the adaptive non - linear mapping normalization method, dynamically performs normalization according to the data distribution characteristics adaptively. Its formula is: Among them, is the normalized data, α i is the adjustment coefficient for each dimension, β is the non - linear mapping exponent, and X t-i is the data at the past moment.

4. An intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 1, characterized in that, A feature construction module that extracts infection risk features based on the preprocessed data, including: Constructing a patient physiological abnormality degree based on the patients' physiological indicators, where the patients' physiological indicators include: body temperature, heart rate, respiratory rate, blood oxygen saturation, white blood cell count, C-reactive protein, and lactate concentration; Constructing an environmental pollution index based on the ward environmental monitoring data, where the ward environmental monitoring data includes: air microbial concentration, PM2.5 / PM10 particulate matter concentration, carbon dioxide concentration, relative humidity, and room temperature; Constructing a medical staff behavior deviation degree based on the medical staff's operation behavior data, where the medical staff's operation behavior data includes: hand hygiene compliance rate, nursing operation time, alcohol disinfection amount, and glove replacement frequency.

5. An intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 4, wherein, Construct the physiological abnormality degree of a patient based on the patient's physiological indicators. The calculation formula is as follows: Λ i 、Δ j and Ω k are the adjustment coefficients of the indicators, is the abnormality expression of indicator i, represents the dynamic change amount of the physiological indicators, represents the clinical related risk degree of each indicator, β i 、γ j and δ k are the weight indices, and N, M, and L are the numbers of indicators.

6. The intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 4, wherein, Construct an environmental pollution index based on the ward environmental monitoring data, and its calculation logic is as follows: Among them, E env (t) is the environmental pollution index, A(t) is the air microbial concentration at time t, and A ref is the air microbial concentration under normal conditions, H(t) is the relative humidity at time t, and H thr is the relative humidity under normal conditions, λ1, α1, and β1 are adjustment parameters, PM(t) is the PM2.5 / PM10 particulate matter concentration, T(t) is the room temperature at time t, and T norm is the room temperature under normal conditions, δ1, μ1, and η1 are adjustment parameters, CO2(t) is the carbon dioxide concentration at time t, is the carbon dioxide concentration within the safe range, ρ1, ξ1, and θ1 are adjustment parameters, and γ1 and κ1 are weight coefficients.

7. An intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 4, characterized in that, Construct the deviation degree of medical staff's behavior based on the operation behavior data of medical staff, and its calculation logic is as follows: Among them, D care (t) is the deviation degree of medical staff's behavior, W(t) is the hand hygiene compliance rate, C(t) is the nursing operation time, S(t) is the amount of alcohol disinfection, G(t) is the glove replacement frequency, G norm , C thr and W ref are the operation indicators under normal conditions, α, β, δ, μ, λ and η are adjustment parameters, and γ is the weight coefficient.

8. An intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 1, characterized in that, Constructing a patient infection risk assessment model, including: Obtaining the multi-modal signals after historical synchronization and the corresponding patient infection risks, constructing infection risk features based on the multi-modal signals after historical synchronization, and setting the infection risk features and the corresponding patient infection risk index as a data set; Dividing the data set into a training set and a validation set; Training a long short-term memory network using the training set, and completing the training when the loss function converges or reaches the maximum number of iterations. Using the validation set to verify the trained long short-term memory network. The input of the long short-term memory network is the infection risk feature, and the output is the real-time patient infection risk index; Setting the long short-term memory network that passes the verification as the patient infection risk assessment model.

9. The intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 8, wherein, The infection risk index, whose calculation formula is where R risk (t) is the infection risk index, D care (t) is the deviation degree of medical staff's behavior, E env (t) is the environmental pollution index, F(t) is the physiological abnormality degree of the patient, σ(·) is the activation function, and α3, β3, γ3, λ3, δ3 and μ3 are adjustment parameters.

10. The intelligent early warning system for preventing and controlling infections in a burn ward based on big data according to claim 1, characterized in that, Setting up a multi-level early warning mechanism based on the infection risk index, including: Set up a multi-level early warning mechanism according to the infection risk index, and the early warning levels are divided as follows: Early warning level = where τ1 and τ2 are preset risk thresholds. When the early warning level is low risk, normal monitoring is carried out without special intervention; when the early warning level is medium risk, the monitoring frequency is increased to remind medical staff to optimize operations; when the early warning level is high risk, an immediate warning is issued and interventions are taken.

Images