Hospital infection risk dynamic assessment method based on environmental monitoring data

Abstract

The invention relates to a hospital infection risk dynamic assessment method based on environmental monitoring data, and belongs to the technical field of infection risk assessment, and the method comprises the following steps: S1, collecting environmental parameters of a hospital key area in real time through a distributed Internet of Things sensor network, and synchronously integrating a personnel flow thermodynamic diagram of a UWB positioning system and HIS clinical operation data; s2, constructing a rule base containing 12-dimensional coupling rules; s3, performing risk assessment based on an environment-driven infection propagation tree model EITT, firstly calculating a pathogen exposure dose, then dynamically allocating propagation path weights according to a real-time environment state, and finally generating a risk thermodynamic diagram fused with a space-time dimension; and S4, setting a three-level risk value threshold value, starting corresponding prevention and control measures when the risk value reaches the corresponding threshold value, realizing linkage with the environment regulation and control equipment through an OPC-UA protocol, and performing feedback evaluation on an execution effect. According to the invention, the infection risk can be dynamically monitored, response levels can be automatically divided according to infection conditions, and a coping strategy is given.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of infection risk assessment, and in particular to a hospital infection risk dynamic assessment method based on environmental monitoring data. BACKGROUND

[0002] As one of the core challenges in the field of medical safety, the prevention and control effect of hospital infection is directly related to the patient cure rate, medical quality and medical cost. In recent years, with the intensification of drug-resistant bacteria transmission, the popularization of invasive medical operations and the frequent occurrence of public health emergencies, the limitations of traditional infection risk assessment technology have become increasingly prominent.

[0003] The current mainstream hospital infection risk assessment method (such as manual account statistics and monthly report analysis) relies on the periodic aggregation of historical data, and the evaluation period usually lasts for several weeks or even months, which is often discovered through retrospective analysis after the occurrence of infection events, resulting in delayed warning and missed optimal intervention window. According to the data of China Hospital Infection Management Annual Report, under the static assessment mode, the average identification time of hospital infection outbreak event is 72 hours, which is more than 48 hours behind the ideal intervention time.

[0004] The existing technical solution stops at the output of risk assessment results, and does not form a dynamic feedback link with the hospital environment control equipment such as intelligent fresh air system, ultraviolet disinfection robot and automatic hand sanitizer, so that even if high-risk areas are found through evaluation, the prevention and control equipment needs to be started by manual operation, which exists response delay SUMMARY The purpose of the present application is to solve the problems in the background art, and to provide a hospital infection risk dynamic assessment method based on environmental monitoring data, which can dynamically monitor the infection risk and automatically divide the response level according to the infection situation, and give the coping strategy.

[0005] The technical scheme of the present application: a hospital infection risk dynamic assessment method based on environmental monitoring data, comprising the following steps: S1, real-time acquisition of environmental parameters of key areas in the hospital through a distributed Internet of Things sensor network, synchronous integration of personnel flow heat map of UWB positioning system and HIS clinical operation data; S2, cleaning, denoising and missing value filling of the collected multi-source data, time and space alignment of environmental data and clinical events by using dynamic time warping algorithm DTW and Gaussian process regression, and construction of a rule base containing 12-dimensional coupling rules; S3, risk assessment based on environment-driven infection transmission tree model EITT, first calculating the pathogenic exposure dose, then dynamically allocating the transmission path weight according to the real-time environmental state, and finally generating a risk heat map integrating time and space dimensions; S4, set a three-level risk value threshold, when the risk value reaches the corresponding threshold, start the corresponding prevention and control measures, realize linkage with the environmental control equipment through OPC-UA protocol, and feedback evaluation on the execution effect.

[0006] Preferably, the distributed Internet of Things sensor network adopts a redundant design, and automatically switches to a backup node when a single node fails; the environmental parameters include temperature and humidity, CO2 concentration, ≥0.3 μm particle concentration, high-frequency contact surface ATP value, and negative pressure room pressure difference, the data acquisition frequency is 1 Hz, and the accuracy of the environmental parameters includes temperature and humidity ±0.5℃, CO2 concentration ±50ppm, high-frequency contact surface ATP value detection limit 5RLU, and negative pressure room pressure difference range ±60Pa.

[0007] Preferably, the 12-dimensional coupling rule base is constructed by combining machine learning with expert experience, and when the personnel density increases by 10%, the surface disinfection frequency needs to be ≥2 times / hour.

[0008] Preferably, the propagation path weight is adjusted according to the change of the environmental parameter threshold, and the calculation formula of the pathogenic exposure dose is: Dose=∫[C_air·V_inhale+C_surface·S_contact]dt Wherein, V_inhale is calculated by infrared thermal imaging, S_contact is a contact area parameter.

[0009] Preferably, the adjustment mechanism of the propagation path weight includes that when the negative pressure room pressure difference is <5Pa, the air transmission weight is increased from 0.45 to 0.92.

[0010] Preferably, the accuracy of the risk thermodynamic diagram is 1m×1m, which can intuitively show the infection risk distribution of each area.

[0011] Preferably, the three-level risk value threshold and the corresponding prevention and control measures are: when R∈[0.6, 0.8), push the warning information to the regional responsible nurse terminal and attach the intervention suggestion; when R∈[0.8, 0.9), automatically dispatch the disinfection robot to the target area to perform spot disinfection; when R≥0.9, block the high-risk area and start whole-house fumigation disinfection, and trigger the emergency response process of the hospital infection department.

[0012] Preferably, the environmental control equipment includes an ultraviolet disinfection robot and an intelligent fresh air system, and when the ultraviolet disinfection robot performs disinfection, the intelligent fresh air system can synchronously increase the number of air changes.

[0013] Preferably, the feedback evaluation on the execution effect in step S4 includes collecting verification data at three time nodes of 5 minutes, 30 minutes and 2 hours after the intervention, and the verification data includes air bacterial colony number and surface ATP value.

[0014] Preferably, during the period of public health emergencies, the response threshold is automatically lowered, and the monitoring range is expanded to fever clinics, emergency buffer zones and other areas, and the infection diffusion path is predicted through multi-region risk heat map comparison analysis.

[0015] Compared with the prior art, the application has the following beneficial technical effects: In the application, full-dimensional data input is realized through a distributed sensor network and multi-system integration, information barriers are broken through data processing technology, dynamic risk quantification is realized with the aid of an EITT model, and finally a prevention and control closed loop is formed through three-level response and equipment linkage. Each link is connected, forming a complete technical chain from monitoring to intervention, solving the problems of lagging of traditional evaluation methods and data island, realizing dynamic management of the whole process of hospital infection risk, providing systematic technical support for active prevention and control, and improving the real-time performance and accuracy of infection risk assessment as a whole. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 Structure schematic diagram of the embodiment in the application DETAILED DESCRIPTION In order to make the above-mentioned purposes, features and advantages of the application more apparent and easy to understand, the specific embodiments of the application will be described in detail below with reference to the drawings of the specification.

[0018] In the following description, many specific details are set forth in order to provide a thorough understanding of the application, but the application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the application, therefore the application is not limited to the specific embodiments disclosed below.

[0019] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the application. In this specification, "in one embodiment" does not mean the same embodiment, nor does it mean that the embodiment is single or selectively excluded from other embodiments.

[0020] Thirdly, the application is described in detail in combination with the schematic diagram, in the detailed description of the embodiments of the application, for the convenience of description, the sectional view of the device structure will be partially enlarged without the general proportion, and the schematic diagram is only an example, which should not limit the scope of protection of the application herein. In addition, the three-dimensional spatial dimensions including length, width and depth should be included in actual production.

[0021] Embodiment one As Figure 1 shown, the hospital infection risk dynamic assessment method based on environmental monitoring data proposed by the application includes the following steps: S1, real-time collection of environmental parameters of key areas in the hospital through a distributed Internet of Things sensor network, synchronous integration of personnel flow heat maps of a UWB positioning system and HIS clinical operation data; S2, cleaning, denoising and missing value filling of the collected multi-source data, realization of spatio-temporal alignment of environmental data and clinical events by using a dynamic time warping algorithm DTW and a Gaussian process regression, and construction of a rule base containing 12-dimensional coupling rules; S3, risk assessment based on an environment-driven infection transmission tree model EITT, calculation of pathogen exposure dose, dynamic allocation of transmission path weight according to real-time environmental state, and finally generation of a risk heat map integrating the spatio-temporal dimension; S4, setting of three-level risk value thresholds, starting of corresponding prevention and control measures when the risk value reaches the corresponding threshold, linkage with environmental control equipment through an OPC-UA protocol, and feedback evaluation of the execution effect.

[0022] The distributed Internet of Things sensor network adopts a redundant design, and automatically switches to a backup node when a single node fails; the environmental parameters include temperature and humidity, CO2 concentration, concentration of particles greater than 0.3 microns, ATP value of high-frequency contact surface and pressure difference of negative pressure ward, the data collection frequency is 1Hz, and the accuracy of the environmental parameters includes temperature and humidity ±0.5℃, CO2 concentration ±50ppm, ATP value detection limit of high-frequency contact surface 5RLU, and negative pressure ward pressure difference range ±60Pa.

[0023] The introduction of a redundant design sensor network in the data collection link ensures seamless switching in case of failure, ensures data continuity, 1Hz high-frequency sampling and accurate parameter accuracy, ensures that environmental transient changes such as pressure difference fluctuations when the operating room door is opened can be captured, provides a high-quality data basis for subsequent evaluation, avoids data interruption caused by sensor failure, and improves the reliability of data collection; high-precision parameters ensure that environmental changes are accurately captured, solve the evaluation error problem caused by traditional extensive collection, and provide accurate input for risk quantification.

[0024] 12-dimensional coupling rule base is constructed by combining machine learning and expert experience. When the personnel density increases by 10%, the surface disinfection frequency needs to be ≥2 times / hour. The construction of the 12-dimensional coupling rule base combines data-driven and expert experience. Strong association rules such as personnel density increasing by 10% and surface contamination rate increasing by 25% are mined from historical data through machine learning such as Apriori algorithm, and then combined with the requirements of hospital infection experts on disinfection specifications such as surface disinfection frequency ≥2 times / hour to form a directly executable decision logic. The 2-dimensional coupling rule base is constructed around the key influencing factors of hospital infection transmission, covering multi-dimensional correlation relationships such as environmental parameters, personnel behavior, clinical operation, etc., including: Personnel density and surface disinfection frequency, CO2 concentration and air exchange frequency, temperature and humidity and pathogen survival time, operation time and air bacterial colony count, high-frequency contact surface ATP value and disinfection intensity, negative pressure ward pressure difference and personnel protection level, ≥0.3 μm particle concentration and laminar purification level, patient transfer terminal disinfection time, number of visitors and disinfection interval, instrument use frequency and sterilization cycle, ultraviolet disinfection time and environmental illumination, and fresh air system operation time and filter replacement cycle. Based on historical infection data of the hospital, the support (≥8%) and confidence (≥90%) between each factor are calculated through association rule algorithm, and the rules that are significant in data but meaningless in clinic are removed through expert verification of data mining results. Finally, the 12-dimensional coupling rule base is obtained. When the risk assessment triggers an alarm, the rule base directly outputs the matching intervention scheme, avoiding the subjectivity of manual decision-making. Through quarterly iteration and emergency rule supplement, it can quickly respond to new infection risks such as drug-resistant bacteria transmission and sudden infectious diseases, and solve the problem of insufficient scene adaptation of traditional static rule base.

[0025] In this embodiment, full-dimensional data input is achieved through distributed sensor network and multi-system integration, information barriers are broken through data processing technology, dynamic risk quantification is achieved with the help of EITT model, and finally a prevention and control closed loop is formed through three-level response and equipment linkage. Each link is closely connected to form a complete technical chain from monitoring to intervention, solving the problems of lagging of traditional evaluation methods and data island, and realizing the dynamic management of hospital infection risk throughout the whole process. It provides systematic technical support for active prevention and control, and improves the real-time and accuracy of infection risk assessment as a whole.

[0026] Embodiment two As shown in Figure 1 The hospital infection risk dynamic evaluation method based on environmental monitoring data proposed by the application adjusts the transmission path weight according to the change of the environmental parameter threshold, and the calculation formula of the pathogen exposure dose is: Dose=∫[C_air·V_inhale+C_surface·S_contact]dt Wherein, V_inhale is calculated by infrared thermal imaging, S_contact is a contact area parameter.

[0027] The adjustment mechanism of the transmission path weight includes that when the negative pressure room pressure difference is less than 5 Pa, the air transmission weight is increased from 0.45 to 0.92.

[0028] The precision of the risk heat map is 1 m*1 m, and the infection risk distribution of each area can be intuitively displayed. The risk heat map is based on the inverse distance weighted (IDW) interpolation algorithm, and the risk value R of the discrete evaluation point is calculated to the 1 m*1 m grid. Through color mapping, such as red representing R>=0.8, the spatio-temporal visualization of risk is realized, the distribution and change of the high-risk area are intuitively displayed, the hospital infection personnel can quickly locate the high-risk area such as the CO2 over-standard area of the outpatient waiting area, the problem that the traditional table or static chart is difficult to accurately locate is solved, and the regional control efficiency is improved.

[0029] In the embodiment, the pathogen exposure dose calculation is based on the dose-response theory, and the pathogen exposure amount of air and contact transmission is dynamically accumulated through the integral formula. Among them, C_air air pathogen concentration is calculated by the calibrated optical particle counter, V_inhale inhalation volume is calculated based on the respiratory mechanics model of infrared thermal imaging, and S_contact contact area is preset according to the object type, realizing the quantitative integration of multi-path exposure, infection risk conversion, breaking through the limitation of traditional qualitative evaluation, and making the risk assessment more scientific; meanwhile, considering the double paths of air and contact transmission, the real infection risk is comprehensively reflected.

[0030] The transmission path weight adopts a dynamic adjustment mechanism, and the mapping of environmental parameters and weights is realized based on fuzzy logic reasoning. For example, when the negative pressure room pressure difference is less than 5 Pa, the air transmission barrier is invalid, and through a triangular membership function, the parameter state is mapped to a weight value from 0.45 to 0.92, so that the evaluation model adapts to real-time environmental changes, solves the problem that the fixed weight of the traditional model cannot reflect the dynamic influence of the environment, makes the risk assessment more in line with the actual transmission scene, and improves the accuracy of the evaluation results under different environmental conditions.

[0031] Embodiment three As shown in Figure 1 Compared with embodiment one or embodiment two, in the hospital infection risk dynamic evaluation method based on environmental monitoring data provided by the application, the three-level risk value threshold and the corresponding prevention and control measures are: when R belongs to [0.6, 0.8), the warning information is pushed to the terminal of the regional responsible nurse and the intervention suggestion is attached; when R belongs to [0.8, 0.9), the disinfection robot is automatically dispatched to the target area to perform spot sterilization and disinfection; and when R is greater than or equal to 0.9, the high-risk area is blocked and whole-house fumigation disinfection is started, and the emergency response process of the hospital infection department is triggered.

[0032] The environmental regulation device comprises an ultraviolet disinfection robot and an intelligent fresh air system, and the intelligent fresh air system can synchronously increase the air exchange frequency when the ultraviolet disinfection robot performs disinfection.

[0033] The feedback evaluation of the execution effect in step S4 comprises collecting verification data at three time nodes of 5 minutes, 30 minutes and 2 hours after the intervention, and the verification data comprises air bacterial colony counts and surface ATP values.

[0034] Further comprising automatically reducing the response threshold and expanding the monitoring range to areas such as fever clinics and emergency buffer zones during the period of public health emergencies, and comparing and analyzing the risk heat maps of multiple areas to predict the infection spread path.

[0035] In this embodiment, the three-level risk response mechanism avoids the resource mismatch problem of the traditional binary alarm high risk / low risk, and according to the risk level, the prevention and control resources are invested in stages, which not only ensures that the high-risk areas are strongly intervened, but also avoids the over-prevention and control of the low-risk areas, and improves the utilization efficiency of prevention and control resources.

[0036] The equipment cooperates to improve the prevention and control effect, such as quickly reducing the concentration of air pathogens by timely air exchange after disinfection; at the same time, the effect of independent operation of the equipment is avoided, such as the residual pathogens caused by not opening the fresh air during disinfection, which improves the synergy and effectiveness of the prevention and control measures.

[0037] The execution effect feedback realizes closed-loop optimization through multi-time node verification, and the immediate effect at 5 minutes after intervention, the short-term effect at 30 minutes, and the continuous effect at 2 hours are collected. Air bacterial colony counts and surface ATP values are collected, the data is input into the evaluation model after standardization, the risk reduction rate RRR is calculated, the effectiveness of the prevention and control measures is verified in real time, the blindness after intervention is avoided, and the measures are iteratively optimized through continuous feedback, such as adjusting the disinfection time, forming a virtuous cycle of evaluation-execution-optimization, and continuously improving the prevention and control effect.

[0038] The emergency public health event response mechanism is realized through dynamic threshold adjustment and monitoring range expansion, improves the adaptability of the system to special scenes, realizes early warning and early intervention in the event of an epidemic and other emergencies, expands the monitoring range and path prediction function, provides data support for emergency command, and reduces the risk of large-scale infection spread The embodiments of the application are described in detail above in combination with the drawings, but the application is not limited thereto, and various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the purpose of the application.

Claims

1. A method for dynamically assessing hospital infection risk based on environmental monitoring data, characterized by: The following steps are involved: S1. Real-time collection of environmental parameters in key areas of the hospital through a distributed IoT sensor network, synchronously integrating the personnel flow heat map of the UWB positioning system with HIS clinical operation data; S2. Clean, denoise, and fill missing values ​​in the collected multi-source data. Use the dynamic time warping algorithm (DTW) and Gaussian process regression to achieve spatiotemporal alignment between environmental data and clinical events, and build a rule library containing 12-dimensional coupling rules. S3. Perform risk assessment based on the environment-driven infection transmission tree model (EITT). First, the pathogen exposure dose is calculated, then the transmission path weight is dynamically assigned based on the real-time environmental status, and finally a risk heat map is generated that integrates the spatiotemporal dimensions. S4. Set three-level risk value thresholds. When the risk value reaches the corresponding threshold, initiate corresponding prevention and control measures, realize linkage with environmental control equipment through the OPC-UA protocol, and provide feedback and evaluation on the execution effect.

2. A hospital infection risk dynamic assessment method based on environmental monitoring data according to claim 1, characterized in that: The distributed IoT sensor network adopts a redundant design and automatically switches to a backup node when a single node fails. Environmental parameters include temperature and humidity, CO2 concentration, particle concentration ≥0.3μm, ATP value of high-frequency contact surfaces and negative pressure ward pressure difference. The data acquisition frequency is 1Hz. The accuracy of environmental parameters includes temperature and humidity ±0.5℃, CO2 concentration ±50ppm, high-frequency contact surface ATP value detection limit 5RLU, and negative pressure ward pressure difference range ±60Pa.

3. A hospital infection risk dynamic assessment method based on environmental monitoring data according to claim 2, characterized in that: The 12-dimensional coupling rule base is constructed by combining machine learning with expert experience. When the population density increases by 10%, the surface disinfection frequency needs to be ≥2 times / hour.

4. A method for dynamic assessment of hospital infection risk based on environmental monitoring data according to claim 3, characterized in that: The transmission path weight is adjusted as the environmental parameter threshold changes, and the formula for calculating the pathogen exposure dose is: Dose=∫[C_air·V_inhale+C_surface·S_contact]dt Among them, V_inhale is calculated by infrared thermal imaging, S_contact is the contact area parameter.

5. A hospital infection risk dynamic assessment method based on environmental monitoring data according to claim 4, characterized in that: The adjustment mechanism of the transmission path weight includes increasing the airborne transmission weight from 0.45 to 0.92 when the pressure difference in the negative pressure ward is <5Pa.

6. A method for dynamic assessment of hospital infection risk based on environmental monitoring data according to claim 5, characterized in that: The accuracy of the risk heat map is 1m×1m, which can intuitively display the infection risk distribution in each area.

7. A method for dynamic assessment of hospital infection risk based on environmental monitoring data according to claim 6, characterized in that: The third-level risk value threshold and corresponding prevention and control measures are: when R∈[0.6,0.8), the warning information is pushed to the terminal of the regional responsible nurse with intervention suggestions; when R∈[0.8,0.9), the disinfection robot is automatically dispatched to the target area to perform fixed-point disinfection; When R≥0.9, the high-risk area will be blocked and whole-house fumigation disinfection will be initiated, and the hospital infection department's emergency response process will be triggered.

8. A method for dynamic assessment of hospital infection risk based on environmental monitoring data according to claim 7, characterized in that: Environmental control equipment includes ultraviolet disinfection robots and intelligent fresh air systems. When the ultraviolet disinfection robots are performing disinfection, the intelligent fresh air system can simultaneously increase the ventilation rate.

9. A method for dynamic assessment of hospital infection risk based on environmental monitoring data according to claim 8, characterized in that: The feedback evaluation of the implementation effect in step S4 includes collecting verification data at three time points: 5 minutes, 30 minutes, and 2 hours after the intervention. The verification data includes the number of airborne colonies and the surface ATP value.

10. A method for dynamic assessment of hospital infection risk based on environmental monitoring data according to claim 9, characterized in that: It also includes automatically lowering the response threshold during public health emergencies, expanding the monitoring scope to areas such as fever clinics and emergency buffer zones, and predicting the infection spread path through comparative analysis of multi-region risk heat maps.

Images