Big data-based laminar ward risk assessment system for hematology department

Abstract

The invention relates to the technical field of laminar ward risk assessment, in particular to a hematology department laminar ward risk assessment system based on big data, and the system comprises an infection propagation prediction module which collects the position and movement data of each person in a ward, serializes the data, calculates the activity mode and interaction frequency of each person, and transmits the data to a database; and carrying out probability prediction on the potential cross infection event, and outputting a potential infection propagation graph. According to the invention, by collecting and analyzing the position and movement data of the personnel in the ward and combining the interaction frequency to evaluate the probability of cross infection, higher-precision infection propagation prediction can be realized, the data of the environment sensor and the health condition of the patient are further integrated, and the air quality, the temperature and humidity and the vital signs of the patient can be dynamically monitored; the association between the environment and the health state is accurately evaluated, the real-time evaluation of the infection risk is effectively supported, the accuracy of risk management is optimized through the identification and weight adjustment of key risk factors, and the safety guarantee of the ward environment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of laminar flow ward risk assessment, and particularly to a risk assessment system for hematology laminar flow wards based on big data. Background Art

[0002] The technical field of laminar flow ward risk assessment involves using advanced data analysis tools and algorithms to monitor and manage environmental and patient safety risks in hospital laminar flow wards. Laminar flow wards, especially those for hematology, require extremely high air purity to reduce the risk of infection and provide a sterile environment. The development in this technical field focuses on real-time data monitoring, optimization and implementation of infection control strategies, and automatic adjustment of environmental control systems. By integrating data from multiple sensors, such as air quality, temperature, humidity monitoring, and information on the health status of patients, the system can assess and warn of potential risks, enabling medical staff to take timely action to ensure patient safety.

[0003] Among them, the risk assessment system for hematology laminar flow wards refers to a specially designed system for assessing possible health and safety risks in hematology laminar flow wards. This system analyzes big data and uses complex algorithms to predict and manage possible infection risks and other related risks. Its main uses include improving patient safety, optimizing the quality of the ward environment, and reducing the occurrence of hospital infections through an early warning mechanism, thereby enhancing the quality of medical services and the treatment effect of patients.

[0004] The existing technologies are relatively passive in risk prediction and environmental regulation, lacking systematic analysis of the activities and interactions of people in the ward, resulting in limited prediction of cross-infection risks. Environmental control mostly relies on adjusting preset parameters and is difficult to respond quickly according to real-time data changes, prone to problems such as being unable to adjust in time due to sudden changes in air quality or humidity, thus increasing the hidden danger of infection transmission. The lack of a precise early warning mechanism leads to the inability of medical staff to intervene in a timely manner when the risk increases, increasing the infection risk of patients in the ward and making it difficult to ensure environmental safety. Summary of the Invention

[0005] The object of the present invention is to solve the drawbacks existing in the prior art, and to propose a risk assessment system for hematology laminar flow wards based on big data.

[0006] To achieve the above object, the present invention adopts the following technical solutions: The risk assessment system for hematology laminar flow wards based on big data includes:

[0007] The infection transmission prediction module collects the location and movement data of each person in the ward, serializes the data, calculates the activity patterns and interaction frequencies of each person, predicts the probability of potential cross-infection events, and outputs a potential infection transmission map;

[0008] Based on the potential infection transmission map, the environmental monitoring integration module collects environmental sensor data and patients' health monitoring data, including air quality, temperature, humidity, and vital signs, mines the correlation between environmental factors and patients' health status, estimates the probability of infection transmission, and outputs real-time environmental risk assessment data;

[0009] The risk matrix optimization module uses the real-time environmental risk assessment data to identify key risk factors, adjusts the weights in the matrix, corresponding to the current environment and patients' health status, and outputs an optimized risk weight matrix;

[0010] The decision support output module calculates the correlation between risk factors and patients' health status in the ward based on the optimized risk weight matrix, analyzes the impact of risk factors on infection transmission, and outputs the assessment results of the impact of risk factors.

[0011] The potential infection transmission map includes activity patterns, interaction frequencies, and cross-infection probabilities. The real-time environmental risk assessment data includes air quality, temperature, humidity, and vital sign correlations. The optimized risk weight matrix specifically includes adjusted weights, environmental parameters, and health status. The assessment results of the impact of risk factors are specifically the correlation between risk factors and patients' health status and the impact on infection transmission.

[0012] As a further solution of the present invention, the step of calculating the activity pattern and interaction frequency of each person is specifically as follows:

[0013] Continuously collect through sensors deployed in the ward to obtain the position data of each person in the ward, record the position coordinates and moving speed of each person in real time, and obtain the original position and movement dataset;

[0014] Based on the original position and movement dataset, serialize the position data. Each data point includes a timestamp, position coordinates, and moving speed. Perform time series analysis on the data to generate a serialized position dataset;

[0015] Analyze the serialized position dataset, calculate the moving frequency and staying time of each person, using the formula:

[0016]

[0017] Generate activity patterns;

[0018] where P i represents the average activity intensity of individual i, d i,t is the moving distance of individual i at time t, and T is the total monitoring time;

[0019] Based on the activity pattern, using the formula:

[0020]

[0021] Generate interaction frequency data by calculating the average interaction frequency between individuals i and j;

[0022] Among them, F ij represents the average interaction frequency between individuals i and j, x i,t and y i,t are coordinate positions, and N is the total number of time points monitored.

[0023] As a further solution of the present invention, the steps for obtaining the potential infection transmission map are specifically as follows:

[0024] Extract the interaction frequency of each pair of individuals from the interaction frequency data, combine it with the activity pattern data of each pair of individuals, calculate the cross-infection probability between each pair of individuals, and generate a preliminary infection probability matrix;

[0025] With reference to the dynamic change of the distance between individuals, refine the preliminary infection probability matrix, and by adjusting the weight of the interaction distance, use the formula:

[0026]

[0027] Calculate the cross-infection probability between individuals i and j, and generate an updated infection probability matrix;

[0028] Among them, P infect (i,j) represents the cross-infection probability between individuals i and j, F ij represents the average interaction frequency between individuals i and j, obtained from the interaction frequency data, indicating the frequency of interaction between the two people, β represents the transmission coefficient, used to adjust the influence of the interaction frequency and distance on the infection probability, D ij is the minimum distance calculated between the two individuals;

[0029] Perform normalization processing on the updated infection probability matrix to ensure that all probability values are between 0 and 1, and perform iterative statistical analysis to generate a normalized infection probability matrix;

[0030] Based on the normalized infection probability matrix, simulate the transmission path of the virus in the personnel network by means of graph theory to generate a potential infection transmission map.

[0031] As a further solution of the present invention, the steps for mining the correlation between environmental factors and the health status of patients are specifically as follows:

[0032] Based on the potential infection transmission map, continuously collect air quality indicators, temperature and humidity data in the ward through environmental sensors, and ensure that each data point is attached with a time stamp to obtain an environmental monitoring data set;

[0033] Based on the potential infection transmission map, synchronously collect the patient's vital sign data, including heart rate, blood pressure, and body temperature. Each data point is attached with a timestamp. Align the vital sign data with the environmental data set by comparing the timestamps to obtain the patient health data set;

[0034] Conduct statistical analysis on the environmental monitoring data set and the patient health data set, using the correlation analysis formula:

[0035]

[0036] Calculate the correlation coefficient between the environmental variables and the vital signs, and generate the environmental and health correlation analysis result;

[0037] Among them, E k represents the environmental monitoring value at the k-th collection time point, and H k represents the patient health monitoring value at the corresponding time point. and are the averages of the environmental and health monitoring values at all collection time points, and n is the total number of collection time points.

[0038] As a further solution of the present invention, the steps for obtaining the real-time environmental risk assessment data are specifically as follows:

[0039] Extract the correlation coefficient from the environmental and health correlation analysis result, evaluate the impact of each parameter based on the real-time monitored environmental data, calculate the risk contribution degrees of temperature and humidity, and obtain the environmental parameter risk contribution data;

[0040] Combine the environmental parameter risk contribution data and the updated patient health status data, and apply the formula:

[0041]

[0042] Calculate to obtain the infection transmission probability data;

[0043] Among them, P infect represents the probability of infection transmission, calculates the infection risk under the target environmental factors and patient health status, α represents the weight of the correlation coefficient C between the environment and health data, and the weight adjusts the influence degree of the correlation between the environmental data and the health status on the infection probability. β i is the weight coefficient of the i-th environmental parameter, and the coefficient adjusts the contribution degrees of multiple environmental parameters including temperature, humidity, etc. in the infection risk assessment. R i represents the risk contribution value of the i-th environmental parameter, measures the direct contribution of the environmental parameter to the infection risk, and m is the number of reference environmental parameters;

[0044] Based on the infection transmission probability data, comprehensively referring to the current personnel density and movement patterns in the ward, real-time environmental risk assessment data is generated.

[0045] As a further solution of the present invention, the steps for obtaining the optimized risk weight matrix are specifically as follows:

[0046] Extract key risk factors from the real-time environmental risk assessment data, analyze the degree of impact of each factor on the health of patients, call the correlation parameters of multiple risk factors for calculation, obtain the initial risk weight of each factor, and obtain the initial risk weight list;

[0047] Based on the initial risk weight list, dynamically adjust the risk weights according to the current environmental data and the health status of patients, call the adjustment coefficient and the original risk weight for calculation, execute the weighted model, and use the formula:

[0048]

[0049] Calculate the risk weight data to generate an adjusted risk weight data set;

[0050] Among them, W i is the adjusted weight of the i-th risk factor, R i is the original risk weight, γ i is the adjustment coefficient based on the environmental data, H i is the influence degree of the health data, δ is the adjustment coefficient based on the health status, and η is the adjustment flexibility coefficient;

[0051] Integrate the adjusted risk weight data into the risk assessment matrix, call the risk weight data and the risk assessment matrix for combination, update the matrix structure to reflect the actual risk of the current environment and the health status of patients, and output the optimized risk weight matrix.

[0052] As a further solution of the present invention, the steps for obtaining the evaluation result of the influence of the risk factor are specifically as follows:

[0053] Call the optimized risk weight matrix, use statistical methods to calculate the correlation between multiple risk factors in the matrix and the health status data of patients, and use the formula:

[0054]

[0055] Generate the correlation data between the risk factor and the health status;

[0056] Among them, C i represents the correlation score of the i-th risk factor, w ij is the weight of the i-th risk factor in the matrix for the j-th health index, and h j is the actual value of the j-th health index;

[0057] Analyze the data on the correlation between the risk factors and the health status, determine which risk factors have a significant impact on the health status and infection transmission of patients, perform calculations to obtain the contribution degree of each risk factor to infection transmission, and use the formula:

[0058] I i = λ·C i

[0059] Generate a preliminary assessment of the impact of risk factors on infection transmission;

[0060] where, I i represents the impact score of the i-th risk factor on infection transmission, and λ is the standardization coefficient;

[0061] Integrate the preliminary assessments of the impacts of all the risk factors on infection transmission, and perform a weighted average according to the criticality of each score to obtain the risk factor impact assessment result.

[0062] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0063] In the present invention, by collecting and analyzing the position and movement data of the personnel in the ward and combining the interaction frequency to evaluate the probability of cross-infection, it is helpful to achieve a higher-precision prediction of infection transmission. Further integrating the environmental sensor data and the patient's health status can dynamically monitor the air quality, temperature and humidity, and the patient's vital signs, accurately evaluate the correlation between the environment and the health status, and effectively support the real-time assessment of the infection risk. By identifying and adjusting the weights of the key risk factors, the accuracy of risk management is optimized, and the safety guarantee of the ward environment is improved. Integrating this series of dynamic regulation means, the system provides an efficient early warning mechanism, significantly reduces the infection incidence rate, and improves the patient treatment effect and medical quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is the system flowchart of the present invention;

[0065] Figure 2 is the step flowchart for calculating the activity pattern and interaction frequency of each person in the present invention;

[0066] Figure 3 is the step flowchart for obtaining the potential infection transmission map of the present invention;

[0067] Figure 4 is the step flowchart for mining the correlation between environmental factors and the patient's health status in the present invention;

[0068] Figure 5 is the step flowchart for obtaining the real-time environmental risk assessment data of the present invention;

[0069] Figure 6 Flow chart of the acquisition steps of the optimized risk weight matrix of the present invention;

[0070] Figure 7 Flow chart of the acquisition steps of the evaluation result of the influence of the risk factors of the present invention. Detailed implementation manners

[0071] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0072] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more unless otherwise specifically defined.

[0073] Embodiment:

[0074] Please refer to Figure 1 , the laminar flow ward risk assessment system based on big data includes:

[0075] The infection transmission prediction module collects the position and movement data of each person in the ward, serializes the data, calculates the activity patterns and interaction frequencies of each person, predicts the probability of potential cross-infection events, and outputs a potential infection transmission map;

[0076] The environmental monitoring integration module collects environmental sensor data and the health monitoring data of patients based on the potential infection transmission map, including air quality, temperature, humidity and vital signs, mines the correlation between environmental factors and the health status of patients, estimates the probability of infection transmission, and outputs real-time environmental risk assessment data;

[0077] The risk matrix optimization module uses the real-time environmental risk assessment data to identify key risk factors, adjusts the weights in the matrix, corresponding to the current environment and the health status of patients, and outputs an optimized risk weight matrix;

[0078] The decision support output module calculates the correlation between the risk factors in the ward and the health status of patients based on the optimized risk weight matrix, analyzes the influence of the risk factors on infection transmission, and outputs the evaluation result of the influence of the risk factors.

[0079] The potential infection transmission map includes activity patterns, interaction frequencies, and cross-infection probabilities. The real-time environmental risk assessment data includes air quality, temperature, humidity, and vital sign correlations. The optimized risk weight matrix specifically includes adjusted weights, environmental parameters, and health conditions. The impact assessment result of risk factors is specifically the correlation between risk factors and the patient's health status and the impact of infection transmission.

[0080] Please refer to Figure 2 , the steps for calculating the activity pattern and interaction frequency of each person are specifically as follows:

[0081] Continuously collect through sensors deployed in the room to obtain the position data of each person in the ward, and record the position coordinates and moving speed of each person in real time to obtain the original position and movement data set;

[0082] Through sensors deployed in the room, continuously collect the position coordinates and moving speed of each person in the ward. The sensors have positioning functions, can record data in real time, and transmit it to the central database in real time through wireless networks. Each data point includes a timestamp, position coordinates, and moving speed. After the data points are serialized, it is convenient for subsequent time series analysis to generate the original position and movement data set, which will be used as the basis for analyzing the activity pattern and interaction frequency of each person.

[0083] Based on the original position and movement data set, serialize the position data. Each data point includes a timestamp, position coordinates, and moving speed. Perform time series analysis on the data to generate a serialized position data set;

[0084] The original position data set is serialized through time series analysis. The key to serialization is to mark each data point with time, including timestamp, position coordinates, and moving speed, so that the data can be arranged and compared according to time order in subsequent analysis. The serialized position data set provides a basis for activity pattern calculation and interaction frequency analysis, ensuring the availability of data and the accuracy of analysis.

[0085] Analyze the serialized position data set and calculate the moving frequency and staying time of each person using the formula:

[0086]

[0087] Generate the activity pattern;

[0088] Among them, P i represents the average activity intensity of individual i, d i,t is the moving distance of individual i at time t, and T is the total monitoring time;

[0089] Formula:

[0090]

[0091] The benefit of the formula is that through the calculation of the average activity intensity, it can effectively quantify the movement characteristics of each person during the monitoring time, providing a standardized activity index for the subsequent calculation of the interaction frequency.

[0092] Detailed explanation of the formula and the derivation process of the formula calculation:

[0093] Set the position of a certain person i within a monitoring time T = 5 minutes, and record the moving distance d i,t per minute as: 2, 3, 2, 4, and 1 meter respectively. Then, according to the formula, the average moving intensity P of this person is calculated i , and the calculation process is as follows:

[0094]

[0095] The result shows that the average moving intensity of the person during the monitoring time is 2.4 meters per minute, which is used as a basic parameter for the subsequent calculation of the interaction frequency to evaluate the activity levels of different individuals.

[0096] Based on the activity pattern, the formula:

[0097]

[0098] is used to calculate the average interaction frequency between persons i and j to generate interaction frequency data;

[0099] where F ij represents the average interaction frequency between individuals i and j, x i,t and y i,t are coordinate positions, and N is the total number of monitoring time points.

[0100] Formula:

[0101]

[0102] The benefit of the formula is that by using the relative positions at each time point, it captures the distance changes between two persons in a weighted form, which can enhance the sensitivity of the interaction frequency and help detect cases of frequent proximity within a short time, reducing misjudgments.

[0103] Detailed explanation of the formula and the derivation process of the formula calculation:

[0104] Set the position information of persons i and j during the monitoring time as:

[0105] x i = [2, 3, 4, 3, 2], y i = [1, 1, 2, 2, 1], x j = [1, 2, 2, 2, 3], y j= [2, 2, 2, 3, 3], and a total of N = 5 data points are recorded within the time period. Calculate item by item according to the formula:

[0106]

[0107] The calculation result is F ij = 0.5175. The result shows that the average interaction frequency between the two people within the specified time helps to evaluate the actual contact opportunities within the monitoring window.

[0108] Please refer to Figure 3 , and the specific steps for obtaining the potential infection transmission map are as follows:

[0109] Extract the interaction frequency of each pair of people from the interaction frequency data, combine it with the activity pattern data of each pair of people, calculate the cross-infection probability between each pair of people, and generate a preliminary infection probability matrix;

[0110] Extract the interaction frequency of each pair of people from the interaction frequency data, calculate the cross-infection probability between each pair of people. First, it is necessary to sort out and analyze the activity pattern data, calculate the encounter frequency and distance between each pair of people under different conditions, determine the action patterns of each person and the dynamic distance between them through the collected data. The data will be used as the basis for calculating the infection probability. Use statistical methods to process the activity pattern data to obtain the encounter probability between each pair of people under different time and space conditions. This probability is accurately calculated by dynamically monitoring the position changes of each person and combining the time stamps, ensuring the accuracy and reliability of the infection probability calculation, and obtaining a preliminary infection probability matrix.

[0111] Refer to the dynamic change of the distance between people, refine the preliminary infection probability matrix, and by adjusting the weight of the interaction distance, use the formula:

[0112]

[0113] Calculate the cross-infection probability between person i and j, and generate an updated infection probability matrix;

[0114] Among them, P infect (i, j) represents the cross-infection probability between person i and j, F ij represents the average interaction frequency between person i and j, obtained from the interaction frequency data, indicating the frequency of interaction between the two people. β represents the transmission coefficient, used to adjust the influence of interaction frequency and distance on the infection probability, and D ij is the minimum distance calculated between the two individuals;

[0115] Formula:

[0116]

[0117] The advantage of the formula is that it comprehensively refers to the interaction frequency and the minimum distance between people, increasing the sensitivity of the model to the actual transmission risk, so as to more accurately predict the probability of cross-infection.

[0118] Detailed explanation of the formula and the derivation process of formula calculation:

[0119] The set monitoring data shows that the interaction frequency F between person i and j ij is 0.5 times per hour, and the minimum distance D ij is 2 meters. The transmission coefficient β is set to 0.03 and substituted into the formula for calculation:

[0120] P infect (i,j) = 1 - e -0.03·0.5·2

[0121] = 1 - e -0.03

[0122] = 1 - 0.97045

[0123] ≈ 0.02955

[0124] This result indicates that under the current set parameter conditions, the cross-infection probability between person i and j is approximately 2.955%, indicating that under interactions with lower frequency and larger distances, the infection probability is lower. This helps hospital managers evaluate and adjust the personnel interaction strategy to reduce the infection risk.

[0125] Normalize the updated infection probability matrix to ensure that all probability values are between 0 and 1, and perform iterative statistical analysis to generate a normalized infection probability matrix;

[0126] For the process of generating a normalized infection probability matrix through iterative statistical analysis, first, it is necessary to obtain the transmission probability between each node from the original infection probability data. Set the initial probability values by monitoring various disease transmission events and the corresponding transmission frequencies and intensities. Each node represents an individual, and the edges between nodes represent the potential transmission paths between these individuals. By applying normalization techniques, all probability values are adjusted to be between 0 and 1 to ensure that each probability value can reflect the true level of relative transmission risk. The maximum-minimum normalization method is used in the normalization process, that is, each value is subtracted by the minimum value and then divided by the difference between the maximum value and the minimum value to standardize all probability values, so that the matrix maintains numerical stability and the effectiveness of comparison during the iteration process. This process is iterated until it converges to a stable probability distribution.

[0127] Based on the normalized infection probability matrix, simulate the transmission path of the virus in the personnel network through graph theory methods to generate a potential infection transmission graph.

[0128] Finally, in the process of constructing a potential infection transmission graph using network analysis methods, it is first necessary to transform the normalized infection probability matrix into a graph theory model. This transformation involves mapping the data points in the probability matrix to network nodes and edges. Each node represents a person, and the weight of each edge represents the infection probability between two people. Through this mapping, the potential infection paths between people can be visually displayed. This graphical representation method not only helps to intuitively analyze the potential risks of infection transmission but also facilitates iterative analysis, including identifying risk nodes or groups, so as to formulate more effective preventive measures. The generated potential infection transmission graph provides an effective visual tool for hospital management.

[0129] Please refer to Figure 4 , and the steps for mining the correlation between environmental factors and the patient's health status are specifically as follows:

[0130] Based on the potential infection transmission graph, continuously collect air quality indicators, temperature, and humidity data in the ward through environmental sensors, and ensure that each data point is attached with a timestamp to obtain an environmental monitoring dataset;

[0131] Through the continuous monitoring of the environmental sensor system, the air quality indicators, temperature, and humidity data collected in the ward not only need to be kept updated but also ensure that each data point has a timestamp. This process involves the regular calibration of environmental monitoring equipment and the application of time synchronization technology to ensure the consistency of the time tags of all collected environmental data. This is crucial for the subsequent analysis and use of data. The accuracy of data collection directly affects the reliability of environmental data analysis and the overall effectiveness of the environmental monitoring system. Through these measures, an environmental monitoring dataset can be systematically collected and generated, and this dataset will serve as the basic data source for evaluating the impact of ward environmental changes on patient health.

[0132] Based on the potential infection transmission graph, synchronously collect patient vital sign data, including heart rate, blood pressure, and body temperature. Each data point is attached with a timestamp, and the vital sign data is aligned with the environmental dataset by comparing the timestamps to obtain a patient health dataset;

[0133] The synchronous collection of patient vital sign data includes heart rate, blood pressure, and body temperature. The data collection involves using medical monitoring devices and aligning with the timestamps of environmental data. This requires implementing strict time synchronization measures during the data collection process to ensure that each vital sign data point can accurately match the environmental data point, which is crucial for subsequent data analysis. Through timestamp alignment, it can be ensured that when analyzing the impact of environmental changes on patient health, each vital sign indicator can accurately reflect its relationship with environmental variables. This synchronization mechanism not only improves the accuracy of data processing but also enhances the practical value of the dataset, thereby generating a patient health dataset, which will be used for correlation analysis with the environmental monitoring dataset.

[0134] Perform statistical analysis on the environmental monitoring dataset and the patient health dataset, using the correlation analysis formula:

[0135]

[0136] Calculate the correlation coefficient between environmental variables and vital signs, and generate the environmental and health correlation analysis results;

[0137] Among them, E k represents the environmental monitoring value at the kth collection time point, and H k represents the patient health monitoring value at the corresponding time point. and are the means of the environmental and health monitoring values at all collection time points, and n is the total number of collection time points.

[0138] Formula:

[0139]

[0140] The benefit of the formula is to provide a scientific basis for subsequent medical monitoring and environmental management by quantifying the direct relationship between environmental data and patient health data.

[0141] Detailed explanation of the formula and the formula calculation derivation process:

[0142] Set that within a certain period of time, the monitored environmental temperature data is E = {22, 23, 21, 22, 24}, and the corresponding patient heart rate data within the same period is H = {78, 76, 77, 80, 79}.

[0143] Calculate the average value of the data:

[0144]

[0145] and

[0146]

[0147] Calculate the numerator part of the covariance:

[0148] -0.4 * 0 + 0.6 * (-2) + (-1.4) * (-1) + (-0.4) * 2 + 1.6 * 1 = 0.8

[0149] Calculate the variance of E and the variance of H:

[0150]

[0151] Calculate the correlation coefficient obtained:

[0152]

[0153] The results show that the correlation between the environmental temperature and the patient's heart rate is low, which means that the change in environmental temperature has a small direct impact on the patient's heart rate in the dataset, thus providing crucial data support for medical monitoring.

[0154] Please refer to Figure 5 , and the specific steps for obtaining real-time environmental risk assessment data are as follows:

[0155] Extract the correlation coefficient from the results of the environmental and health correlation analysis, evaluate the impact of each parameter based on the real-time monitored environmental data, calculate the risk contribution degrees of temperature and humidity, and obtain the environmental parameter risk contribution data;

[0156] According to the results of the environmental and health correlation analysis, extract the correlation coefficient, identify the impact of differential environmental parameters, collect temperature and humidity parameters based on the real-time monitored data, use data acquisition instruments for environmental monitoring to ensure the accuracy of the parameters, analyze the contribution degree of each parameter to the risk of infection transmission, adopt a systematic quantitative analysis method, generate the corresponding risk contribution data to support subsequent calculations, and judge the key environmental factors by comparing multiple data, and summarize to obtain the current environmental parameter risk contribution data.

[0157] Combine the environmental parameter risk contribution data and the updated patient health status data, and apply the formula:

[0158]

[0159] Calculate the infection transmission probability data;

[0160] Among them, P infect represents the probability of infection transmission, which calculates the infection risk under the target environmental factors and the patient's health status. α represents the weight of the correlation coefficient C between the environment and health data, and the weight adjusts the influence degree of the correlation between environmental data and health status on the infection probability. β i is the weight coefficient of the i-th environmental parameter, and the coefficient adjusts the contribution degrees of multiple environmental parameters including temperature, humidity, etc. in the infection risk assessment. Ri represents the risk contribution value of the i-th environmental parameter, measuring the direct contribution of the environmental parameter to the infection risk, and m is the number of reference environmental parameters;

[0161] Formula:

[0162]

[0163] The advantage of the formula is that by comprehensively considering the weights and correlation coefficients of the reference environmental parameters, it can more accurately evaluate the infection transmission probability and provide a basis for real-time risk management.

[0164] Detailed explanation of the formula and the derivation process of formula calculation:

[0165] Set parameters α = 0.5, C = 0.7 (obtained through correlation analysis), the environmental parameter weights β1 = 0.3 (temperature), β2 = 0.2 (humidity), where the temperature R1 = 25 °C (real-time monitoring data) and the humidity R2 = 60% (real-time monitoring data). Substitute the parameters into the formula for calculation:

[0166] P infect = 0.5·0.7 + 0.3·25 + 0.2·60

[0167] P infect = 0.35 + 7.5 + 12 = 19.85

[0168] The result shows that the infection transmission probability is 19.85, indicating a relatively high transmission risk under the current environmental parameters.

[0169] Based on the infection transmission probability data, comprehensively referring to the current personnel density and movement patterns in the ward, real-time environmental risk assessment data is generated.

[0170] Based on the infection transmission probability data, comprehensively referring to the current personnel density and movement patterns in the ward, real-time environmental risk assessment data is generated to support the immediate management decision-making in the ward. In this process, the analysis of personnel movement data relies on real-time video monitoring and movement tracking technologies, and the calculation of personnel density is automatically updated through admission and discharge records. Integrating dynamic data, potential hotspots of infection transmission are predicted through algorithms to guide the hospital management to take immediate and preventive intervention measures.

[0171] Please refer to Figure 6 , the specific steps for obtaining the optimized risk weight matrix are as follows:

[0172] Extract key risk factors from the real-time environmental risk assessment data, analyze the degree of impact of each factor on the patient's health, call the correlation parameters of multiple risk factors for calculation, obtain the initial risk weight of each factor, and get the initial risk weight list;

[0173] The key risk factors extracted from real-time environmental risk assessment data include the speed of environmental change and patient density. Each risk factor needs to be assigned a weight according to its impact on patient health. These weights are obtained through regression analysis of past data. The initial risk weights of each factor are determined through analysis. The weight of each risk factor reflects the potential threat level to the patient's health status. This data will support the subsequent adjustment of the risk assessment matrix, provide data support for hospital management, and ensure a rapid response in case of emergencies.

[0174] Based on the initial risk weight list, dynamically adjust the risk weights according to the current environmental data and the patient's health status. Call the adjustment coefficient to calculate with the original risk weight, execute the weighted model, using the formula:

[0175]

[0176] Calculate the risk weight data to generate an adjusted risk weight data set;

[0177] where, W i is the adjusted weight of the i-th risk factor, R i is the original risk weight, γ i is the adjustment coefficient based on environmental data, H i is the impact degree of health data, δ is the adjustment coefficient based on health status, and η is the adjustment flexibility coefficient;

[0178] Formula:

[0179]

[0180] The advantage of the formula is to dynamically adjust the weights of environmental risk factors, match the hospital environment and the patient's health status, and enhance the sensitivity of the model to outliers and improve the accuracy of weight adjustment by adding square root and division operations.

[0181] Detailed explanation of the formula and the derivation process of formula calculation:

[0182] Set R i = 0.2, representing the initial risk weight of a certain risk factor, γ i = 0.05, representing the environmental data adjustment coefficient, δ = 0.03, representing the impact coefficient of health data, H i = 0.8, representing the impact degree of health data, η = 1.1, representing the adjustment flexibility coefficient. First, calculate the values inside the square root and the denominator:

[0183]

[0184] Then multiply the two to get the W i value:

[0185] W i = 0.4472·0.048 = 0.0215

[0186] The results show that under the given environmental and health conditions, the adjusted weight of the risk factor is 0.0215, indicating that the risk factor has a relatively small impact on the overall risk assessment and is helpful for controlling and managing environmental risks.

[0187] Integrate the adjusted risk weight data into the risk assessment matrix, call the risk weight data and combine it with the risk assessment matrix, update the matrix structure to reflect the actual risks of the current environment and the patient's health status, and output the optimized risk weight matrix.

[0188] The process of integrating the adjusted risk weight data into the risk assessment matrix involves multiple steps. First, the weight data of multiple risk factors needs to be checked against the existing matrix data. After verifying the data consistency, the corresponding values in the matrix are adjusted according to the current risk weights. This process requires comprehensive reference to the changes in the current environmental data and the patient's health status. Each data update affects the weight distribution in the matrix. The output optimized risk weight matrix not only reflects the current environmental and health status but also provides a prediction basis for future estimated environmental changes or deteriorating health conditions, making the matrix more dynamic and sensitive to the actual situation on-site, ensuring patient safety and optimizing hospital resource allocation.

[0189] Please refer to Figure 7 , and the specific steps for obtaining the assessment results of the impact of risk factors are as follows:

[0190] Call the optimized risk weight matrix, use statistical methods to calculate the correlation between multiple risk factors in the matrix and the patient's health status data, using the formula:

[0191]

[0192] Generate the correlation data between risk factors and health status;

[0193] Among them, C i represents the correlation score of the i-th risk factor, w ij is the weight of the i-th risk factor in the matrix for the j-th health indicator, and h j is the actual value of the j-th health indicator;

[0194] Formula:

[0195]

[0196] Detailed explanation of the formula and the derivation process of the formula calculation: This formula is designed to calculate the correlation between each risk factor and the patient's health status. Among them, C iRepresents the correlation score of the i-th risk factor, w ij Is the weight of the risk factor for the j-th health indicator, h j Is the actual value of the j-th health indicator. Three health indicators are set for monitoring: heart rate, blood pressure, and body temperature.

[0197] The data sources are as follows:

[0198] h1 (heart rate): Monitored by medical equipment. The normal heart rate range for general adults is 60 - 100 bpm, and the set monitoring value is 85 bpm.

[0199] h2 (blood pressure): The normal systolic blood pressure is 120 mmHg, and the set monitoring value is 120 mmHg.

[0200] h3 (body temperature): The normal body temperature is 36.5℃ to 37.5℃, and the set monitoring value is 36.8℃.

[0201] The weight w ij Is set based on the sensitivity of each health indicator to the impact of the condition. The following is an example setting:

[0202] w i1 : The impact weight of heart rate on the risk factor is 0.3.

[0203] w i2 : The impact weight of blood pressure on the risk factor is 0.4.

[0204] w i3 : The impact weight of body temperature on the risk factor is 0.3.

[0205] Calculation process:

[0206] C i =(0.3×85)+(0.4×120)+(0.3×36.8)

[0207] C i =25.5 + 48 + 11.04 = 84.54

[0208] The results show that considering the heart rate, blood pressure, and body temperature of the current patient, the correlation score of the i-th risk factor is 84.54. This indicates that there is a moderate correlation between the risk factor and the current health status of the patient, and based on this score, the impact degree of the risk factor on the patient can be evaluated.

[0209] Analyze the correlation data between risk factors and health status, determine which risk factors have a significant impact on the patient's health status and infection transmission, and calculate the contribution degree of each risk factor to infection transmission using the formula:

[0210] I i =λ·C i

[0211] Generate a preliminary assessment of the impact of risk factors on infection transmission;

[0212] where I i represents the impact score of the i-th risk factor on infection transmission, and λ is the standardization coefficient;

[0213] Formula:

[0214] I i = λ·C i

[0215] Detailed explanation of the formula and the derivation process of the formula calculation: The formula is used to calculate the impact score of each risk factor on infection transmission. Among them, I i represents the impact score of the i-th risk factor, and λ is the standardization coefficient used to adjust the relevance score C i to match the impact score range, and C i is the risk factor relevance score obtained from the previous steps.

[0216] The numerical sources are as follows:

[0217] C i : Set C i to 84.54 from the previous calculation.

[0218] λ: Set to 0.05. The coefficient is optimized based on the previous data of infection transmission and the risk assessment model so that the impact score can reflect the actual transmission risk.

[0219] Calculation process:

[0220] I i = 0.05 × 84.54

[0221] I i = 4.227

[0222] The results show that, comprehensively referring to the relevance between the i-th risk factor and the patient's health status, its impact score on infection transmission is 4.227. This score indicates the degree of contribution of the risk factor to infection transmission in the current ward environment and is a quantitative description of its potential influence.

[0223] Integrate the preliminary assessments of the impacts of all risk factors on infection transmission, and perform a weighted average according to the criticality of each score to obtain the risk factor impact assessment result.

[0224] According to the risk factor impact scores, iteratively analyze the contribution of multiple risk factors to infection transmission, which requires combining the previously obtained correlation data with the infection model. First, determine the parameters of the infection model, including the baseline infection probability and patient susceptibility, which are obtained from previous data analysis. For each risk factor, weighted calculations will be performed using the adjusted correlation scores. Referring to the actual environmental changes within the ward, the weight parameters need to be dynamically adjusted according to real-time monitoring data. The applicability of the weight parameters needs to be re-evaluated for each calculation to ensure the accuracy of the evaluation results. This comprehensive analysis will output a quantitative assessment of the impact of each risk factor on infection transmission, providing a powerful risk management tool for clinical practice.

[0225] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the relevant art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as they do not depart from the technical solution content of the present invention, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A risk assessment system for a laminar flow ward in the hematology department based on big data, characterized in that, The system includes: The infection transmission prediction module collects the location and movement data of each person in the ward, serializes the data, calculates the activity patterns and interaction frequencies of each person, predicts the probability of potential cross-infection events, and outputs a potential infection transmission map; The environmental monitoring integration module collects environmental sensor data and the health monitoring data of patients based on the potential infection transmission map, including air quality, temperature, humidity, and vital signs, mines the correlation between environmental factors and the health status of patients, estimates the probability of infection transmission, and outputs real-time environmental risk assessment data; The risk matrix optimization module uses the real-time environmental risk assessment data to identify key risk factors, adjusts the weights in the matrix, corresponding to the current environment and the health status of patients, and outputs an optimized risk weight matrix; The decision support output module calculates the correlation between the risk factors in the ward and the health status of patients based on the optimized risk weight matrix, analyzes the impact of risk factors on infection transmission, and outputs the evaluation results of the impact of risk factors.

2. The risk assessment system for a laminar flow ward in the hematology department based on big data according to claim 1, wherein, The potential infection transmission map includes activity patterns, interaction frequencies, and cross-infection probabilities. The real-time environmental risk assessment data includes air quality, temperature, humidity, and the correlation of vital signs. The optimized risk weight matrix specifically includes adjusted weights, environmental parameters, and health status. The evaluation results of the impact of risk factors are specifically the correlation between risk factors and the health status of patients and the impact on infection transmission.

3. The risk assessment system for a hematology laminar flow ward based on big data according to claim 2, wherein The steps of calculating the activity patterns and interaction frequencies of each person are specifically as follows: Continuously collect through sensors deployed in the ward to obtain the location data of each person in the ward, and record the location coordinates and movement speed of each person in real time to obtain the original location and movement data set; Based on the original location and movement data set, serialize the location data. Each data point includes a timestamp, location coordinates, and movement speed, and perform time series analysis on the data to generate a serialized location data set; Analyze the serialized location data set, calculate the movement frequency and residence time of each person, using the formula: Generate activity patterns; Among them, P i represents the average activity intensity of individual i, and d i,t is the moving distance of individual i at time t, and T is the total monitoring time; Based on the activity patterns, using the formula: Calculate the average interaction frequency between person i and person j to generate interaction frequency data; Among them, F ij represents the average interaction frequency between individuals i and j, x i,t and y i,t are coordinate positions, and N is the total number of time points monitored.

4. The risk assessment system for a laminar flow ward in the hematology department based on big data according to claim 3, wherein The steps of obtaining the potential infection transmission map are specifically as follows: Extract the interaction frequencies of each pair of people from the interaction frequency data, combine with the activity pattern data of each pair of people, calculate the cross-infection probability between each pair of people, and generate a preliminary infection probability matrix; With reference to the dynamic change of the distance between people, refine the preliminary infection probability matrix, and by adjusting the weight of the interaction distance, using the formula: Calculate the cross-infection probability between person i and person j to generate an updated infection probability matrix; Among them, P infect (i, j) represents the cross-infection probability between person i and person j, F ij represents the average interaction frequency between person i and person j, obtained from the interaction frequency data, indicating the frequency of interaction between the two people. β represents the transmission coefficient, used to adjust the influence of interaction frequency and distance on the infection probability, D ij is the minimum distance calculated between two individuals; Perform normalization processing on the updated infection probability matrix to ensure that all probability values are between 0 and 1, and perform iterative statistical analysis to generate a normalized infection probability matrix; Based on the normalized infection probability matrix, simulate the transmission path of the virus in the personnel network through graph theory methods to generate a potential infection transmission map.

5. The risk assessment system for a hematology laminar flow ward based on big data according to claim 4, wherein The steps of mining the correlation between environmental factors and the health status of patients are specifically as follows: Based on the potential infection transmission map, continuously collect air quality indicators, temperature, and humidity data in the ward through environmental sensors, while ensuring that each data point is attached with a timestamp, to obtain an environmental monitoring dataset; Based on the potential infection transmission map, synchronously collect patient vital sign data, including heart rate, blood pressure, and body temperature. Each data point is attached with a timestamp. Align the vital sign data with the environmental dataset by comparing timestamps to obtain a patient health dataset; Conduct statistical analysis on the environmental monitoring dataset and the patient health dataset, using the correlation analysis formula: Calculate the correlation coefficient between environmental variables and vital signs, and generate the environmental and health correlation analysis result; Among them, E k represents the environmental monitoring value at the k-th collection time point, and H k represents the patient health monitoring value at the corresponding time point. And are the means of the environmental and health monitoring values at all collection time points, and n is the total number of collection time points.

6. The risk assessment system for a laminar flow ward in the hematology department based on big data according to claim 5, wherein The specific steps for obtaining the real-time environmental risk assessment data are as follows: Extract the correlation coefficient from the environmental and health correlation analysis result, evaluate the impact of each parameter based on the real-time monitored environmental data, calculate the risk contribution degree of temperature and humidity, and obtain the environmental parameter risk contribution data; Combine the environmental parameter risk contribution data and the updated patient health status data, and apply the formula: Calculate to obtain the infection transmission probability data; Among them, P infect represents the probability of infection transmission, which calculates the infection risk under the target environmental factors and the patient's health status. α represents the weight of the correlation coefficient C between the environment and health data, and the weight adjusts the influence degree of the correlation between environmental data and health status on the infection probability. β i is the weight coefficient of the i-th environmental parameter, and the coefficient adjusts the contribution degrees of multiple environmental parameters including temperature, humidity, etc. in the infection risk assessment. R i represents the risk contribution value of the i-th environmental parameter, which measures the direct contribution of the environmental parameter to the infection risk. m is the number of reference environmental parameters; Based on the infection transmission probability data, comprehensively refer to the density and movement patterns of the current personnel in the ward to generate real-time environmental risk assessment data.

7. The risk assessment system for a hematology laminar flow ward based on big data according to claim 6, wherein The specific steps for obtaining the optimized risk weight matrix are as follows: Extract the key risk factors from the real-time environmental risk assessment data, analyze the degree of influence of each factor on patient health, call the correlation parameters of multiple risk factors for calculation, obtain the initial risk weight of each factor, and get the initial risk weight list; Based on the initial risk weight list, dynamically adjust the risk weight according to the current environmental data and the patient's health condition. Call the adjustment coefficient to calculate with the original risk weight, execute the weighted model, and use the formula: Calculate the risk weight data and generate the adjusted risk weight dataset; Among them, W i is the adjusted weight of the i-th risk factor, R i is the original risk weight, γ i is the adjustment coefficient based on environmental data, H i is the influence degree of health data, δ is the adjustment coefficient based on health status, and η is the adjustment flexibility coefficient; Integrate the adjusted risk weight data into the risk assessment matrix, call the risk weight data to combine with the risk assessment matrix, update the matrix structure to reflect the actual risk of the current environment and patient health condition, and output the optimized risk weight matrix.

8. The risk assessment system for the laminar flow ward in the hematology department based on big data according to claim 7, wherein The specific steps for obtaining the risk factor impact assessment result are as follows: Call the optimized risk weight matrix, use statistical methods to calculate the correlation between multiple risk factors in the matrix and the patient health condition data, and use the formula: Generate the risk factor and health condition correlation data; Among them, C i represents the correlation score of the i-th risk factor, and w ij is the weight of the i-th risk factor on the j-th health indicator in the matrix, and h j is the actual value of the j-th health indicator; Analyze the risk factor and health condition correlation data, determine which risk factors have a significant impact on the patient's health condition and infection transmission, and execute the calculation of the infection transmission contribution degree of each risk factor, using the formula: I i = λ·C i Generate a preliminary assessment of the impact of risk factors on infection transmission; Among them, I i represents the impact score of the i-th risk factor on the spread of infection, and λ is the normalization coefficient; Combine all the preliminary assessments of the impact of risk factors on infection transmission, and perform a weighted average according to the criticality of each score to obtain the risk factor impact assessment result.

Images