A method for disinfection optimization process for a care environment

Abstract

The present application relates to a disinfection optimization processing method for a nursing environment, and relates to the field of process optimization, comprising: according to a microorganism concentration sequence array, predicting a microorganism concentration prediction distribution after a future time; according to an activity path and an activity frequency of a patient, optimizing the microorganism concentration prediction distribution to obtain a corrected microorganism concentration prediction distribution; combining physical sign data, illness information and a historical activity range of the patient to set a atomization concentration constraint; according to the corrected microorganism concentration prediction distribution, performing disinfection scheme optimization with the multiple atomization concentration constraints as limits to output an optimal disinfection scheme. Through the present application, the technical problem of the traditional method being unable to set matching disinfection parameters according to the current environment state of an infectious disease department and the disinfection precision being insufficient can be solved; precise, efficient and safe disinfection management can be realized, potential harm to the patient caused by excessive disinfection can be reduced, disinfection quality and patient comfort can be effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of process optimization, in particular to a disinfection optimization processing method for nursing environment. BACKGROUND

[0002] In the nursing process of modern hospitals, especially in the infectious department, environmental disinfection has always been a key link to prevent nosocomial infection and cross infection. With the gradual improvement of hospital infection control standards, the disinfection quality has an important influence on the health and treatment effect of patients.

[0003] However, the traditional disinfection methods often fail to consider the real-time changing environmental conditions such as microbial concentration, air circulation, patient activity, etc. These methods usually adopt fixed disinfection frequency and intensity, ignoring the actual concentration changes of microorganisms at each time point and spatial area and the individual needs of patients, which has the problem of insufficient precision. SUMMARY

[0004] The present application provides a disinfection optimization processing method for nursing environment to solve the technical problem of insufficient disinfection precision of traditional disinfection methods which cannot set matching disinfection parameters according to the current environmental state of the infectious department.

[0005] The technical solution of the present application to solve the above technical problem is as follows: the present application provides a disinfection optimization processing method for nursing environment, comprising: monitoring and obtaining a plurality of microbial concentration sequences of a plurality of monitoring points in the covered area of the infectious department, performing interpolation prediction of a plurality of unknown points, and generating a microbial concentration sequence array; according to the microbial concentration sequence array, predicting and obtaining a microbial concentration prediction distribution after a future time; according to the activity path and activity frequency of the patients in the covered area of the infectious department, optimizing the microbial concentration prediction distribution to obtain a corrected microbial concentration prediction distribution; combining the sign data, illness information and historical activity range of a plurality of patients, setting a plurality of atomization concentration constraints; according to the corrected microbial concentration prediction distribution, performing disinfection scheme optimization with the plurality of atomization concentration constraints as the limit, outputting an optimal disinfection scheme, and controlling the disinfection robot to execute the disinfection control after the future time.

[0006] Preferably, the disinfection optimization processing method for nursing environment further comprises: periodically monitoring a plurality of monitoring points in the covered area of the infectious department by a microbial monitoring system to generate a plurality of microbial concentration sequences; based on the Kriging interpolation principle, performing interpolation prediction of a plurality of unknown points according to the plurality of microbial concentration sequences to generate a plurality of microbial concentration prediction sequences; constructing a microbial concentration sequence array based on the plurality of microbial concentration sequences and a plurality of microbial concentration prediction sequences.

[0007] Preferably, the disinfection optimization processing method for a care environment further comprises: synchronously monitoring environment parameters of a plurality of monitoring points to generate a plurality of environment parameter sequences, wherein the environment parameters include temperature, humidity, and airflow direction; mapping and compensating the plurality of microorganism concentration prediction sequences according to the plurality of environment parameter sequences to obtain a plurality of compensated microorganism concentration prediction sequences; and constructing a microorganism concentration sequence array based on the plurality of microorganism concentration sequences and the plurality of compensated microorganism concentration prediction sequences.

[0008] Preferably, the disinfection optimization processing method for a care environment further comprises: combining position information of the plurality of monitoring points and a plurality of unknown points, respectively constructing an environment parameter sequence array and a microorganism concentration prediction sequence array according to the plurality of environment parameter sequences and the plurality of microorganism concentration prediction sequences; collecting a sample environment parameter sequence array set, a sample microorganism concentration prediction sequence array set, and a plurality of sample compensated microorganism concentration prediction sequence sets, and performing supervised training on a feedforward neural network until the model converges to obtain a concentration compensation analysis plug-in; and inputting the environment parameter sequence array and the microorganism concentration prediction sequence array into the concentration compensation analysis plug-in to output a plurality of compensated microorganism concentration prediction sequences.

[0009] Preferably, the disinfection optimization processing method for a care environment further comprises: collecting a sample microorganism concentration sequence array set and a sample environment parameter sequence array set according to microorganism monitoring records of an infectious disease department coverage area, and obtaining sample predicted microorganism concentration arrays of different sample microorganism concentration sequence arrays and sample environment parameter sequence arrays at the same historical time to obtain a sample predicted microorganism concentration array set; training a feedforward neural network using the sample microorganism concentration sequence array set, the sample environment parameter sequence array set, and the sample predicted microorganism concentration array set to construct a concentration prediction plug-in; and using the concentration prediction plug-in to predict according to the microorganism concentration sequence array and the environment parameter sequence array to obtain a predicted microorganism concentration array; and fitting a concentration distribution according to the predicted microorganism concentration array to generate a microorganism concentration prediction distribution.

[0010] Preferably, the disinfection optimization processing method for a care environment further comprises: monitoring a plurality of activity paths and a plurality of activity frequencies of a plurality of patients; determining a plurality of activity ranges according to the activity paths, determining a plurality of microorganism concentration enhancement coefficients according to the activity frequencies, and sequentially determining a plurality of activity ranges and a plurality of microorganism concentration enhancement coefficients of the plurality of patients; and optimizing and enhancing the microorganism concentration prediction distribution according to the plurality of activity ranges and the plurality of microorganism concentration enhancement coefficients to obtain a corrected microorganism concentration prediction distribution.

[0011] Preferably, the disinfection optimization processing method for a care environment further comprises: randomly selecting first patient sign data, first condition information and first historical activity range; performing tolerable atomization concentration analysis according to the disinfectant type, the first sign data and the first condition information, and predicting a first maximum atomization concentration; setting a first atomization concentration constraint according to the first historical activity range and the first maximum atomization concentration, and adding to the plurality of atomization concentration constraints.

[0012] Preferably, the disinfection optimization processing method for a care environment further comprises: collecting a sample sign data set and a sample condition information set with the disinfectant type as a constraint, and obtaining a maximum atomization concentration tolerable by a patient under different sample sign data and sample condition information, set as a sample atomization concentration, to obtain a sample atomization concentration set; constructing an atomization concentration analysis model based on machine learning with the sample sign data set, the sample condition information set and the sample atomization concentration set as training data; and performing tolerable atomization concentration analysis according to the disinfectant type, the first sign data and the first condition information by using the atomization concentration analysis model, and outputting the first maximum atomization concentration.

[0013] Preferably, the disinfection optimization processing method for a care environment further comprises: performing disinfection atomization concentration configuration based on the corrected microorganism concentration prediction distribution, and obtaining an atomization concentration distribution; adjusting the disinfection atomization concentration distribution according to the plurality of atomization concentration constraints to obtain an optimized atomization concentration distribution, wherein if a maximum atomization concentration in the atomization concentration constraint is greater than or equal to a disinfection atomization concentration in a same activity range, no adjustment is performed, and if the maximum atomization concentration in the atomization concentration constraint is less than the disinfection atomization concentration in the same activity range, the maximum atomization concentration is replaced; and generating an optimal disinfection scheme according to a predetermined disinfection path and the optimized atomization concentration distribution.

[0014] The beneficial effects of the present application are: by monitoring a plurality of microorganism concentration sequences of a plurality of monitoring points in the infection department coverage area, interpolating and predicting a plurality of unknown points to generate a microorganism concentration sequence array; then, according to the microorganism concentration sequence array, a microorganism concentration prediction distribution after a future time is obtained; further, according to the activity path and activity frequency of the patients in the infection department coverage area, the microorganism concentration prediction distribution is optimized to obtain a corrected microorganism concentration prediction distribution; on the other hand, combining the sign data, illness information and historical activity range of a plurality of patients, a plurality of atomization concentration constraints are set; finally, taking the plurality of atomization concentration constraints as the limit, the disinfection scheme optimization is carried out according to the corrected microorganism concentration prediction distribution, and the optimal disinfection scheme is output, and the disinfection robot is controlled to execute the disinfection control after the future time. That is, through real-time monitoring, dynamic prediction and individualized optimization, precise, efficient and safe disinfection management can be realized, and the potential harm of excessive disinfection to patients is reduced, and the disinfection quality of the hospital infection department and the comfort of the patients are effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 A flowchart of a disinfection optimization processing method for a nursing environment provided by the present application is shown.

[0016] Figure 2 A flowchart of generating a microorganism concentration sequence array in a disinfection optimization processing method for a nursing environment provided by the present application is shown. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0018] In the description of the present application, the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0019] In the description of the present application, the term "for example" is used to mean "serving as an example, instance, or illustration." Any embodiment described as "for example" is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for purposes of explanation, specific details are set forth to provide a thorough understanding of the present application. It will be apparent to one skilled in the art, however, that the present application can be practiced without using these specific details. In other instances, well-known structures and processes are not elaborated in order not to obscure the description of the present application with unnecessary details. Thus, the present application is not intended to be limited by the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0020] As shown in the embodiments, Figure 1 The embodiments of the present application provide a disinfection optimization processing method for a nursing environment, which specifically comprises the following steps:

[0021] S10: monitoring and acquiring a plurality of microbial concentration sequences of a plurality of monitoring points in the infectious disease department coverage area, performing interpolation prediction of a plurality of unknown points, and generating a microbial concentration sequence array.

[0022] Further, as shown in the embodiments, Figure 2 The step S10 of the present application further comprises:

[0023] S11: periodically monitoring a plurality of monitoring points in the infectious disease department coverage area through a microbial monitoring system, and generating a plurality of microbial concentration sequences; and S12: based on the Kriging interpolation principle, performing interpolation prediction of a plurality of unknown points according to the plurality of microbial concentration sequences, and generating a plurality of microbial concentration prediction sequences.

[0024] Specifically, the microbial concentration in the infectious disease department environment needs to be measured in real time or periodically by a monitoring system, which usually includes a sensor and a microbial sampling device, and can accurately measure the concentration of bacteria, viruses, fungi and other microorganisms in the air, and collect data according to a certain time interval.

[0025] First, according to the specific situation of the infectious disease department (such as the number of wards, air circulation, patient activity area, etc.), a number of monitoring points are selected within the covered area. The layout of these monitoring points usually takes into account the differences in microbial concentration in different areas, as well as the functional needs of the ward, such as ward area, corridor, nurse station, public area, etc., which may have different monitoring needs. Then, the monitoring system will perform regular data collection according to the set time interval (such as every 10 minutes). Each collection will record the microbial concentration data of each monitoring point, which is usually stored in the form of a time series. Each monitoring point generates a microbial concentration value after each measurement. Over time, the concentration data at multiple time points can form a microbial concentration sequence. For example, if monitoring point A measures every 10 minutes, after 4 hours of monitoring, 24 microbial concentration data points will be obtained, which constitute the concentration sequence of monitoring point A. These microbial concentration data will serve as the basis for subsequent analysis, resulting in a number of microbial concentration sequences.

[0026] Kriging interpolation is a spatial statistics-based interpolation method that not only considers the data of the measurement points, but also combines the spatial correlation (i.e., spatial autocorrelation) between the measurement points. The core idea of the Kriging method is to estimate the value of unknown points through known monitoring points and spatial models. This method is very useful in environmental monitoring because environmental data (such as microbial concentration) usually have spatial correlation, and the concentration values in adjacent areas are often similar. Through Kriging interpolation, the blank areas due to uneven distribution of monitoring points or lack of monitoring in some areas can be effectively filled. First, use the microbial concentration sequences of multiple monitoring points collected in step S11 as the basis data for interpolation. Each monitoring point's microbial concentration sequence contains the concentration changes at different times. Through these data, the spatial relationship between different monitoring points can be established. Then, according to the Kriging interpolation principle, for each unknown point (i.e., unmonitored area or time period), the spatial autocorrelation between it and the known monitoring points is calculated, and the microbial concentration value of the unknown point is estimated using a spatial model (usually based on a semi-variance function). The semi-variance function is a function used to measure the similarity between spatial points in Kriging interpolation, which usually depends on the characteristics of geographic space, such as distance, variance, etc. Kriging method calculates the optimal weight of the predicted value through semi-variance function, thus obtaining the predicted concentration value of each unknown point. For each unmonitored point, the Kriging algorithm outputs a microbial concentration prediction value. By combining the prediction values of multiple unknown points, a complete microbial concentration prediction sequence can be generated. The prediction sequence not only includes the actual data of the monitored points, but also contains the data of the unknown points calculated by interpolation.

[0027] By using the Kriging interpolation algorithm, the concentration of the unmonitored area or time point is interpolated and predicted according to the microbial concentration sequence, and a plurality of microbial concentration prediction sequences are generated; this process provides a scientific prediction of the concentration of the unmonitored area by considering the spatial correlation and spatial distribution of the data, ensuring that the global microbial concentration trend can be accurately grasped in actual disinfection.

[0028] S13: Constructing a microbial concentration sequence array based on the plurality of microbial concentration sequences and the plurality of microbial concentration prediction sequences.

[0029] Further, the step S13 of the present application further comprises:

[0030] S131: Synchronously monitoring the environmental parameters of the plurality of monitoring points to generate a plurality of environmental parameter sequences, wherein the environmental parameters include temperature, humidity and airflow direction.

[0031] Specifically, by synchronously monitoring the environmental parameters, more background information can be provided for subsequent microbial concentration prediction and disinfection optimization. Environmental parameters such as temperature, humidity and airflow direction have important influence on the growth, spread and disinfection effect of microorganisms, therefore, real-time acquisition of these environmental data is crucial for precise control of environmental disinfection. Similar to the microbial monitoring system, the environmental parameter monitoring points should be reasonably set according to the spatial layout of the infectious disease department area. These monitoring points should cover multiple key areas within the infectious disease department, especially places that are easily affected by pathogen transmission (such as wards, corridors, nurse stations, etc.). Reasonable monitoring point layout can ensure representative environmental data.

[0032] To ensure the real-time and accuracy of data, environmental parameters need to be monitored simultaneously with microbial concentration monitoring. This means that each monitoring of microbial concentration requires simultaneous collection of data such as temperature, humidity, and airflow direction. Synchronous monitoring ensures that fluctuations in microbial concentration under different environmental conditions can be reflected in a timely manner and that a correlation between the two can be established. Environmental parameters (such as temperature, humidity, and airflow direction) are collected at different time points at each monitoring point, forming a sequence of environmental parameters. Temperature directly affects the growth rate and activity of microorganisms. Generally speaking, high temperatures help inactivate many pathogens, while low temperatures may promote the survival and spread of some pathogens. Therefore, temperature monitoring helps understand the impact of the environment on microorganisms. Humidity has a significant impact on the distribution and spread of airborne microorganisms. Excessive humidity can promote the growth of mold and bacteria, while low humidity can increase fine particulate matter in the air, affecting respiratory health. Controlling humidity within a certain range can help reduce the concentration of airborne microorganisms. Air circulation is a key factor in controlling the spread of airborne pathogens. The direction and speed of airflow determine the transmission path of microorganisms indoors. A good ventilation system can effectively reduce the residence time and concentration of microorganisms in the air, preventing cross-infection. For example, if the temperature is measured every hour, 24 temperature data points can be obtained in a day, forming a temperature sequence. Similarly, humidity and airflow direction will also generate corresponding sequences based on the sampling frequency. These sequences not only provide environmental data for each monitoring point at different times but also provide valuable auxiliary information on changes in microbial concentrations. By analyzing these environmental parameter sequences, we can understand the potential impact of environmental conditions on microbial concentrations and provide data support for subsequent disinfection strategy optimization.

[0033] By synchronously monitoring environmental parameters within the infectious department (such as temperature, humidity, and airflow direction), more background information can be provided for predicting changes in microbial concentrations and optimizing disinfection plans; by collecting and analyzing environmental parameter sequences, the impact of environmental conditions on microbial concentrations can be revealed, thereby providing a scientific basis for future disinfection strategies.

[0034] S132: Mapping and compensating the multiple microorganism concentration prediction sequences according to the multiple environmental parameter sequences to obtain multiple compensated microorganism concentration prediction sequences.

[0035] Furthermore, step S132 of the present invention further includes:

[0036] S1321: combine the position information of the plurality of monitoring points and the plurality of unknown points, and construct an environmental parameter sequence array and a microorganism concentration prediction sequence array according to the plurality of environmental parameter sequences and the plurality of microorganism concentration prediction sequences, respectively; S1322: collect a sample environmental parameter sequence array set, a sample microorganism concentration prediction sequence array set, and a plurality of sample compensation microorganism concentration prediction sequence sets, and perform supervised training on the feedforward neural network until the model converges to obtain a concentration compensation analysis plug-in; S1323: input the environmental parameter sequence array and the microorganism concentration prediction sequence array into the concentration compensation analysis plug-in, and output a plurality of compensation microorganism concentration prediction sequences.

[0037] Specifically, the position information of the plurality of monitoring points is combined, and an environmental parameter sequence array is constructed according to the plurality of environmental parameter sequences, that is, the environmental parameter sequences of different monitoring points are integrated together to form an environmental parameter sequence array. This array contains environmental parameter data of all monitoring points, and these data are integrated according to the time dimension. For example, assuming that there are N monitoring points and each monitoring point has T time point environmental data, the final environmental parameter sequence array will be an N x T matrix, wherein each row represents an environmental parameter data sequence of a monitoring point, and each column represents environmental parameters of all monitoring points at the same time point. On the other hand, a microorganism concentration prediction sequence array is constructed according to the position information of the plurality of unknown points and the plurality of microorganism concentration prediction sequences.

[0038] First, a sample environmental parameter sequence array set is collected, wherein the environmental parameter data of each sample includes temperature, humidity, air flow direction and the like. These data are arranged in time sequence and can reflect environmental changes at different times and positions. Then, for each environmental parameter sample, a microorganism concentration prediction sequence array set is collected. These concentration data reflect the microorganism concentration distribution at different time points and positions, and are the key input for training the neural network. Then, sample compensation microorganism concentration prediction sequences corresponding to the sample environmental parameter sequence array and the sample microorganism concentration prediction sequence array are collected. Compensation prediction is based on the correction of environmental parameters and original concentration prediction, and the purpose is to compensate for prediction errors caused by changes in environmental parameters or model bias.

[0039] Then, the sample environmental parameter sequence array and the sample microorganism concentration prediction sequence array are input, and the sample compensated microorganism concentration prediction sequence is supervised, and these collected sample data sets (environmental parameters, microorganism concentration prediction sequence, compensated concentration prediction sequence) are input into the feedforward neural network as a training set for supervised learning; the neural network model continuously adjusts the weight and bias parameters by comparing the error between the true data and the prediction results to achieve optimal prediction performance; the goal of the neural network is to learn how environmental parameters affect microorganism concentration and obtain a concentration compensation analysis plug-in that can compensate for concentration through training; the training process continues until the error of the model converges, indicating that the model has found the best parameter configuration and can effectively predict microorganism concentration from environmental data and perform compensation. After training, the concentration compensation analysis plug-in obtained as an independent module can be used for compensation analysis of new environmental parameters and microorganism concentration prediction.

[0040] Finally, the environmental parameter sequence array and the microorganism concentration prediction sequence array are input into the concentration compensation analysis plug-in, and based on the previously trained feedforward neural network, the concentration compensation analysis is performed. The plug-in applies the neural network model to perform compensation calculation based on the input environmental parameters and predicted concentration, and outputs the compensated microorganism concentration prediction value. This compensation process can correct the concentration prediction deviation caused by environmental factor changes, measurement errors, or other unknown factors. After concentration compensation analysis, the plug-in outputs multiple compensated microorganism concentration prediction sequences. These compensated sequences have higher accuracy than the original prediction sequences and can more accurately reflect the actual microorganism concentration distribution. The output compensated concentration sequence will be used to further adjust the disinfection strategy to ensure that the disinfection intensity and frequency match the actual microorganism concentration, avoiding excessive disinfection or insufficient disinfection.

[0041] By using the plug-in to compensate for the input environmental and concentration data, the compensated microorganism concentration prediction sequence is generated. These compensated data can provide more accurate basis for optimizing the disinfection scheme, ensuring that the disinfection process is more accurate, safe, and effective.

[0042] S133: Construct a microorganism concentration sequence array based on the plurality of microorganism concentration sequences and the plurality of compensated microorganism concentration prediction sequences.

[0043] Specifically, the plurality of microorganism concentration sequences and the plurality of compensated microorganism concentration prediction sequences are merged to generate a microorganism concentration sequence array. By merging and optimizing the original data and the compensated data, a comprehensive concentration prediction picture is formed, which provides more accurate basis for subsequent disinfection schemes and enables precise and efficient environmental disinfection management.

[0044] S20: predicting a microbial concentration prediction distribution after a future time according to the microbial concentration sequence array.

[0045] Further, the step S20 of the present application further comprises:

[0046] S21: collecting a sample microbial concentration sequence array set and a sample environmental parameter sequence array set according to the microbial monitoring records of the infection department coverage area, and obtaining a sample predicted microbial concentration array after the same historical time of different sample microbial concentration sequence arrays and sample environmental parameter sequence arrays, to obtain a sample predicted microbial concentration array set; S22: training a feedforward neural network using the sample microbial concentration sequence array set, the sample environmental parameter sequence array set, and the sample predicted microbial concentration array set, to construct a concentration prediction plug-in; S23: using the concentration prediction plug-in to make predictions according to the microbial concentration sequence array and the environmental parameter sequence array, to obtain a predicted microbial concentration array; S24: fitting the concentration distribution according to the predicted microbial concentration array, to generate a microbial concentration prediction distribution.

[0047] Specifically, the microbial monitoring records of the infection department coverage area are obtained, which are data obtained by real-time monitoring and recording of microbial concentrations at each monitoring point in the infection department. These records usually include the microbial concentrations (such as the concentrations of bacteria, viruses, or fungi) of each monitoring point at different time periods. Microbial monitoring data is the core data for environmental disinfection optimization, and can reflect the degree of environmental pollution, the type of pollutants, and the microbial distribution state at different time periods. According to the microbial monitoring records, a sample microbial concentration sequence array set is collected, which is a set composed of microbial concentration sequences of multiple samples (i.e., multiple monitoring points). Each sample concentration sequence is a time series data reflecting the change of microbial concentration at a monitoring point at multiple time points. By collecting these sequences, the microbial concentration fluctuations in different areas of the infection department can be comprehensively understood. Together with the microbial concentration data, environmental parameters (such as temperature, humidity, air flow direction, etc.) are also monitored regularly. These sample environmental parameter sequence array sets are sequence data obtained by monitoring different environmental factors. Environmental parameters directly affect the growth, spread, and concentration change of microorganisms, and therefore they are essential factors in predicting microbial concentrations. Environmental parameter sequence arrays reflect the changes of environmental conditions at each monitoring point over time. On the other hand, the microbial concentration arrays after the same historical time (such as 1 hour later) of different sample microbial concentration sequence arrays and sample environmental parameter sequence arrays are obtained, which are set as sample predicted microbial concentration arrays, to obtain a sample predicted microbial concentration array set.

[0048] The feedforward neural network is a basic neural network structure, which is usually used for regression or classification tasks. In this scenario, the feedforward neural network is used to establish a prediction model by historical microbial concentration data, environmental parameter data, and prediction data. Further, a sample microbial concentration sequence array and a sample environmental parameter sequence array are inputted, and a sample predicted microbial concentration array is supervised. The sample microbial concentration sequence array set, the sample environmental parameter sequence array set, and the sample predicted microbial concentration array set are used as training data to supervise the training of the feedforward neural network. The training process of the neural network includes forward propagation of input data and error back propagation. By continuously adjusting the weights and biases in the network, the output of the network is optimized to minimize the prediction error. Iterative training is performed until the loss function converges, and a concentration prediction plug-in is obtained. The concentration prediction plug-in is a trained neural network model that can predict microbial concentrations in real time when given new input data.

[0049] By training the feedforward neural network, historical microbial concentration sequences, environmental parameter data, and predicted concentration data are inputted into the network, and the neural network model is trained so that the model can predict microbial concentrations in future periods. After this training process, the network model is constructed into a concentration prediction plug-in that can automatically predict and adjust microbial concentrations when new data is inputted, providing real-time and accurate support for disinfection decisions.

[0050] Then, the microbial concentration sequence array and the environmental parameter sequence array are inputted into the concentration prediction plug-in for prediction, i.e., the future microbial concentration is predicted according to the inputted historical data and environmental conditions, and a predicted microbial concentration array is outputted. Each data point in the predicted microbial concentration array corresponds to a microbial concentration value at a specific location (monitoring point) and time point. These predicted values provide a scientific basis for subsequent disinfection decisions. Next, concentration distribution fitting is performed according to the predicted microbial concentration array, i.e., the concentration values in the predicted microbial concentration array are combined with environmental parameters, spatial locations of monitoring points, and time factors to generate a comprehensive concentration distribution. The concentration data is interpolated, smoothed, or optimized to make the concentration data show a reasonable spatial distribution pattern in the entire region. In particular, if the concentration prediction values at some locations or time points deviate from the expected values, the fitting process will adjust these values to make them more consistent with the actual situation, generating a microbial concentration prediction distribution. The microbial concentration prediction distribution is the fitted concentration data, which shows the microbial concentration distribution of different monitoring points at future time points. This distribution is not just a single concentration value, but a global view that can intuitively reflect how microbial concentrations change in space and time.

[0051] S30: optimizing the microorganism concentration prediction distribution according to the activity path and the activity frequency of the patient in the infection department coverage area, to obtain a corrected microorganism concentration prediction distribution.

[0052] Further, the step S30 of the present application further comprises:

[0053] S31: monitoring and obtaining a plurality of activity paths and a plurality of activity frequencies of a plurality of patients; S32: determining the activity range according to the activity path, and determining the microorganism concentration enhancement coefficient according to the activity frequency analysis, and sequentially analyzing and determining a plurality of activity ranges and a plurality of microorganism concentration enhancement coefficients of a plurality of patients; S33: optimizing and enhancing the microorganism concentration prediction distribution according to the plurality of activity ranges and the plurality of microorganism concentration enhancement coefficients, to obtain a corrected microorganism concentration prediction distribution.

[0054] Specifically, a plurality of activity paths and a plurality of activity frequencies of a plurality of patients are monitored and obtained. In the infection department, the activity path of the patient refers to the moving route or the location of the patient in the hospital or department. The activity range of the patient can include the bed area, the washroom, the corridor, etc. The activity trajectory of the patient is recorded by a sensor (such as RFID, GPS positioning system or other indoor positioning technology) or a patient handheld device. These path data can provide the specific location and moving trend of the patient's activity. The activity frequency refers to the frequency of the patient's activity in a unit time. These data can be collected by a sensor or regular check, which helps to analyze the activity intensity of the patient. By monitoring the activity path and the frequency of a plurality of patients, it can be more accurately understood how the patient affects the spread of microorganisms in the infection department. The activity path and the frequency of the patient are important influencing factors when calculating the microorganism concentration distribution.

[0055] Next, the activity range is determined according to the activity path, which refers to the area where the patient actually moves in the hospital, and is usually derived from the patient's activity path. For some patients, the activity range may be relatively limited (such as moving only near the bed), while for other patients, it may spread throughout the entire infectious disease department area. The microorganism concentration enhancement coefficient is determined according to the activity frequency analysis. The microorganism concentration enhancement coefficient is a coefficient that measures the impact of patient activity on the microorganism concentration in the surrounding environment. Patients who move frequently may cause an increase in the microorganism concentration in the surrounding area, especially in areas with strong air flow. If a patient moves frequently, the microorganism concentration may be enhanced due to air flow, breathing, or emissions, etc. Therefore, an enhancement coefficient needs to be set to reflect this impact. By analyzing the patient's activity frequency, the corresponding enhancement coefficient can be calculated, which can quantify the impact of patient activity on the microorganism concentration in the environment. By determining the activity range and calculating the microorganism concentration enhancement coefficient, it can provide a basis for subsequent optimization and adjustment of microorganism concentration. Patient activity will affect the microorganism concentration, especially for patients who move frequently, which may cause an increase in the microorganism concentration in the air or on the ground.

[0056] After obtaining the activity range and enhancement coefficient of each patient, the next step is to optimize the microorganism concentration prediction distribution based on this information. Since patient activity will affect the microorganism concentration in the surrounding area, it is necessary to adjust the distribution of microorganism concentration according to the activity range and frequency of each patient. Patients with a larger activity range will affect a wider area of microorganism concentration, while patients with a high activity frequency may have a greater impact on the surrounding environment in a short period of time. By applying the activity range and microorganism concentration enhancement coefficient, these factors can be incorporated into the prediction model to update the original microorganism concentration distribution. For example, in areas with high activity frequency, the microorganism concentration will be relatively higher, so the concentration in this area needs to be appropriately enhanced. During the optimization process, the predicted microorganism concentration distribution becomes more accurate, reflecting the real impact of patient activity on the microorganism concentration in the environment. The corrected microorganism concentration prediction distribution is obtained. Through this process, the original microorganism concentration prediction can be adjusted according to the patient's activity information, thereby improving the accuracy of disinfection strategies.

[0057] The microorganism concentration prediction distribution is optimized according to the activity range and enhancement coefficient of the patient, and the corrected microorganism concentration prediction distribution is obtained. This process effectively combines the patient's activity, making the disinfection scheme more accurate and personalized, and optimizing the environmental disinfection strategy in the infectious disease department.

[0058] S40: Set multiple atomization concentration constraints in combination with the physical data, illness information, and historical activity range of multiple patients.

[0059] Further, the step S40 of the present application further comprises:

[0060] S41: randomly select first sign data, first illness information and first historical activity range of a first patient.

[0061] Specifically, any patient is randomly selected as the first patient from the plurality of patients, and the first sign data, the first illness information and the first historical activity range of the first patient are obtained, wherein the first sign data includes the current health status, body temperature, pulse, blood pressure and other physiological data of the patient, which can help understand the health status of the patient, especially whether in the acute phase of infection, for example, fever, decreased blood pressure, increased heart rate and other information may indicate that the patient's immune system is challenged and needs special attention; the first illness information includes the disease type, illness stage (such as whether in the acute phase of infection or recovery period) and treatment plan of the patient, which is very important for determining the patient's immunity, viral load and sensitivity to disinfectant, for example, patients receiving immunosuppressive therapy may need higher intensity disinfection, while patients with strong immune system may not need excessive disinfection. The first historical activity range refers to the specific area of the patient's previous activity in the hospital, which can reflect the patient's potential impact on the environment, and these data help determine which areas may need more disinfection resources, for example, if the first patient often activities in the area with strong air flow, these areas may need more attention and disinfection.

[0062] S42: performing tolerable atomization concentration analysis according to the disinfectant type, the first sign data and the first illness information, and predicting to obtain the first maximum atomization concentration.

[0063] Further, the step S42 of the present application further comprises:

[0064] S421: collecting a sample sign data set and a sample illness information set with the disinfectant type as a constraint, and obtaining the maximum atomization concentration that the patient can tolerate under different sample sign data and sample illness information, which is set as a sample atomization concentration, to obtain a sample atomization concentration set; S422: taking the sample sign data set, the sample illness information set and the sample atomization concentration set as training data, and constructing an atomization concentration analysis model based on machine learning; S423: using the atomization concentration analysis model to perform tolerable atomization concentration analysis according to the disinfectant type, the first sign data and the first illness information, and outputting the first maximum atomization concentration.

[0065] Specifically, first, the type of disinfectant is obtained, which determines its effective ingredients, concentration, volatility, and other characteristics. Different types of disinfectants (such as chlorine-based disinfectants, hydrogen peroxide, alcohol, etc.) have different effects on the environment and the human body, so the maximum atomized concentration that the patient can tolerate must be set according to the selected disinfectant; different disinfectants can cause different degrees of irritation or harm to the human body, especially for patients with weak immune systems (such as patients undergoing chemotherapy), so it is crucial to select the appropriate type of disinfectant and set the maximum atomized concentration that the patient can tolerate based on it. Next, with the disinfectant type as a constraint, sample sign data sets and sample condition information sets are collected; and the maximum atomized concentration that the patient can tolerate under different sample signs and sample condition information is collected, that is, according to the patient's sign and condition data, the tolerance of the patient to the disinfectant can be estimated, and through historical data, medical literature, expert opinions or clinical trial data, the maximum concentration of disinfectant that the patient can tolerate under different health conditions (such as weak immune system or normal state) can be determined, for example, for patients with weak and sick, the concentration of atomized disinfectant may need to be reduced to prevent irritation of the respiratory tract or cause other discomfort, and the maximum atomized concentration may also be related to factors such as the patient's age, weight, and severity of illness. For young and healthy patients, they can tolerate higher concentrations of disinfectant atomization, while older or seriously ill patients need lower concentrations. Each sample patient will have a maximum atomized concentration that they can tolerate, and these concentration values will be used as sample atomized concentrations to form a sample atomized concentration set, which contains the maximum tolerance concentration of different patients under different signs and condition conditions. This atomized concentration set will be the basis for subsequent disinfection strategy optimization and adjustment, used to calculate and control the disinfectant concentration that each patient can tolerate, to ensure effective disinfection without causing discomfort. By reasonably setting the maximum atomized concentration that each patient can tolerate, excessive disinfection can be avoided, and respiratory discomfort or other physical reactions caused by high concentration can be reduced, ensuring patient comfort.

[0066] A machine learning algorithm, such as random forest, is used to build a nebulization concentration analysis model. This model will be able to predict the maximum nebulization concentration that a patient can tolerate by inputting the patient's physical data, illness information, and other features; then, using the sample physical data set, sample illness information set, and sample nebulization concentration set as training data, the goal of the training process is to enable the model to identify the relationship between different patients' physiological characteristics, illness information, and their tolerable nebulization concentration. After training is complete, the model can predict the nebulization concentration based on different input data (such as physical signs, illness, etc.), and obtain the nebulization concentration analysis model. Further, the disinfectant type, the first physical data, and the first illness information are input into the nebulization concentration analysis model for tolerable nebulization concentration analysis, and the first maximum nebulization concentration is output. That is, the maximum nebulization disinfectant concentration that the patient can tolerate under the current physical and illness conditions. This concentration value will be used as a constraint condition in the subsequent disinfection scheme to ensure that the patient's tolerance range is not exceeded during disinfection, while achieving sufficient disinfection effect.

[0067] By establishing a machine learning model, the nebulization concentration can be personalized according to the patient's physical signs, illness, disinfectant type, and other information, which can ensure that each patient receives the most appropriate disinfectant concentration and avoid excessive or insufficient disinfection; by accurately predicting the maximum nebulization concentration that the patient can tolerate, respiratory tract irritation and other health problems caused by excessively high concentration can be effectively avoided, ensuring patient comfort and safety; under the premise of safety, by adjusting the nebulization concentration, environmental disinfection can be more efficient, disinfection effect can be improved, and risks caused by excessive disinfection can be reduced.

[0068] S43: Set a first nebulization concentration constraint according to the first historical activity range and the first maximum nebulization concentration, and add it to the plurality of nebulization concentration constraints.

[0069] Specifically, the first historical activity range and the first maximum nebulization concentration are used as the first nebulization concentration constraint, which ensures that the nebulization disinfection concentration does not exceed the patient's tolerance within the patient's activity range; finally, multiple nebulization concentration constraints are obtained by using the same method.

[0070] By combining the patient's historical activity range and maximum nebulization concentration, a customized disinfection concentration constraint can be established for each patient to ensure that the maximum disinfection effect is achieved while avoiding harm or discomfort to the patient; this personalized constraint based on the patient's activity range and maximum tolerance concentration can more accurately control the disinfection process, avoid excessive disinfection in certain areas of the environment, and reduce waste of disinfectants in the air and environment; by setting reasonable nebulization concentration constraints, fine management of the disinfection process can be achieved without affecting the disinfection quality, ensuring that the disinfection intensity of each area matches the patient's needs.

[0071] S50: Constrain the disinfection scheme optimization according to the plurality of atomization concentration constraints, output the optimal disinfection scheme, and control the disinfection robot to perform disinfection control after the future time point based on the corrected microbial concentration prediction distribution.

[0072] Further, the step S50 of the present application further comprises:

[0073] S51: Configure the disinfection atomization concentration based on the corrected microbial concentration prediction distribution, and obtain the atomization concentration distribution; S52: Adjust the disinfection atomization concentration distribution according to the plurality of atomization concentration constraints, and obtain an optimized atomization concentration distribution, wherein if the maximum atomization concentration in the atomization concentration constraint is greater than or equal to the disinfection atomization concentration in the same activity range, no adjustment is made, and if the maximum atomization concentration in the atomization concentration constraint is less than the disinfection atomization concentration in the same activity range, the disinfection atomization concentration is replaced with the maximum atomization concentration; S53: Generate an optimal disinfection scheme according to the predetermined disinfection path and the optimized atomization concentration distribution.

[0074] Specifically, configuring the disinfection atomization concentration based on the corrected microbial concentration prediction distribution means determining the required disinfection concentration for different regions and time points according to the microbial concentration prediction distribution. This process uses microbial concentration data to guide the setting of disinfection atomization concentration, to ensure that the disinfectant can reach an effective concentration for microorganisms, while avoiding excessive or insufficient disinfection. This process involves matching the required microbial concentration with the corresponding disinfection atomization concentration, and configuring the disinfection atomization concentration throughout the infectious disease department area. In the disinfection process of the infectious disease department, it is necessary to ensure that the concentration of the disinfectant matches the distribution of the microbial concentration, to maximize the disinfection effect.

[0075] Then, after determining the preliminary disinfection atomization concentration distribution, further adjustments need to be made according to the individual constraints of the patients (such as the maximum tolerable concentration), to ensure that the disinfection process is both effective and does not cause discomfort or harm to the patients; that is, for each patient's atomization concentration constraint, check the disinfection atomization concentration in the corresponding activity range, and if the disinfection concentration in that area exceeds the maximum tolerable concentration of the patient (i.e. the atomization concentration constraint), adjust it. If the maximum atomization concentration is greater than or equal to the disinfection concentration in the activity range, no adjustment is needed, and the disinfection concentration remains unchanged; if the maximum atomization concentration is less than the disinfection concentration in the activity range, adjust the disinfection concentration to the maximum tolerable concentration, so that the concentration of the disinfectant is within the range that the patient can tolerate, reducing the discomfort or potential harm to the patient. Finally, through these adjustments, an optimized disinfection atomization concentration distribution is generated, ensuring that each patient can be disinfected within the maximum safe concentration range, while maintaining the disinfection effect.

[0076] The predetermined disinfection path refers to the order and manner of disinfection operation, and ensures that each area is fully disinfected, especially areas with frequent patient activities or high-risk areas; based on the optimized atomization concentration distribution, a final disinfection scheme, i.e., an optimal disinfection scheme, is generated in combination with the predetermined disinfection path. Through the combination of microorganism concentration prediction, environmental parameters, patient activity path and individualized atomization concentration constraints, a precise disinfection plan can be generated to avoid excessive disinfection or missed disinfection areas, thereby improving overall disinfection efficiency; by avoiding exposure to high-concentration disinfectants, respiratory irritation and other patient discomfort can be reduced, and it is ensured that the disinfection process does not impose additional burden or health risks on patients.

[0077] The disinfection optimization processing method for a nursing environment provided by the embodiment of the present application has at least the following technical effects:

[0078] By monitoring a plurality of microorganism concentration sequences of a plurality of monitoring points in the infectious disease department coverage area, interpolation prediction of a plurality of unknown points is performed to generate a microorganism concentration sequence array; then, a microorganism concentration prediction distribution after a future time is obtained based on the microorganism concentration sequence array; further, the microorganism concentration prediction distribution is optimized based on the activity path and activity frequency of patients in the infectious disease department coverage area to obtain a corrected microorganism concentration prediction distribution; on the other hand, a plurality of atomization concentration constraints are set in combination with the sign data, illness information and historical activity range of a plurality of patients; finally, the optimal disinfection scheme is output by optimizing the disinfection scheme based on the corrected microorganism concentration prediction distribution with the plurality of atomization concentration constraints as the limit, and the disinfection robot is controlled to perform disinfection control after the future time. That is, through real-time monitoring, dynamic prediction and individualized optimization, precise, efficient and safe disinfection management can be realized, while reducing the potential harm of excessive disinfection to patients, effectively improving the disinfection quality of the infectious disease department of the hospital and the comfort of patients.

[0079] Although the preferred embodiments of the present application have been described, those skilled in the art can make further changes and modifications to these embodiments once they know the basic inventive concept.

[0080] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A disinfection optimization treatment method for a nursing environment, characterized in that the method include: Monitor and obtain several microbial concentration sequences of several monitoring points within the area covered by the infectious disease department, perform interpolation predictions on multiple unknown points, and generate a microbial concentration sequence array; According to the microorganism concentration sequence array, predicting and obtaining a predicted distribution of microorganism concentrations at a future time; Optimizing the predicted microbial concentration distribution based on the activity paths and activity frequencies of patients in the area covered by the infectious disease department to obtain a corrected predicted microbial concentration distribution; Multiple atomization concentration constraints are set based on multiple patients' vital sign data, medical condition information, and historical activity ranges. The vital sign data includes at least current health status, body temperature, pulse, and blood pressure. The medical condition information includes disease type, stage, and treatment plan. The historical activity range refers to the patient's historical activity area in the hospital. Taking the multiple atomization concentration constraints as restrictions, optimizing the disinfection plan according to the corrected microbial concentration predicted distribution, outputting the optimal disinfection plan, and controlling the disinfection robot to perform disinfection control after the future time; The method of predicting and obtaining a predicted distribution of microbial concentrations at a future time based on the microbial concentration sequence array includes: According to the microbial monitoring records of the area covered by the infectious disease department, a sample microbial concentration sequence array set and a sample environmental parameter sequence array set are collected, and sample predicted microbial concentration arrays of different sample microbial concentration sequence arrays and sample environmental parameter sequence arrays after the same historical moment are obtained to obtain a sample predicted microbial concentration array set; Using the sample microbial concentration sequence array set, the sample environmental parameter sequence array set, and the sample predicted microbial concentration array set, a feedforward neural network is trained to construct a concentration prediction plug-in; Using the concentration prediction plug-in, prediction is performed based on the microbial concentration sequence array and the environmental parameter sequence array to obtain a predicted microbial concentration array; Performing concentration distribution fitting based on the predicted microbial concentration array to generate a predicted microbial concentration distribution; Among them, multiple atomization concentration constraints are set by combining the vital signs data, disease information and historical activity range of multiple patients, including: Randomly selecting the first vital sign data, the first condition information, and the first historical activity range of the first patient; Performing a tolerable atomization concentration analysis based on the disinfectant type, the first vital sign data, and the first condition information, and predicting and obtaining a first maximum atomization concentration; setting a first atomization concentration constraint according to the first historical activity range and the first maximum atomization concentration, and adding the constraint to the plurality of atomization concentration constraints; The tolerable atomization concentration analysis is performed based on the disinfectant type, the first vital sign data, and the first condition information, including: Taking the disinfectant type as a constraint, a sample vital sign data set and a sample condition information set are collected, and the maximum aerosol concentration that the patient can tolerate under different sample vital sign data and sample condition information is obtained, set as the sample aerosol concentration, to obtain a sample aerosol concentration set; Using the sample vital sign data set, sample condition information set, and sample aerosol concentration set as training data, an aerosol concentration analysis model is constructed based on machine learning; The atomization concentration analysis model is used to perform a tolerable atomization concentration analysis based on the disinfectant type, the first vital sign data, and the first condition information, and a first maximum atomization concentration is output.

2. A disinfection optimization treatment method for a nursing environment according to claim 1, characterized in that: The monitoring obtains several microbial concentration sequences of several monitoring points in the area covered by the infectious disease department, performs interpolation prediction on multiple unknown points, and generates a microbial concentration sequence array, including: Through the microbial monitoring system, several monitoring points within the area covered by the infectious disease department are regularly monitored to generate several microbial concentration sequences; Based on the Kriging interpolation principle, interpolation prediction of multiple unknown points is performed according to the multiple microbial concentration sequences to generate multiple microbial concentration prediction sequences; A microorganism concentration sequence array is constructed based on the several microorganism concentration sequences and the plurality of microorganism concentration prediction sequences.

3. A disinfection optimization treatment method for a nursing environment according to claim 2, characterized in that: A microbial concentration sequence array is constructed based on the plurality of microbial concentration sequences and the plurality of microbial concentration prediction sequences, comprising: Synchronously monitor and obtain environmental parameters of several monitoring points to generate several environmental parameter sequences, where the environmental parameters include temperature, humidity and airflow direction; mapping and compensating the plurality of microorganism concentration prediction sequences according to the plurality of environmental parameter sequences to obtain a plurality of compensated microorganism concentration prediction sequences; A microorganism concentration sequence array is constructed based on the several microorganism concentration sequences and the multiple compensated microorganism concentration prediction sequences.

4. A disinfection optimization treatment method for a nursing environment according to claim 3, characterized in that: Mapping and compensating the plurality of microorganism concentration prediction sequences according to the plurality of environmental parameter sequences to obtain a plurality of compensated microorganism concentration prediction sequences, including: Combining the location information of the plurality of monitoring points and the plurality of unknown points, and constructing an environmental parameter sequence array and a microorganism concentration prediction sequence array respectively according to the plurality of environmental parameter sequences and the plurality of microorganism concentration prediction sequences; Collecting a sample environmental parameter sequence array set, a sample microbial concentration prediction sequence array set, and multiple sample compensation microbial concentration prediction sequence sets, and performing supervised training on a feedforward neural network until the model converges to obtain a concentration compensation analysis plug-in; The environmental parameter sequence array and the microorganism concentration prediction sequence array are input into the concentration compensation analysis plug-in, and a plurality of compensated microorganism concentration prediction sequences are output.

5. A disinfection optimization treatment method for a nursing environment according to claim 1, characterized in that: According to the activity paths and activity frequencies of patients in the area covered by the infectious disease department, the predicted distribution of microbial concentration is optimized to obtain a corrected predicted distribution of microbial concentration, including: Monitor and obtain multiple activity paths and multiple activity frequencies of multiple patients; Determine the activity range based on the activity path, determine the microbial concentration enhancement coefficient based on the activity frequency analysis, and sequentially analyze and determine multiple activity ranges and multiple microbial concentration enhancement coefficients for multiple patients; The microorganism concentration predicted distribution is optimized and enhanced according to the multiple activity ranges and the multiple microorganism concentration enhancement coefficients to obtain a corrected microorganism concentration predicted distribution.

6. A disinfection optimization treatment method for a nursing environment according to claim 1, characterized in that: The disinfection scheme is optimized based on the predicted distribution of the corrected microbial concentration and the multiple atomization concentration constraints, and an optimal disinfection scheme is output, including: Performing disinfection atomization concentration configuration based on the corrected microbial concentration predicted distribution to obtain an atomization concentration distribution; Adjusting the disinfection atomization concentration distribution according to the multiple atomization concentration constraints to obtain an optimized atomization concentration distribution, wherein if the maximum atomization concentration in the atomization concentration constraint is greater than or equal to the disinfection atomization concentration within the same activity range, no adjustment is performed; if the maximum atomization concentration in the atomization concentration constraint is less than the disinfection atomization concentration within the same activity range, the maximum atomization concentration is used for replacement; An optimal disinfection plan is generated according to the predetermined disinfection path and the optimized atomization concentration distribution.

Images