Medical institution disinfection quality multi-dimensional data monitoring and analyzing system

Abstract

The invention relates to the technical field of medical institution disinfection, and discloses a medical institution disinfection quality multi-dimensional data monitoring and analyzing system. The system comprises a multi-source disinfection data acquisition module, an intelligent analysis processing module and a disinfection control execution module. The multi-source disinfection data acquisition module forms a multi-modal monitoring signal through the heterogeneous data aggregation unit; the intelligent analysis processing module receives the signal, the evaluation unit outputs a disinfection effect evaluation signal and a potential risk early warning signal, the parameter optimization unit obtains an optimal disinfection parameter combination, and the dynamic regulation and control unit outputs a disinfection operation threshold value; in the disinfection control execution module, an adaptive disinfection execution unit adjusts operation parameters, an equipment cooperative control unit adjusts associated equipment parameters and collects feedback, and a man-machine interaction and early warning sub-module outputs a comprehensive safety index. According to the system, multi-source convergence, intelligent analysis and equipment cooperative control of disinfection data are realized, and a comprehensive solution is provided for monitoring, regulating and controlling the disinfection quality of a medical institution.

Description

Technical Field

[0001] This invention relates to the field of disinfection technology in medical institutions, specifically a multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions. Background Technology

[0002] Currently, data collection in medical institution disinfection work faces significant limitations. In most scenarios, disinfection data comes from a single source, often relying on only a few types of equipment or manual recording, making it difficult to cover all kinds of key information in the disinfection process. Data generated by different devices has different formats and standards, forming information silos that cannot be effectively aggregated and integrated, making it difficult to grasp the overall picture of the disinfection work.

[0003] In evaluating disinfection effectiveness, existing methods largely rely on manual sampling or single-indicator testing, resulting in assessments that are both delayed and incomplete. They fail to reflect the real-time effects of disinfection operations and make it difficult to comprehensively determine whether disinfection has achieved its intended goals. Potential risks are often only identified after problems occur, lacking early warning capabilities and leaving risk control in a reactive state.

[0004] The setting of disinfection parameters also has shortcomings. Most are based on empirical values ​​or fixed standards, failing to consider the actual differences in different environments and equipment conditions, making it difficult to achieve dynamic optimization of parameters. The determination of disinfection operation thresholds lacks a flexible adjustment mechanism and cannot be adapted to real-time monitoring data, which may lead to incomplete disinfection or waste of resources.

[0005] Regarding equipment coordination, there is a lack of effective linkage mechanisms between various disinfection devices and related equipment. They operate independently, making it difficult to achieve a synergistic effect. Feedback information on equipment operating status cannot be collected and utilized in a timely manner, affecting the overall control of the disinfection process. At the same time, the interaction between medical staff and the disinfection system is not intuitive or efficient enough, making it impossible to obtain comprehensive safety information in real time, thus limiting the ability to perceive and handle potential risks. Summary of the Invention

[0006] The purpose of this invention is to provide a multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides a multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions, the system comprising:

[0008] A multi-source disinfection data acquisition module, comprising a heterogeneous data aggregation unit, wherein the heterogeneous data aggregation unit generates multimodal monitoring signals;

[0009] The intelligent analysis and processing module receives the multimodal monitoring signals and includes an evaluation unit, a parameter optimization unit, and a dynamic control unit. The evaluation unit processes the multimodal monitoring signals and outputs a disinfection effect evaluation signal and a potential risk warning signal. The parameter optimization unit processes the multimodal monitoring signals and the disinfection effect evaluation signal to obtain the optimal combination of disinfection parameters. The dynamic control unit defines safe disinfection constraints based on the multimodal monitoring signals and outputs a disinfection operation threshold.

[0010] The disinfection control execution module includes an adaptive disinfection execution unit, an equipment collaborative control unit, and a human-machine interaction and early warning submodule. The adaptive disinfection execution unit adjusts the disinfection operation parameters according to the optimal disinfection parameter combination. The equipment collaborative control unit adjusts the operating parameters of associated equipment according to the disinfection operation threshold and the multimodal monitoring signals, and collects real-time status feedback signals. The human-machine interaction and early warning submodule processes the potential risk early warning signals and the disinfection effect evaluation signals and outputs a comprehensive safety index.

[0011] Preferably, the multi-source disinfection data acquisition module includes a distributed sensor array and a mobile detection terminal. The multi-source disinfection data acquisition module is deployed in the treatment area, disinfection supply center, and on the surface of medical devices, and collects raw monitoring signals. The mobile detection terminal acquires microbial load data of key areas through portable detection devices and forms mobile monitoring signals. The heterogeneous data aggregation unit receives the raw monitoring signals and the mobile monitoring signals to achieve cross-device data fusion and form the multimodal monitoring signals. The multimodal monitoring signals include evaluation modal signals, optimization modal signals, and regulation modal signals.

[0012] Preferably, the evaluation unit includes a disinfection effect evaluation submodule based on a fuzzy comprehensive evaluation model and a grey relational analysis model. The evaluation modal signal includes disinfectant concentration data, action time data, ambient temperature and humidity data, equipment operating power data, and historical disinfection pass rate data. The disinfectant concentration data, action time data, ambient temperature and humidity data, equipment operating power data, and historical disinfection pass rate data are input into the fuzzy comprehensive evaluation model and output the disinfection effect evaluation signal. The disinfectant concentration data, ambient temperature and humidity data, and equipment operating power data are fused through an improved grey relational analysis model and the potential risk warning signal is output.

[0013] Preferably, the fuzzy comprehensive evaluation model includes a data preprocessing submodule, a feature weight allocation submodule, a comprehensive evaluation submodule, and an evaluation result visualization submodule. The data preprocessing submodule performs sliding window filtering on the environmental temperature and humidity data and extracts the fluctuation features that satisfy the significant range of the disinfection effect as environmental interference signals. The feature weight allocation submodule uses the analytic hierarchy process (AHP) weighted entropy weight method to process the disinfectant concentration data, the action time data, and the equipment operating power data and outputs the data to the comprehensive evaluation submodule. The comprehensive evaluation submodule defines the safety index calculation rules. The evaluation result visualization submodule generates a dynamic heatmap of the safety index based on a four-color matrix, using matrix blocks of different colors to represent the safety index of different monitoring points.

[0014] Preferably, the parameter optimization unit processes the optimized modal signal and the disinfection effect evaluation signal through a disinfection parameter-effect mapping model, outputs initial disinfection parameters, and uses a particle swarm optimization algorithm to iteratively optimize the initial disinfection parameters to obtain the optimal combination of disinfection parameters. The optimized modal signal includes a disinfection object material signal and a load type signal. The disinfection parameter-effect mapping model is constructed using the disinfection object material signal, the disinfection effect evaluation signal, and the load type signal. The disinfection parameter-effect mapping model determines the direction of parameter adjustment through multi-factor coupling analysis.

[0015] Preferably, the parameter optimization unit further includes a disinfection scheme database, a load feature identification submodule, and a parameter dynamic matching submodule. The disinfection scheme database stores parameter configuration records of historical disinfection operations. The load feature identification submodule classifies the disinfection difficulty level according to the load type signal and the disinfection object material signal. The parameter dynamic matching submodule obtains the optimal disinfection parameter combination through a particle swarm optimization algorithm based on the disinfection difficulty level, the disinfection parameter-effect mapping model, and the disinfection object material signal.

[0016] Preferably, the dynamic control unit defines safe disinfection constraints based on the control mode signal, outputs a disinfection operation threshold, and dynamically adjusts the disinfection operation threshold to ensure it is not lower than the safe operating benchmark value through an adaptive PID control algorithm. The control mode signal includes a real-time microbial monitoring signal and a person density signal in the disinfection area. The real-time microbial monitoring signal and the person density signal in the disinfection area define the safe disinfection constraints. The disinfection operation threshold is calculated by comprehensively considering the microbial load and the frequency of personnel contact. The adaptive PID control algorithm dynamically adjusts the disinfectant supply rate and the equipment running time to ensure that the disinfection operation threshold is not lower than the safe operating benchmark value.

[0017] Preferably, the human-computer interaction and early warning submodule includes a 3D scene visualization platform and a multi-level early warning engine. The 3D scene visualization platform receives the multimodal monitoring signals and the real-time status feedback signals and dynamically renders the disinfection effect distribution process. The multi-level early warning engine processes the potential risk early warning signals and the disinfection effect evaluation signals according to the improved analytic hierarchy process and outputs a comprehensive safety index. When the comprehensive safety index is lower than the safety index threshold, the human-computer interaction and early warning submodule controls the adaptive disinfection execution unit to perform enhanced disinfection operations. The 3D scene visualization platform supports multi-view switching analysis, can display the spatial relationship between disinfection blind spots and key areas in real time, and simulates the effect coverage and resource consumption under different disinfection schemes based on Agent modeling.

[0018] Preferably, the multi-level early warning engine includes a three-level early warning mechanism with color states: blue, yellow, and red. When the pass rate in the disinfection effect evaluation signal drops to a first pass rate threshold, the blue early warning state is activated, increasing the environmental monitoring frequency to an enhanced monitoring frequency. When the pass rate in the disinfection effect evaluation signal drops to a second pass rate threshold, routine medical operations are suspended, and the adaptive disinfection execution unit is controlled to perform enhanced disinfection operations. When the pass rate in the disinfection effect evaluation signal drops to a third pass rate threshold, the multi-level early warning engine issues an emergency isolation signal and controls the medical institution's multi-dimensional data monitoring and analysis system for disinfection quality to initiate a comprehensive disinfection procedure.

[0019] Preferably, the optimal disinfection parameter combination includes disinfectant concentration, duration of action, equipment operating power, and disinfection area priority. The adaptive disinfection execution unit includes an intelligent disinfection equipment cluster and an integrated online concentration monitoring device. The intelligent disinfection equipment cluster can identify the optimal disinfection parameter combination and adjust the agent supply and action time as needed. The integrated online concentration monitoring device receives the optimal disinfection parameter combination and adjusts the effective component content of the disinfectant in real time based on near-infrared spectroscopy analysis. The real-time status feedback signal includes equipment operating temperature, remaining agent, ambient illuminance, surface cleanliness, and microbial residue.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] The heterogeneous data aggregation unit in the multi-source disinfection data acquisition module can integrate disinfection-related data from multiple sources and in different formats to form multimodal monitoring signals. This breaks through the limitations of traditional data acquisition and allows for a comprehensive presentation of various information during the disinfection process. The multimodal monitoring signals provide rich and comprehensive basic information for subsequent analysis and processing, enabling a more three-dimensional understanding of disinfection work.

[0022] The evaluation unit in the intelligent analysis and processing module processes multimodal monitoring signals and outputs disinfection effect evaluation signals and potential risk warning signals. This changes the previous situation of delayed and one-sided evaluation, enabling a more timely and comprehensive reflection of disinfection effectiveness and potential risks. The parameter optimization unit combines multimodal monitoring signals and disinfection effect evaluation signals to derive the optimal combination of disinfection parameters. This allows parameter settings to be dynamically generated based on actual data, rather than relying on fixed standards or experience, thus adapting to disinfection needs in different scenarios. The dynamic control unit defines safe disinfection constraints based on multimodal monitoring signals and outputs disinfection operation thresholds, making threshold determination more flexible and targeted, and allowing adjustments as monitoring data changes.

[0023] The adaptive disinfection execution unit in the disinfection control execution module adjusts operating parameters based on the optimal combination of disinfection parameters, enabling flexible changes in disinfection operations to adapt to different disinfection needs. The equipment coordination control unit adjusts the operating parameters of associated equipment based on disinfection operation thresholds and multimodal monitoring signals, and collects real-time status feedback signals, promoting linkage between devices, making equipment operation more coordinated, and improving the overall operational efficiency of the disinfection system. The human-machine interaction and early warning submodule processes potential risk warning signals and disinfection effect evaluation signals and outputs a comprehensive safety index, allowing medical staff to intuitively and conveniently obtain disinfection-related safety information, enhancing their perception and control over the disinfection process.

[0024] The various modules of the system work together to form a complete closed loop from data acquisition, analysis and processing to execution control. This makes disinfection work no longer a simple superposition of isolated links, but an organically linked whole process in which each link supports and influences each other, working together to improve the quality of disinfection. Attached Figure Description

[0025] Figure 1 This is a schematic diagram illustrating the working principle of the multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions as described in this invention.

[0026] Figure 2 A flowchart illustrating the operation of the multi-source disinfection data acquisition module;

[0027] Figure 3 Flowchart for the fuzzy comprehensive evaluation model;

[0028] Figure 4 This is a flowchart of the threshold adjustment for the dynamic control unit;

[0029] Figure 5 This is a flowchart illustrating the operation of the human-computer interaction and early warning submodule. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Please see Figure 1 This invention provides a multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions, the system comprising:

[0032] The multi-source disinfection data acquisition module includes a heterogeneous data aggregation unit, which processes the collected data to form multimodal monitoring signals. The intelligent analysis and processing module receives these multimodal monitoring signals; its internal evaluation unit processes the signals and outputs disinfection effectiveness evaluation signals and potential risk warning signals. The parameter optimization unit combines the multimodal monitoring signals and disinfection effectiveness evaluation signals to calculate the optimal combination of disinfection parameters. The dynamic control unit determines safe disinfection constraints based on the multimodal monitoring signals and outputs disinfection operation thresholds. The adaptive disinfection execution unit in the disinfection control execution module adjusts the disinfection operation parameters according to the optimal combination of disinfection parameters. The equipment collaborative control unit adjusts the operating parameters of associated equipment based on the disinfection operation thresholds and multimodal monitoring signals, while simultaneously collecting real-time status feedback signals. The human-machine interaction and early warning submodule processes the potential risk warning signals and disinfection effectiveness evaluation signals and outputs a comprehensive safety index.

[0033] Example 1: See Figure 2 The multi-source disinfection data acquisition module comprises a distributed sensor array and mobile detection terminals. These components are distributed according to a pre-defined deployment plan in multiple key areas of the medical institution. The treatment area includes outpatient clinics, wards, operating rooms, and other places where patients directly receive medical services. The disinfection supply center, as the core area for centralized cleaning, disinfection, and sterilization of medical devices, and various medical device surfaces are all equipped with related sensing devices from the distributed sensor array. The distributed sensor array collects raw monitoring signals in real time through various sensors. These raw monitoring signals include, but are not limited to, data on disinfectant concentration changes in different areas, disinfection operation duration, temperature and humidity data in the treatment environment, and power data of the disinfection equipment during operation.

[0034] Mobile testing terminals, carried by medical staff or specialized testing personnel, conduct targeted testing on key areas using portable devices. These key areas include operating table surfaces in operating rooms, contact surfaces of instruments and equipment in ICU wards, and sterile storage areas in the central sterile supply department. The portable devices can quickly acquire microbial load data from these areas, thus generating mobile monitoring signals. These mobile monitoring signals complement the raw monitoring signals collected by the distributed sensor array in terms of data type and coverage. The raw monitoring signals focus on large-scale, continuous monitoring of routine parameters, while the mobile monitoring signals focus on key data such as the specific quantity of microorganisms in key areas.

[0035] The heterogeneous data aggregation unit, as the core processing component of the multi-source disinfection data acquisition module, receives raw monitoring signals from the distributed sensor array and mobile monitoring signals from the mobile detection terminal. Since the raw and mobile monitoring signals originate from different devices, their data formats, transmission protocols, and sampling frequencies differ. The heterogeneous data aggregation unit achieves cross-device data fusion through data conversion, format standardization, and time synchronization. During the data fusion process, data from different sources are verified and filtered to remove outliers and invalid data, ensuring the accuracy and consistency of the fused data and ultimately forming a multimodal monitoring signal.

[0036] Multimodal monitoring signals are categorized into evaluation modal signals, optimization modal signals, and regulation modal signals based on their role in subsequent processing. Evaluation modal signals are primarily used for calculations within the evaluation unit of the intelligent analysis and processing module, and include disinfectant concentration data, action time data, environmental temperature and humidity data, equipment operating power data, and historical disinfection pass rate data. The historical disinfection pass rate data is obtained by statistically analyzing past disinfection operation results, recording the disinfection pass rates in different time periods and areas.

[0037] The evaluation unit of the intelligent analysis and processing module consists of a disinfection effect evaluation sub-module based on a fuzzy comprehensive evaluation model and a grey relational analysis model. Data from the evaluation modal signal are input into the fuzzy comprehensive evaluation model. This model first standardizes the input data, converting data of different magnitudes and units into unified evaluation indicators. Then, based on the degree of influence of each data point on the disinfection effect, corresponding weights are assigned. The disinfection effect is then comprehensively evaluated through fuzzy matrix operations, ultimately outputting a disinfection effect evaluation signal. This signal reflects the degree to which the current disinfection operation meets the standards and its overall level.

[0038] The improved grey relational analysis model is mainly used for early warning of potential risks. It receives disinfectant concentration data, ambient temperature and humidity data, and equipment operating power data from the evaluation modal signal. By calculating the correlation between these data, it analyzes the potential impact of changes in these data on the disinfection effect, identifying combinations of factors that may lead to a decrease in disinfection effectiveness and pose potential risks. For example, when the disinfectant concentration is low and the ambient temperature is high, grey relational analysis can reveal that this combination may increase the risk of microbial residue, thus outputting a potential risk warning signal. This signal can alert relevant personnel to potential problems such as incomplete disinfection.

[0039] Throughout the process, the multi-source disinfection data acquisition module ensured the comprehensiveness and diversity of data sources, the heterogeneous data aggregation unit realized the effective fusion of different types of data, and provided complete and accurate multimodal monitoring signals for subsequent intelligent analysis and processing. The evaluation unit then conducted in-depth analysis of these signals through different models and output relevant signals that reflect the disinfection effect and potential risks, providing data support and analytical basis for the disinfection quality monitoring and management of medical institutions.

[0040] Example 2: See Figure 3 The multi-source disinfection data acquisition module is equipped with a distributed sensor array and mobile detection terminals. These components are distributed in the treatment area, the disinfection supply center, and on the surface of medical devices to collect raw monitoring signals. The mobile detection terminals acquire microbial load data of key areas through portable detection devices, forming mobile monitoring signals. The heterogeneous data aggregation unit receives the raw monitoring signals and mobile monitoring signals, and through cross-device data fusion processing, forms a multimodal monitoring signal that includes evaluation modal signals, optimization modal signals, and regulation modal signals.

[0041] In the intelligent analysis and processing module, the fuzzy comprehensive evaluation model of the evaluation unit includes a data preprocessing submodule, a feature weight allocation submodule, a comprehensive evaluation submodule, and an evaluation result visualization submodule. The data preprocessing submodule uses a sliding window filter for environmental temperature and humidity data. This process performs sliding calculations on continuously collected temperature and humidity data according to a set time window length, filtering out outliers with excessively large instantaneous fluctuations and extracting fluctuation features within the range significantly affecting the disinfection effect. These fluctuation features are defined as environmental interference signals. For example, in high-temperature and high-humidity environments, continuous fluctuations in temperature and humidity may affect the stability of the effective components of the disinfectant; related fluctuation data is extracted as environmental interference signals.

[0042] The feature weight allocation submodule uses an entropy weighting method weighted by the analytic hierarchy process (AHP) to process disinfectant concentration data, action time data, and equipment operating power data. The AHP constructs a judgment matrix to compare the importance of each data point in the disinfection effect pairwise, determining preliminary weights. The entropy weighting method calculates information entropy based on the data's inherent dispersion, obtaining objective weights. The weights obtained from these two methods are combined to form the final feature weights, and the processed data is then sent to the comprehensive evaluation submodule.

[0043] The comprehensive evaluation submodule predefines safety index calculation rules. These rules are based on the characteristic weights and actual values ​​of various data points, and derive the safety index through multi-factor comprehensive calculation. For example, when the disinfectant concentration is within the standard range and the contact time is sufficient, the corresponding safety index will increase accordingly; if the equipment's operating power is unstable, the safety index will be affected to some extent. The comprehensive evaluation submodule calculates the input data according to these rules, generating relevant results reflecting the current disinfection status.

[0044] The assessment results visualization submodule generates a dynamic heatmap of safety indices based on a four-color matrix. Each color corresponds to a different safety index range: green represents a high safety index, yellow a medium safety index, orange a low safety index, and red a very low safety index. Different colored matrix blocks correspond to different monitoring points on the heatmap. The distribution and changes in color visually display the safety index status of each monitoring point, enabling relevant personnel to quickly grasp the disinfection status of different areas.

[0045] The parameter optimization unit processes the optimization modal signals and disinfection effect evaluation signals using a disinfection parameter-effect mapping model. The optimization modal signals include the material signal of the object to be disinfected and the load type signal. The material signal reflects the material properties of the surface of the medical device or treatment area, such as metal, plastic, or fabric; the load type signal reflects the degree and type of contamination of the items to be disinfected, such as organic or inorganic contamination. The parameter optimization unit combines these signals with the disinfection effect evaluation signal to construct a disinfection parameter-effect mapping model. This model, through multi-factor coupling analysis, studies the correlation between parameters such as disinfectant concentration, contact time, and equipment power and the disinfection effect, determining the direction for parameter adjustment.

[0046] After obtaining the initial disinfection parameters, the parameter optimization unit uses a particle swarm optimization algorithm to iteratively optimize these parameters. The particle swarm optimization algorithm simulates information sharing and cooperation among individuals in a swarm; each particle represents a set of disinfection parameter combinations. By continuously updating the particle's position and velocity, it searches for the optimal solution in the parameter space. During the iteration process, the fitness of the particles is evaluated based on the feedback from the disinfection effect assessment signal. Parameter combinations with lower fitness are eliminated, while those with higher fitness are retained and optimized. After multiple iterations, the optimal disinfection parameter combination is obtained.

[0047] Throughout the process, the fuzzy comprehensive evaluation model processes and analyzes various data to generate an assessment result that reflects the disinfection effect, which is then presented in the form of an intuitive heat map. The parameter optimization unit builds a model based on multi-source signals and continuously optimizes disinfection parameters in conjunction with intelligent algorithms, making the disinfection operation more in line with actual needs. The two work together to provide a systematic processing flow for disinfection quality monitoring and parameter adjustment in medical institutions.

[0048] Example 3: See Figure 4 The distributed sensor array and mobile detection terminals of the multi-source disinfection data acquisition module are distributed in a preset manner in the treatment area, disinfection supply center, and on the surface of medical devices. The distributed sensor array continuously collects raw monitoring signals, while the mobile detection terminals acquire microbial load data of key areas through portable detection devices to form mobile monitoring signals. After receiving the two types of signals, the heterogeneous data aggregation unit performs cross-device data fusion processing to generate a multimodal monitoring signal that includes evaluation modal signals, optimization modal signals, and regulation modal signals. The optimization modal signals cover the material signal of the disinfection object and the load type signal, while the regulation modal signals include real-time microbial monitoring signals and the population density signal of the disinfection area.

[0049] The parameter optimization unit comprises a disinfection scheme database, a load feature recognition submodule, and a parameter dynamic matching submodule. The disinfection scheme database stores detailed parameter configuration records for all past disinfection operations, covering information such as disinfectant concentration, contact time, and equipment operating power used under different disinfection objects and environmental conditions. These records are stored chronologically and categorized by disinfection scenario for easy retrieval and recall. The load feature recognition submodule receives load type and disinfection object material signals from the optimization modal signals. The load type signal reflects the degree of contamination and type of contaminants on the items to be disinfected, while the disinfection object material signal indicates whether the object is metal, plastic, rubber, or fabric. Based on these signals, the load feature recognition submodule classifies the disinfection difficulty into multiple levels; for example, light contamination on metal instrument surfaces is level one, and heavy organic contamination on fabric items is level five. Different levels correspond to different processing complexities.

[0050] During runtime, the parameter dynamic matching submodule first references historical parameter configuration records in the disinfection scheme database that are similar to the current disinfection difficulty level and the material of the object being disinfected. Combined with the parameter adjustment direction output by the disinfection parameter-effect mapping model, an initial parameter range is determined. Subsequently, a particle swarm optimization algorithm is used to optimize within this parameter range. The algorithm treats each possible parameter combination as a particle, iteratively updating the particle's position and velocity to move the particle towards a better parameter combination. During the iteration process, the search direction of the particles is continuously adjusted based on the feedback from the disinfection effect evaluation signal, ultimately converging to obtain the optimal disinfection parameter combination.

[0051] After receiving the control mode signal, the dynamic control unit begins to define safe disinfection constraints. Real-time microbial monitoring signals provide data on the quantity and types of microorganisms in the current area, while the population density signal in the disinfection area reflects the number and frequency of people moving within that area per unit time. The dynamic control unit combines these two types of signals to determine the basic requirements that the disinfection operation must meet under the current conditions—that is, the safe disinfection constraints. For example, when the real-time microbial quantity in a treatment area exceeds the standard and the population density is high, the safe disinfection constraints will be increased accordingly.

[0052] The disinfection threshold is determined by comprehensively calculating the microbial load and the frequency of personnel contact, using the following formula:

[0053] T = α × M + β × F

[0054] Where T represents the disinfection operation threshold, M represents the microbial load, F represents the personnel contact frequency, and α and β represent the weighting coefficients of the microbial load and personnel contact frequency, respectively.

[0055] An adaptive PID control algorithm is used to dynamically adjust the disinfection operation threshold to ensure it does not fall below the safe operating baseline. This algorithm collects real-time data on the deviation between the disinfection operation threshold and the safe operating baseline, as well as the rate of change of this deviation. Following proportional, integral, and derivative control laws, it calculates the control variable that needs adjustment. This control variable acts on the disinfectant supply system and the disinfection equipment runtime control system. When the disinfection operation threshold is lower than the safe operating baseline, the algorithm increases the disinfectant supply rate and extends the equipment runtime; when the threshold is higher than the baseline by a certain degree, it appropriately reduces the supply rate and shortens the runtime. Through this dynamic adjustment, the disinfection operation threshold is consistently maintained above the safe operating baseline.

[0056] Throughout the process, the parameter optimization unit integrates historical data and real-time signals, combined with intelligent algorithms, to derive the most suitable disinfection parameters for the current situation. Meanwhile, the dynamic control unit ensures that the disinfection intensity meets safety requirements based on real-time monitoring of microorganisms and personnel movement. Working in tandem, these two units enable precise disinfection operations targeting different objects and contamination levels, while also adjusting to real-time environmental changes, forming a flexible and comprehensive disinfection control system.

[0057] Example 4: See Figure 5 The intelligent analysis and processing module receives multimodal monitoring signals from the multi-source disinfection data acquisition module. These signals include evaluation mode signals, optimization mode signals, and control mode signals. The evaluation mode signals include disinfectant concentration data, action time data, ambient temperature and humidity data, equipment operating power data, and historical disinfection pass rate data. These data are input into the evaluation unit for processing, outputting disinfection effect evaluation signals and potential risk warning signals. The parameter optimization unit obtains the optimal combination of disinfection parameters based on the optimization mode signals and disinfection effect evaluation signals, while the dynamic control unit outputs disinfection operation thresholds based on the control mode signals.

[0058] The human-computer interaction and early warning submodule within the disinfection control execution module includes a 3D scene visualization platform and a multi-level early warning engine. The 3D scene visualization platform receives multimodal monitoring signals and real-time status feedback signals from the equipment collaborative control unit. These real-time status feedback signals include information such as equipment operating temperature, remaining disinfectant levels, ambient illumination, surface cleanliness, and microbial residue levels. Based on these signals, the platform constructs a 3D virtual scene of the medical institution, accurately recreating the spatial layout of the treatment area, disinfection supply center, and medical devices. Dynamic rendering technology is used to display the distribution of disinfection effects in different areas in real time. For example, during disinfection in the operating room, the platform uses varying transparency or color gradients to show the diffusion range of the disinfectant in the air, the degree of coverage on instrument surfaces, and the changes in microbial residue levels over time.

[0059] The 3D scene visualization platform supports multi-view switching analysis. A top-down view allows observation of the disinfection coverage of the entire treatment area, a side view shows the disinfection effect on the surfaces of high-level equipment, and a magnified view focuses on the disinfection status of small parts such as surgical instrument joints. Simultaneously, the platform can display the spatial relationship between disinfection blind spots and key areas in real time. For example, areas behind surgical lights that are not fully covered by disinfectant are marked and their location is displayed in relation to the surrounding aseptic operating areas. Based on Agent modeling technology, the platform can simulate the effect coverage and resource consumption under different disinfection schemes. Each Agent represents a disinfection operation unit, and the simulation displays data such as the amount of disinfectant used, equipment energy consumption, disinfection completion time, and the corresponding disinfection effect coverage under different parameter combinations.

[0060] The multi-level early warning engine employs an improved analytic hierarchy process (AHP) to handle potential risk warning signals and disinfection effectiveness assessment signals. The improved AHP constructs multi-level judgment matrices to weight and comprehensively calculate signals from different sources, ultimately outputting a comprehensive safety index. When the comprehensive safety index falls below a safety index threshold, the human-computer interaction and early warning submodule sends instructions to the adaptive disinfection execution unit to perform enhanced disinfection operations, such as increasing disinfectant concentration or extending the contact time.

[0061] The multi-level early warning engine features a three-tiered warning system with color-coded alerts: blue, yellow, and red. When the pass rate in the disinfection effectiveness assessment signal drops to the first pass rate threshold, the system activates the blue alert. At this point, the environmental monitoring interval is shortened from once per hour to once every 15 minutes to more frequently capture changes in environmental parameters. When the pass rate drops to the second pass rate threshold, the system suspends routine medical procedures, such as routine outpatient examinations and non-emergency surgeries, and controls the adaptive disinfection execution unit to perform enhanced disinfection operations. These enhanced operations include increasing the operating power of disinfection equipment and expanding the disinfection area. When the pass rate drops to the third pass rate threshold, the multi-level early warning engine issues an emergency isolation signal. This signal triggers the access control system to close the entrances and exits of the relevant area and simultaneously controls the entire system to initiate a comprehensive disinfection program. This comprehensive disinfection program covers multiple rounds of all-around disinfection of the air, object surfaces, and medical devices, and prohibits unauthorized personnel from entering the area.

[0062] Example 5: The intelligent analysis and processing module receives multimodal monitoring signals generated by the multi-source disinfection data acquisition module. After processing by the evaluation unit, parameter optimization unit, and dynamic control unit, it outputs disinfection effect evaluation signals, potential risk warning signals, optimal disinfection parameter combinations, and disinfection operation thresholds. The adaptive disinfection execution unit, equipment collaborative control unit, and human-machine interaction and early warning submodule in the disinfection control execution module perform corresponding operations based on these output signals.

[0063] The optimal combination of disinfection parameters includes disinfectant concentration, duration of action, equipment operating power, and disinfection area priority. Disinfectant concentration is determined based on the specific object being disinfected and the degree of contamination; for example, the concentration for high-risk surgical instruments is higher than that for general medical instruments. Duration of action is related to the type and concentration of the disinfectant, as well as the resistance of the microorganisms; some stubborn microorganisms require longer action time to be effectively killed. Equipment operating power must match the atomization and heating requirements of the disinfectant to ensure that the disinfectant acts on the object being disinfected in the best possible way. Disinfection area priority is determined based on patient density and the risk level of medical procedures within the area; operating rooms and ICUs are typically set as the highest priority and receive priority for disinfection resources.

[0064] The adaptive disinfection execution unit consists of an intelligent disinfection equipment cluster and an integrated online concentration monitoring device. The intelligent disinfection equipment cluster includes various devices such as atomizing disinfection machines, ultraviolet disinfection carts, and high-temperature sterilizers. These devices identify the optimal combination of disinfection parameters through internal control chips and automatically adjust the agent supply and treatment time according to the parameter requirements. For example, when receiving the optimal parameter combination for an infectious disease ward, the atomizing disinfection machine will increase the amount of disinfectant sprayed and extend the operating time to ensure that every corner of the ward is exposed to a sufficient concentration of disinfectant; the ultraviolet disinfection cart will adjust the number of lamps turned on and the irradiation duration according to the parameters to achieve the corresponding disinfection effect.

[0065] The integrated online concentration monitoring device is installed near the disinfectant delivery pipeline or the area of ​​action of the disinfection equipment. After receiving the concentration standard from the optimal disinfection parameter combination, it monitors the content of the active ingredient in the disinfectant in real time based on near-infrared spectroscopy. Near-infrared spectroscopy involves irradiating the disinfectant with near-infrared light of a specific wavelength and analyzing the concentration of the active ingredient based on the spectral characteristics of the reflected light. When the concentration deviates from the standard value, the device automatically adjusts the mixing ratio or dosage of the disinfectant to maintain the content of the active ingredient within the optimal range. For example, when the active ingredient degrades due to environmental factors during storage or transportation, the monitoring device will promptly detect and adjust, replenishing with a high-concentration mother liquor to restore the set concentration.

[0066] The equipment collaborative control unit collects real-time status feedback signals during operation. These signals include: equipment operating temperature (the temperature of the disinfection equipment itself and the ambient temperature of the area it operates in); reagent remaining quantity (the amount of various disinfectants remaining for timely replenishment); ambient illuminance (primarily for light-dependent disinfection methods, such as the light intensity during ultraviolet disinfection); surface cleanliness (detecting the amount of dirt residue on the object's surface using optical sensors); and microbial residue (obtained through rapid test strips or sensors to determine the number of microorganisms remaining after disinfection). These real-time status feedback signals are transmitted to the intelligent analysis and processing module and the human-machine interaction and early warning submodule, serving as the basis for parameter optimization, dynamic control, and early warning prompts.

[0067] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0068] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions, characterized in that, include: A multi-source disinfection data acquisition module, comprising a heterogeneous data aggregation unit, wherein the heterogeneous data aggregation unit generates multimodal monitoring signals; The intelligent analysis and processing module receives the multimodal monitoring signals and includes an evaluation unit, a parameter optimization unit, and a dynamic control unit. The evaluation unit processes the multimodal monitoring signals and outputs a disinfection effect evaluation signal and a potential risk warning signal. The parameter optimization unit processes the multimodal monitoring signals and the disinfection effect evaluation signal to obtain the optimal combination of disinfection parameters. The dynamic control unit defines safe disinfection constraints based on the multimodal monitoring signals and outputs a disinfection operation threshold. The disinfection control execution module includes an adaptive disinfection execution unit, an equipment collaborative control unit, and a human-machine interaction and early warning submodule. The adaptive disinfection execution unit adjusts the disinfection operation parameters according to the optimal disinfection parameter combination. The equipment collaborative control unit adjusts the operating parameters of associated equipment according to the disinfection operation threshold and the multimodal monitoring signals, and collects real-time status feedback signals. The human-machine interaction and early warning submodule processes the potential risk early warning signals and the disinfection effect evaluation signals and outputs a comprehensive safety index.

2. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 1, characterized in that, The multi-source disinfection data acquisition module includes a distributed sensor array and a mobile detection terminal. The multi-source disinfection data acquisition module is deployed in the treatment area, disinfection supply center, and on the surface of medical devices to collect raw monitoring signals. The mobile detection terminal acquires microbial load data of key areas through portable detection devices and forms mobile monitoring signals. The heterogeneous data aggregation unit receives the raw monitoring signals and the mobile monitoring signals to achieve cross-device data fusion and form the multimodal monitoring signals. The multimodal monitoring signals include evaluation modal signals, optimization modal signals, and regulation modal signals.

3. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 2, characterized in that, The evaluation unit includes a disinfection effect evaluation submodule based on a fuzzy comprehensive evaluation model and a grey relational analysis model. The evaluation modal signal includes disinfectant concentration data, action time data, ambient temperature and humidity data, equipment operating power data, and historical disinfection pass rate data. The disinfectant concentration data, action time data, ambient temperature and humidity data, equipment operating power data, and historical disinfection pass rate data are input into the fuzzy comprehensive evaluation model and output the disinfection effect evaluation signal. The disinfectant concentration data, ambient temperature and humidity data, and equipment operating power data are fused through an improved grey relational analysis model and the potential risk warning signal is output.

4. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 3, characterized in that, The fuzzy comprehensive evaluation model includes a data preprocessing submodule, a feature weight allocation submodule, a comprehensive evaluation submodule, and an evaluation result visualization submodule. The data preprocessing submodule performs sliding window filtering on the environmental temperature and humidity data and extracts the fluctuation features that satisfy the significant range of the disinfection effect as environmental interference signals. The feature weight allocation submodule uses the analytic hierarchy process (AHP) weighted entropy weight method to process the disinfectant concentration data, the action time data, and the equipment operating power data and outputs the data to the comprehensive evaluation submodule. The comprehensive evaluation submodule defines the safety index calculation rules. The evaluation result visualization submodule generates a dynamic heatmap of the safety index based on a four-color matrix, using matrix blocks of different colors to represent the safety index of different monitoring points.

5. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 2, characterized in that, The parameter optimization unit processes the optimized modal signal and the disinfection effect evaluation signal through a disinfection parameter-effect mapping model, outputs initial disinfection parameters, and uses a particle swarm optimization algorithm to iteratively optimize the initial disinfection parameters to obtain the optimal combination of disinfection parameters. The optimized modal signal includes the disinfection object material signal and the load type signal. The disinfection parameter-effect mapping model is constructed using the disinfection object material signal, the disinfection effect evaluation signal, and the load type signal. The disinfection parameter-effect mapping model determines the direction of parameter adjustment through multi-factor coupling analysis.

6. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 5, characterized in that, The parameter optimization unit further includes a disinfection scheme database, a load feature identification submodule, and a parameter dynamic matching submodule. The disinfection scheme database stores parameter configuration records of historical disinfection operations. The load feature identification submodule classifies the disinfection difficulty level according to the load type signal and the disinfection object material signal. The parameter dynamic matching submodule obtains the optimal disinfection parameter combination through a particle swarm optimization algorithm based on the disinfection difficulty level, the disinfection parameter-effect mapping model, and the disinfection object material signal.

7. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 2, characterized in that, The dynamic control unit defines safe disinfection constraints based on the control mode signal, outputs a disinfection operation threshold, and dynamically adjusts the disinfection operation threshold to ensure it is not lower than the safe operation benchmark value through an adaptive PID control algorithm. The control mode signal includes real-time microbial monitoring signals and human density signals in the disinfection area. The real-time microbial monitoring signals and human density signals in the disinfection area define the safe disinfection constraints. The disinfection operation threshold is calculated by comprehensively considering the microbial load and the frequency of human contact. The adaptive PID control algorithm dynamically adjusts the disinfectant supply rate and equipment operating time to ensure that the disinfection operation threshold is not lower than the safe operation benchmark value.

8. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 1, characterized in that, The human-computer interaction and early warning submodule includes a 3D scene visualization platform and a multi-level early warning engine. The 3D scene visualization platform receives the multimodal monitoring signals and the real-time status feedback signals and dynamically renders the disinfection effect distribution process. The multi-level early warning engine processes the potential risk early warning signals and the disinfection effect evaluation signals according to the improved analytic hierarchy process and outputs a comprehensive safety index. When the comprehensive safety index is lower than the safety index threshold, the human-computer interaction and early warning submodule controls the adaptive disinfection execution unit to perform enhanced disinfection operations. The 3D scene visualization platform supports multi-view switching analysis, can display the spatial relationship between disinfection blind spots and key areas in real time, and simulates the effect coverage and resource consumption under different disinfection schemes based on Agent modeling.

9. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 8, characterized in that, The multi-level early warning engine includes a three-level early warning mechanism with color-coded states: blue, yellow, and red. When the pass rate in the disinfection effect evaluation signal drops to the first pass rate threshold, the blue early warning state is activated, increasing the environmental monitoring frequency to an enhanced monitoring frequency. When the pass rate in the disinfection effect evaluation signal drops to the second pass rate threshold, routine medical procedures are suspended, and the adaptive disinfection execution unit is controlled to perform enhanced disinfection. When the pass rate in the disinfection effect evaluation signal drops to the third pass rate threshold, the multi-level early warning engine issues an emergency isolation signal and controls the medical institution's multi-dimensional data monitoring and analysis system for disinfection quality to initiate a comprehensive disinfection procedure.

10. The multi-dimensional data monitoring and analysis system for disinfection quality in medical institutions according to claim 1, characterized in that, The optimal disinfection parameter combination includes disinfectant concentration, duration of action, equipment operating power, and disinfection area priority. The adaptive disinfection execution unit includes an intelligent disinfection equipment cluster and an integrated online concentration monitoring device. The intelligent disinfection equipment cluster can identify the optimal disinfection parameter combination and adjust the agent supply and action time as needed. The integrated online concentration monitoring device receives the optimal disinfection parameter combination and adjusts the effective ingredient content of the disinfectant in real time based on near-infrared spectroscopy analysis. The real-time status feedback signal includes equipment operating temperature, remaining agent level, ambient illuminance, surface cleanliness, and microbial residue.

Images