Method and system for evaluating outpatient medical medium exposure and pressure stress

Abstract

The invention relates to a method and system for evaluating outpatient medical care medium exposure and pressure stress. The method comprises the following steps: acquiring an original data stream, and binding the original data stream with a preset outpatient service space grid to obtain an original multi-modal data stream; based on an exposure event dynamics attribute dictionary, performing exposure event identification on the original multi-modal data stream to obtain an exposure event sequence; based on the exposure event sequence, calculating the exposure load of each outpatient service area to obtain a dynamic exposure load curve; inputting the dynamic exposure load curve and the toughness factor score corresponding to the outpatient service area into a pre-trained risk prediction model to obtain a group pressure stress risk level; and based on the group pressure stress risk level and the dynamic exposure load curve, summarizing the assessment results of the outpatient service areas to obtain an outpatient service environment pressure stress assessment report. According to the method, the high-voltage area in the outpatient service scene can be quickly positioned, and resource configuration is optimized.

Description

Technical Field

[0001] This invention belongs to the field of intelligent behavior assessment, and in particular relates to a method and system for assessing outpatient medical care media exposure and stress response. Background Technology

[0002] As occupational health management in medical institutions develops towards refinement and prevention, IoT sensing, wearable devices, and multi-source data analysis technologies are maturing. These technologies enable continuous and objective monitoring of the work environment, driving the evolution of stress management from macro-intervention to micro-early warning. Traditional outpatient stress assessment mainly relies on periodic anonymous questionnaires or post-event interviews. This approach collects and statistically analyzes subjective feelings retrospectively. Its processing logic is static and summary-based, aiming to depict a general stress profile over a period of time. However, this traditional questionnaire method has several significant problems. First, it relies on subjective memory, making it susceptible to recency effects and emotional biases, resulting in insufficient data timeliness and accuracy. Second, the assessment results are static cross-sections, failing to reveal the true process of how stress dynamically accumulates and evolves in specific work scenarios. More importantly, traditional methods treat stress as a collection of personal perceptions, and their analysis and improvement suggestions often point to individual stress resistance or a general workload, making it difficult to accurately pinpoint objective, improveable deficiencies in the work system and environmental design that lead to stress. Therefore, it cannot provide managers with clear, actionable, and targeted improvement guidelines. Summary of the Invention

[0003] Therefore, it is necessary to provide a method for assessing outpatient healthcare media exposure and stress response that can accurately pinpoint objective work system defects and environmental design flaws that lead to stress, in order to address the aforementioned technical problems.

[0004] Firstly, this application provides a method for assessing outpatient healthcare media exposure and stress, including:

[0005] The raw data stream is acquired and bound to a pre-defined outpatient space grid to obtain the raw multimodal data stream, which includes system log data, environmental indicator data, and emergency event data.

[0006] Based on the exposure event dynamics attribute dictionary, exposure events are identified in the original multimodal data stream to obtain the exposure event sequence;

[0007] Based on the exposure event sequence, the exposure load of each outpatient area is calculated to obtain a dynamic exposure load curve;

[0008] The dynamic exposure load curve and the corresponding outpatient area resilience factor score are input into a pre-trained risk prediction model to obtain the group stress risk level.

[0009] Based on the group stress risk level and dynamic exposure load curve, the assessment results of each outpatient area are summarized to obtain an outpatient environmental stress assessment report.

[0010] Furthermore, based on the exposure event dynamics attribute dictionary, exposure events are identified in the original multimodal data stream to obtain an exposure event sequence, including:

[0011] The system log data in the original multimodal data stream is matched with the digital event rules in the exposure event dynamics attribute dictionary to obtain digital exposure events;

[0012] Environmental exposure events were obtained by performing sliding time window statistics on environmental indicator data; the environmental indicator data included decibel values ​​and human traffic flow.

[0013] By directly mapping the label field in the emergency event data to the standardized event type in the exposure event dynamics attribute dictionary, proactively reported exposure events can be obtained.

[0014] Based on the dynamic attribute dictionary of exposure events, basic impact values ​​and decay half-lives are assigned to digital exposure events, environmental exposure events, and actively reported exposure events to obtain parameterized exposure events.

[0015] Based on timestamps, parameterized exposure events are sorted by time to obtain an exposure event sequence.

[0016] Furthermore, by performing sliding time window statistics on environmental indicator data, environmental exposure events are obtained, including:

[0017] The environmental indicator data is cleaned to obtain time-series environmental signal data;

[0018] Based on a sliding time window, statistical characteristics of time-series environmental signal data are calculated within each sliding time window to obtain a signal feature vector; the statistical characteristics include intensity characteristics, stability characteristics, and trend characteristics.

[0019] Based on preset environmental threshold rules, the signal feature vector is mapped to the corresponding event type to obtain the initial environmental exposure event;

[0020] Based on the statistical characteristics of the initial environmental exposure event from start time to end time, the event intensity index is calculated.

[0021] The environmental exposure event is obtained by integrating the initial environmental exposure event and the event intensity index.

[0022] Furthermore, based on the exposure event sequence, the exposure load for each outpatient area is calculated to obtain a dynamic exposure load curve, including:

[0023] Based on the exposure event sequence, the residual impact value of each exposure event in the outpatient area is calculated using the following formula:

[0024]

[0025]

[0026] in, To expose the remaining impact of the event at the current calculation time t, The base impact value is given, where t is the current calculation time. The time when the event occurred. It is the attenuation constant. The decay half-life;

[0027] Based on the residual impact value, all exposure events within the outpatient area are summed to obtain the basic dynamic exposure load value;

[0028] Calculate the stacking reinforcement coefficient based on the exposure event sequence and preset stacking rules;

[0029] The dynamic exposure load value at the current time is calculated based on the basic dynamic exposure load value and the stacking enhancement coefficient.

[0030] By integrating the current dynamic exposure load value and the historical dynamic exposure load value, a dynamic exposure load curve is obtained.

[0031] Furthermore, based on the exposure event sequence and preset stacking rules, the stacking reinforcement coefficient is calculated, including:

[0032] The candidate event set is obtained by filtering events within a preset decay time window from the exposure event sequence;

[0033] The candidate event set is matched with the stacking rules to identify the stacking pattern of the candidate event set and obtain a list of stacking pattern instances. The stacking patterns include the same type of dense pattern, the high impact value continuous pattern, and the related event chain pattern.

[0034] For each stacking mode instance in the stacking mode instance list, calculate the single-mode strength value, and based on the single-mode strength value, obtain the basic stacking coefficient through a saturation function mapping;

[0035] Remove duplicate stacked pattern instances from the stacked pattern instance list to obtain pure stacked pattern instances;

[0036] The stacking enhancement coefficient is obtained by weighted summing of the basic stacking coefficients corresponding to the pure stacking mode instances.

[0037] Furthermore, the resilience factor score was obtained through the following method:

[0038] Acquire aggregated assessment data from historical assessment periods, and calculate the core resilience dimensions of the outpatient area based on the aggregated assessment data to obtain the basic dimension indicator values; the core resilience dimensions include resource adequacy, process robustness, and support perception.

[0039] The basic dimension index values ​​are normalized to obtain normalized index values, and based on the standardization function, the normalized index values ​​are converted into scores to obtain the resilience index score.

[0040] Based on the core resilience dimension, a weighted average of the resilience index scores is used to obtain the core dimension score.

[0041] The resilience factor score is obtained by weighted summation of the core dimension scores; the resilience factor score is used to characterize the outpatient area's buffering and recovery capacity against media exposure.

[0042] Secondly, this application also provides a system for assessing outpatient healthcare media exposure and stress, comprising:

[0043] The matching module is used to acquire the raw data stream and bind the raw data stream to the preset outpatient space grid to obtain the raw multimodal data stream; the raw multimodal data stream includes system log data, environmental indicator data and emergency event data;

[0044] The identification module is used to identify exposure events in the original multimodal data stream based on the exposure event dynamics attribute dictionary, and obtain the exposure event sequence.

[0045] The load module is used to calculate the exposure load of each outpatient area based on the exposure event sequence, and obtain a dynamic exposure load curve;

[0046] The prediction module is used to input the dynamic exposure load curve and the corresponding outpatient area resilience factor score into the pre-trained risk prediction model to obtain the group stress risk level;

[0047] The reporting module is used to summarize the assessment results of each outpatient area based on the group stress risk level and dynamic exposure load curve, and generate an outpatient environmental stress assessment report.

[0048] Thirdly, this application also provides a computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement any step of the method provided in the first aspect of this application.

[0049] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any step of the method provided in the first aspect of this application.

[0050] The aforementioned method and system for assessing media exposure and stress response in outpatient healthcare workers involves acquiring raw data streams and binding them to a pre-defined outpatient spatial grid to obtain raw multimodal data streams. These raw multimodal data streams include system log data, environmental indicator data, and emergency event data. Based on an exposure event dynamics attribute dictionary, exposure events are identified within the raw multimodal data streams to obtain exposure event sequences. Based on these sequences, the exposure load for each outpatient area is calculated, resulting in a dynamic exposure load curve. The dynamic exposure load curve and the corresponding outpatient area's resilience factor score are input into a pre-trained risk prediction model to obtain a group stress response risk level. Based on the group stress response risk level and the dynamic exposure load curve, the assessment results for each outpatient area are summarized to generate an outpatient environmental stress response assessment report. This system enables real-time dynamic calculation of the stress load from media exposure for outpatient healthcare workers, accurately pinpointing objective system defects and environmental design flaws that lead to stress, and revealing the true process of how stress dynamically accumulates and evolves with events in specific work scenarios. This improves the accuracy of stress assessment and provides a basis for subsequent targeted improvements. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of a process for assessing outpatient healthcare media exposure and stress stress according to an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of a structure for evaluating outpatient healthcare media exposure and stress response system according to an embodiment of the present invention; Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] In one embodiment, such as Figure 1As shown, a method for assessing media exposure and stress in outpatient healthcare is provided. This embodiment illustrates the method applied to a terminal, but it is understood that the method can also be applied to a server, or to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0056] Step 101: Obtain the raw data stream and bind the raw data stream to the preset outpatient space grid to obtain the raw multimodal data stream; the raw multimodal data stream includes system log data, environmental indicator data and emergency event data.

[0057] The raw data stream refers to the unprocessed data continuously generated from various hospital information systems. This data is mixed and lacks clear structured classification. The outpatient spatial grid refers to a predefined coordinate system that divides the entire outpatient department into multiple independent areas, with each grid cell corresponding to a specific physical space. The raw multimodal data stream refers to the data set formed by binding the raw data stream to the outpatient spatial grid, which is already classified according to spatial location. Multimodality is reflected in the inclusion of several data types that need to be distinguished in subsequent steps. The terminal receives data in real time from all connected data sources, including the hospital information system's database, environmental sensors, and event reporting applications. For each received data piece, it reads the inherent or additional location identification information and matches the location identifier with various cells in the outpatient spatial grid to determine which specific outpatient area the data belongs to. After binding, the originally mixed data is organized into data sets separated by different outpatient areas. Within each area's data set, various types of raw data are still contained. For example, digital system logs are event logs that are extracted in real time from interfaces such as hospital information systems, electronic medical records, and internal communication software, where personal identification information has been removed but departmental areas and timestamps are retained; environmental sensor signals are sensors deployed in various areas of the outpatient department that transmit anonymized environmental indicator data in real time, such as average decibel values ​​and patient flow counts; proactive event reporting is a way for medical staff to proactively mark difficult or emergency events anonymously or by selecting areas through lightweight applications.

[0058] Step 102: Based on the exposure event dynamics attribute dictionary, perform exposure event identification on the original multimodal data stream to obtain the exposure event sequence.

[0059] Specifically, the Exposure Event Dynamics Attribute Dictionary is a predefined rule base or knowledge base that specifies what data patterns can be identified as an exposure event and defines the core characteristics of each type of event. An exposure event sequence refers to a list of all exposure events identified within an outpatient area, arranged in chronological order of their occurrence. The terminal reads data from a specific outpatient area in the raw multimodal data stream, scans the data according to the rules set in the Exposure Event Dynamics Attribute Dictionary, identifies data patterns that match the event characteristics, records the timestamp of each identified event and the event type information obtained from the dictionary, and sorts all identified events within the area from earliest to latest timestamp.

[0060] Step 103: Based on the exposure event sequence, calculate the exposure load for each outpatient area to obtain the dynamic exposure load curve.

[0061] Specifically, exposure load is a quantitative indicator used to represent the total pressure exerted by all exposure events on an outpatient area at a specific moment. A dynamic exposure load curve refers to a curve that changes over time by calculating the exposure load value of an area at regular intervals over a continuous period and then connecting all these load values ​​at those time points. The terminal selects a current calculation moment, reads the sequence of exposure events for the target area, and calculates the residual impact of each event in the sequence at the current moment. The impact of an event does not disappear instantly but persists for a period and gradually weakens. The residual impact of all exposure events in the area at the current moment is combined to obtain a total value, which is the exposure load at that moment. This calculation process is repeated continuously, pushing the current calculation moment forward. Each calculation yields a load value for that moment. Connecting different moments with their corresponding load values ​​on a time-load coordinate graph forms the dynamic exposure load curve.

[0062] Step 104: Input the dynamic exposure load curve and the corresponding outpatient area resilience factor score into the pre-trained risk prediction model to obtain the group stress risk level.

[0063] The resilience factor score is a pre-assessed score that measures the resilience and stress-mitigation capabilities of a particular outpatient area. This score represents the area's intrinsic attributes, such as staffing, process efficiency, and team support. The pre-trained risk prediction model is an algorithmic model trained on extensive historical data, possessing analytical and predictive capabilities. The group stress risk level is a conclusive level provided by the model after analysis, used to evaluate the current level of stress risk faced by the healthcare staff in the area. The terminal selects an outpatient area, using its dynamic exposure load curve and the area's inherent resilience factor score as a complete set of input information. This input information is fed into the pre-trained risk prediction model. Internally, the model analyzes the complex relationship between external stress and internal resilience based on patterns learned from historical data, performs comprehensive calculations, and outputs a clear level result: the group stress risk level. Optionally, the current dynamic exposure load curve of the target area, especially its recent shape, such as the upward slope, peak level, duration of high load, and the inherent resilience factor score of the area, can be input into the pre-trained prediction model. The model will then calculate and output the probability value of the medical staff in the area being in a state of high stress in the current and short term. Based on the preset probability threshold, the probability value will be converted into a clear risk level.

[0064] Step 105: Based on the group stress risk level and dynamic exposure load curve, summarize the assessment results of each outpatient area to obtain the outpatient environmental stress assessment report.

[0065] The Outpatient Environmental Stress Assessment Report is a comprehensive document that summarizes the analysis results of all assessed outpatient areas. The terminal collects assessment results from all outpatient areas, primarily including the group stress risk level and dynamic exposure load curve for each area. These results are integrated for a global analysis to identify high-risk areas, common characteristics of their stress load curves, and the distribution of risk throughout the hospital. The integrated analysis results are then presented in a clear and easy-to-understand format as a complete Outpatient Environmental Stress Assessment Report. Optionally, the assessment results from each area can be summarized and analyzed in depth. The report generation module will generate a stress risk heat map of all outpatient areas, visually displaying high-risk clusters; for each high-risk area, its main stress sources are diagnosed; the dynamic characteristics of the risk are analyzed; and all analyses are integrated to form a structured report.

[0066] This embodiment provides a method for assessing media exposure and stress response of outpatient medical staff. It acquires raw data streams and binds these streams to a pre-defined outpatient spatial grid to obtain raw multimodal data streams. These raw multimodal data streams include system log data, environmental indicator data, and emergency event data. Based on an exposure event dynamics attribute dictionary, exposure events are identified within the raw multimodal data streams to obtain exposure event sequences. Based on these sequences, the exposure load for each outpatient area is calculated, resulting in a dynamic exposure load curve. The dynamic exposure load curve and the corresponding outpatient area's resilience factor score are input into a pre-trained risk prediction model to obtain a group stress response risk level. Based on the group stress response risk level and the dynamic exposure load curve, the assessment results for each outpatient area are summarized to obtain an outpatient environmental stress response assessment report. Through these methods, the stress load of outpatient medical staff exposed to media can be dynamically calculated in real time, accurately locating objective work system defects and environmental design flaws that lead to stress. This reveals the true process of how stress dynamically accumulates and evolves with events in specific work scenarios, thereby improving the accuracy of stress assessment and providing a basis for subsequent targeted improvements.

[0067] In one embodiment, based on an exposure event dynamics attribute dictionary, exposure events are identified in the original multimodal data stream to obtain an exposure event sequence, including:

[0068] Step 201: Match the system log data in the original multimodal data stream with the digital event rules in the exposure event dynamics attribute dictionary to obtain digital exposure events.

[0069] System log data refers to records automatically generated from software platforms such as hospital information systems, medical equipment, and triage systems. The content typically includes status information, error codes, operation logs, and timestamps. Digital event rules are a series of predefined logical judgment conditions in the exposure event dynamics attribute dictionary. These rules are used to identify meaningful abnormal or stress event patterns from the chaotic system logs. A digital exposure event is a specific event identified after successfully matching system log data with digital event rules. The event initially includes basic information such as event type and occurrence time. The terminal extracts system log data belonging to a specific outpatient area from the raw multimodal data stream. Simultaneously, it reads all digital event rules from the exposure event dynamics attribute dictionary, uses these rules to scan and judge the log data, checking whether the record sequence, frequency, and content in the logs meet the conditions set by a certain rule. Once a rule's condition is met, a digital exposure event is triggered. This event is marked as the type corresponding to the rule, and the timestamp corresponding to the log record that meets the condition is recorded as the event occurrence time.

[0070] Step 202: Perform sliding time window statistics on environmental indicator data to obtain environmental exposure events; environmental indicator data includes decibel values ​​and pedestrian traffic.

[0071] Specifically, environmental indicator data refers to physical quantity data that continuously changes over time, collected by various sensors deployed in the outpatient area, mainly including decibel levels and pedestrian flow. Sliding time window statistics is a data processing method that defines a fixed-length time window that slides continuously along the time axis. At each window position, the data within the window is calculated. An environmental exposure event refers to an event identified when the statistical value exceeds a preset threshold after performing sliding time window statistics on environmental indicator data. Optionally, this includes continuous high-noise events or instantaneous peak pedestrian flow events. The terminal sets a statistical window and a sliding step size. For each time window, it calculates the statistical characteristics of the environmental data within the window, including the average decibel level, the standard deviation of the decibel level, and the slope of pedestrian flow change. The calculated statistical characteristics are compared with preset threshold rules. When the conditions are met, an environmental exposure event is generated, and the start time and event type of the event are recorded.

[0072] Step 203: Directly map the label field in the emergency event data to the standardized event type in the exposure event dynamics attribute dictionary to obtain the proactively reported exposure event.

[0073] Specifically, emergency event data refers to data recorded through manual channels or specific systems that describes conflict or abnormal situations. Each data entry typically contains a descriptive tag field. The tag field is a text field in the emergency event data used to briefly describe the nature of the event. Standardized event types are a predefined, unified, and unambiguous list of event categories in the exposure event dynamics attribute dictionary. Actively reported exposure events refer to exposure events identified by directly mapping the tag fields of the emergency event data; these represent stress events actively perceived and reported by healthcare personnel. The terminal reads the tag field content from the emergency event data and simultaneously loads the standardized event type list from the dictionary. A mapping table is established to associate common tag content with standard event types. For each piece of emergency event data, based on the tag, the corresponding standardized event type is found by querying the mapping table, and an actively reported exposure event is created using this type. The event occurrence time is extracted from the reported information.

[0074] Step 204: Based on the exposure event dynamics attribute dictionary, assign basic impact values ​​and decay half-lives to digital exposure events, environmental exposure events, and actively reported exposure events to obtain parameterized exposure events.

[0075] Specifically, the baseline impact value is a quantified numerical value representing the initial stress exerted on healthcare workers by a certain type of exposure event immediately upon its occurrence. The decay half-life is a time duration representing the time required for the stress impact of the event to decay to half of its initial value. Parameterized exposure events refer to event objects that have been assigned the two key kinetic parameters—baseline impact value and decay half-life—for various types of exposure events, enabling quantifiable calculation and comparison of the events. The terminal retrieves the type of each event identified in the previous steps, and queries the exposure event kinetic attribute dictionary using the event type as the key. The dictionary predefines the baseline impact value and decay half-life corresponding to each event type. The parameter values ​​are assigned to the corresponding events, transforming each event from a simple type label into a complete object containing three elements: event type, baseline impact value, and decay half-life.

[0076] Step 205: Based on the timestamp, sort the parameterized exposure events by time to obtain the exposure event sequence.

[0077] Here, the timestamp refers to the specific time of occurrence recorded by each parameterized exposure event object. The exposure event sequence refers to an ordered list formed by arranging all parameterized exposure events within a clinic area according to their timestamps from earliest to latest. The terminal collects all parameterized exposure events belonging to the same clinic area, reads the timestamp from each event object, and uses a sorting algorithm to arrange all events in chronological order of occurrence. After sorting, the exposure event sequence for that area within a specific time period is obtained.

[0078] This embodiment organizes discrete, disordered events into an ordered historical record arranged chronologically, accurately reflecting the distribution of stress events on the timeline. It transforms qualitative events into mathematically operable ones, making subsequent stress perception calculations based on media exposure possible.

[0079] In one embodiment, environmental indicator data is statistically analyzed using a sliding time window to obtain environmental exposure events, including:

[0080] Step 301: Clean the environmental indicator data to obtain time-series environmental signal data.

[0081] Data cleaning refers to a series of operations that preprocess raw data to correct, remove, or rectify errors, omissions, anomalies, or inconsistent formats. Time-series environmental signal data refers to the clean, well-organized, and chronologically ordered sequence of environmental indicator data obtained after data cleaning. Each data point contains a timestamp and multiple corresponding environmental indicator values. The terminal checks the environmental indicator data stream. For temporary data gaps caused by sensor malfunctions or data transmission interruptions, the average or interpolation of data from before and after the data point is used to fill in the gaps. Abnormal data points that deviate significantly from the normal range due to transient interference are identified and processed. Smoothing is achieved using moving average filtering, or extreme noise can be directly removed. This ensures that all data has a uniform timestamp format and consistent numerical units, forming a well-organized data stream.

[0082] Step 302: Based on the sliding time window, calculate the statistical characteristics of the time-series environmental signal data within each sliding time window to obtain the signal feature vector; the statistical characteristics include intensity characteristics, stability characteristics and trend characteristics.

[0083] Specifically, a sliding time window is a fixed-length observation interval that slides continuously along a time axis. Statistical features are quantitative indicators used to describe the distribution and variation patterns of data within the window. The signal feature vector is a mathematical vector, with each dimension corresponding to a calculated statistical feature value. The terminal sets the window length and sliding step size, places the first window starting from the beginning of the time series data, extracts environmental signal data corresponding to all time points covered by the current window, forming a subset of data to be analyzed, and calculates the arithmetic mean of all data points within the window to characterize the overall signal level, the standard deviation of all data points within the window to quantify the intensity of signal fluctuations, and performs linear regression analysis on the data point sequence within the window to calculate its slope. A positive slope indicates an upward trend, a negative slope indicates a downward trend, and the absolute value of the slope reflects the rate of change. The calculated mean, standard deviation, and slope values ​​are combined in a fixed order to form a three-dimensional signal feature vector, which is bound to the end timestamp of the current window. The window slides forward by the step size, repeating the calculation to generate a series of signal feature vectors arranged in chronological order for the entire time series data stream.

[0084] Step 303: Based on the preset environmental threshold rules, map the signal feature vector to the corresponding event type to obtain the initial environmental exposure event.

[0085] Specifically, environmental threshold rules are a series of predefined logical judgment conditions. An initial environmental exposure event refers to a preliminary event identified when a signal feature vector satisfies a certain environmental threshold rule. The terminal loads a pre-defined environmental threshold rule library from a configuration file. Each rule explicitly defines the triggering condition and the corresponding event type. In chronological order, each newly generated signal feature vector is matched against all rules in the rule library. When a feature vector first satisfies the triggering condition of a rule, an initial environmental exposure event object is created. This object records the event type, determined by the triggering rule; the start time, which is the start timestamp of the current window; and the triggering feature vector, which is the value of the feature vector that satisfies the condition. Subsequent feature vectors are then checked; as long as subsequent vectors still satisfy the event's duration condition, the event is considered ongoing, and the latest valid time of the event is updated. For example, a simple finite state machine is run for each type of potential event to determine the start and end of the event. When the condition is first met, the event is marked as a candidate for start; when the condition is continuously met for a minimum duration, the event is confirmed to have started; and when the condition is no longer met and continues for a certain period, the event is marked as ended.

[0086] Step 304: Calculate the event intensity index based on the statistical characteristics of the initial environmental exposure event from start time to end time.

[0087] The event intensity index is a quantifiable value used to measure the severity of an initial environmental exposure event throughout its duration. When the terminal detects that subsequent feature vectors no longer meet the event's duration conditions, it determines the event has ended, recording the end timestamp of the last window that met the conditions as the event's end time. The complete time range of the event is determined. All corresponding signal feature vectors within this complete time range are retrieved from storage, forming an event feature vector sequence. Based on the event type, a predefined intensity calculation algorithm is invoked to aggregate and calculate the event feature vector sequence, yielding the final event intensity index value. For example, for noise events, the intensity index is the average decibel value during the event's duration, or the product of the square root of the average decibel value's duration combined with intensity and duration; for pedestrian events, the intensity index is the peak pedestrian density during the event's duration.

[0088] Step 305: Integrate the initial environmental exposure events and event intensity indicators to obtain the environmental exposure events.

[0089] The environmental exposure event is a complete exposure event object. The terminal merges the initial environmental exposure event object, which includes the event type, start time, end time, and event intensity index, to create a new, complete environmental exposure event object. Its attributes include: event type, start time, end time, and intensity index. This completes the conversion from signal to event and outputs an objective description of the event.

[0090] This embodiment uses threshold judgment to formally identify the numerical characteristics representing environmental conditions as a business event with a clear type and start time, adding a magnitude dimension to the event, enabling quantitative comparison of the severity of different events, and improving the accuracy of outpatient media exposure assessment.

[0091] In one embodiment, based on the exposure event sequence, the exposure load of each outpatient area is calculated to obtain a dynamic exposure load curve, including:

[0092] Step 401: Based on the exposure event sequence, calculate the residual impact value of each exposure event in the outpatient area using the following formula:

[0093]

[0094]

[0095] in, To expose the remaining impact of the event at the current calculation time t, The base impact value is given, where t is the current calculation time. The time when the event occurred. It is the attenuation constant. This refers to the decay half-life.

[0096] The residual impact value refers to the remaining pressure impact of an exposure event at the current calculation moment. It is a quantified value representing the residual historical impact of the event at the current moment. The decay constant is a physical quantity used to calculate the rate of impact decay, determined by the decay half-life. The decay half-life refers to the time required for the base impact value of an exposure event to decay to half. The terminal selects a current calculation moment and iterates through each parameterized exposure event in the exposure event sequence of the outpatient area. For each event, it reads its preset base impact value, event occurrence time, and decay half-life. For each event, it calculates its specific decay constant. This formula ensures that after the half-life, the impact value decays to exactly half of the base value. An exponential decay formula is used for calculation, simulating the natural process of pressure impact decaying exponentially over time. It is the length of time that has elapsed since the event occurred, and the remaining impact value of each event in the sequence is calculated at the current moment.

[0097] Step 402: Based on the residual impact value, sum all exposure events within the outpatient area to obtain the basic dynamic exposure load value.

[0098] Specifically, the baseline dynamic exposure load refers to the sum of the residual impact values ​​of all exposure events within the outpatient area at the current calculation time. It represents the total stress level experienced by the area without considering the interactions between events. The terminal collects the residual impact values ​​of all exposure events belonging to the outpatient area, and the sum of all collected residual impact values ​​is the baseline dynamic exposure load value of the area at the current time.

[0099] Step 403: Calculate the stacking reinforcement coefficient based on the exposure event sequence and the preset stacking rules.

[0100] Specifically, stacking rules are pre-defined logical conditions used to identify specific event occurrence patterns. They describe the temporal density or correlation of events and may amplify the overall pressure. The stacking amplification coefficient is a value greater than or equal to 1, used to quantify the pressure amplification effect caused by event stacking patterns. A coefficient of 1 indicates no amplification, while a larger coefficient indicates a stronger amplification effect. The terminal sets a backtracking time window, filters events occurring within this window from the exposed event sequence, matches candidate events against the stacking rule base, and calculates a pattern strength value for each identified stacking pattern instance based on the number, intensity, and temporal density of events it contains. This strength value is mapped to a stacking coefficient using a pre-defined saturation function to prevent the coefficient from increasing indefinitely. Events repeatedly calculated in pattern recognition are removed to obtain pattern instances. The stacking coefficients corresponding to the pure pattern instances are then weighted and summed to obtain the stacking amplification coefficient.

[0101] Step 404: Calculate the dynamic exposure load value at the current time based on the basic dynamic exposure load value and the stacking enhancement coefficient.

[0102] The current dynamic exposure load value refers to the final, corrected total exposure load value of the outpatient area at the current moment, calculated after considering the event stacking amplification effect. The terminal multiplies the baseline load value by the stacking coefficient to obtain the final dynamic exposure load value corrected for the stacking effect, and sets reasonable upper and lower limits for the result to prevent extreme values ​​from occurring.

[0103] Step 405: Integrate the current dynamic exposure load value and the historical dynamic exposure load value to obtain the dynamic exposure load curve.

[0104] The dynamic exposure load curve is a curve with time on the horizontal axis and dynamic exposure load value on the vertical axis, visually displaying the continuous trajectory of the exposure load of an outpatient area over time. The terminal stores the current dynamic exposure load value and its corresponding timestamp as a data point in the database. It then shifts the current calculation time forward by a fixed time interval to obtain a new current time, uses this new time as the current calculation time, and re-executes the calculation to obtain the dynamic exposure load value for the next time point, storing it again. This cyclical calculation continues. When needed for display or use, all data points within a specified time range are retrieved from the database, and the data points are connected in chronological order on the coordinate system to form the dynamic exposure load curve.

[0105] This embodiment synthesizes a series of load snapshots at discrete time points into a continuous and visualized pressure change trend line, thereby achieving dynamic and continuous monitoring of the pressure level in the outpatient area and improving the accuracy of the assessment.

[0106] In one embodiment, a stacking reinforcement coefficient is calculated based on the exposure event sequence and preset stacking rules, including:

[0107] Step 501: Select events within a preset decay time window from the exposed event sequence to obtain a candidate event set.

[0108] The decay time window is a preset time period used for backtracking. Its length is usually determined based on the typical decay half-life of the event. Its purpose is to capture recent events that still have a significant impact at the current moment and may have a cumulative effect with other events. The candidate event set refers to the set of all exposed events that occurred before the current calculation time and are within this decay time window, selected from the complete exposed event sequence. Using the current calculation time as a reference point, the terminal calculates the start time of the window according to the preset decay time window length. It traverses each event object in the exposed event sequence in chronological order. For each event, its timestamp is read, and it is determined whether the event must have occurred before the current moment and be within the backtracking time window. All event objects that meet the conditions are selected and added to a temporary set. After filtering, a list of all eligible events with temporal information is generated, i.e., the candidate event set. This set will serve as the basis for subsequent pattern recognition.

[0109] Step 502: Match the candidate event set with the stacking rules, identify the stacking pattern of the candidate event set, and obtain a list of stacking pattern instances; the stacking patterns include the same type dense pattern, the high impact value continuous pattern, and the associated event chain pattern.

[0110] Specifically, stacking rules are predefined pattern templates used to describe specific spatiotemporal relationships between events. Stacking patterns are the specific pattern types defined by the rules, mainly including: dense patterns of the same type, referring to multiple events of the same type occurring consecutively within a very short period; high-impact value consecutive patterns, referring to two or more events with very high basic impact values ​​occurring consecutively, regardless of their type; and related event chain patterns, referring to the identification of event sequences with causal or logical connections. A stacking pattern instance refers to a specific combination of events identified in the candidate event set that conforms to a certain stacking rule. A stacking pattern instance list refers to a list of all identified stacking pattern instances. The terminal loads a predefined stacked rule base. Each rule is implemented as an executable judgment logic. Optionally, for dense patterns of the same type, candidate events are grouped by type. For each group of events of the same type, they are sorted by time and scanned using a sliding window. If the number of events within a short time window exceeds a threshold, it is identified as a pattern instance. For continuous high-impact patterns, the event type is ignored, and all events are scanned in chronological order. If the base impact values ​​of two or more consecutive events exceed the high-impact threshold, it is identified as a pattern instance. For chained event patterns, a predefined event cause-effect graph is maintained. The candidate event set is searched to find if there is a path where each event is a potential cause of the next event in the graph. A variant of the graph traversal algorithm is used for implementation. Whenever a pattern is identified, a stacked pattern instance object is created. All identified pattern instance objects are added to a dynamically growing list of stacked pattern instances.

[0111] Step 503: Calculate the single-mode strength value for each stacking mode instance in the stacking mode instance list, and obtain the basic stacking coefficient based on the single-mode strength value through saturation function mapping.

[0112] Specifically, the single-mode strength value is an intermediate calculated value used to measure the strength of the reinforcement effect within a single stacked mode instance. The calculation method depends on the mode type. For dense modes, it is the product of the number of events and the time density; for high-impact modes, it is the sum of the base impact values ​​of all contained events. The saturation function is a specific mathematical function whose characteristic is that as the input value increases, the output value gradually approaches an upper limit, preventing the reinforcement effect from being infinitely amplified. The base stacking coefficient refers to a reinforcement coefficient normalized to a reasonable range after mapping the single-mode strength value through the saturation function. For each mode instance, the terminal calls a specific strength calculation function based on its type. For dense modes of the same type, the strength value is calculated as the product of the number of events / time span and the average base impact value, considering both density and strength. For high-impact continuous modes, the strength value is the sum of the base impact values ​​of consecutive events. The strength value of a chain of related events is the product of the geometric mean of the base impact values ​​of all events in the chain and the chain length. The geometric mean better reflects the overall vulnerability of the chain. The calculated raw strength value is input into a preset saturation function to prevent extreme events from causing the reinforcement coefficient to increase infinitely, ensuring system stability.

[0113] Step 504: Remove duplicate stacked mode instances from the stacked mode instance list to obtain pure stacked mode instances.

[0114] Duplicate stacked pattern instances refer to pattern instances identified by different rules that share most or all of the same events. A group of events can be identified as a dense pattern of the same type, and also as a high-impact continuous pattern due to its high impact value. Pure stacked pattern instances refer to a list of instances remaining after removing duplicate instances from the stacked pattern instance list, where the reinforcement effect contributed by each event is calculated only once, and there are no duplicates. The terminal establishes a mapping table to record which stacked pattern instances reference each event in the candidate event set. It checks the mapping table; if an event is referenced by multiple pattern instances, it is determined that there is a duplication or conflict between these pattern instances. A predefined conflict resolution strategy is used to decide which instance to keep. Priorities are assigned to different types of patterns. For example, associated event chain patterns have the highest priority, followed by high-impact continuous patterns, and dense patterns of the same type have the lowest priority. When multiple instances conflict, the instance with the highest priority is kept. Based on the conflict resolution result, all instances determined to be duplicates are removed from the original list, generating a pure stacked pattern instance list, ensuring that the events contained in each instance in the list are no longer shared by other instances in the list.

[0115] Step 505: The basic stacking coefficients corresponding to the pure stacking mode instances are weighted and summed to obtain the stacking enhancement coefficients.

[0116] The stacking amplification coefficient is a value greater than or equal to 1, representing the overall pressure amplification effect caused by the stacking patterns of all recent events. The terminal reads the preset weight values ​​for each type of stacking pattern from the configuration file. The weights are set based on domain knowledge and reflect the relative contribution of different types of patterns to the pressure amplification. It iterates through the list of pure stacking pattern instances, and for each instance in the list, it obtains the basic stacking coefficient and the corresponding weight to calculate the weighted contribution. The weighted contributions of all instances are added together to obtain the stacking amplification coefficient.

[0117] This embodiment integrates the reinforcing effects of all non-repeating stacking patterns to generate a comprehensive amplification factor, which is used to correct the base exposure load value. This reflects the key psychological principle of stress superposition in the mathematical model, thereby improving the accuracy of the assessment.

[0118] In one embodiment, the toughness factor score is obtained by the following method:

[0119] Step 601: Obtain the aggregated assessment data within the historical assessment period, and calculate the indicators of the core resilience dimension of the outpatient area based on the aggregated assessment data to obtain the basic dimension indicator values; the core resilience dimension includes the resource adequacy dimension, the process robustness dimension, and the support perception dimension.

[0120] The historical assessment period is a pre-defined past timeframe used to collect and aggregate assessment data. Choosing a sufficiently long period ensures stable and statistically significant data, avoiding the impact of short-term fluctuations. The aggregated assessment data refers to data collected and initially summarized from various hospital information systems within the historical assessment period, serving as the foundation for calculating resilience indicators. The core resilience dimensions assess the resilience of the outpatient area in three basic ways: resource adequacy focuses on the availability and matching of tangible resources such as human resources, equipment, and supplies; process robustness focuses on the efficiency, standardization, and ability to cope with disruptions in workflows; and perceived support focuses on intangible factors such as team atmosphere, management support, and psychological safety. The basic dimension indicator values ​​refer to the raw quantitative values ​​obtained by processing the aggregated assessment data using specific formulas for each core resilience dimension; these values ​​typically have actual physical meaning. The terminal extracts raw data from relevant data sources within the historical evaluation period, performs data cleaning to ensure data integrity and accuracy, and calculates indicators by dimension. For example, for the resource adequacy dimension, it calculates indicators such as the ratio of peak nurse-to-patient flow, the average availability rate of key equipment, and the timely replenishment rate of commonly used consumables; for the process robustness dimension, it calculates indicators such as the standard deviation of average patient waiting time, the average completion time of the medical order processing process, and the number of abnormal interruptions caused by process issues; for the support perception dimension, it processes anonymous employee survey data to calculate indicators such as the average score of team collaboration satisfaction, the perception score of superior support, and the resource availability score for work pressure relief. It may also analyze the sentiment tendency of keywords in internal communication software and calculate one or more basic dimension indicator values ​​for each dimension.

[0121] Step 602: Normalize the basic dimension index values ​​to obtain normalized index values, and convert the normalized index values ​​into scores based on the standardization function to obtain the resilience index score.

[0122] Specifically, normalization is a data preprocessing technique aimed at eliminating the influence of different indicators due to variations in units and value ranges, mapping all indicator values ​​to a unified, dimensionless range. A normalized indicator value is a number between 0 and 1 after normalization. A standardization function is a predefined mathematical function used to map normalized indicator values ​​to a more intuitive fractional scale. This function is non-linear to reflect the marginal effect of indicator value changes on resilience. The resilience index score refers to the score obtained after normalization and standardization transformation of each basic indicator. For each basic dimension indicator value, the terminal uses an appropriate normalization method based on the indicator's nature, inputting the normalized indicator value into the standardization function. This function is typically an S-curve function or a variant thereof. The function's purpose is to ensure that the score changes significantly when the indicator value varies within the middle range, and changes gradually when the indicator value approaches extremes, conforming to the law of diminishing marginal utility. The output is the transformed score, i.e., the resilience index score.

[0123] Step 603: Based on the core resilience dimension, perform a weighted average of the resilience index scores to obtain the core dimension score.

[0124] Specifically, weighted averaging is a method of calculating averages. Each value in the calculation is assigned a weight, representing its importance. The result is the sum of the products of each value and its weight, divided by the sum of the weights. Core dimension scoring refers to the comprehensive score representing the overall resilience level of each core resilience dimension, obtained by weighted averaging of all resilience indicator scores under that dimension. The terminal assigns weights to the resilience indicator scores under each core resilience dimension. These weights are set based on domain knowledge and reflect the importance of the indicator to its dimension. For each core dimension, its score is calculated. Since the sum of the weights is normalized to 1, a direct weighted sum is calculated, resulting in three independent core dimension scores, representing the resilience levels in resources, processes, and support, respectively.

[0125] Step 604: The core dimension scores are weighted and summed to obtain the resilience factor score; the resilience factor score is used to characterize the outpatient area's buffering and recovery capacity against media exposure.

[0126] The resilience factor score is a comprehensive score derived from the weighted sum of scores across three core dimensions. It is used to comprehensively characterize the outpatient area's ability to buffer and recover from stressful events. The terminal assigns weights to the three core dimensions, representing the relative importance of resources, processes, and support to overall resilience. The determination of these weights requires in-depth domain research; for example, process robustness is considered the cornerstone and has the highest weight; perceived support impacts long-term health and has the next highest weight; and resource adequacy is fundamental and has the lowest weight. All weights are summed to 1, and the final resilience factor score is calculated through this weighted summation.

[0127] This embodiment uses a resilience factor score to concisely summarize the inherent stress resistance of an outpatient area, enabling stress assessment to simultaneously consider internal resistance, making it more scientific and comprehensive, and improving the reliability of the assessment.

[0128] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0129] Based on the same inventive concept, this application also provides a system for assessing outpatient healthcare media exposure and stress, which implements the aforementioned method for assessing outpatient healthcare media exposure and stress. The solution provided by this system is similar to the implementation described in the above method. Therefore, the specific limitations of one or more embodiments of the system for assessing outpatient healthcare media exposure and stress provided below can be found in the limitations of the method for assessing outpatient healthcare media exposure and stress described above, and will not be repeated here.

[0130] In one exemplary embodiment, such as Figure 2 As shown, a system 700 for assessing outpatient healthcare media exposure and stress is provided, comprising:

[0131] The matching module 701 is used to acquire the raw data stream and bind the raw data stream to the preset outpatient space grid to obtain the raw multimodal data stream; the raw multimodal data stream includes system log data, environmental indicator data and emergency event data;

[0132] The identification module 702 is used to identify exposure events in the original multimodal data stream based on the exposure event dynamics attribute dictionary to obtain an exposure event sequence;

[0133] The load module 703 is used to calculate the exposure load of each outpatient area based on the exposure event sequence, and obtain a dynamic exposure load curve;

[0134] Prediction module 704 is used to input the dynamic exposure load curve and the corresponding outpatient area resilience factor score into the pre-trained risk prediction model to obtain the group stress risk level;

[0135] Report module 705 is used to summarize the assessment results of each outpatient area based on the group stress risk level and dynamic exposure load curve, and obtain an outpatient environmental stress assessment report.

[0136] Furthermore, the identification module 702 is also used for:

[0137] The system log data in the original multimodal data stream is matched with the digital event rules in the exposure event dynamics attribute dictionary to obtain digital exposure events;

[0138] Environmental exposure events were obtained by performing sliding time window statistics on environmental indicator data; the environmental indicator data included decibel values ​​and human traffic flow.

[0139] By directly mapping the label field in the emergency event data to the standardized event type in the exposure event dynamics attribute dictionary, proactively reported exposure events can be obtained.

[0140] Based on the dynamic attribute dictionary of exposure events, basic impact values ​​and decay half-lives are assigned to digital exposure events, environmental exposure events, and actively reported exposure events to obtain parameterized exposure events.

[0141] Based on timestamps, parameterized exposure events are sorted by time to obtain an exposure event sequence.

[0142] Furthermore, the identification module 702 is also used for:

[0143] The environmental indicator data is cleaned to obtain time-series environmental signal data;

[0144] Based on a sliding time window, statistical characteristics of time-series environmental signal data are calculated within each sliding time window to obtain a signal feature vector; the statistical characteristics include intensity characteristics, stability characteristics, and trend characteristics.

[0145] Based on preset environmental threshold rules, the signal feature vector is mapped to the corresponding event type to obtain the initial environmental exposure event;

[0146] Based on the statistical characteristics of the initial environmental exposure event from start time to end time, the event intensity index is calculated.

[0147] The environmental exposure event is obtained by integrating the initial environmental exposure event and the event intensity index.

[0148] Furthermore, the load module 703 is also used for:

[0149] Based on the exposure event sequence, the residual impact value of each exposure event in the outpatient area is calculated using the following formula:

[0150]

[0151]

[0152] in, To expose the remaining impact of the event at the current calculation time t, The base impact value is given, where t is the current calculation time. The time when the event occurred. It is the attenuation constant. The decay half-life;

[0153] Based on the residual impact value, all exposure events within the outpatient area are summed to obtain the basic dynamic exposure load value;

[0154] Calculate the stacking reinforcement coefficient based on the exposure event sequence and preset stacking rules;

[0155] The dynamic exposure load value at the current time is calculated based on the basic dynamic exposure load value and the stacking enhancement coefficient.

[0156] By integrating the current dynamic exposure load value and the historical dynamic exposure load value, a dynamic exposure load curve is obtained.

[0157] Furthermore, the load module 703 is also used for:

[0158] The candidate event set is obtained by filtering events within a preset decay time window from the exposure event sequence;

[0159] The candidate event set is matched with the stacking rules to identify the stacking pattern of the candidate event set and obtain a list of stacking pattern instances. The stacking patterns include the same type of dense pattern, the high impact value continuous pattern, and the related event chain pattern.

[0160] For each stacking mode instance in the stacking mode instance list, calculate the single-mode strength value, and based on the single-mode strength value, obtain the basic stacking coefficient through a saturation function mapping;

[0161] Remove duplicate stacked pattern instances from the stacked pattern instance list to obtain pure stacked pattern instances;

[0162] The stacking enhancement coefficient is obtained by weighted summing of the basic stacking coefficients corresponding to the pure stacking mode instances.

[0163] Furthermore, the system also includes a resilience module for:

[0164] Acquire aggregated assessment data from historical assessment periods, and calculate the core resilience dimensions of the outpatient area based on the aggregated assessment data to obtain the basic dimension indicator values; the core resilience dimensions include resource adequacy, process robustness, and support perception.

[0165] The basic dimension index values ​​are normalized to obtain normalized index values, and based on the standardization function, the normalized index values ​​are converted into scores to obtain the resilience index score.

[0166] Based on the core resilience dimension, a weighted average of the resilience index scores is used to obtain the core dimension score.

[0167] The resilience factor score is obtained by weighted summation of the core dimension scores; the resilience factor score is used to characterize the outpatient area's buffering and recovery capacity against media exposure.

[0168] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of an outpatient healthcare media exposure and stress response method as described above.

[0169] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0170] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0171] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A method for assessing outpatient healthcare media exposure and stress, characterized in that, The method includes: The raw data stream is acquired and bound to a preset outpatient space grid to obtain a raw multimodal data stream; the raw multimodal data stream includes system log data, environmental indicator data, and conflict event data; Based on the exposure event dynamics attribute dictionary, exposure events are identified in the original multimodal data stream to obtain an exposure event sequence; Based on the exposure event sequence, the exposure load of each outpatient area is calculated to obtain a dynamic exposure load curve; The dynamic exposure load curve and the corresponding resilience factor score of the outpatient area are input into a pre-trained risk prediction model to obtain the group stress risk level. Based on the group stress risk level and the dynamic exposure load curve, the assessment results of each outpatient area are summarized to obtain an outpatient environmental stress assessment report.

2. The method according to claim 1, characterized in that, The exposure event sequence is obtained by identifying exposure events in the original multimodal data stream based on the exposure event dynamics attribute dictionary, including: The system log data in the original multimodal data stream is matched with the digital event rules in the exposure event dynamics attribute dictionary to obtain digital exposure events; Environmental exposure events are obtained by performing sliding time window statistics on the environmental indicator data; the environmental indicator data includes decibel values ​​and pedestrian traffic. The label field in the conflict event data is directly mapped to the standardized event type in the exposure event dynamics attribute dictionary to obtain the proactively reported exposure event; Based on the exposure event dynamics attribute dictionary, the digital exposure event, the environmental exposure event, and the actively reported exposure event are assigned a basic impact value and a decay half-life to obtain parameterized exposure events; Based on the timestamp, the parameterized exposure events are sorted by time to obtain the exposure event sequence.

3. The method according to claim 2, characterized in that, The process of performing sliding time window statistics on the environmental indicator data to obtain environmental exposure events includes: The environmental indicator data is cleaned to obtain time-series environmental signal data; Based on a sliding time window, statistical features of the time-series environmental signal data are calculated within each sliding time window to obtain a signal feature vector; the statistical features include intensity features, stability features, and trend features. Based on preset environmental threshold rules, the signal feature vector is mapped to the corresponding event type to obtain the initial environmental exposure event; Based on the statistical characteristics of the initial environmental exposure event from start time to end time, an event intensity index is calculated; The environmental exposure event is obtained by integrating the initial environmental exposure event and the event intensity index.

4. The method according to claim 1, characterized in that, The process of calculating the exposure load for each outpatient area based on the exposure event sequence to obtain a dynamic exposure load curve includes: Based on the exposure event sequence, the residual impact value of each exposure event in the outpatient area is calculated using the following formula: in, To expose the remaining impact of the event at the current calculation time t, The base impact value is given, where t is the current calculation time. The time when the event occurred. It is the attenuation constant. The decay half-life; Based on the remaining impact value, all the exposure events in the outpatient area are summed to obtain the basic dynamic exposure load value; Based on the exposure event sequence and the preset stacking rules, the stacking reinforcement coefficient is calculated; Based on the basic dynamic exposure load value and the stacking enhancement coefficient, calculate the dynamic exposure load value at the current time; The dynamic exposure load curve is obtained by integrating the current dynamic exposure load value and the historical dynamic exposure load values.

5. The method according to claim 4, characterized in that, The calculation of the stacking reinforcement coefficient based on the exposure event sequence and preset stacking rules includes: From the exposure event sequence, events within a preset decay time window are selected to obtain a candidate event set; The candidate event set is matched with the stacking rules to identify the stacking pattern of the candidate event set and obtain a stacking pattern instance list; the stacking patterns include the same type dense pattern, the high impact value continuous pattern, and the associated event chain pattern. For each stacking mode instance in the stacking mode instance list, a single-mode strength value is calculated, and based on the single-mode strength value, a basic stacking coefficient is obtained through a saturation function mapping. Remove duplicate stacking pattern instances from the stacking pattern instance list to obtain pure stacking pattern instances; The stacking enhancement coefficient is obtained by weighted summing of the basic stacking coefficients corresponding to the pure stacking mode instance.

6. The method according to claim 1, characterized in that, The toughness factor score was obtained through the following method: Acquire aggregated assessment data from historical assessment periods, and calculate the core resilience dimension indicators of the outpatient area based on the aggregated assessment data to obtain the basic dimension indicator values; the core resilience dimension includes resource adequacy dimension, process robustness dimension, and support perception dimension; The basic dimension index values ​​are normalized to obtain normalized index values, and based on the standardization function, the normalized index values ​​are converted into scores to obtain the resilience index score. Based on the core resilience dimensions, the resilience index scores are weighted and averaged to obtain the core dimension score. The resilience factor score is obtained by weighted summation of the scores of the core dimensions. The resilience factor score is used to characterize the outpatient area's ability to buffer and recover from media exposure.

7. A system for assessing outpatient healthcare media exposure and stress, characterized in that, The system includes: The matching module is used to acquire the raw data stream and bind the raw data stream to a preset outpatient space grid to obtain the raw multimodal data stream; the raw multimodal data stream includes system log data, environmental indicator data, and conflict event data; The identification module is used to identify exposure events in the original multimodal data stream based on the exposure event dynamics attribute dictionary, and obtain an exposure event sequence. The load module is used to calculate the exposure load of each outpatient area based on the exposure event sequence, and obtain a dynamic exposure load curve; The prediction module is used to input the dynamic exposure load curve and the corresponding resilience factor score of the outpatient area into a pre-trained risk prediction model to obtain the group stress risk level. The reporting module is used to summarize the assessment results of each outpatient area based on the group stress risk level and the dynamic exposure load curve, and obtain an outpatient environmental stress assessment report.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images