Real-time monitoring system and method for nursing pressure based on multimodal data

Abstract

The present invention provides a real-time monitoring system and method for nursing pressure based on multimodal data, and the system includes a data acquisition module, an intelligent analysis module, an analysis module, and a real-time feedback module. The data acquisition module synchronously collects the physiological data, behavioral activity data, and environmental context data of the nursing staff and forms a nursing signal; the analysis module receives the nursing signal from the data acquisition module, and receives the dynamic context information from the intelligent analysis module, and generates a pressure state signal containing a stress event assessment; the real-time feedback module receives the pressure state signal containing a stress event assessment from the analysis module, and generates and outputs a real-time feedback signal based on the pressure state signal. The present invention provides a real-time monitoring system and method for nursing pressure based on multimodal data, which can solve the problem that traditional wearable devices only collect physiological behavioral data and cannot be associated with specific nursing scenarios such as drug administration operations and patient first aid.

Description

Technical Field

[0001] The present invention relates to the field of medical health information technology, and in particular to a real-time monitoring system and method for nursing pressure based on multimodal data. Background Art

[0002] Nursing is a high-pressure, core healthcare system. World Health Organization statistics show that the burnout rate among nurses exceeds 35%. Existing stress monitoring technologies have significant limitations: mainstream solutions rely on wearable devices to collect physiological indicators such as heart rate variability. However, the frequent movement involved in nursing work leads to significant motion artifacts. For example, an accelerated heart rate while carrying a patient can be misinterpreted as psychological stress. Behavioral monitoring technologies use accelerometers to analyze gait, but they cannot distinguish between normal brisk walking and rapid movements caused by anxiety. Environmental sensors record noise levels but are not linked to nursing tasks—the same decibel level can have vastly different stress impacts in routine inspections and emergency situations. A more prominent problem lies in data silos: physiological monitoring devices, hospital information systems, and environmental sensor networks operate independently, making stress assessments detached from the actual work environment. For example, when existing systems detect a peak in electrical skin charge, they cannot automatically track whether the nurse was handling chemotherapy drugs or addressing questions from a family member at that moment.

[0003] Current academic research attempts multimodal fusion, but two major bottlenecks remain. First, it uses a simple weighted average to fuse ECG, gait, and acoustic features, failing to establish a dynamic mapping between nursing operation phases and stress indicators. This results in subtle hand tremors in night shift medication dispensing scenarios being overlooked by general algorithms. Second, it relies on a centralized cloud computing architecture, resulting in a delay of over three minutes between data collection and feedback, missing the optimal intervention window for stressful events. While existing patents propose integrating stress assessment with electronic medical records, these only statically link patient illnesses with historical stress data, failing to capture the real-time contextual shifts as nurses move around the ward. Marketed products, such as the "Nursing Angel Bracelet," only provide vibration alerts for excessive heart rate, failing to distinguish between physiological and psychological stress sources. Therefore, there is an urgent need to develop a real-time monitoring system that is deeply embedded in the nursing workflow and dynamically links context, physiology, and behavior to accurately identify the causes of stressful events and implement tiered interventions. Summary of the Invention

[0004] In view of the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a real-time monitoring system and method for nursing pressure based on multimodal data, which is used to solve the problem that traditional wearable devices only collect physiological behavioral data and cannot be associated with specific nursing scenarios such as medication operations and patient first aid. The present invention uses multi-source sensors to synchronously collect physiological data, behavioral data, and environmental data in nursing operations; the intelligent analysis module parses the hospital information system and dynamically generates three-dimensional situational information such as task type, regional risk, and patient criticality index; the analysis module establishes nursing stage-stress characterization association rules: monitoring the correlation between hand tremor and heart rate variability during the drug preparation stage, and analyzing the matching degree of gait and environmental channels during the patient transfer stage; when the situation is dynamically associated with abnormal physiological behavior, a graded pressure event marker is triggered, and precise intervention is implemented through tactile feedback equipment and the administrator terminal.

[0005] The present invention provides a real-time monitoring system for nursing pressure based on multimodal data, comprising:

[0006] Data acquisition module, which synchronously collects the physiological data, behavioral activity data and environmental situation data of the nursing staff and forms nursing signals;

[0007] Intelligent analysis module, which receives nursing signals from the data acquisition module and extracts dynamic situational information related to the current working status of the nursing staff;

[0008] The analysis module receives nursing signals from the data acquisition module and dynamic context information from the intelligent analysis module. Through the context perception and stress event association model based on the nursing workflow, the analysis module dynamically associates and integrates physiological data, behavioral activity data, and dynamic context information to generate a stress state signal that includes a stress event assessment. The stress event association model establishes a spatiotemporal correlation map centered on the nursing workflow stage. During the medication operation period, it focuses on monitoring the correlation between hand tremor behavioral data and heart rate variability physiological data. During the patient transfer stage, it strengthens the analysis of the matching degree between gait urgency behavioral data and the width of the environmental channel.

[0009] The real-time feedback module receives the pressure state signal containing the pressure event evaluation from the analysis module, and generates and outputs a real-time feedback signal based on the pressure state signal.

[0010] In one embodiment of the present invention, the data acquisition module includes a wearable sensing unit and an environmental perception unit integrated into a nursing staff badge; the wearable sensing unit continuously collects electrocardiogram signals and galvanic skin response signals through flexible bioelectrodes to form physiological data, and captures limb movement trajectories and posture changes through an inertial measurement unit to form behavioral activity data; the environmental perception unit collects sound pressure characteristics of the working area through distributed noise sensors to form environmental context data, and receives nursing task lists and patient critical value alarm records pushed by the hospital information system in real time through a near-field communication interface; the wearable sensing unit and the environmental perception unit fuse the three types of data into nursing signals through a timestamp synchronization mechanism.

[0011] In one embodiment of the present invention, the context-aware operations performed by the intelligent analysis module include: parsing the hospital information system data stream from the nursing signal to identify the types of tasks currently performed by the nursing staff, such as drug preparation, critical care, and family communication; combining the positioning beacon coordinates with the ward electronic fence information to determine whether it is in a pressure-graded area such as the surgical preparation area, isolation ward or nurse station; obtaining the severity index and special nursing needs mark of the current case in real time by associating with the patient's electronic medical record; and finally integrating the task type, the regional risk level associated with the nursing signal, and the case severity index into a dynamic context information vector.

[0012] In one embodiment of the present invention, the stress event association model in the analysis module operates according to the following process: establishing a spatiotemporal association map centered on the nursing workflow stage, focusing on monitoring the correlation between hand tremor behavioral data and heart rate variability physiological data during the medication operation period, and strengthening the analysis of the matching degree between gait urgency behavioral data and environmental channel width during the patient transfer stage; when a sudden increase in skin conductance occurs in a critically ill patient monitoring scenario accompanied by abnormal respiratory rate, a first-level stress event marker is triggered; when a sustained high heart rate and irregular activity path are detected in a cross-ward continuous processing call ring scenario, a second-level stress event marker is triggered; and finally, a stress state signal carrying a stress level label and event cause classification is generated.

[0013] In one embodiment of the present invention, a context-weighted mechanism is introduced into the dynamic association fusion process: the weight coefficient of the insufficient ambient light factor is automatically increased in the night shift workflow stage, and the correction parameter of the behavioral data caused by the restraint of protective equipment is added in the infectious disease isolation area workflow stage; when the system recognizes that nursing staff continuously handle high-risk drug preparation tasks, a high-sensitivity analysis mode is activated for the subtle hand tremor behavior data collected during the same period; when the communication context of the patient's family continues to be extremely high, the voice emotion feature extraction algorithm is activated to enhance the analysis dimension of the environmental context data.

[0014] In one embodiment of the present invention, the real-time feedback module includes a multi-level output channel: when the pressure status signal contains a first-level stress event marker, the tactile belt worn by the caregiver is driven to generate a low-frequency vibration prompt, and a deep breathing guidance animation is projected onto their smart glasses; when the signal contains a second-level stress event marker, a predetermined soothing audio is played through the nurse station broadcast system, and an encrypted early warning text is sent to the nursing team leader terminal at the same time; all feedback signals are accompanied by a stress event cause code, among which events related to drug verification errors trigger yellow visual warnings, and events related to sudden changes in the patient's condition trigger red visual warnings.

[0015] In one embodiment of the present invention, the analysis module has a built-in personalized calibration engine: baseline physiological parameters are automatically collected during the low-stress period of morning shift handover for nursing staff to establish an individualized resting heart rate and gait smoothness baseline; when an abnormal ECG signal occurs in the midday workflow, the person's baseline data is called to perform drift compensation calculations; by continuously learning the stress response pattern during the shift cycle, the assessment threshold coefficients for the night shift and the day shift are dynamically adjusted; and a personalized stress sensitivity factor report is generated weekly for optimizing the associated model parameters.

[0016] In one embodiment of the present invention, the stress event assessment includes a cause tracing function: when the system detects that an abnormal ECG signal and an abnormal positioning data occur simultaneously, it automatically traces back the workflow context data of the previous ten minutes. If it is found that the person handles three ward emergency calls in succession and passes through a noisy corridor environment, a stress event report with dual attribution of environmental interference and task overload is generated; when the behavioral data shows repeated returns to the pharmacy, the failed medication collection record of the drug management system is associated to form a process-blocking stress event analysis chain.

[0017] In one embodiment of the present invention, dynamic situational information includes team collaboration indicators: the number of other caregivers within a five-meter range is obtained through Bluetooth beacon ranging, and the team response delay time is calculated in combination with voice activity detection; when the team personnel density in the critical patient transfer scenario is lower than the threshold and the call response times out, the situational pressure weight coefficient is automatically increased; a team collaboration loss mark is added to the stress state signal and a request for reinforcement is recommended.

[0018] The present invention also provides a real-time monitoring method for nursing pressure based on multimodal data, comprising:

[0019] S1: Synchronously collect the physiological data, behavioral activity data and environmental situation data of the nursing staff and form nursing signals;

[0020] S2: Receives nursing signals from the data acquisition module and extracts dynamic situational information related to the nursing staff's current working status;

[0021] S3: Receives nursing signals from the data acquisition module and dynamic situational information from the intelligent analysis module. Through the situational awareness and stress event association model based on the nursing workflow, it dynamically associates and integrates physiological data, behavioral activity data, and dynamic situational information to generate a stress state signal that includes stress event assessment.

[0022] S4: Receive a pressure state signal including a pressure event assessment from the analysis module, and generate and output a real-time feedback signal based on the pressure state signal.

[0023] The present invention provides a real-time nursing pressure monitoring system and method based on multimodal data. The system synchronously collects physiological data, behavioral data, and environmental data in nursing operations through multi-source sensors. The intelligent analysis module parses the hospital information system and dynamically generates three-dimensional situational information such as task type, regional risk, and patient criticality index. The analysis module establishes nursing stage-stress characterization association rules: monitoring the correlation between hand tremor and heart rate variability during the drug preparation stage, and analyzing the matching degree between gait and environmental channels during the patient transfer stage. When the situation is dynamically associated with abnormal physiological behavior, a graded stress event marker is triggered, and precise intervention is implemented through tactile feedback equipment and the administrator terminal. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 This is the system architecture diagram of the real-time monitoring system for nursing stress based on multimodal data;

[0026] Figure 2 A schematic diagram showing the workflow of a real-time monitoring system for nursing stress based on multimodal data;

[0027] Figure 3 Flowchart of the method for real-time monitoring of nursing stress based on multimodal data. DETAILED DESCRIPTION

[0028] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0029] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0030] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0031] See Figure 1-3 , which shows the real-time nursing stress monitoring system and method based on multimodal data of the present invention. The real-time nursing stress monitoring system based on multimodal data of the present invention includes a data acquisition module, an intelligent analysis module, an analysis module, and a real-time feedback module. The data acquisition module synchronously collects the physiological data, behavioral activity data, and environmental context data of the nursing staff and forms a nursing signal; the data acquisition module synchronously collects the physiological data, behavioral activity data, and environmental context data of the nursing staff and forms a nursing signal; the analysis module receives the nursing signal from the data acquisition module and the dynamic context information from the intelligent analysis module. Through the context perception and stress event association model based on the nursing workflow, the physiological data, behavioral activity data, and dynamic context information are dynamically associated and integrated to generate a stress state signal containing a stress event assessment; the stress event association model establishes a spatiotemporal correlation map centered on the nursing workflow stage, focusing on monitoring the correlation between hand tremor behavioral data and heart rate variability physiological data during the medication operation period, and strengthening the analysis of the matching degree between gait urgency behavioral data and environmental channel width during the patient transfer stage; the real-time feedback module receives the stress state signal containing a stress event assessment from the analysis module, and generates and outputs a real-time feedback signal based on the stress state signal.

[0032] like Figure 1As shown, the real-time monitoring system for nursing pressure based on multimodal data includes a collaborative operation mechanism of a data acquisition module, an intelligent analysis module, an analysis module and a real-time feedback module. The data acquisition module realizes the synchronous capture of three types of data through embedded hardware. The flexible fabric electrode embedded in the nursing staff badge collects single-lead ECG signals at a sampling rate of 256 Hz. The electrode adopts a silver-silver chloride coating and an elastic fiber base to ensure that the skin contact impedance is stable below ten kilo-ohms. The ECG signal is processed by an instrument amplifier to obtain waveform data with a bandwidth of 0.5 to 100 Hz. The nine-axis inertial measurement unit integrated on the side wall of the badge collects three-axis acceleration, three-axis angular velocity and three-axis magnetic raw data at a frequency of 100 Hz, and solves the trunk pitch angle and roll angle through the quaternion fusion algorithm. The operation stability index is calculated in combination with the hand motion covariance matrix. Environmental data collection is achieved through a distributed micro-electromechanical system microphone array, whose frequency response range covers 100 Hz to 10 kHz, and the sound pressure level measurement accuracy reaches plus or minus two decibels. At the same time, the hospital information system interface uses the medical information exchange standard protocol to parse the patient admission update and medical order execution message stream in real time, and extract key fields including nursing task type code, patient early warning score value and ward electronic fence coordinates. The above physiological data, behavioral activity data and environmental context data are wirelessly transmitted to the edge gateway node via a low-power wide area network, and are synchronized and aligned at the millisecond level by the hardware timestamp engine based on the precision clock protocol, and finally encapsulated into a structured nursing signal data packet, which contains a unified time scale reference, 256 electrocardiogram waveform fragments, 15-dimensional behavioral feature vectors and context tuple triple data entities.

[0033] Furthermore, the hardware architecture of the data acquisition module is physically based on the wearable sensor unit and the environmental perception unit. The wearable sensor unit adopts a three-layer printed circuit board stacking design of 40 mm by 60 mm. The physiological sensor layer integrates the ECG signal conditioning front-end chip, providing an input impedance greater than 100 gigaohms and a common mode rejection ratio of 110 decibels. It cooperates with the constant voltage 0.5 volt excitation circuit to achieve accurate measurement of skin conductance in the range of 1 to 100 microSiemens. The behavioral perception layer is equipped with a nine-axis motion processing chip to output raw sensor data through the integrated circuit bus, and uses the gradient descent method to fuse the accelerometer and gyroscope data to generate the posture quaternion, and calculates the hand operation tremor index within a 30-second window in real time, which is defined as the arithmetic mean of the standard deviation of the three-axis angular velocity. At the same time, the moving path efficiency parameter is derived as the ratio of the actual displacement distance to the path integral length. The environmental perception unit consists of four microphone nodes deployed on the ceiling of the ward, which are calculated through the beamforming algorithm. The method separates human voices from ambient noise within a three-meter radius. Acoustic feature extraction includes the root mean square value of the sound pressure level and the zero-crossing rate statistic in a 500-millisecond time window. The hospital information system interface module parses the patient management event field in the medical information exchange standard message, extracts the ward classification code from the patient admission message, obtains the early warning score value from the patient status update message, and parses the current nursing task status code from the doctor's order execution message. Data synchronization uses the precision time protocol to control the master-slave clock deviation to within one microsecond. The final generated nursing signal data packet is encapsulated in the protocol buffer format and contains a 64-bit timestamp field, a repeated floating-point ECG waveform array, a nested behavior data structure, and a context tuple container. The behavior data structure stores the three-dimensional acceleration peak, pitch angle variation range, and hand stability index. The context tuple container contains the task type enumeration value, the regional risk level flag associated with the nursing signal, and the patient criticality score floating-point number.

[0034] Specifically, the intelligent analysis module executes the dynamic situational information generation process at the edge computing node. It first performs semantic analysis on the medical text stream input by the hospital information system, and uses the biomedical pre-trained language model to perform word segmentation and entity recognition operations on the nursing records. For example, from the sentence "At 15:02, perform a subcutaneous injection of 10 units of insulin on three beds", it extracts entity elements such as the action subject insulin, the administration route subcutaneous injection, the target three beds, and the dosage parameter 10 units. At the same time, the ultra-wideband positioning system provides the three-dimensional spatial coordinates of the nursing staff at a refresh rate of 30 Hz, and determines the current area attributes by matching the spatial topological relationship with the ward electronic fence database. When the coordinates fall into the polygonal geographic fence of the chemotherapy dispensing room, the high-risk operation flag is activated. The patient's critical condition assessment is calculated by a dynamic weight formula, which integrates 30% early warning score weight, 50% ventilator dependence mark weight and 20% pain score weight to output critical condition. The re-indexing and spatiotemporal alignment engine is introduced into the process of generating situational information, which associates and maps the parsed nursing task entities with the positioning trajectory along the time axis. For example, if it is determined that the nurse is at the coordinate point of bed three at a certain moment and there is an insulin injection task record in the same period, it is confirmed as a medication operation situation. The dynamic situational information finally output is a structured vector updated every 200 milliseconds. The vector contains data entities of six dimensions: the task type field stores enumeration codes such as medication operations or patient transfers, the regional risk level field associated with the nursing signal records classification indexes such as general wards or isolation areas, the patient criticality index field stores floating-point values ​​ranging from zero to ten, the task duration field counts the current number of continuous operations in seconds, the operation constraint field marks the status of equipment such as sterile gloves or protective masks, and the team collaboration field stores the number of other nurses within a five-meter range. The vector is directly passed to the analysis module for real-time processing through the zero-copy shared memory mechanism.

[0035] In one embodiment of the present invention, the stress event association model in the analysis module achieves accurate discrimination by constructing a spatiotemporal association map. The map establishes a multidimensional analysis space with the nursing workflow stage as the spatiotemporal coordinate axis, establishes a dynamic coupling model of hand motion trajectory and physiological indicators in the medication operation stage, captures the original data of the wrist's three-axis angular velocity through a nine-axis inertial measurement unit, extracts the tremor power spectrum density characteristic values ​​in the three to eight Hz frequency band through four-layer wavelet decomposition, synchronously analyzes the RR interval standard deviation of the electrocardiogram signal as a representation of heart rate variability, and uses the sliding window Pearson correlation coefficient to calculate the dynamic correlation between tremor energy and the low-frequency band energy of heart rate variability. When the correlation coefficient exceeds 0.7 within five consecutive sampling windows (window width of eight seconds) and the standard deviation of heart rate variability is lower than 30% of the baseline value, an operation anxiety event marker is triggered; in the patient transfer stage, environmental constraints and behavioral pattern analysis are integrated, and the displacement vector direction angle is calculated based on the position update data of the ultra-wideband positioning system at 30 times per second. , combined with the building information model database to obtain the current channel width parameters, define the gait fitness function as a nonlinear mapping relationship between the coefficient of variation and the channel width, and generate a path compression warning signal when the function output value exceeds the dynamic threshold and lasts for more than ten seconds; initiate a multimodal anomaly detection protocol in the critical care scenario, perform first-order differential processing on the skin conductance signal to capture the sudden change point, when the differential value exceeds three times the standard deviation and the respiratory rate is detected to be greater than 25 times per minute, the first-level stress event marker is activated in combination with the condition that the patient's early warning score is greater than seven points; construct a behavioral pattern portrait in the cross-ward task processing stage, and calculate the ratio of the return distance to the straight-line distance within five minutes through the path integral algorithm. When the ratio is greater than 2.5 and the ECG signal shows a sustained heart rate of more than one hundred and ten beats per minute, the second-level stress event marker is triggered. The final generated stress state signal includes four-level stress level coding, six types of event cause classification labels, and a confidence score of zero point zero to one point zero.

[0036] like Figure 2As shown, the dynamic correlation fusion process adopts a context-adaptive weighting mechanism to achieve precise analysis, automatically adjusts the multimodal data weight coefficient for different workflow stages, increases the weight of ambient light sensor data during night shifts to 1.4 times the base value, and increases the compensation coefficient of 0.3 for the correction parameter of behavioral data caused by the restraint of protective equipment during operation in the infectious disease isolation area; when the system recognizes the chemotherapy drug preparation task (task code CHEMO), it starts the high-sensitivity analysis mode to expand the hand micro-tremor monitoring frequency band to 2 to 12 Hz, and shortens the sampling time window to 0.5 seconds to capture transient anomalies; the patient's family communication situation timeout triggers the voice emotion enhancement protocol, and the beamforming technology of the distributed microphone array is used to extract the dominant voice stream within a radius of three meters, and the Mel-frequency cepstral coefficients are used to Pitch period joint analysis technology identifies features of accelerated speech speed and increased pitch, generates an emotional tension index from zero to ten, and incorporates environmental context data as a new dimension; a cross-module data verification mechanism ensures analysis reliability. When physiological data shows a sudden increase in skin conductance but behavioral data does not detect abnormal movement, environmental sensors are automatically called to detect whether the ambient temperature exceeds 28 degrees Celsius or whether sudden noise exceeds 70 decibels; if the associated model outputs a medication error risk warning, the hand operation trajectory data of the previous two minutes is forced to be traced back, and the kinematic inverse solution is used to verify whether the bottle grasping angle deviates from the standard value by more than 15 degrees. The final dynamic weight matrix is ​​updated every 200 milliseconds and includes physiological data weights, behavioral data weights, environmental data weights, and the four-tuple parameters of the new dimension.

[0037] Furthermore, the real-time feedback module is equipped with multi-level output channels to achieve precise intervention. When the pressure state signal contains a first-level event marker and the confidence level exceeds 0.8, the tactile belt worn by the nurse is driven to generate a 5 Hz intermittent vibration mode (vibration for three seconds with an interval of two seconds), and a deep breathing guidance animation is simultaneously projected within the 30-degree field of view of the augmented reality glasses. The animation includes a three-dimensional model of lung expansion and an eight-second breathing rhythm light band; when the signal contains a second-level event marker and the confidence level is greater than 0.6, the color temperature of the ambient light at the nurse station is switched to a warm white color of 4,700 Kelvin through the digital addressable lighting interface. The illumination gradient adjustment takes no more than three seconds. At the same time, an event summary message encrypted by the Advanced Encryption Standard algorithm is sent to the nursing team leader's handheld terminal. The message contains the pressure event type code, The grid coordinates of the location of occurrence and the duration of the event; all feedback signals are attached with a visual code of the cause of the event; drug verification error events trigger a yellow pulse border warning in the augmented reality glasses, which lasts for six seconds and is accompanied by a flashing medicine bottle icon; sudden changes in the patient's condition trigger a red full-screen fading warning, with a synchronous superimposed ECG symbol fluctuation animation; for continuous standing operation exceeding the limit (more than forty-five minutes), the system automatically retrieves the nearest cushioned seat position and generates a three-dimensional navigation path, indicating the turning direction through the intensity difference of the vibration motors on the left and right sides of the tactile belt; the multi-channel feedback protocol sets a priority arbitration mechanism. When level one and level two events occur at the same time, tactile feedback and breathing guidance are executed first, the ambient light adjustment is delayed by five seconds, and the manager's alarm message is sent immediately under bandwidth conditions.

[0038] like Figure 3 The figure shows a real-time monitoring method for nursing stress based on multimodal data according to the present invention. S1: synchronously collect the physiological data, behavioral activity data and environmental context data of the nursing staff and form a nursing signal; S2: receive the nursing signal from the data acquisition module, and extract dynamic context information related to the current working status of the nursing staff; S3: receive the nursing signal from the data acquisition module and the dynamic context information from the intelligent analysis module, and dynamically associate and fuse the physiological data, behavioral activity data and the dynamic context information through a context perception and stress event association model based on the nursing workflow to generate a stress state signal containing a stress event assessment; S4: receive the stress state signal containing a stress event assessment from the analysis module, and generate and output a real-time feedback signal based on the stress state signal.

[0039] Furthermore, the analysis module has a built-in personalized calibration engine to achieve dynamic optimization. The baseline parameter collection protocol is started during the low-stress period of the morning shift handover of the nursing staff (ten minutes after the system automatically recognizes the sign-in), and the ECG signal in the resting state is continuously monitored for three minutes. The average heart rate is calculated as the individualized baseline value and the standard deviation range is stored. The torso angular velocity data in the standing posture is collected synchronously through the inertial measurement unit to establish a gait stability baseline matrix. When abnormal heart rate fluctuations occur in the daytime workflow (exceeding the baseline value by 20% for 30 consecutive seconds), the morning baseline data of the person is called to calculate the drift compensation. The compensation formula is corrected heart rate = real-time heart rate × (morning average heart rate / group average heart rate); the shift mode adaptation mechanism dynamically adjusts the evaluation threshold, and the night shift period (10 pm to 6 am) The heart rate variability threshold for patients receiving chemotherapy (at 1:00 p.m.) was raised by 0.2 times, and the criteria for determining gait disorders were relaxed by 15 percent. The continuous learning module updated model parameters through stress response pattern analysis. When the system detected that a nurse had persistent hand tremors but no abnormal physiological indicators when handling chemotherapy drugs, the person was automatically labeled as behaviorally sensitive and the split gain of the tremor feature in the decision tree was reduced. A personalized stress sensitivity factor report was generated every week, including three-dimensional data: physiological indicator offset statistics, behavioral characteristic abnormality frequency, and environmental factor response intensity. This report drove the association model to perform incremental parameter optimization and used the stochastic gradient descent method to update the classifier weight matrix. The optimization process was completed at the edge computing node and took no more than 15 minutes, ensuring that the system could adapt to the stress representation patterns of different nurses.

[0040] Specifically, dynamic situational information expands the dimension of team collaboration indicators, realizes collaborative status perception through low-power Bluetooth beacon network, and has a built-in Bluetooth 5.0 chipset in the nursing staff badge, broadcasting the device identifier containing the role code at one-second intervals. The number of collaborative personnel is counted when the signal strength indication value of other badges received within a five-meter radius is greater than negative seventy decibels milliwatts; the voice activity detection subsystem is implemented through a distributed microphone array, and a double-threshold endpoint detection algorithm is used to identify valid voice segments. The team response delay time is defined as the time difference between the first frame of voice triggered by the call bell and the first frame of voice answered by the nearest nurse; the collaborative pressure analysis model is started in the critical patient transfer scenario. When the transfer task is activated (scenario code PT_TRANSFER), the team personnel density parameters (number of nurses per unit area) and the key node response time are monitored in real time: if Bluetooth ranging is detected to be less than two collaborative personnel within three meters and the call is If the response to the ventilator disconnection alarm times out for twenty seconds, the situational stress weight coefficient will be automatically increased by fifty percent. A team collaboration gap marker (labeled TEAM_GAP) will be added to the stress state signal, and a reinforcement request message will be generated and sent to the nursing scheduling system. The message includes the required professional qualifications (such as respiratory therapists), the expected arrival time window (within sixty seconds), and the optimal path recommendation (based on the indoor navigation database). The long-term collaboration effectiveness evaluation module constructs a team stress profile and counts the average number of collaboration gap events per nurse per week. When the value of a certain person exceeds two standard deviations of the team mean, a team support sensitivity factor is added to the personalized calibration engine, and its decision tree splitting threshold is lowered by fifteen percent. All collaboration data is encapsulated as a situational information substructure, which includes three sets of dynamic parameters: real-time collaboration personnel count, average response delay seconds, and collaboration demand satisfaction rate, which are updated synchronously with the main situation vector every two hundred milliseconds.

[0041] The present invention's real-time nursing pressure monitoring system and method based on multimodal data synchronously collects physiological data, behavioral data, and environmental data in nursing operations through multi-source sensors; the intelligent analysis module parses the hospital information system and dynamically generates three-dimensional situational information such as task type, regional risk, and patient criticality index; the analysis module establishes nursing stage-stress characterization association rules: monitoring the correlation between hand tremor and heart rate variability during the drug preparation stage, and analyzing the matching degree between gait and environmental channels during the patient transfer stage; when the situation is dynamically associated with abnormal physiological behavior, a graded stress event marker is triggered, and precise intervention is implemented through tactile feedback equipment and the administrator terminal.

[0042] Therefore, the real-time nursing pressure monitoring system and method based on multimodal data of the present invention solves the problem that traditional wearable devices only collect physiological behavior data and cannot be associated with specific nursing situations such as medication operations and patient first aid.

[0043] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. A real-time monitoring system for nursing pressure based on multimodal data, characterized by: include: A data acquisition module, which synchronously collects the physiological data, behavioral activity data and environmental situation data of the nursing staff and forms a nursing signal; an intelligent analysis module, which receives the nursing signal from the data acquisition module and extracts dynamic situation information related to the current working status of the nursing staff; The context-aware operations performed by the intelligent analysis module include: parsing the hospital information system data stream from nursing signals to identify the type of task currently being performed by the nursing staff, which includes medication preparation, critical care monitoring, and family communication; combining the positioning beacon coordinates with the ward's electronic fence information to determine the pressure classification area the nurse is in, which includes the surgical preparation area, isolation ward, and nurse station; obtaining the severity index and special nursing needs mark of the current case in real time by linking with the patient's electronic medical record; and finally integrating the task type, the regional risk level associated with the nursing signal, and the case severity index into a dynamic context information vector. An analysis module is provided, wherein the analysis module receives the nursing signal from the data acquisition module and the dynamic situation information from the intelligent analysis module, and dynamically associates and fuses the physiological data, behavioral activity data and the dynamic situation information through a situation perception and stress event association model based on the nursing workflow to generate a stress state signal containing a stress event assessment. The stress event association model establishes a spatiotemporal association map with the nursing workflow stage as the axis, focuses on monitoring the correlation between hand tremor behavior data and heart rate variability physiological data during the medication operation period, and analyzes the matching degree between gait urgency behavior data and environmental channel width during the patient transfer stage; the stress event association model in the analysis module operates according to the following process: establishes a spatiotemporal association map with the nursing workflow stage as the axis, focuses on monitoring the correlation between hand tremor behavior data and heart rate variability physiological data during the medication operation period, and analyzes the matching degree between gait urgency behavior data and environmental channel width during the patient transfer stage. During the patient transfer phase, the matching degree between the gait urgency behavioral data and the width of the environmental channel is analyzed; when a sudden increase in skin conductance and abnormal respiratory rate occur in the critical patient monitoring scenario, a first-level stress event marker is triggered; when a continuous high heart rate and irregular activity path are detected in the cross-ward continuous processing call ring scenario, a second-level stress event marker is triggered; and finally a stress state signal carrying a stress level label and event cause classification is generated; the dynamic association fusion introduces a contextual weighting mechanism: the weight coefficient of the insufficient ambient light factor is automatically increased in the night shift workflow phase, and the correction parameter of the behavioral data caused by the restraint of protective equipment is added in the infectious disease isolation area workflow phase; when the system recognizes that the nursing staff is continuously handling high-risk drug preparation tasks, a high-sensitivity analysis mode is activated for the subtle hand tremor behavioral data collected during the same period; when the patient family communication scenario continues to timeout, the voice emotion feature extraction algorithm is activated to enhance the environmental context data analysis dimension; A real-time feedback module receives the pressure state signal including the pressure event assessment from the analysis module, and generates and outputs a real-time feedback signal based on the pressure state signal.

2. The real-time monitoring system for nursing pressure based on multimodal data according to claim 1 is characterized in that: The data acquisition module includes a wearable sensor unit and an environmental perception unit integrated into the nursing staff badge; the wearable sensor unit continuously collects electrocardiogram signals and skin electrical response signals through flexible bioelectrodes to form physiological data, and captures limb movement trajectories and posture changes through an inertial measurement unit to form behavioral activity data; the environmental perception unit collects sound pressure characteristics of the working area through distributed noise sensors to form environmental context data, and receives nursing task lists and patient critical value alarm records pushed by the hospital information system in real time through a near-field communication interface; the wearable sensor unit and the environmental perception unit fuse the three types of data into nursing signals through a timestamp synchronization mechanism.

3. The real-time monitoring system for nursing pressure based on multimodal data according to claim 1, characterized in that: The real-time feedback module includes a multi-level output channel: when the pressure state signal contains a first-level pressure event marker, it drives the tactile belt worn by the caregiver to generate a low-frequency vibration prompt and projects a deep breathing guidance animation onto their smart glasses; When the signal contains a secondary stress event marker, a predetermined soothing audio is played through the nurse station broadcast system, and an encrypted warning text is sent to the nursing team leader's terminal at the same time; all feedback signals are accompanied by a stress event cause code, among which events related to drug verification errors trigger yellow visual alerts, and events related to sudden changes in the patient's condition trigger red visual alerts.

4. The real-time monitoring system for nursing pressure based on multimodal data according to claim 1 is characterized in that: The analysis module has a built-in personalized calibration engine: it automatically collects baseline physiological parameters during the low-stress period of morning shift handover for nursing staff to establish a personalized resting heart rate and gait stability baseline; When an ECG signal abnormality occurs during the lunchtime workflow, the nurse's baseline data is used to calculate drift compensation. By continuously learning the stress response pattern during the shift cycle, the assessment threshold coefficients for the night and day shifts are dynamically adjusted. A personalized stress sensitivity factor report is generated weekly to optimize the correlation model parameters.

5. The real-time monitoring system for nursing pressure based on multimodal data according to claim 1 is characterized in that: The stress event assessment includes a cause tracing function: when the system detects that an abnormal ECG signal and an abnormal positioning data occur simultaneously, it automatically traces back the workflow scenario data of the previous ten minutes. If it is found that the person handles three ward emergency calls in succession and passes through a noisy corridor environment, a stress event report with dual attribution of environmental interference and task overload is generated; when the behavioral data shows repeated returns to the pharmacy, the failed medication collection record of the drug management system is associated to form a process-blocking stress event analysis chain.

6. The real-time monitoring system for nursing pressure based on multimodal data according to claim 1, characterized in that: The dynamic situational information includes team collaboration indicators: obtaining the number of other caregivers within a five-meter range through Bluetooth beacon ranging, and calculating the team response delay time in combination with voice activity detection; automatically increasing the critical patient transfer situation pressure weight coefficient when the team personnel density in the critical patient transfer situation is lower than the threshold and the call response times out; adding a team collaboration loss mark to the stress state signal and suggesting a request for reinforcement.

7. A method for real-time monitoring of nursing pressure using the real-time monitoring system for nursing pressure based on multimodal data according to any one of claims 1 to 6, characterized in that: include: S1: Synchronously collect the physiological data, behavioral activity data and environmental situation data of the nursing staff and form nursing signals; S2: receiving the nursing signal from the data acquisition module, and extracting dynamic situation information related to the current working status of the nursing staff; S3: receiving the nursing signal from the data acquisition module and the dynamic situation information from the intelligent analysis module, dynamically correlating and fusing the physiological data, behavioral activity data, and the dynamic situation information through a situation perception and stress event association model based on the nursing workflow, and generating a stress state signal including a stress event assessment; S4: Receive the pressure state signal including the pressure event assessment from the multimodal fusion analysis module, and generate and output a real-time feedback signal based on the pressure state signal.

Images