Multi-dimensional physiological signal monitoring system for psychological stress of nursing personnel

Abstract

The invention discloses a multi-dimensional physiological signal monitoring system for mental stress of nursing personnel. The system comprises a signal acquisition module, a context acquisition module, a processing analysis module and an output module. The signal acquisition module is used for acquiring multi-dimensional physiological data such as an electrocardiosignal, a motion signal and an environment signal of a nurse in real time; the context acquisition module acquires a clinical operation log from a hospital information system; the processing analysis module is used for distinguishing psychological pressure from physiological interference through a multi-modal signal fusion filtering technology, intelligently associating an effective pressure event with clinical operation and determining the context of a pressure source; and the output module is used for generating a personalized pressure analysis report and a management instrument panel containing the clinical work context. The problems that in the prior art, pressure monitoring specificity is poor, and a pressure source is unknown are solved, accurate monitoring and intelligent source tracing of the psychological pressure of nursing personnel in a real clinical environment are achieved, and a scientific basis is provided for individual pressure management and organization health intervention.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of medical health information technology, and specifically relates to a multi-dimensional physiological signal monitoring system for psychological stress of nursing staff. BACKGROUND

[0002] In the field of medical health management, nursing staff, as the core force of the medical service system, has long been under high-intensity and high-load work pressure. Continuous housing pressure not only leads to psychological health problems such as job burnout, anxiety and depression of nursing staff, but also directly affects the quality of nursing and patient safety. Therefore, establishing an effective nursing staff psychological stress monitoring and intervention mechanism has become an important issue to improve the quality of medical services.

[0003] At present, the monitoring of the psychological stress of nursing staff mainly relies on two technical approaches of subjective evaluation scale and physiological signal monitoring. The subjective evaluation scale is evaluated by regular questionnaire survey, and this method has obvious limitations: the evaluation results are easily affected by the subjective will of the surveyed person, the data real-time performance is poor, and frequent questionnaire filling increases the work burden of nursing staff.

[0004] In the aspect of objective monitoring, the existing technology mainly adopts a single physiological signal-based stress evaluation method. Common ones are: 1) heart rate variability analysis based on electrocardiogram signal, which monitors the balance state of the autonomic nervous system to evaluate the stress level; 2) skin conductance change monitoring based on skin electric response, which reflects the emotional arousal degree; 3) muscle tension detection based on electromyography signal, which indirectly infers the psychological stress state.

[0005] However, these single physiological signal monitoring methods have serious defects in real clinical environment: Firstly, the physiological signal change is non-specific, and the reduction of heart rate variability may be caused by psychological stress or physical activity or environmental factors, resulting in low signal-to-noise ratio of the monitoring result; Secondly, the existing technology cannot associate the stress physiological signal with the specific clinical work scene, and it is difficult to provide targeted intervention basis; Thirdly, the traditional method only focuses on the instantaneous intensity of stress, and lacks dynamic assessment of the continuous impact of stress.

[0006] Therefore, there is an urgent need for a stress monitoring solution that can adapt to real clinical environment, accurately distinguish the nature of stress, intelligently identify the source of stress, and provide continuous care. SUMMARY

[0007] The purpose of this invention is to provide a multi-dimensional physiological signal monitoring system for the psychological stress of nursing staff. This invention solves the problems of poor specificity and unclear stress sources in existing technologies, and realizes accurate monitoring and intelligent source tracing of the psychological stress of nursing staff in a real clinical environment, providing a scientific basis for individual stress management and organizational health intervention.

[0008] The technical solution adopted in this invention is as follows: A multi-dimensional physiological signal monitoring system for nursing staff psychological stress includes: The signal acquisition module is used to acquire multi-dimensional physiological signals of nursing staff in real time, including at least electrocardiogram signals and motion signals; The context acquisition module is used to obtain the clinical operation log data of the nursing staff from the hospital information system; The processing and analysis module is configured to perform the following operations: identify preliminary stress events based on the electrocardiogram signal; filter the preliminary stress events based on the motion signal to distinguish between psychological stress events and physiological interference events; and perform time correlation analysis between the filtered psychological stress events and the clinical operation log data to determine the clinical work context corresponding to the psychological stress event. The output module is used to output stress analysis results that include the clinical work context.

[0009] Preferably, the signal acquisition module is further configured to acquire environmental signals, and the processing and analysis module is further configured to perform auxiliary filtering processing on the preliminary pressure event based on the environmental signals; wherein, the environmental signals include ambient light intensity signals and / or ambient temperature signals.

[0010] Preferably, the filtering operation based on motion signals and environmental signals includes: Calculate the statistical characteristic values ​​of the motion signal within a preset time window; Calculate the gradient of environmental signal changes within the same time window; When the statistical feature value exceeds the first preset threshold and / or the change gradient exceeds the second preset threshold, the corresponding preliminary stress event is marked as a physiological disturbance event and filtered.

[0011] Preferably, the processing and analysis module is further configured as follows: After identifying a psychological stress event, the recovery trajectory of physiological signals is continuously monitored over a preset time period; the degree of persistent psychological impact of the psychological stress event on nursing staff is assessed based on the recovery trajectory; wherein, the stress analysis results also include assessment information on the degree of persistent psychological impact.

[0012] Preferably, the operation for assessing the degree of persistent psychological impact includes: Establish an individual dynamic recovery baseline based on the individual nursing staff's historical stress event recovery data; The recovery trajectory from the current psychological stress event is compared with the individual's dynamic recovery baseline to generate a stress resilience index; The degree of persistent psychological impact is determined based on the stress resilience index.

[0013] Preferably, when the pressure recovery index is lower than a preset threshold, the output module is also used to generate and output active intervention prompt information.

[0014] Preferably, the time correlation analysis operation includes: Determine the timing of the psychologically stressful event; Retrospectively review all clinical operation records within a preset time window prior to the time of occurrence; The most relevant clinical work contexts are identified based on the temporal proximity and operational importance of the operation log and the stress event.

[0015] Preferably, when there are records with operation status of failure or abnormality within the preset time window, such records are preferentially identified as the corresponding clinical work context.

[0016] Preferably, the pressure analysis results output by the output module include: Personal stress insight reports, which visualize the relationship between psychological stressful events and clinical work context; and / or anonymous aggregated management dashboards that display stress distribution characteristics based on population data statistics.

[0017] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the functions of the processing and analysis module and the output module of the system described above.

[0018] The beneficial effects of this invention are as follows: The nursing staff psychological stress monitoring system of this invention achieves a qualitative leap from traditional monitoring to intelligent insight through technological innovations in multimodal signal fusion, intelligent association with clinical context, and dynamic stress assessment. Its beneficial effects are specifically reflected in the following three progressive levels: Traditional stress monitoring technologies rely on single physiological indicators such as electrocardiograms, resulting in serious non-specificity issues. Changes in heart rate variability may originate from emotional fluctuations, physical activity, or sudden environmental changes, leading to extremely low signal-to-noise ratios in real clinical scenarios. This invention innovatively solves this industry challenge through multimodal fusion filtering of motion and environmental signals. For the first time, this invention possesses the intelligent judgment capability to distinguish between psychological stress and physiological disturbances. By calculating the statistical characteristic values ​​of motion signals and the gradient of environmental signal changes, and setting dual filtering thresholds, the system can accurately identify physical activities such as running and rapid walking, as well as physiological disturbances caused by environmental factors such as sudden changes in light and temperature, thereby achieving precise extraction of psychological stress events. This innovation enables the system to maintain high reliability and accuracy in dynamic and complex clinical working environments, overcoming the limitation of traditional technologies being effective only in controlled environments.

[0019] Existing technologies can only provide quantitative values ​​of stress levels, failing to reveal the source and context of stress, resulting in poor data operability. This invention achieves deep semantic interpretation of stress events through intelligent association of clinical operation logs, establishing an intelligent mapping relationship between physiological signals and clinical work. Through time window backtracking and priority matching algorithms, the system can automatically associate stress events with specific clinical operations (such as failed intravenous punctures, delays in processing medical orders, etc.), providing a clear work context for stress generation. This innovation transforms stress data from abstract numerical values ​​into behavioral records with clear clinical significance, not only helping nursing staff accurately identify individual stress patterns but also providing empirical evidence for medical institutions to optimize processes and allocate resources.

[0020] Traditional techniques focus only on the instantaneous intensity of stress, failing to assess its lasting impact and an individual's resilience. This invention, by introducing a stress resilience index and personalized baseline assessment, enables proactive management of mental health, quantifying the subsequent effects of stressful events and an individual's psychological resilience. By establishing a dynamic personal recovery baseline and continuously monitoring the recovery trajectory of physiological signals, the system can accurately identify individuals who recover slowly and are susceptible to the lasting effects of stress, thus enabling early warning and targeted intervention. This innovation transforms the system from a passive monitoring tool into a proactive health management partner, forming a complete "monitoring-assessment-intervention" care loop by promptly delivering personalized stress reduction suggestions.

[0021] In summary, this invention, through multi-layered technological innovation, produces a series of synergistic and amplified system-level effects. It not only surpasses the performance limitations of traditional monitoring technologies but also achieves breakthroughs in clinical applicability and humanistic care. The system outputs no longer isolated stress values ​​but a complete insight encompassing the source of stress, its degree of impact, and improvement suggestions, providing strong technical support for improving the well-being of nursing staff and the quality of medical care. This multi-dimensional improvement, from technical performance to system value, fully demonstrates the significant advancements and unexpected technical effects of this invention compared to existing technologies. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the system of the present invention; Figure 2 This is a schematic diagram of the system architecture of the present invention; Figure 3 This is a schematic diagram of the monitoring method of the present invention.

[0023] In the diagram, 1 is the signal acquisition module; 2 is the context acquisition module; 3 is the processing and analysis module; 31 is the filtering unit; 32 is the recovery assessment unit; 33 is the context association unit; 4 is the output module; 41 is the intervention prompt unit; 42 is the personal report generation unit; and 43 is the management dashboard generation unit. Detailed Implementation

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments, but this is not intended to limit the scope of protection of the present invention.

[0025] like Figure 1 and 2 As shown, this invention proposes a multi-dimensional physiological signal monitoring system for the psychological stress of nursing staff, including: ① Signal acquisition module 1 is used to acquire multi-dimensional physiological signals of nursing staff in real time. The multi-dimensional physiological signals include at least electrocardiogram signals and motion signals. Signal acquisition module 1 is also used to acquire environmental signals. The processing and analysis module 3 is further configured to perform auxiliary filtering processing on preliminary stress events based on environmental signals. The environmental signals include ambient light intensity signals and / or ambient temperature signals.

[0026] Specifically, to address the shortcomings of existing single-physiological signal monitoring methods in real clinical environments, such as susceptibility to interference, low signal-to-noise ratio, and inability to distinguish between psychological stress and physiological stress responses, this invention features a multimodal enhancement design for the signal acquisition module. This module not only acquires electrocardiogram (ECG) and motion signals but also integrates an ambient light sensor and an ambient temperature sensor to obtain ambient light intensity and temperature signals.

[0027] The rationale behind this design lies in the non-specific nature of nurses' autonomic nervous system responses. Changes in heart rate variability in electrocardiogram (ECG) signals can stem from psychological stress such as emotional fluctuations, or be triggered by sudden changes in environmental factors. For example, rapidly moving from a dimly lit nurses' station to a brightly lit emergency ward can cause pupil constriction, leading to a temporary increase in heart rate and a decrease in heart rate variability. Similarly, traversing ward areas with different temperature zones can cause similar physiological signal changes due to thermoregulation. Relying solely on ECG signals would prevent the system from distinguishing between these physiological disturbances caused by sudden environmental changes and the psychological stress events being monitored, leading to misjudgments and significantly reducing the accuracy and clinical reference value of the output results.

[0028] Based on this, one of the core improvements of this invention lies in the fact that the processing and analysis module is further configured to perform auxiliary filtering processing on the initially identified stress events using the environmental signals. Specifically, the filtering unit in the processing and analysis module is programmed to synchronously analyze the statistical characteristic values ​​of the motion signal, the gradient of ambient light intensity change, and the gradient of ambient temperature change within a preset time window. When the statistical characteristic value exceeds a preset motion threshold, and / or the gradient of the environmental signal change exceeds its respective preset environmental threshold, the filtering unit determines that the initial stress event is a physiological disturbance event and filters it out.

[0029] The consequences and benefits of this design are significant and multi-layered. First, by introducing environmental signals and performing multimodal fusion filtering, the system achieves more refined differentiation of physiological signal sources, greatly improving the signal-to-noise ratio and recognition specificity of the target signal (psychological stress events). Second, this method achieves seamless and objective context awareness; without active input from nursing staff, the system can automatically identify and eliminate physiological interference caused by environmental factors, enhancing the system's practicality and user experience. Finally, this design enables the entire monitoring system to adapt to the complex and ever-changing environment of real clinical work, fundamentally improving its robustness and reliability in uncontrolled, highly dynamic work scenarios, laying a solid foundation for outputting stress analysis results with high clinical credibility.

[0030] ② Context acquisition module 2 is used to obtain clinical operation log data of nursing staff from the hospital information system; Specifically, this module is used to automatically retrieve clinical operation log data from the hospital information system (HIS) and correlate it with physiological signal monitoring results over time, thereby giving stress events a clear clinical work context.

[0031] Simple physiological signals such as electrocardiograms and electromyograms can only reflect the arousal level of the autonomic nervous system, i.e., the "intensity" of stress, but cannot reveal the "source" and "nature" of stress. Stress in nursing work is highly contextual, and its roots are deeply embedded in specific clinical tasks, such as performing difficult procedures, processing urgent medical orders, or responding to inquiries from patients' families. If the physiological indicators of stressful events cannot be correlated with the specific work content that triggered the event, the output of the entire monitoring system will lack actionable guidance and cannot support personalized stress management and systematic process optimization.

[0032] Based on this, the present invention constructs a data bridge between the physiological information world and the clinical work world through a context acquisition module. This module, as a software interface service, preferably uses an application programming interface conforming to medical information exchange standards to retrieve the target nurse's clinical operation logs from the HIS in a read-only and secure manner, either periodically or triggered. The acquired operation logs are time-series structured data, with key fields including at least the operation timestamp, operator identification, operation type, operation object (such as the patient, medication, or nursing item involved), and operation execution result status (such as success, failure, warning).

[0033] The output of this module serves as a key input to the processing and analysis module. The context association unit within the processing and analysis module is configured to execute the following core logic: When a valid psychological stress event is identified, it uses the event's occurrence time as a baseline, traces back a preset time window (e.g., moving forward three minutes), and retrieves all clinical operation records within that time period from the log data provided by the context acquisition module. Subsequently, based on preset priority rules (e.g., prioritizing records with operation statuses of "failure" or "warning," followed by records with a "completed" status closest to the stress event's timestamp), it determines the clinical work context most likely related to the stress event, thereby completing the intelligent labeling of the stress source.

[0034] This design significantly enhances the system's intelligence and output value. By linking anonymous physiological signal fluctuations to concrete clinical tasks, the system achieves a qualitative interpretation of stressful events, with its output evolving from "high stress level" to "high stress level due to difficulties encountered while performing a specific operation." This not only enables individual nurses to accurately identify their own stress patterns and weaknesses but also provides healthcare administrators with empirical data support for process reengineering, skills training, and resource optimization (e.g., identifying a particular software module or nursing procedure that commonly triggers high stress). Ultimately, the context acquisition module transforms this invention from a simple physiological parameter recording tool into a stress management and organizational optimization system capable of deeply integrating into clinical business logic and supporting decision-making.

[0035] ③ Processing and Analysis Module 3: To achieve accurate monitoring and in-depth analysis of nursing staff's psychological stress and overcome the shortcomings of existing technologies such as poor signal specificity, unclear stress sources, and lack of impact assessment, the processing and analysis module 3 of this invention is configured to execute a coherent and intelligent multi-level analysis process. As the core computing engine of the system, this module, through the collaborative work of its internal units, transforms raw physiological signals and clinical context data into highly clinically significant stress insights.

[0036] The processing and analysis module 3 first identifies preliminary stress events based on the electrocardiogram (ECG) signal received from the signal acquisition module 1. Given that ECG signals are susceptible to interference from various non-psychological factors, the built-in filtering unit 31 of module 3 is activated to execute an advanced filtering algorithm. This algorithm calculates the statistical characteristic values ​​of the motion signal (such as acceleration amplitude values ​​at the 75th percentile) and the gradient of environmental signals (such as ambient light intensity and temperature) within a preset time window, and compares these characteristic values ​​with preset thresholds. When the statistical characteristic value exceeds a first preset threshold (indicating medium-to-high intensity exercise) and / or the gradient exceeds a second preset threshold (indicating a sudden change in environment), it is determined that the preliminary stress event originates from physiological interference rather than psychological stress, and is filtered out. This filtering step is a key technical prerequisite for ensuring the accuracy of subsequent analysis, effectively improving the system's signal-to-noise ratio and anti-interference capability.

[0037] Valid psychological stress events confirmed by the filtering unit 31 are transmitted to the context association unit 33. This unit is configured to perform precise location of the stress source. Its operation process includes: first, determining the time point of occurrence of the psychological stress event; then, tracing back to a preset time window before that time point, obtaining all clinical operation records from the hospital information system (HIS) via the context acquisition module 2; finally, based on the temporal proximity and operational importance of the operation record and the stress event, determining the most relevant clinical work context through preset priority rules. A key optimization configuration is that when there are records with an operation status of "failure" or "abnormality" within the preset time window, these records are prioritized as the corresponding clinical work context. The reason for this design is that "failure" or "abnormality" events in clinical operations are usually high-intensity, immediate stress sources, and their causal correlation with psychological stress events is significantly stronger than that of successful routine operations. Using this priority rule can more realistically and accurately reflect the essential causes of stress, thereby achieving intelligent tracing and characterization of stress events and solving the core problem of "unknown source" of stress.

[0038] To further assess the long-term effects of stress, the processing and analysis module 3 also includes a recovery assessment unit 32. This unit continuously monitors the physiological signal recovery trajectory within a preset time period (e.g., within 60 minutes of the event) after identifying a psychological stress event. The core innovation of this unit lies in introducing a personalized assessment model, which establishes a dynamic personal recovery baseline based on the individual caregiver's historical stress event recovery data. By comparing the recovery trajectory of the current stress event with this dynamic personal recovery baseline, a quantitative stress resilience index is generated, which is used to determine the degree of persistent psychological impact of the event. This design is based on the recognition that the harm of stress lies not only in its instantaneous intensity but also in the persistence of its effects; ignoring the recovery process makes it impossible to comprehensively assess mental health risks. The introduction of this unit allows the system to evolve from instantaneous monitoring of stress intensity to long-term assessment of stress tolerance and psychological resilience, providing crucial evidence for early intervention.

[0039] In summary, the processing and analysis module 3, through a progressive processing flow of filtering, correlation, and evaluation, produces two key technical consequences and beneficial effects: First, it outputs semantically rich stress analysis results containing clear clinical context and information on the assessment of lasting impact, rather than simple numerical values, greatly improving the interpretability and operability of the results. Second, through multimodal filtering and personalized baseline comparison, it ensures that the analysis results output by the system in real and complex clinical environments have high accuracy and reliability. Ultimately, this module elevates the present invention from a simple physiological parameter recording system into an intelligent analysis system capable of deeply understanding clinical business logic and supporting individualized stress management and organizational health intervention.

[0040] ④ Output module 4, which serves as the final value output interface of the system, is responsible for transforming the semantically rich analysis results generated by the processing and analysis module 3 into highly operable stress analysis results that are tailored to different users and support different decision-making scenarios.

[0041] The output module 4 includes an intervention prompt unit 41, a personal report generation unit 42, and a management dashboard generation unit 43.

[0042] Intervention prompt unit 41 is configured to provide a real-time or near-real-time proactive intervention mechanism. Specifically, the technical solution involves: continuously monitoring the stress resilience index generated by the recovery assessment unit 32; when the stress resilience index falls below a preset personalized threshold, it is determined that the caregiver is currently experiencing persistent psychological impact and requires immediate intervention; subsequently, the unit automatically generates and outputs a proactive intervention prompt message. This prompt message is pushed non-invasively to the caregiver's terminal device (such as a mobile nursing terminal or authorized smart wearable device), and its content includes specific, lightweight stress relief suggestions, such as guiding short breathing exercises or suggesting a short rest. The technical effect of this design is that it achieves a leap from passive monitoring to proactive support, extending the system's service loop from "analysis" to "immediate intervention," significantly improving the system's timeliness and support value.

[0043] The individual report generation unit 42 is configured to generate personal stress insight reports to support nurses' self-awareness and health management. The technical solution involves periodically (e.g., daily or weekly) aggregating all psychological stress event data related to a specific nurse, processed by the analysis module 3. Each event data point includes stress intensity, clinical work context, and stress resilience index. Subsequently, this unit uses visualization techniques (such as timeline charts and heatmaps) to demonstrate the correlation between psychological stress events and their corresponding clinical work context. Access to this report is strictly limited to the individual nurse, ensuring their privacy. The technical effect of this design is that it transforms abstract physiological data into concrete and easily understood personal work reviews, empowering nurses to identify their own stress patterns and thus engage in targeted self-regulation, demonstrating the system's "individual empowerment" value.

[0044] The management dashboard generation unit 43 is configured to generate anonymized aggregated management dashboards to support organizational health management and process optimization decisions by management. Its core technical solution lies in data anonymization and aggregation processing: before data input, all personally identifiable information is first removed; then, based on group data (such as by ward, shift, or job position), statistics are calculated to determine indicators such as the frequency of stress events, average intensity, and average stress resilience index; finally, the dashboard displays the stress distribution characteristics obtained from these group data statistics, such as rankings of high-stress clinical tasks and heatmaps of high-stress work periods. This design offers two technical benefits: firstly, it strictly adheres to data privacy and ethical standards, avoiding evaluation of individuals; secondly, it provides nursing managers with a macro-level, objective data view, accurately identifying systemic bottlenecks and organizational risk points in workflows, providing empirical evidence for human resource allocation, process reengineering, and targeted training, thus realizing the "organizational optimization" value of data.

[0045] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the functions of the processing and analysis module 3 and the output module 4 of the system.

[0046] Further, see Figure 3 A multi-dimensional physiological signal monitoring method for psychological stress among nursing staff, the specific steps of which are as follows: Phase 1: Synchronous Acquisition of Multi-Source Data Physiological signal acquisition: Smart devices worn by nursing staff (name badges or wristbands) continuously and synchronously acquire multi-dimensional physiological signals. Electrocardiogram (ECG) signals are acquired using medical-grade electrodes to capture raw ECG waveforms for subsequent calculation of heart rate variability, a core indicator for assessing stress levels. Motion signals are acquired using a triaxial accelerometer to collect body activity data, used to identify states such as running and brisk walking. Environmental signals are acquired using integrated sensors to collect ambient light intensity and temperature, used to detect sudden changes in the environment. All signals are timestamped with high precision to ensure data synchronization.

[0047] Clinical context acquisition: The context acquisition module retrieves nurses' clinical operation logs periodically or triggered by the standard interface of the hospital information system (HIS). The log content includes at least the operation time, operation type (e.g., medication administration, puncture), operation object (patient / medication), and operation result (success / failure).

[0048] Phase Two: Core Intelligent Processing and Analysis This stage is completed by the processing and analysis module, which is the "brain" of the entire system. Specifically, it includes: Preliminary stress event identification: The system analyzes the incoming electrocardiogram signals in real time. When it detects that indicators such as heart rate variability (HRV) have deteriorated significantly in a short period of time, it automatically marks a preliminary stress event and records the time of its occurrence.

[0049] Multimodal Filtering and Verification: Upon initial event triggering, the filtering unit immediately activates to distinguish between genuine and false events. It simultaneously analyzes motion and environmental signals within a time window before and after the event: if the motion signal characteristic value exceeds a preset threshold, it is determined to be motion interference. If the gradient of environmental signal changes (such as sudden changes in light intensity) exceeds a preset threshold, it is determined to be environmental interference. If either of these conditions is met, the event is marked as a "physiological interference event" and filtered out. If neither threshold is exceeded, it is confirmed as a valid psychological stress event and proceeds to the next step of analysis.

[0050] Intelligent Clinical Context Association: For valid psychological stress events, the context association unit is activated. Using that time point as a baseline, it traces back a certain period and searches the HIS logs for all operation records within that time window. The unit uses priority rules for intelligent matching. Rule 1: Prioritize associating records with an operation status of "failed" or "abnormal".

[0051] Rule 2: If there are no anomalies, then associate the critical operation that is closest to that time point in time.

[0052] At this point, the system has clearly labeled the source of the stress event, for example: "Failed venipuncture led to increased pressure".

[0053] Post-stress assessment: The recovery assessment unit begins monitoring the aftereffects of the event. It continuously monitors the recovery trajectory of physiological signals over a period of time following the stressful event. The unit compares the current recovery rate with the individual caregiver's historical recovery baseline to calculate a stress resilience index. This index quantifies the degree of sustained impact of the stress.

[0054] Phase 3: Multi-dimensional Results Output and Application This stage is completed by the output module, which transforms the analysis results into actionable insights. Specifically, it includes: When the system determines that the resilience index of a stress event is too low, the intervention prompt unit will automatically generate a gentle suggestion (such as guiding deep breathing) and push it to the caregiver's terminal device to provide immediate care.

[0055] The individual report generation unit generates a private, visual report for each caregiver on a regular basis (e.g., daily / weekly). The report clearly shows their stressful events, sources, and recovery progress in a timeline or other format, helping them with self-awareness and management.

[0056] The management dashboard generation unit performs anonymized aggregation analysis on all data to generate dashboards for managers. These dashboards display macro-level insights such as group stress distribution and ranking of high-stress tasks, providing data support for process optimization and organizational health management.

[0057] Furthermore, this invention has been tested in real-world clinical scenarios. These cases systematically demonstrate that this invention not only achieves precise monitoring at the technical level, but also produces unexpected and significant effects on individual care and organizational management.

[0058] Case 1: After hearing an emergency call in the ward, Nurse A rushed to the ward to participate in the rescue. At the same time, Nurse B, whose critically ill patient was unstable, sat quietly at the nurses' station, filled with anxiety.

[0059] The smart devices worn by nurses A and B continuously collected electrocardiogram (ECG) and three-dimensional acceleration (3D) signals. The system analyzed heart rate variability (HRV) in real time and found that both nurses' HRV indices significantly deteriorated within a short period (RMSSD values ​​decreased by more than 30%). The system generated a preliminary stress event for each nurse and recorded the precise timestamp T1. The system activated its filtering unit to analyze data within a 30-second time window before and after T1: For nurse A, the acceleration signal was analyzed, and the 75th percentile of its vector amplitude within the window was calculated. The results showed that it was consistently higher than the preset exercise threshold (2.5g), indicating that she was in a high-intensity running state. The system determined that this physiological response was mainly caused by exercise, so it marked this event as "physiological interference" and filtered it out. For nurse B, the analysis showed that her acceleration signal was stable (amplitude far below the exercise threshold), and there were no sudden changes in ambient light intensity or temperature. The system confirmed this as a valid psychological stress event. The system reviewed nurse B's HIS operation records in the minutes before T1 and found that she had just submitted a condition assessment record for this critically ill patient. The system associated the stress event with the context: "In a sedentary state, possibly related to continuous monitoring of a critically ill patient's condition." Nurse B's personal stress report recorded one valid event, while Nurse A's data was filtered and no stress record was generated, effectively avoiding false alarms. This invention, through multimodal filtering technology, successfully distinguishes between physiological stress and psychological stress, ensuring the specificity and accuracy of the monitoring results.

[0060] Case 2: Nurse C attempted to perform intravenous puncture on a patient with poor vascular conditions. The first attempt failed, but the second attempt was successful.

[0061] During the puncture procedure, the system detected a sudden drop in Nurse C's HRV, generating an initial stress event. Upon inspection, the motion signal was stable and there was no environmental interference; therefore, it was determined to be a valid psychological stress event, timestamped T2. The context acquisition module retrieved all of Nurse C's operation logs from the HIS for the three minutes prior to time T2. The context association unit identified two relevant records: T2-1 minute: Venous puncture procedure, site: dorsal vein of the hand, result: failed. T2-30 seconds: Venous puncture procedure, site: forearm vein, result: successful. Based on the rule of "prioritizing association with operation failures," the system links stress events with "failed intravenous puncture" records. In Nurse C's personal stress insight report, this event is clearly marked as: "Significantly elevated stress level during failed dorsal hand vein puncture." This invention, through intelligent contextual association, precisely links stress signals to specific clinical operation results, revealing "operation failure" as a key stressor and giving stress analysis direct and clear guiding significance.

[0062] Case 3: The head nurse of an internal medicine ward found that the night shift nurses had low morale and generally reported high stress, but the reason was unclear.

[0063] The system anonymously collects data on valid psychological stress events from all night shift nurses in the ward (10 PM to 6 AM), including event time, intensity, and associated clinical context. The management dashboard generation unit performs clustering analysis on the data, finding that "new admission patient assessment document entry" had the highest frequency of stress events, accounting for 38% of all events. These events are concentrated during the evening peak admission period (10 PM - 12 AM). Events related to this task generally have a low average stress resilience index, indicating that the stress from this task is persistent. On the management dashboard, the "new admission assessment document entry" task ranks first in the "high-stress task ranking" and is highlighted in a prominent color. The head nurse visually identifies systemic problems through the dashboard, inferring that the complex design and numerous fields of the currently used electronic admission assessment forms are the main reason for the concentrated stress during night shifts. This invention, through anonymized aggregation analysis of group data, accurately identifies systemic bottlenecks in the workflow from scattered individual stress events, providing empirical evidence for management to optimize processes and improve organizational health.

[0064] Case 4: Nurse D was still subjected to unwarranted accusations from the patient's family after patiently explaining the situation. Her emotions were severely affected. Even though half an hour had passed since the conflict, she still felt uneasy, which affected her subsequent work.

[0065] The system detected a sharp drop in HRV (Heart Rate of Return) when Nurse D was complained about (T3). Combined with stable movement and environmental signals, this was confirmed as a high-intensity, effective psychological stress event, associated with the context of "encountering misunderstanding and accusations from a patient's family." Following the T3 event, the recovery assessment unit continuously monitored her HRV data for 60 minutes and plotted a recovery curve. The system retrieved the nurse's recent stress event data and calculated her personal dynamic baseline for the time it typically takes for her HRV to recover to a calm level after stress, which is approximately 20 minutes. After this complaint event, her HRV took 55 minutes to recover to baseline. The system calculated her stress resilience index as 20 minutes / 55 minutes ≈ 0.36 (an index far below 1, indicating slow recovery). In her personal weekly report, this event was specifically noted next to the note: "This event had a prolonged impact on you, and your recovery speed was lower than usual. We recommend focusing on emotional recovery." This invention, through long-term monitoring and comparison with a personalized baseline, achieves a quantitative assessment of the persistent impact of stress, helping to identify the risk of psychological fatigue and burnout early, and enabling a leap from instantaneous monitoring to long-term health management.

[0066] Case 5: Nurse E had just successfully participated in the resuscitation of a suffocating patient. Despite the successful treatment, she was under great mental stress, her hands were trembling slightly, and she was unable to calm down immediately to do her paperwork.

[0067] The system has recorded this rescue as a high-stress event and entered the post-event monitoring phase. The recovery assessment unit detected that 15 minutes after the event, her HRV recovery was slow, with the recovery trajectory far below her personal baseline; the system calculated a real-time resilience index of only 0.25. The intervention prompt unit has a built-in rule: IF Resilience Index < 0.5 THEN Generate an active intervention prompt. When the condition is met, the system is triggered. A non-invasive, gentle suggestion pops up on Nurse E's handheld PDA: "You are still under stress. Try this: Inhale deeply for 4 seconds, hold your breath for 2 seconds, exhale slowly for 6 seconds, repeat several times. [Start Practice] [Will Remind You Later]." Nurse E clicks "Start Practice," and the interface guides her through simple breathing adjustments. One minute later, she feels her tense nerves relax. This invention integrates monitoring, assessment, and intervention, forming a complete proactive health support closed loop. It provides timely and humane micro-interventions when most needed, seamlessly integrating technological care into the workflow, reflecting the system's high level of intelligence and humanistic care.

[0068] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-dimensional physiological signal monitoring system for psychological stress among nursing staff, characterized in that, include: The signal acquisition module (1) is used to acquire multi-dimensional physiological signals of nursing staff in real time, wherein the multi-dimensional physiological signals include at least electrocardiogram signals and motion signals; The context acquisition module (2) is used to acquire the clinical operation log data of the nursing staff from the hospital information system; The processing and analysis module (3) is configured to perform the following operations: identify preliminary stress events based on the electrocardiogram signal; filter the preliminary stress events based on the motion signal to distinguish between psychological stress events and physiological disturbance events; The filtered and confirmed psychological stress events are correlated with the clinical operation log data over time to determine the corresponding clinical work context of the psychological stress event. Output module (4) is used to output stress analysis results containing the clinical work context.

2. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 1, characterized in that, The signal acquisition module (1) is also used to acquire environmental signals, and the processing and analysis module (3) is further configured to perform auxiliary filtering processing on the preliminary pressure event based on the environmental signals; wherein, the environmental signals include ambient light intensity signals and / or ambient temperature signals.

3. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 2, characterized in that, The processing and analysis module (3) includes a filtering unit (31), which is configured to perform the following operations: Calculate the statistical characteristic values ​​of the motion signal within a preset time window; Calculate the gradient of environmental signal changes within the same time window; When the statistical feature value exceeds the first preset threshold and / or the change gradient exceeds the second preset threshold, the corresponding preliminary stress event is marked as a physiological disturbance event and filtered.

4. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 1, characterized in that, The processing and analysis module (3) further includes a recovery evaluation unit (32), which is configured to perform the following operations: After identifying a psychological stress event, the recovery trajectory of physiological signals is continuously monitored over a preset time period. The extent of the lasting psychological impact of the psychological stress event on nursing staff is assessed based on the recovery trajectory. The stress analysis results also include assessment information on the degree of persistent psychological impact.

5. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 4, characterized in that, The recovery assessment unit (32) is specifically configured as follows: Establish an individual dynamic recovery baseline based on the individual nursing staff's historical stress event recovery data; The recovery trajectory from the current psychological stress event is compared with the individual's dynamic recovery baseline to generate a stress resilience index; The degree of persistent psychological impact is determined based on the stress resilience index.

6. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 5, characterized in that, The output module (4) further includes an intervention prompt unit (41). When the pressure recovery index is lower than a preset threshold, the intervention prompt unit (41) is configured to generate and output active intervention prompt information.

7. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 1, characterized in that, The processing and analysis module (3) further includes a context association unit (33), which is configured to perform the following operations: Determine the timing of the psychologically stressful event; Retrospectively review all clinical operation records within a preset time window prior to the time of occurrence; The most relevant clinical work contexts are identified based on the temporal proximity and operational importance of the operation log and the stress event.

8. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 7, characterized in that, The context association unit (33) is further configured to: when there is a record with an operation status of failure or abnormal within the preset time window, prioritize identifying such record as the corresponding clinical work context.

9. The multi-dimensional physiological signal monitoring system for nursing staff psychological stress according to claim 1, characterized in that, The output module (4) includes: The individual report generation unit (42) is used to generate an individual stress insight report, which visually demonstrates the relationship between psychological stress events and clinical work context; and / or the management dashboard generation unit (43) is used to generate an anonymous aggregated management dashboard that displays stress distribution characteristics based on population data statistics.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the functions of the processing and analysis module (3) and the output module (4) of the system as described in any one of claims 1 to 9.

Images