An intelligent monitoring system for respiratory diseases in children

Abstract

This invention relates to the field of intelligent monitoring technology, and more particularly to an intelligent monitoring system for pediatric respiratory diseases. The system includes an acquisition module, an identification module, an anomaly determination module, a correlation determination module, an adjustment module, an early warning module, and a correction module. This invention acquires multidimensional physiological parameters in real time and accurately identifies stress events and moments of respiratory abnormalities based on their temporal changes. It constructs a causal correlation model from psychological stress to physiological deterioration and calculates a stress-asthma index. Simultaneously, when the stress-asthma correlation index increases, a preset interval threshold is adjusted to improve monitoring sensitivity. An early warning is issued based on the comparison between the adjusted correlation index and the preset threshold, while intervention suggestions are output and the preset threshold is dynamically corrected. This effectively solves the problems of inaccurate risk warning and untimely intervention in non-medical environments due to the lack of a causal correlation quantification mechanism and fixed thresholds.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, and in particular to an intelligent monitoring system for respiratory diseases in children. Background Technology

[0002] With the continued rise in the incidence of childhood asthma, non-medical environments such as homes and childcare facilities have become the main venues for daily health management. Studies have shown that psychological stress, such as emotional fluctuations, can significantly increase the risk of asthma attacks through sympathetic nerve excitation. However, existing monitoring methods mostly rely on static threshold determination of single respiratory parameters, which makes it difficult to quantify the intrinsic relationship between stress and respiratory symptoms, and also fails to adaptively adjust monitoring sensitivity according to individual differences and disease evolution. Therefore, how to capture the causal chain of psychological stress-induced respiratory abnormalities in real time among multidimensional physiological parameters and achieve dynamic optimization of thresholds has become a key challenge in improving the effectiveness of home-based management of childhood asthma.

[0003] Chinese Patent Publication No. CN120052840A discloses a smart asthma management and monitoring device for children. The device includes: a physiological parameter monitoring module, an environmental monitoring module, a medication management module, and a data management module. The physiological parameter monitoring module monitors asthma-related physiological parameters in children. The environmental monitoring module monitors environmental parameters. The medication management module records and reminds users of asthma medication use. The data management module collects, analyzes, and displays children's physiological parameters, environmental parameters, and medication information, and supports asthma condition assessment and early warning of disease trends. The physiological parameter monitoring module includes a respiratory rate monitoring unit, a blood oxygen monitoring unit, and a heart rate monitoring unit. The respiratory rate monitoring unit includes a pressure sensor that monitors the child's respiratory rate by detecting pressure changes during the expansion and contraction of the child's chest cavity. The blood oxygen monitoring unit includes an optical sensor that monitors the child's blood oxygen information. The heart rate monitoring unit includes an optical sensor that monitors the child's heart rate information. The sensors used in the respiratory rate monitoring unit are integrated into a wearable device outside the chest cavity, while the sensors used in the blood oxygen monitoring unit and the heart rate monitoring unit are integrated into a wearable device on the wrist.

[0004] Therefore, the existing technology has the following problems: the device relies on the mean within a fixed time window to calculate the severity of asthma, which easily ignores the short-term dynamic fluctuations of parameters, resulting in insufficient sensitivity in capturing the prodromal period of acute attacks; the device relies on a preset general threshold for anomaly detection, which is prone to misjudgment or omission due to individual differences and growth and development; the device relies on static weights for multi-parameter weighting, which is prone to difficulty in accurately identifying the intrinsic relationship between psychological stress and respiratory symptoms due to the lack of targeted modeling of the causal path of stress-induced asthma. Summary of the Invention

[0005] Therefore, the present invention provides an intelligent monitoring system for children's respiratory diseases, which overcomes the problems in the prior art that the lack of a causal correlation quantification mechanism and fixed thresholds make it difficult to accurately warn of risks and intervene in a timely manner in non-medical environments by dynamically quantifying the causal relationship between stress and asthma and adaptively optimizing the monitoring threshold.

[0006] To achieve the above objectives, the present invention provides an intelligent monitoring system for pediatric respiratory diseases, comprising: The acquisition module is used to acquire the intensity of skin conductance, heart interval, respiratory rate, blood oxygen saturation, and crying frequency during the daily activities of children with asthma. The identification module is used to identify stress events and record stress moments based on the temporal changes in the crying frequency, the cardiac interval, and the intensity of skin conductance, as well as a preset interval threshold. An anomaly determination module is used to determine the degree of respiratory anomaly based on the respiratory rate and the blood oxygen saturation and to record the time of the anomaly. The association determination module is used to determine the stress asthma association index based on the geometric mean characteristics of time matching degree, stress intensity coefficient and individual sensitivity coefficient, wherein the time matching degree is determined based on the stress breathing interval, the stress breathing interval is determined based on the stress time and the abnormal time, the individual sensitivity coefficient is determined based on the stress event and the degree of respiratory abnormality, and the stress intensity coefficient is determined based on the heartbeat interval and the skin conductance activity intensity. An adjustment module is used to adjust the preset interval threshold based on the changing characteristics of the stress asthma association index; The early warning module is used to issue an early warning and output intervention suggestions when the adjusted and redefined stress asthma association index is greater than a preset index threshold. The correction module is used to correct the preset index threshold according to the intervention effectiveness, wherein the intervention effectiveness is determined based on the degree of respiratory abnormality, the cardiac interval, and the intensity of skin conductance within a preset correction judgment period after the implementation of the intervention recommendation.

[0007] Furthermore, the identification module includes: The starting determination unit is used to determine the interval start time based on the abnormal duration of the cardiac interval relative to the preset interval threshold, and to determine the crying start time based on the abnormal duration of the crying frequency. A skin conductance determination unit is used to determine the skin conductance initiation time based on the minimum value of the skin conductance activity intensity according to a threshold comparison result of the average skin conductance rate, wherein the average skin conductance rate is determined based on the instantaneous change of the skin conductance activity intensity. The identification unit is used to identify the stress event and record the stress moment based on the relevant characteristics of the crying frequency, the heartbeat interval and the skin conductance intensity within the candidate stress window, wherein the candidate stress window is determined based on the start time of the interval, the start time of the crying and the start time of the skin conductance.

[0008] Furthermore, the identification unit includes: A recording subunit is used to record the time window in which the start time of the interval is earlier than the start time of the skin conductance, and the start time of the skin conductance is earlier than the start time of the crying, as the candidate stress window; The relevant determination subunit is used to determine the cardiorespiratory correlation degree and the heart sound correlation degree respectively based on the correlation characteristics between the skin conductance intensity and the crying frequency and the heartbeat interval within the candidate pressure window. The identification subunit is used to identify the occurrence of the stress event based on the threshold comparison results of the strong cardiac correlation degree and the cardiac voice correlation degree, and to record the start time of the interval as the stress time.

[0009] Furthermore, the anomaly determination module includes: The normalization calculation unit is used to normalize the respiratory rate and blood oxygen saturation within a preset time period to obtain several respiratory normalization values ​​and several blood oxygen normalization values. An anomaly determination unit is used to determine several degrees of respiratory anomaly based on the geometric mean of the respiratory normalization value and the blood oxygen normalization value at each time point. A time determination unit is used to determine the abnormal time based on the threshold comparison result of the respiratory abnormality.

[0010] Furthermore, the association determination module includes: A matching determination unit is used to determine the time matching degree based on the historical interval and the pressure breathing interval, wherein the historical interval is determined based on the pressure breathing interval within the previous preset monitoring period, and the pressure breathing interval is determined based on the interval between the pressure moment and the abnormal moment. A coefficient determination unit is used to determine the pressure intensity coefficient based on the cardiac interval and the skin conductance intensity, and to determine the individual sensitivity coefficient based on the pressure intensity coefficient and the degree of respiratory abnormality. The association determination unit is used to determine the stress asthma association index based on the geometric mean of the time matching degree, the stress intensity coefficient, and the individual sensitivity coefficient.

[0011] Furthermore, the coefficient determination unit includes: The amplitude determination subunit is used to determine the amplitude of interval shortening and the amplitude of intensity increase based on the relative deviation of the cardiac interval and the intensity of skin electrical activity within a preset time range after the pressure moment. An intensity determination subunit is used to determine the pressure intensity coefficient based on the geometric mean of the interval amplitude normalization value and the intensity amplitude normalization value, wherein the interval amplitude normalization value is determined based on the interval shortening amplitude, and the intensity normalization value is determined based on the intensity increase amplitude. An individual determination subunit is used to determine the individual sensitivity coefficient based on the ratio of the maximum value of the respiratory abnormality after the pressure time within the preset monitoring period to the corresponding pressure intensity coefficient.

[0012] Furthermore, the adjustment module includes: The change determination unit is used to determine the rate of change of the index based on the slope of the linear fitting line of the pressure asthma-related index within the next preset adjustment period. A variation determination unit is used to determine the coefficient of variation of the pressure asthma-related index based on the coefficient of variation of the index within the preset adjustment period when the rate of change of the index is positive. An adjustment unit is used to adjust the preset interval threshold according to the exponential coefficient of variation.

[0013] Furthermore, the adjustment unit includes: Adjust the judgment subunit, which is used to determine the index abnormality based on the threshold comparison result of the index variation coefficient; An adjustment subunit is used to reduce the preset interval threshold based on the determination result of exponential anomaly and the relative deviation between the exponential coefficient of variation and the preset variation threshold.

[0014] Furthermore, the correction module includes: An energy efficiency determination unit is used to determine the energy efficiency of the intervention based on the degree of improvement of the respiratory abnormality, the cardiac interval, and the skin conductance intensity within the preset correction judgment period after the implementation of the intervention recommendation. The correction unit is used to reduce the preset index threshold based on the threshold comparison result of the intervention energy efficiency and the relative deviation between the intervention energy efficiency and the preset energy efficiency threshold.

[0015] Furthermore, the energy efficiency determination unit includes: A breathing determination subunit is used to calculate the relative deviation between the breathing abnormality degree and the minimum value of the breathing abnormality degree within the preset correction determination period, so as to obtain the breathing improvement degree. The heartbeat determination subunit is used to calculate the relative deviation between the average value of the heartbeat interval within the preset correction determination period and the preset interval threshold, so as to obtain the interval recovery degree. The skin conductance determination subunit is used to calculate the relative deviation between the maximum value and the average value of the skin conductance activity intensity within the preset correction determination period, so as to obtain the skin conductance drop amplitude. An energy efficiency determination subunit is used to determine the minimum value among the respiratory improvement, the inter-period recovery, and the skin conductance decline as the intervention energy efficiency.

[0016] Compared with existing technologies, the advantages of this invention lie in its ability to accurately identify stress events and abnormal breathing moments by real-time acquisition of multidimensional physiological parameters such as skin conductance intensity, heart rate, respiratory rate, blood oxygen saturation, and crying frequency, and to construct a causal relationship model from psychological stress to physiological deterioration based on their temporal changes. By calculating the stress-asthma correlation index using time matching degree, stress intensity coefficient, and individual sensitivity coefficient, the intrinsic pattern of asthma induced by emotional fluctuations is quantified. When the index continuously rises, the preset interval threshold is adjusted to improve monitoring sensitivity. Based on the comparison between the adjusted correlation index and the preset threshold, an early warning is issued and intervention suggestions are output only when the threshold is exceeded, effectively balancing sensitivity and specificity. Furthermore, the preset threshold is dynamically adjusted according to the improvement in children's physiological indicators after intervention, achieving adaptive optimization from monitoring and early warning to feedback on intervention effects. This effectively solves the problems of inaccurate early warning and untimely intervention in non-medical environments due to the lack of a causal relationship quantification mechanism and fixed thresholds.

[0017] Furthermore, by setting effective duration requirements, the shortened heartbeat interval or increased crying frequency must be maintained for a certain duration to be considered a valid initiation event, thus effectively filtering out false triggers caused by instantaneous sensor noise, single abnormal heartbeats, or brief environmental interference. Simultaneously, considering the physiological characteristics of skin conductance activity, the mean skin conductance rate is used to accurately locate the starting point of sympathetic nerve activation, avoiding misjudgments caused by individual baseline differences. By constructing a candidate stress window based on the temporal relationship of the three initiation times—the earliest occurrence of shortened heartbeat interval, followed by an increase in skin conductance activity, and the latest increase in crying frequency—the physiological logic of autonomic nerve activation preceding behavioral expression is reflected, eliminating accidental co-occurrences that do not conform to the stress response pattern. At the same time, the intensity of synergistic changes among parameters is quantified using the Pearson correlation coefficient, expanding the verification dimension from a single temporal sequence to the overall consistency of change trends. The occurrence and timing of stress events are confirmed based on threshold comparisons of synergistic change intensity, improving the accuracy and reliability of stress event identification.

[0018] Furthermore, by normalizing respiratory rate and blood oxygen saturation, the influence of individual differences, age characteristics, and baseline drift on the raw data can be eliminated, providing a unified benchmark for physiological indicators across different children and time periods. Determining the degree of respiratory abnormality using geometric mean effectively balances the contributions of changes in both indicators; any significant deviation in either indicator is amplified by the sensitivity of the geometric mean, thus enabling earlier detection of abnormal signs in the respiratory system. Finally, the moment when the first respiratory abnormality exceeds the preset abnormality threshold and meets the required duration is defined as the abnormal moment, eliminating transient noise or brief physiological fluctuations and ensuring that the recorded abnormal moment has genuine clinical prodromal significance, thereby improving the accuracy of respiratory abnormality identification.

[0019] Furthermore, by employing a Gaussian decay function in calculating the time-matching degree and determining the decay factor based on the 75th percentile of the ascending sequence of all pressure respiratory intervals within a preset monitoring period, the time-matching degree can smoothly reflect the nonlinear relationship between temporal proximity and the strength of causal association. Simultaneously, the decay factor dynamically determined based on individual historical data ensures the adaptability of the matching degree calculation, allowing most physiologically significant associated events to receive higher matching weights. When determining the pressure intensity coefficient, the variation amplitudes of the heartbeat interval and skin conductance intensity are first extracted centered on the pressure moment. These two amplitudes are then normalized before calculating the pressure intensity coefficient. The influence of individual baseline differences and sensor range is eliminated through relative deviation, making the stress amplitudes of different individuals and different events comparable. Normalization based on historical mean reflects the intensity concept relative to the individual's usual response level, allowing physiological changes of the same amplitude to represent stronger stress in individuals with normally mild responses. By calculating the ratio of the maximum respiratory abnormality to the stress intensity coefficient of the event when determining the individual sensitivity coefficient, the degree of respiratory response induced by a unit stress intensity is quantified, enabling the sensitivity coefficient to accurately reflect an individual's susceptibility to respiratory abnormalities under the same stress intensity. By calculating the stress-asthma association index, which simultaneously incorporates the temporal proximity of the event, the intensity of the current stress, and the individual's inherent reaction tendency, three orthogonal dimensions are used to reflect the risk of stress-induced asthma.

[0020] Furthermore, by calculating the exponential rate of change, the speed and direction of risk evolution can be quantified. Simultaneously, subsequent processes are only initiated when the exponential rate of change is positive, ensuring that adjustments are triggered only during periods of rising risk. By introducing a volatility index to capture the stability characteristics of risk, an exponential anomaly is identified when the coefficient of variation exceeds a preset threshold. This reflects that the risk is not only trending upward but also in a state of violent fluctuation, indicating a significant increase in the risk of autonomic nervous system dysfunction and asthma attacks. By reducing the preset interval threshold based on the relative deviation between the exponential coefficient of variation and the preset threshold of variation, the adjustment magnitude is proportional to the degree of volatility anomaly. This ensures that threshold adjustments occur only when risk exists, and that the adjustment amount matches the degree of risk anomaly, achieving an effective balance between risk response and system stability.

[0021] Furthermore, calculating the improvement in respiratory function objectively reflects the effectiveness of interventions in alleviating core asthma symptoms. When calculating the interval recovery rate, comparing post-intervention heart rate levels with the stress threshold quantifies the degree to which autonomic nervous function returns to normal from a stress state, demonstrating the intervention's effect on restoring physiological homeostasis. Calculating the skin conductance decline amplitude captures the dissipation of emotional stress; the difference between the maximum and average values ​​reflects the severity of the decline in skin conductance from its peak, avoiding interference from individual differences in skin conductance baseline. By defining the minimum value among respiratory improvement, interval recovery rate, and skin conductance decline amplitude as the intervention efficacy, the overall effect of the intervention is limited to the dimension with the worst improvement. Only when all three dimensions—respiratory symptoms, autonomic nervous system function, and emotional stress—achieve some improvement can the intervention be considered truly effective. By reducing the preset index threshold when the intervention efficacy is lower than the preset efficacy threshold, the magnitude of the threshold reduction is proportional to the degree of insufficient intervention effect, triggering an early warning. This achieves deep coupling between intervention efficacy evaluation and system parameter adaptation. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the intelligent monitoring system for pediatric respiratory diseases in this embodiment; Figure 2 This is a logic diagram for determining abnormal moments by the moment determination unit in this embodiment; Figure 3 The judgment logic diagram for judging index anomalies in the judgment subunit has been adjusted for this embodiment; Figure 4 This is a logic diagram for determining whether to reduce the preset exponential threshold in the correction unit of this embodiment. Detailed Implementation

[0023] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0024] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0025] Please see Figure 1 The diagram shown is a schematic of an intelligent monitoring system for pediatric respiratory diseases according to this embodiment. This embodiment of an intelligent monitoring system for pediatric respiratory diseases includes: The acquisition module is used to acquire the intensity of skin conductance, heart interval, respiratory rate, blood oxygen saturation, and crying frequency during the daily activities of children with asthma. The identification module, which is connected to the acquisition module, is used to identify stress events and record stress moments based on the temporal changes of the crying frequency, the heartbeat interval, and the intensity of the skin conductance activity, as well as a preset interval threshold. An anomaly determination module, which is connected to the acquisition module, is used to determine the degree of respiratory anomaly based on the respiratory rate and the blood oxygen saturation and record the time of the anomaly. The association determination module, which is connected to the acquisition module, the identification module, and the anomaly determination module, is used to determine the stress asthma association index based on the geometric mean characteristics of time matching degree, pressure intensity coefficient, and individual sensitivity coefficient. The time matching degree is determined based on the pressure breathing interval, the pressure breathing interval is determined based on the pressure time and the abnormal time, the individual sensitivity coefficient is determined based on the pressure event and the degree of respiratory abnormality, and the pressure intensity coefficient is determined based on the heartbeat interval and the skin conductance activity intensity. An adjustment module, which is connected to the association determination module, is used to adjust the preset interval threshold based on the change characteristics of the stress asthma association index. The early warning module is connected to the association determination module and the adjustment module respectively, and is used to issue an early warning and output intervention suggestions when the re-determined stress asthma association index after adjustment is greater than a preset index threshold. The correction module, which is connected to the acquisition module, the early warning module and the anomaly determination module respectively, is used to correct the preset index threshold according to the intervention effectiveness, wherein the intervention effectiveness is determined based on the respiratory abnormality, the cardiac interval and the skin conductance intensity within the preset correction judgment period after the implementation of the intervention recommendation.

[0026] In this embodiment, the intelligent monitoring system for pediatric respiratory diseases is applied to the daily health management of asthmatic children in non-medical environments such as homes or childcare facilities. Asthmatic children are more susceptible to stress factors such as emotional fluctuations, crying, tension, or external stimuli in these environments. Psychological or physiological stress can induce or worsen airway spasm through sympathetic nerve excitation and changes in respiratory rhythm, thereby increasing the risk of acute asthma attacks. Therefore, this embodiment adopts a cloud-based collaborative architecture, integrating data acquisition functionality into a wearable device worn by the child. This device is responsible for acquiring raw physiological data such as skin conductance intensity, heart rate, respiratory rate, blood oxygen saturation, and crying frequency in real time, and performing preliminary filtering, noise reduction, and compression on the data. The pre-processed data is uploaded to a cloud server in real time via a wireless communication module. Cloud-based computing services perform time-series analysis and fusion modeling of the multidimensional physiological data, dynamically quantifying the intrinsic correlation between psychological stress and respiratory status, and adaptively optimizing monitoring sensitivity based on individual historical data characteristics, thereby generating personalized health assessments and risk warnings. The cloud server pushes the analysis results and corresponding intervention guidance to the caregiver's terminal in real time, while storing all data in a cloud database to support continuous iteration and optimization of the analysis model. By combining the lightweight data collection of the front-end wearable device with the high-performance computing of the back-end cloud platform, the comfort and battery life of children's daily wear are ensured, enabling accurate monitoring and intelligent management of childhood asthma risk in non-medical environments.

[0027] In this embodiment, the acquisition module continuously monitors physiological and behavioral data in children's daily life scenarios to achieve early identification and timely intervention of stress-induced asthma abnormalities. Skin conductance intensity is a core indicator for measuring psychological stress, reflecting the level of sweat gland activity caused by sympathetic nerve excitation. It can be obtained by applying a constant voltage or specific frequency AC excitation, such as full-wave AC, to flexible dry electrodes attached to the wrist or palm, and measuring changes in skin impedance or conductance. Modern wearable devices often use AC excitation to reduce electrode polarization drift and improve signal stability. The cardiac interval is the time interval between two consecutive heartbeats. Physiologically, it usually refers to the number of milliseconds between the starting points of two consecutive heartbeats, such as the R-wave peak on an electrocardiogram or the systolic peak of a photoplethysmogram. It is obtained by emitting red and infrared light onto the skin using wearable devices such as smart bracelets and detecting changes in reflected light intensity. The time interval between adjacent pulse wave peaks is then extracted using peak detection algorithms such as local maxima and zero-crossing point detection to obtain the cardiac interval. Respiratory rate refers to the number of breaths per minute. The number of breaths can be measured by sensing the tension changes caused by changes in thoracic cavity volume during respiration using strain sensors or piezoresistive sensors in the chest and abdominal belt. Blood oxygen saturation refers to the percentage of oxyhemoglobin in the total hemoglobin in the blood. It is used to assess gas exchange efficiency and the degree of hypoxia and is an important physiological indicator for judging the risk of asthma attacks. It can be measured non-invasively using the photoplethysmography principle. Typically, a dual-wavelength light source of red and infrared light is used to illuminate the fingertips or wrists, and blood oxygen saturation is calculated by detecting changes in reflected or transmitted light intensity. Cry frequency refers to the fundamental frequency change of an infant's cry per unit time. Abnormal increases in this frequency are often related to emotional excitement, stress, or physical discomfort. It can be measured by picking up ambient sound signals using a MEMS microphone, extracting effective audio segments through automatic gain control and endpoint detection, extracting sound features using the Mel-frequency cepstral coefficient algorithm, and finally identifying and quantifying the cry and its intensity through fundamental frequency statistical analysis to obtain the cry frequency.

[0028] In this embodiment, the warning is to alert the caregiver to the child's current risk of an asthma attack. The intervention suggestions are based on the child being under stress and suggest specific measures such as comforting the child, guiding the child to do abdominal breathing exercises, assisting in the use of relief medications, suggesting leaving the current environment or seeking medical attention in a timely manner. The aim is to help the caregiver take timely and effective actions when the risk of an acute asthma attack increases, and to block or reduce possible asthma symptoms.

[0029] The preset interval threshold is a baseline value for the heart interval used to determine stressful events. It is calculated by monitoring children daily for a week, recording their heart interval data during periods without stressful events, and taking the 5th percentile as the lower limit of the stress sensitivity threshold. Combined with the clinical reference range for heart intervals in children, it is typically set between 400 and 600 milliseconds. In this embodiment, it is set to 500 milliseconds to avoid misjudging brief activities or minor emotional fluctuations as stress-related events. The preset index threshold is a critical value for triggering intervention recommendations based on the stress-asthma association index. It depends on the minimum association index of the child exhibiting respiratory abnormalities after a stressful event during the historical monitoring period. The statistical distribution is typically set between 0.6 and 0.8, and in this embodiment, it is set to 0.7. This allows for timely warnings to parents or caregivers and the output of intervention suggestions, enabling early identification of the risk of stress-induced asthma attacks. The preset correction judgment period is the length of the time window for evaluating the effectiveness of intervention measures. It depends on the typical duration of asthma symptoms triggered by the stress event and the physiological response time for the intervention suggestions to take effect. It is typically set between 24 and 72 hours, and in this embodiment, it is set to 48 hours. This allows for continuous tracking of the changes in children's physiological indicators after the stress event occurs, verifying whether the intervention measures have effectively blocked or alleviated the asthma attack process.

[0030] By collecting multidimensional physiological parameters such as skin conductance intensity, heart rate, respiratory rate, blood oxygen saturation, and crying frequency in real time, and accurately identifying stress events and abnormal breathing moments based on their temporal changes, a causal relationship model from psychological stress to physiological deterioration was constructed. A stress-asthma correlation index was calculated using time-matching degree, stress intensity coefficient, and individual sensitivity coefficient, quantifying the intrinsic pattern of asthma induced by emotional fluctuations. When the index continuously increases, a preset interval threshold is adjusted to improve monitoring sensitivity. Based on the comparison between the adjusted correlation index and the preset threshold, an early warning and intervention suggestion are issued only when the threshold is exceeded, effectively balancing sensitivity and specificity. Furthermore, the preset threshold is dynamically adjusted according to the improvement in children's physiological indicators after intervention, achieving adaptive optimization from monitoring and early warning to feedback on intervention effects. This effectively solves the problems of inaccurate risk warning and untimely intervention in non-medical environments due to the lack of a causal relationship quantification mechanism and fixed thresholds.

[0031] Specifically, the identification module includes: The start determination unit is used to record the start time of the interval duration window as the interval start time when the interval duration window of the heartbeat interval is less than the preset interval threshold is greater than the preset window threshold, and to record the start time of the crying duration window as the crying start time when the crying frequency is greater than the preset crying threshold and the crying duration window of the preset window threshold is greater than the preset window threshold. The skin conductance determination unit is used to record the time corresponding to the minimum value of the skin conductance activity intensity within the preset determination window as the skin conductance start time when the average skin conductance rate within the previous preset determination window is greater than the preset rate threshold. The average skin conductance rate is determined based on the average value of the instantaneous change rate of the skin conductance activity intensity. An identification unit, which is connected to the initiation determination unit and the skin conductance determination unit respectively, is used to identify the stress event and record the stress time based on the relevant characteristics of the crying frequency, the heartbeat interval and the skin conductance activity intensity within the candidate stress window, wherein the candidate stress window is determined based on the interval initiation time, the crying initiation time and the skin conductance initiation time.

[0032] The preset window threshold is the minimum duration used to determine the validity of events such as shortened heart interval and increased crying. It is determined by collecting physiological data from the target child during stress-free periods, such as quiet sleep or play. The duration of single fluctuations where the heart interval occasionally falls below the preset threshold, and the duration of brief vocalizations where the crying frequency occasionally exceeds the preset threshold, are statistically analyzed. The upper limit of the duration of these two types of interfering events is taken as the preset window threshold, typically set between 2 and 8 seconds. In this embodiment, it is set to 5 seconds to effectively filter out false triggers caused by instantaneous signal jitter, single abnormal heartbeats, or environmental noise. The preset crying threshold is the frequency benchmark used to determine events such as increased crying. It is determined by collecting crying frequency data during stress-free periods through three days of daily monitoring of the child. The 95th percentile is taken as the boundary for determining abnormal increases. Based on this, its value is typically set between 300Hz and 400Hz. In this embodiment, it is set to 350Hz to accurately identify events caused by stress. An abnormally high frequency of crying caused by emotional excitement or stress; a preset rate threshold is a benchmark value for determining the rise rate of skin conductance stress response events. By statistically analyzing the skin conductance response data of healthy children in emotion-induced experiments, the 25th percentile of the slope distribution of the rise phase of the response is taken as the minimum effective response boundary. Based on this, its value is usually set between 0.03µS / s and 0.07µS / s. In this embodiment, it is set to 0.05µS / s, which can accurately identify the rapid rise trend of skin conductance activity intensity within the preset determination window; the preset determination window is a benchmark value for calculating the mean skin conductance rate and determining the time length of skin conductance stress response events. By analyzing the rise dynamic characteristics of skin conductance activity in clinically collected children's stress events, the average time interval from the start of skin conductance rise to the maximum rise rate is statistically analyzed. Based on this, its value is usually set between 8 seconds and 12 seconds. In this embodiment, it is set to 10 seconds, which can capture the rapid rise trend of skin conductance activity.

[0033] Specifically, the identification unit includes: A recording subunit is used to record the time window in which the start time of the interval is earlier than the start time of the skin conductance, and the start time of the skin conductance is earlier than the start time of the crying, as the candidate stress window; A related determination subunit, which is connected to the recording subunit, is used to calculate the Pearson correlation coefficient between the cardiac interval and the skin conductance intensity based on the candidate pressure window to obtain the cardiothoracic correlation, and to calculate the Pearson correlation coefficient between the cardiac interval and the cry frequency to obtain the heart sound correlation. An identification subunit, connected to the related determination subunit, is used to identify the occurrence of the stress event when the absolute value of the strong cardiac correlation degree and the absolute value of the cardiac voice correlation degree are both greater than a preset correlation threshold, and to record the interval start time as the stress time.

[0034] The preset correlation threshold is a benchmark value used to verify the correlation between the intensity of changes in cardiac interval and skin conductance, and between cardiac interval and crying frequency. Physiological data of 50 clinically diagnosed children with stress-induced asthma were collected during real crying stress events. The actual values ​​of cardiac correlation and heart-voice correlation in each event were calculated. At the same time, spurious correlation values ​​were collected in an equal number of calm periods as controls. Receiver operating characteristic curve analysis was used, and the correlation coefficient corresponding to the maximum Youden index was taken as the optimal classification critical point. Based on this, its value is usually set between 0.5 and 0.7. In this embodiment, it is set to 0.6, which can effectively filter random noise interference while ensuring sensitive identification of real stress events.

[0035] By setting effective duration requirements, the shortened heart interval or increased crying frequency must be maintained for a certain duration to be considered a valid initiation event, thus effectively filtering out false triggers caused by instantaneous sensor noise, single abnormal heartbeats, or brief environmental interference. Simultaneously, considering the physiological characteristics of skin conductance activity, the average skin conductance rate is used to accurately locate the starting point of sympathetic nerve activation, avoiding misjudgments caused by individual baseline differences. The temporal relationship of the three initiation times—the earliest occurrence of shortened heart interval, followed by increased skin conductance activity, and finally the most pronounced crying—is used to construct candidate stress windows, reflecting the physiological logic of autonomic nerve activation preceding behavioral expression, thus eliminating accidental co-occurrences that do not conform to the stress response pattern. Furthermore, the Pearson correlation coefficient is used to quantify the intensity of synergistic changes among parameters, expanding the verification dimension from a single temporal sequence to the overall consistency of change trends. Based on the threshold comparison results of the intensity of synergistic changes, the occurrence and timing of stress events are confirmed, improving the accuracy and reliability of stress event identification.

[0036] Please see Figure 2 As shown, this is the logic diagram for determining abnormal times by the time determination unit in this embodiment. In this embodiment, the abnormal determination module includes: The normalization calculation unit is used to normalize the respiratory rate and blood oxygen saturation within a preset time period to obtain several respiratory normalization values ​​and several blood oxygen normalization values. An anomaly determination unit, which is connected to the normalization calculation unit, is used to calculate the geometric mean of the normalized respiratory value and the normalized blood oxygen value at each time point to obtain several respiratory anomaly degrees. A time determination unit, which is connected to the anomaly determination unit, is used to determine the start time of the duration at which the first respiratory abnormality exceeds a preset abnormal threshold as the abnormal time.

[0037] The preset duration is the length of the time window used to extract respiratory rate and blood oxygen saturation data for normalization. By analyzing the time-series data of asthmatic children collected clinically after exercise-induced or allergen exposure, showing an increase in respiratory rate and a decrease in blood oxygen saturation, the distribution of time intervals from the onset of the abnormality to the peak value is statistically analyzed, and the median is taken as the preset duration. Based on this, its value is usually set between 30 and 90 seconds; in this embodiment, it is set to 60 seconds, which can capture abnormal signals at the initial stage when the respiratory rate shows an upward trend or the blood oxygen saturation shows a downward trend. The preset abnormality threshold is used to determine whether the respiratory abnormality is established. The comprehensive index threshold was established by collecting respiratory rate and blood oxygen saturation data from 100 children diagnosed with asthma within 30 minutes before their previous 12 attacks, calculating their respiratory abnormality sequence, and collecting respiratory abnormality data from an equal number of healthy children during normal activities as a control. Receiver operating characteristic curve analysis was used to determine the optimal classification threshold that best distinguishes between the prodromal and quiescent phases, with the maximization of the Youden index as the criterion. Based on this, the value is usually set between 0.6 and 0.8. In this embodiment, it is set to 0.7, which can effectively avoid misjudgment caused by transient physiological fluctuations while ensuring detection sensitivity.

[0038] By normalizing respiratory rate and blood oxygen saturation, the influence of individual differences, age characteristics, and baseline drift on the raw data can be eliminated, providing a unified benchmark for physiological indicators across different children and time periods. Determining the degree of respiratory abnormality using geometric mean effectively balances the contributions of changes in both indicators; any significant deviation in either indicator is amplified by the sensitivity of the geometric mean, thus enabling earlier detection of respiratory abnormalities. Finally, the moment when the first respiratory abnormality exceeds a preset threshold and meets the required duration is defined as the abnormal moment, eliminating transient noise or brief physiological fluctuations and ensuring that the recorded abnormal moment has genuine clinical prodromal significance, thereby improving the accuracy of respiratory abnormality identification.

[0039] Specifically, the association determination module includes: A matching determination unit is used to determine the time matching degree based on the historical intervals within the previous preset monitoring period and the pressure breathing interval, wherein... Among them, S P It is the time matching degree, T t T0 is the pressure breathing interval at the current moment, and T0 is the historical interval. The historical interval is determined based on the preset percentile of the ascending sequence of pressure breathing intervals in the previous preset monitoring period, and the pressure breathing interval is determined based on the absolute value of the difference between the pressure moment and the abnormal moment. A coefficient determination unit is used to determine the pressure intensity coefficient based on the cardiac interval and the skin conductance intensity, and to determine the individual sensitivity coefficient based on the pressure intensity coefficient and the degree of respiratory abnormality. An association determination unit, which is connected to the matching determination unit and the coefficient determination unit respectively, is used to calculate the geometric mean of the time matching degree, the pressure intensity coefficient and the individual sensitivity coefficient to obtain the pressure asthma association index.

[0040] Specifically, the coefficient determination unit includes: The amplitude determination subunit is used to calculate the relative deviation between the preset interval threshold and the minimum value of the heartbeat interval based on a preset window range centered on the pressure moment, so as to obtain the interval shortening amplitude, and to calculate the relative deviation between the maximum value of the skin conductance activity intensity and the preset intensity threshold, so as to obtain the intensity increase amplitude. An intensity determination subunit, connected to the amplitude determination subunit, is used to determine the pressure intensity coefficient based on the geometric mean of the interval amplitude normalization value and the intensity amplitude normalization value. The interval amplitude normalization value is determined based on the ratio of the interval shortening amplitude to the average of the interval shortening amplitude within the preset monitoring period, and the intensity normalization value is determined based on the ratio of the intensity increase amplitude to the average of the intensity increase amplitude within the preset monitoring period. An individual determination subunit, connected to the intensity determination subunit, is used to calculate the ratio of the maximum value of the respiratory abnormality after the pressure time within the preset monitoring period to its corresponding pressure intensity coefficient, so as to obtain the individual sensitivity coefficient.

[0041] The preset monitoring period is the length of the time window used to statistically analyze historical data. By analyzing the diurnal coefficient of variation of indicators such as heart rate interval, skin conductance, and respiratory abnormality in asthmatic children during daily monitoring, the smallest time span where the coefficient of variation tends to stabilize is taken as the preset monitoring period. Based on this, its value is usually set between 5 and 10 days. In this embodiment, it is set to 7 days, which can provide a statistical benchmark for calculating time matching degree, pressure intensity coefficient, and individual sensitivity coefficient. The preset percentile is used to extract the statistics of historical intervals from the ascending sequence of historical intervals. By arranging all pressure respiratory intervals within the preset monitoring period in ascending order from smallest to largest, since the time interval of 75% of the effective associated events gives most physiologically significant pressure respiratory associations a high time weight, the preset percentile is set to 75% in this embodiment, which can reasonably balance the sensitivity of time matching degree. The preset window range is used for The time interval length for extracting physiological data around stress moments is determined by analyzing the average duration of the heartbeat interval shortening to its minimum value and the average duration of skin activity rising to its peak value during the collected stress events. The upper limit of both is taken as the preset window range. Based on this, the value is usually set between 45 seconds and 90 seconds. In this embodiment, it is set to 60 seconds, which can fully cover the entire process of heartbeat interval shortening and skin activity rising. The preset intensity threshold is the benchmark value used to calculate the relative deviation of skin activity intensity. By extracting skin activity intensity data from all stress-free periods in the previous preset monitoring cycle, the average skin activity level representing an individual's daily calm state is used as the preset intensity threshold. Based on this, the value is usually set between 1.5µS and 3.5µS. In this embodiment, it is set to 2.5µS, which can eliminate the influence of differences in skin activity background among different individuals on the quantification of stress intensity.

[0042] By employing a Gaussian decay function in calculating time-matching scores and determining the decay factor based on the 75th percentile of the ascending sequence of all pressure-respiratory intervals within a preset monitoring period, the time-matching scores can smoothly reflect the nonlinear relationship between temporal proximity and the strength of causal association. Simultaneously, the dynamically determined decay factor based on individual historical data ensures the adaptability of the matching score calculation, allowing most physiologically significant associated events to receive higher matching weights. When determining the pressure intensity coefficient, the amplitude of changes in cardiac interval and skin conductance intensity is first extracted centered on the pressure moment. These two amplitudes are then normalized before calculating the pressure intensity coefficient. Relative deviation eliminates the influence of individual baseline differences and sensor range, making the stress amplitudes of different individuals and different events comparable. Normalization based on historical mean reflects the intensity concept relative to the individual's usual response level, allowing physiological changes of the same amplitude to represent stronger stress in individuals with normally mild responses. By calculating the ratio of the maximum respiratory abnormality to the stress intensity coefficient of the event when determining the individual sensitivity coefficient, the degree of respiratory response induced by a unit stress intensity is quantified, enabling the sensitivity coefficient to accurately reflect an individual's susceptibility to respiratory abnormalities under the same stress intensity. By calculating the stress-asthma association index, which simultaneously incorporates the temporal proximity of the event, the intensity of the current stress, and the individual's inherent reaction tendency, three orthogonal dimensions are used to reflect the risk of stress-induced asthma.

[0043] Specifically, the adjustment module includes: The change determination unit is used to calculate the slope of the linear fitting line of the pressure asthma-related index within the next preset adjustment period to obtain the rate of change of the index. A variation determination unit, connected to the change determination unit, is used to calculate the coefficient of variation of the pressure asthma-related index within the preset adjustment period when the index change rate is positive, so as to obtain the index variation coefficient. An adjustment unit, which is connected to the change determination unit and the variation determination unit respectively, is used to adjust the preset interval threshold according to the exponential variation coefficient.

[0044] Please see Figure 3 As shown, this is the logic diagram for determining the abnormality of the index in the adjustment determination subunit of this embodiment. In this embodiment, the adjustment unit includes: Adjust the judgment subunit, which is used to determine that the index is abnormal when the coefficient of variation of the index is greater than a preset variation threshold; An adjustment subunit, connected to the adjustment determination subunit, is used to reduce the preset interval threshold based on the determination result of exponential anomaly and the relative deviation between the exponential coefficient of variation and the preset variation threshold, where Q'=Q×[1-q×(Z-Z0) / Z0], where Q' is the adjusted preset interval threshold, Q is the original preset interval threshold, q is the preset interval adjustment coefficient, Z is the exponential coefficient of variation, and Z0 is the preset variation threshold.

[0045] The preset adjustment period is the length of the time window used to extract stress-asthma association index data and perform trend analysis. It is based on the actual frequency of stress events in the target children during daily monitoring. Statistical analysis is performed according to the basic requirements for sample size calculated using linear regression and coefficient of variation calculations. That is, a certain number of event data points are needed to obtain a stable slope and coefficient of variation estimate. Therefore, the minimum time span required to meet the sample size requirement is calculated. Based on this, its value is usually set between 48 and 96 hours; in this embodiment, it is set to 72 hours, which allows for the accumulation of sufficient event occurrences within this window to support reliable linear fitting and coefficient of variation calculation. The preset variation threshold is the benchmark value used in the adjustment judgment subunit to determine whether the fluctuation of the stress-asthma association index is abnormal. By collecting stress-asthma association index data from 100 asthmatic children during a 30-day period of stable condition, the coefficient of variation within each 72-hour window is calculated. The 90th percentile is used as the upper limit of normal fluctuation. Simultaneously, the coefficient of variation of the same batch of children within a 72-hour window before the actual attack is collected, and the 10th percentile is taken as the lower limit of abnormal fluctuation. The mean of the two boundaries is used as the preset variation threshold. Based on this, its value is usually set between 0.25 and 0.45. In this embodiment, it is set to 0.35, which can identify unstable states that may indicate an increased risk of asthma attacks. The preset interval adjustment coefficient is a proportional factor used to control the adjustment range of the preset interval threshold. By analyzing the system's response speed and threshold oscillation amplitude to simulated risk events under different adjustment coefficients, the coefficient value that can reduce the preset interval threshold by 5% after three consecutive abnormal judgments without causing excessive oscillations is taken as the preset interval adjustment coefficient. Based on this, its value is usually set between 0.05 and 0.15. In this embodiment, it is set to 0.1, which can match the monitoring sensitivity with the degree of risk abnormality.

[0046] By calculating the exponential rate of change, the speed and direction of risk evolution can be quantified. Furthermore, subsequent processes are only initiated when the exponential rate of change is positive, ensuring that adjustments are triggered only during periods of escalating risk. Introducing a volatility indicator captures the stability characteristics of risk; when the coefficient of variation exceeds a preset threshold, it is considered an exponential anomaly, reflecting not only an upward trend but also a highly volatile and unstable state, indicating a significant increase in the risk of autonomic nervous system dysfunction and asthma attacks. By reducing the preset interval threshold based on the relative deviation between the exponential coefficient of variation and the preset threshold, the adjustment magnitude is proportional to the degree of volatility anomaly. This ensures that threshold adjustments occur only when risk exists, and that the adjustment amount matches the degree of risk anomaly, achieving an effective balance between risk response and system stability.

[0047] Please see Figure 4 As shown, this is the logic diagram for the correction unit to determine the reduction of the preset exponential threshold in this embodiment. In this embodiment, the correction module includes: An energy efficiency determination unit is used to determine the energy efficiency of the intervention based on the degree of improvement of the respiratory abnormality, the cardiac interval, and the skin conductance intensity within the preset correction judgment period after the implementation of the intervention recommendation. A correction unit, connected to the energy efficiency determination unit, is used to reduce the preset index threshold according to the relative deviation between the intervention energy efficiency and the preset energy efficiency threshold when the intervention energy efficiency is less than the preset energy efficiency threshold. U'=U×[1-u×(G0-G) / G0], where U' is the preset index threshold after correction, U is the preset index threshold before correction, u is the preset index correction coefficient, G0 is the preset energy efficiency threshold, and G is the intervention energy efficiency.

[0048] Specifically, the energy efficiency determination unit includes: A breathing determination subunit is used to calculate the relative deviation between the breathing abnormality degree and the minimum value of the breathing abnormality degree within the preset correction determination period, so as to obtain the breathing improvement degree. The heartbeat determination subunit is used to calculate the relative deviation between the average value of the heartbeat interval within the preset correction determination period and the preset interval threshold, so as to obtain the interval recovery degree. The skin conductance determination subunit is used to calculate the relative deviation between the maximum value and the average value of the skin conductance activity intensity within the preset correction determination period, so as to obtain the skin conductance drop amplitude. An energy efficiency determination subunit, which is connected to the breathing determination subunit, the heartbeat determination subunit, and the skin conductance determination subunit, is used to determine the minimum value among the breathing improvement, the interval recovery, and the skin conductance decline as the intervention energy efficiency.

[0049] The preset energy efficiency threshold is a critical benchmark value used to determine whether an intervention is effective. It is calculated by collecting data from 100 children with asthma in 200 previous successful intervention events to prevent asthma attacks. The minimum distribution of respiratory improvement, interval recovery, and skin conductance decline after intervention is calculated, and the 10th percentile is taken as the lower limit of effective intervention. At the same time, the natural fluctuation range of the three indicators of the same group of children during calm periods without risk events is collected, and the 90th percentile is taken as the upper limit of random fluctuation. The mean of the two boundaries is taken as the preset energy efficiency threshold. Based on this, its value is usually set between 0.3 and 0. The value is between 5 and 0.4 in this embodiment, which can determine that the intervention effect is insufficient when the intervention efficiency is lower than this threshold. The preset index correction coefficient is a proportional factor used to control the adjustment range of the preset index threshold. By analyzing the system's response speed and threshold oscillation amplitude to simulated intervention failure events under different correction coefficients, the coefficient value that can reduce the preset index threshold by 7% after 3 consecutive ineffective interventions without causing excessive oscillation is taken as the preset index correction coefficient. Based on this, its value is usually set between 0.1 and 0.3. In this embodiment, it is set to 0.2, which can enable timely early warning in the future.

[0050] Calculating respiratory improvement objectively reflects the intervention's effectiveness in alleviating core asthma symptoms. Calculating interval recovery quantifies the degree to which autonomic nervous function returns to normal from a stress state by comparing post-intervention heart rate levels with a stress threshold, demonstrating the intervention's effect on restoring physiological homeostasis. Calculating the magnitude of skin conductance decline captures the dissipation of emotional stress; the difference between the maximum and average values ​​reflects the severity of the decline from peak skin conductance, avoiding interference from individual differences in skin conductance baseline. By defining the minimum value among respiratory improvement, interval recovery, and skin conductance decline as the intervention efficacy, the overall effect of the intervention is limited to the dimension with the worst improvement. Only when respiratory symptoms, autonomic nervous system function, and emotional stress all show some improvement can the intervention be considered truly effective. By reducing the preset index threshold when the intervention efficacy is lower than the preset efficacy threshold, the magnitude of the threshold reduction is proportional to the degree of insufficient intervention effect, triggering an early warning. This achieves deep coupling between intervention efficacy assessment and system parameter adaptation.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent monitoring system for respiratory diseases in children, characterized in that, include: The acquisition module is used to acquire the intensity of skin conductance, heart interval, respiratory rate, blood oxygen saturation, and crying frequency during the daily activities of children with asthma. The identification module is used to identify stress events and record stress moments based on the temporal changes in the crying frequency, the cardiac interval, and the intensity of skin conductance, as well as a preset interval threshold. An anomaly determination module is used to determine the degree of respiratory anomaly based on the respiratory rate and the blood oxygen saturation and to record the time of the anomaly. The association determination module is used to determine the stress asthma association index based on the geometric mean characteristics of time matching degree, stress intensity coefficient and individual sensitivity coefficient, wherein the time matching degree is determined based on the stress breathing interval, the stress breathing interval is determined based on the stress time and the abnormal time, the individual sensitivity coefficient is determined based on the stress event and the degree of respiratory abnormality, and the stress intensity coefficient is determined based on the heartbeat interval and the skin conductance activity intensity. An adjustment module is used to adjust the preset interval threshold based on the changing characteristics of the stress asthma association index; The early warning module is used to issue an early warning and output intervention suggestions when the adjusted and redefined stress asthma association index is greater than a preset index threshold. The correction module is used to correct the preset index threshold according to the intervention effectiveness, wherein the intervention effectiveness is determined based on the degree of respiratory abnormality, the cardiac interval, and the intensity of skin conductance within a preset correction judgment period after the implementation of the intervention recommendation.

2. The intelligent monitoring system for pediatric respiratory diseases according to claim 1, characterized in that, The identification module includes: The starting determination unit is used to determine the interval start time based on the abnormal duration of the cardiac interval relative to the preset interval threshold, and to determine the crying start time based on the abnormal duration of the crying frequency. A skin conductance determination unit is used to determine the skin conductance initiation time based on the minimum value of the skin conductance activity intensity according to a threshold comparison result of the average skin conductance rate, wherein the average skin conductance rate is determined based on the instantaneous change of the skin conductance activity intensity. The identification unit is used to identify the stress event and record the stress moment based on the relevant characteristics of the crying frequency, the heartbeat interval and the skin conductance intensity within the candidate stress window, wherein the candidate stress window is determined based on the start time of the interval, the start time of the crying and the start time of the skin conductance.

3. The intelligent monitoring system for pediatric respiratory diseases according to claim 2, characterized in that, The identification unit includes: A recording subunit is used to record the time window in which the start time of the interval is earlier than the start time of the skin conductance, and the start time of the skin conductance is earlier than the start time of the crying, as the candidate stress window; The relevant determination subunit is used to determine the cardiorespiratory correlation degree and the heart sound correlation degree respectively based on the correlation characteristics between the skin conductance intensity and the crying frequency and the heartbeat interval within the candidate pressure window. The identification subunit is used to identify the occurrence of the stress event based on the threshold comparison results of the strong cardiac correlation degree and the cardiac voice correlation degree, and to record the start time of the interval as the stress time.

4. The intelligent monitoring system for pediatric respiratory diseases according to claim 3, characterized in that, The anomaly determination module includes: The normalization calculation unit is used to normalize the respiratory rate and blood oxygen saturation within a preset time period to obtain several respiratory normalization values ​​and several blood oxygen normalization values. An anomaly determination unit is used to determine several degrees of respiratory anomaly based on the geometric mean of the respiratory normalization value and the blood oxygen normalization value at each time point. A time determination unit is used to determine the abnormal time based on the threshold comparison result of the respiratory abnormality.

5. The intelligent monitoring system for pediatric respiratory diseases according to claim 4, characterized in that, The association determination module includes: A matching determination unit is used to determine the time matching degree based on the historical interval and the pressure breathing interval, wherein the historical interval is determined based on the pressure breathing interval within the previous preset monitoring period, and the pressure breathing interval is determined based on the interval between the pressure moment and the abnormal moment. A coefficient determination unit is used to determine the pressure intensity coefficient based on the cardiac interval and the skin conductance intensity, and to determine the individual sensitivity coefficient based on the pressure intensity coefficient and the degree of respiratory abnormality. The association determination unit is used to determine the stress asthma association index based on the geometric mean of the time matching degree, the stress intensity coefficient, and the individual sensitivity coefficient.

6. The intelligent monitoring system for pediatric respiratory diseases according to claim 5, characterized in that, The coefficient determination unit includes: The amplitude determination subunit is used to determine the amplitude of interval shortening and the amplitude of intensity increase based on the relative deviation of the cardiac interval and the intensity of skin electrical activity within a preset time range after the pressure moment. An intensity determination subunit is used to determine the pressure intensity coefficient based on the geometric mean of the interval amplitude normalization value and the intensity amplitude normalization value, wherein the interval amplitude normalization value is determined based on the interval shortening amplitude, and the intensity normalization value is determined based on the intensity increase amplitude. An individual determination subunit is used to determine the individual sensitivity coefficient based on the ratio of the maximum value of the respiratory abnormality after the pressure time within the preset monitoring period to the corresponding pressure intensity coefficient.

7. The intelligent monitoring system for pediatric respiratory diseases according to claim 6, characterized in that, The adjustment module includes: The change determination unit is used to determine the rate of change of the index based on the slope of the linear fitting line of the pressure asthma-related index within the next preset adjustment period. A variation determination unit is used to determine the coefficient of variation of the pressure asthma-related index based on the coefficient of variation of the index within the preset adjustment period when the rate of change of the index is positive. An adjustment unit is used to adjust the preset interval threshold according to the exponential coefficient of variation.

8. The intelligent monitoring system for pediatric respiratory diseases according to claim 7, characterized in that, The adjustment unit includes: Adjust the judgment subunit, which is used to determine the index abnormality based on the threshold comparison result of the index variation coefficient; An adjustment subunit is used to reduce the preset interval threshold based on the determination result of exponential anomaly and the relative deviation between the exponential coefficient of variation and the preset variation threshold.

9. The intelligent monitoring system for pediatric respiratory diseases according to claim 8, characterized in that, The correction module includes: An energy efficiency determination unit is used to determine the energy efficiency of the intervention based on the degree of improvement of the respiratory abnormality, the cardiac interval, and the skin conductance intensity within the preset correction judgment period after the implementation of the intervention recommendation. The correction unit is used to reduce the preset index threshold based on the threshold comparison result of the intervention energy efficiency and the relative deviation between the intervention energy efficiency and the preset energy efficiency threshold.

10. The intelligent monitoring system for pediatric respiratory diseases according to claim 9, characterized in that, The energy efficiency determination unit includes: A breathing determination subunit is used to calculate the relative deviation between the breathing abnormality degree and the minimum value of the breathing abnormality degree within the preset correction determination period, so as to obtain the breathing improvement degree. The heartbeat determination subunit is used to calculate the relative deviation between the average value of the heartbeat interval within the preset correction determination period and the preset interval threshold, so as to obtain the interval recovery degree. The skin conductance determination subunit is used to calculate the relative deviation between the maximum value and the average value of the skin conductance activity intensity within the preset correction determination period, so as to obtain the skin conductance drop amplitude. An energy efficiency determination subunit is used to determine the minimum value among the respiratory improvement, the inter-period recovery, and the skin conductance decline as the intervention energy efficiency.

Images