Non-contact sleep apnea syndrome monitoring and early warning device

Abstract

The application discloses a non-contact respiratory sleep syndrome monitoring and early warning device and belongs to the technical field of medical equipment. The non-contact respiratory sleep syndrome monitoring and early warning device comprises a monitoring and sensing module, a data analysis module and an abnormality early warning module. The application solves the problem of certain limitations in the prior art. The application accurately collects respiratory data, blood oxygen data and heart rate data of a patient through infrared sensing function and can determine the quality of data collection, thereby ensuring the accuracy and reliability of the collected data. The collected respiratory data, blood oxygen data and heart rate data are compared with preset warning values to determine whether the patient's respiration, blood oxygen and heart rate are abnormal. The preset warning values can be dynamically adjusted based on the historical monitoring data of the patient, thereby improving the flexibility and accuracy of device monitoring. In addition, the warning sound can be set individually according to the abnormality classification and patient characteristics to wake up the patient and prevent adverse events.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical equipment, in particular to a non-contact respiratory sleep syndrome monitoring and early warning device. BACKGROUND

[0002] Respiratory sleep syndrome is usually manifested as apnea or hypopnea phenomenon during sleep, which leads to a decrease in blood oxygen saturation, seriously affecting the sleep quality and physical health of patients, and long-term can cause a series of complications such as cardiovascular disease, cognitive dysfunction, etc. At present, for the monitoring of respiratory sleep syndrome, the traditional contact monitoring method has certain limitations, which needs to attach many sensors to the patient's body, not only bringing discomfort to the patient, but also the complex line connection increases the difficulty and error of monitoring; Therefore, it does not meet the existing needs, and for this we propose a non-contact respiratory sleep syndrome monitoring and early warning device. SUMMARY

[0003] The purpose of the present application is to provide a non-contact respiratory sleep syndrome monitoring and early warning device, which is placed above the pillow of the patient, and the respiratory frequency, blood oxygen and heart rate of the individual in sleep are monitored by using the infrared sensing function. When the respiration or blood oxygen or heart rate exceeds the warning value, a specific warning sound is sent out in time to wake up the patient, so as to prevent adverse events and solve the problems raised in the above background technology.

[0004] To achieve the above purpose, the present application provides the following technical scheme: a non-contact respiratory sleep syndrome monitoring and early warning device, which is arranged above the pillow of the bed head, comprising a monitoring and sensing module, a data analysis module and an abnormal early warning module;

[0005] The monitoring and sensing module is configured to use the infrared sensing function of the infrared sensor to monitor the respiratory frequency, blood oxygen and heart rate of the individual in sleep in real time, collect the respiratory data, blood oxygen data and heart rate data of the patient, and determine the quality of data collection, while positioning the position of the patient in the pillow area, automatically adjusting the focusing area of the infrared sensing, specifically:

[0006] The standard deviation and mean value of the patient's historical respiratory data and blood oxygen data are obtained, the corresponding ratio coefficient is calculated and compared with the preset difference threshold value, and the initial position offset threshold value is determined as the trigger condition or the initial position offset threshold value is adjusted according to the comparison result;

[0007] The data analysis module is configured to compare and analyze the collected respiratory data, blood oxygen data and heart rate data with the preset warning value respectively, to determine whether the patient's respiration and blood oxygen are abnormal, and to dynamically adjust the warning value based on the patient's historical monitoring data;

[0008] The abnormality early warning module is configured to, once any one of the respiratory rate, blood oxygen, or heart rate appears an abnormal condition, grade the abnormal condition, combine the abnormal grading result with the sleep habit and physiological characteristics of the patient, set the type, volume, frequency, and duration of the warning sound, and then issue a specific warning sound to wake up the patient, and at the same time, synchronize the abnormal grading result, the abnormal data, and the monitoring data to the remote medical platform and the patient family account associated with the device.

[0009] Further, the monitoring and sensing module comprises:

[0010] The infrared sensing module is configured to emit an infrared light beam to the sleeping individual through an infrared sensor, capture the respiratory condition, blood oxygen information, and heart rate information of the individual in the sleep process, and collect the respiratory data, blood oxygen data, and heart rate data of the patient.

[0011] The data determination module is configured to identify the error data in the respiratory data, blood oxygen data, and heart rate data, according to the identification result, extract the data amount information corresponding to the error data in the respiratory data, blood oxygen data, and heart rate data respectively, evaluate the collection quality of the respiratory data, blood oxygen data, and heart rate data, and perform abnormal alarm when the collection quality of the respiratory data, blood oxygen data, and heart rate data is abnormal.

[0012] The intelligent positioning module is configured to use the millimeter wave radar technology to position the position of the patient in the pillow area in real time, and automatically adjust the focusing area of the infrared sensing module when the position of the patient deviates.

[0013] Further, the data determination module comprises:

[0014] The data identification module is configured to pre-analyze the characteristics of the existing normal respiratory data, blood oxygen data, and heart rate data in the numerical range, fluctuation law, and change trend, and establish a data identification model, input the collected respiratory data, blood oxygen data, and heart rate data into the data identification model for comparison, and regard the data not conforming to the data identification model as error data.

[0015] The data extraction module is configured to extract the first data amount information corresponding to the error data in the respiratory data, extract the second data amount information corresponding to the error data in the blood oxygen data, and extract the third data amount information corresponding to the error data in the heart rate data.

[0016] The abnormality determination module is configured to compare the first data amount information, the second data amount information, and the third data amount information with the preset parameter threshold respectively, specifically:

[0017] If the first data amount information is lower than the preset parameter threshold, it is determined that the collection quality of the respiratory data is abnormal, and abnormal alarm is performed.

[0018] If the second data quantity information is lower than the preset parameter threshold, it is determined that the collection quality of blood oxygen data is abnormal, and an abnormal alarm is performed.

[0019] If the third data quantity information is lower than the preset parameter threshold, it is determined that the collection quality of heart rate data is abnormal, and an abnormal alarm is performed.

[0020] Further, the position offset threshold corresponding to the patient position offset for triggering automatic adjustment of the focusing area is set, including:

[0021] The historical respiratory data and historical blood oxygen data corresponding to the patient are retrieved;

[0022] The respiratory data standard deviation and respiratory data mean value corresponding to the patient are obtained according to the historical respiratory data corresponding to the patient;

[0023] The blood oxygen data standard deviation and blood oxygen data mean value corresponding to the patient are obtained according to the historical blood oxygen data corresponding to the patient;

[0024] The respiratory data standard deviation and respiratory data mean value corresponding to the patient are processed by ratio, to obtain a respiratory ratio coefficient;

[0025] The blood oxygen data standard deviation and blood oxygen data mean value corresponding to the patient are processed by ratio, to obtain a blood oxygen ratio coefficient;

[0026] The respiratory ratio coefficient and the blood oxygen ratio coefficient are compared, to obtain a coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient;

[0027] The coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient is compared with a preset difference threshold value;

[0028] When the coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient does not exceed the preset difference threshold value, a preset initial position offset threshold is retrieved as a trigger condition for triggering automatic adjustment of the focusing area;

[0029] When the coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient exceeds the preset difference threshold value, the initial position offset threshold is adjusted.

[0030] Further, when the coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient exceeds the preset difference threshold value, the initial position offset threshold is adjusted, including:

[0031] When the coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient exceeds the preset difference threshold value, the respiratory ratio coefficient and the blood oxygen ratio coefficient are retrieved;

[0032] The historical patient position offset data corresponding to the patient is retrieved;

[0033] acquire a position offset standard deviation and a time interval standard deviation of triggering automatic adjustment of the focus area by using the historical patient position offset data corresponding to the patient;

[0034] normalize the position offset standard deviation and the time interval standard deviation of triggering automatic adjustment of the focus area, and acquire the normalized position offset standard deviation and the time interval standard deviation of triggering automatic adjustment of the focus area;

[0035] adjust the initial position offset threshold by using the normalized position offset standard deviation and the time interval standard deviation of triggering automatic adjustment of the focus area in combination with a respiratory ratio coefficient and an oxygen saturation ratio coefficient, and acquire an adjusted position offset threshold.

[0036] Further, the data analysis module comprises:

[0037] The data processing module is configured to receive and pre-process the respiratory data, the oxygen saturation data, and the heart rate data from the monitoring and sensing module, including removing outliers and filling missing values, while converting the respiratory data, the oxygen saturation data, and the heart rate data into a unified data format, and temporarily storing the pre-processed respiratory data, the oxygen saturation data, and the heart rate data in a cache area.

[0038] The comparative analysis module is configured to store preset respiratory alert values, oxygen saturation alert values, and heart rate alert values, compare and analyze the collected respiratory data with the respiratory alert values, compare and analyze the collected oxygen saturation data with the oxygen saturation alert values, and compare and analyze the collected heart rate data with the heart rate alert values.

[0039] The abnormality judgment module is configured to judge whether the patient's respiration, oxygen saturation, and heart rate are abnormal according to the comparison and analysis results of the comparative analysis module.

[0040] When the respiratory data is higher or lower than the respiratory alert value range, the patient's respiration is abnormal.

[0041] When the oxygen saturation data is lower than the oxygen saturation alert value range, the patient's oxygen saturation is abnormal.

[0042] When the heart rate data is lower or lower than the heart rate alert range, the patient's heart rate is abnormal.

[0043] The dynamic adjustment module is configured to collect and store the patient's long-term historical monitoring data, analyze the change trend and law of the historical monitoring data, and dynamically adjust the alert values according to the analysis results.

[0044] Further, the dynamic adjustment module comprises:

[0045] The data collection module is configured to collect historical monitoring data of the patient for a long time, including time series records of respiratory rate, blood oxygen saturation data and heart rate, and corresponding abnormal event records, and store the collected historical monitoring data in time series;

[0046] The historical data analysis module is configured to determine the normal fluctuation range and change rule of the patient's respiratory, blood oxygen and heart rate data by calculating the mean, standard deviation and variance statistical indicators of the historical monitoring data, and on the other hand, use time series analysis technology to identify the periodicity, trend and abnormal fluctuation mode in the historical monitoring data, and determine the dynamic change trend of the patient's health status;

[0047] The warning value adjustment module is configured to dynamically adjust the respiratory warning value, blood oxygen warning value and heart rate warning value according to the patient health change trend and rule obtained by the patient feature analysis module, and simultaneously feed back the adjusted warning values to the comparative analysis module and the abnormality judgment module to update the judgment basis.

[0048] Further, the abnormality early warning module comprises:

[0049] The abnormality grading module is configured to grade the abnormal respiratory data, blood oxygen data and heart rate data according to the comparative analysis results of the data analysis module and the respiratory warning value, the blood oxygen warning value and the heart rate warning value, and determine different levels according to the severity of the abnormality, including three levels of mild, moderate and severe;

[0050] The patient feature analysis module is configured to collect and analyze the sleep habits and physiological characteristics of the patient, and extract the features of the patient's sleep habits and physiological characteristics through statistical analysis method, wherein the sleep habits are the sleep time, sleep posture, sleep cycle characteristics and easily disturbed time period of the patient, and the physiological characteristics are the age, gender, hearing condition and medical history of the patient;

[0051] The warning sound setting module is configured to set the warning sound parameters individually according to the abnormality grading results and the information provided by the patient feature analysis module, including the type, volume, frequency and duration of the sound;

[0052] The warning playing module is configured to play the corresponding warning sound through the loudspeaker on the device to wake up the patient according to the parameter configuration of the warning sound setting module after the data analysis module determines that the patient has an abnormality;

[0053] The synchronous notification module is configured to synchronize the monitored respiratory data, blood oxygen data and heart rate data and abnormal data to the remote medical platform in real time, so as to ensure that the medical personnel can obtain the patient data in time, and at the same time, send a notification message to the family members through a mobile application or a short message platform associated with the patient's family account, and inform the current situation of the patient.

[0054] Further, the abnormality grading module grades the abnormality into three levels of mild, moderate and severe according to the severity of the abnormality, specifically as follows:

[0055] Mild: the respiratory data is higher or lower than the respiratory warning value range by not more than 5%, and the duration is not more than 1 minute;

[0056] The blood oxygen data is lower than the blood oxygen warning value range by not more than 5%, and the duration is not more than 2 minutes;

[0057] The heart rate data is higher or lower than the heart rate warning value range by not more than 20%, and the duration is not more than 30 seconds;

[0058] Moderate: the respiratory data is higher or lower than the respiratory warning value range by not more than 10%, and the duration is between 1 and 3 minutes;

[0059] The blood oxygen data is lower than the blood oxygen warning value range by not more than 10%, and the duration is between 2 and 5 minutes;

[0060] The heart rate data is higher or lower than the heart rate warning value range by not more than 30%, and the duration is between 30 seconds and 2 minutes;

[0061] Severe: the respiratory data is higher or lower than the respiratory warning value range by more than 10%, and the duration is more than 3 minutes;

[0062] The blood oxygen data is lower than the blood oxygen warning value range by more than 30%, and the duration is more than 3 minutes;

[0063] The heart rate data is higher or lower than the heart rate warning value range by more than 20%, and the duration is not more than 30 seconds.

[0064] Further, the warning sound setting module sets the warning sound parameters individually according to the abnormality grading result and the information provided by the patient characteristic analysis module, specifically as follows:

[0065] Mild: the sound type is a soft prompt tone, the volume is 40-50 decibels, the frequency is to prompt once every 2-3 seconds, and the duration is to last for 1-2 seconds each time;

[0066] Moderate: the sound type is a clear alarm tone, the volume is 60-70 decibels, the frequency is to prompt once every 1-2 seconds, and the duration is to last for 3-5 seconds each time;

[0067] Severe: the sound type is a strong alarm tone, the volume is 80-90 decibels, the frequency is to prompt once every second, and the duration is to last for 5-10 seconds each time.

[0068] Compared with the prior art, the present application has the following advantages:

[0069] 1. This invention uses infrared sensing to accurately capture breathing, blood oxygen, and heart rate information during human sleep, thereby collecting the patient's breathing, blood oxygen, and heart rate data. By identifying erroneous data, the quality of data collection can be judged, thus ensuring the accuracy and reliability of the collected breathing, blood oxygen, and heart rate data, and effectively avoiding misjudgments caused by data quality issues.

[0070] 2. This invention compares and analyzes the collected respiratory data, blood oxygen data, and heart rate data with preset warning values ​​to determine whether the patient's breathing, blood oxygen, and heart rate are abnormal. The preset warning values ​​can be dynamically adjusted based on the patient's historical monitoring data, thereby improving the flexibility and accuracy of the device's monitoring. Once any of the respiratory rate, blood oxygen, or heart rate is abnormal, the abnormality is classified, and the classification results are combined with the patient's sleep habits and physiological characteristics to personalize the warning sound, thereby waking the patient with a specific warning sound and preventing adverse events. Attached Figure Description

[0071] Fig. 1 This is a schematic diagram of the overall structure of the non-contact sleep apnea monitoring and early warning device of the present invention.

[0072] Fig. 2 This is an external view of the non-contact sleep apnea monitoring and early warning device of the present invention. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0074] To address the limitations of existing monitoring methods for sleep apnea syndrome, traditional contact-based methods require attaching numerous sensors to the patient's body. This not only causes discomfort but also increases the difficulty and potential for errors due to complex wiring. Please refer to [link to relevant documentation]. Figs. 1-2 This embodiment provides the following technical solution:

[0075] A non-contact sleep apnea syndrome monitoring and early warning device, the device being installed above the pillow at the head of the bed, includes a monitoring and sensing module, a data analysis module, and an abnormal early warning module;

[0076] The monitoring and sensing module is configured to use the infrared sensing function of the infrared sensor to monitor the respiratory rate, blood oxygen, and heart rate of the individual in sleep in real time, collect the respiratory data, blood oxygen data, and heart rate data of the patient, determine the collection quality of the data, and locate the position of the patient in the pillow area to automatically adjust the focusing area of the infrared sensing, specifically as follows:

[0077] The historical respiratory data and blood oxygen data of the patient are called to obtain the standard deviation and the average value respectively, the corresponding ratio coefficient is calculated and compared with the preset difference threshold value, the comparison result is used to determine whether to call the initial position offset threshold value as a trigger condition or to adjust the initial position offset threshold value;

[0078] The data analysis module is configured to compare and analyze the collected respiratory data, blood oxygen data, and heart rate data with preset warning values respectively, determine whether the patient's respiration and blood oxygen appear abnormal, and dynamically adjust the warning values based on the historical monitoring data of the patient.

[0079] The abnormal warning module is configured to once any one of the respiratory rate, blood oxygen, or heart rate appears an abnormal condition, grade the abnormal condition, combine the abnormal grading result with the sleep habits and physiological characteristics of the patient, set the type, volume, frequency, and duration of the warning sound, and then issue a specific warning sound to wake up the patient, and synchronously transmit the abnormal grading result, abnormal data, and monitoring data to the remote medical platform and the patient family account associated with the device.

[0080] The technical effects of the above technical scheme are as follows: the device is arranged above the pillow at the head of the bed, so that contactless respiratory sleep syndrome monitoring can be achieved, discomfort of the patient is avoided, and the monitoring difficulty and error are reduced, and the operation is simple, without the need for complex professional operation skills, so that the patient can easily use and maintain the device, the compliance of the patient in using the device is improved, the monitoring and sensing module can use the infrared sensor to contactlessly monitor the respiratory frequency, blood oxygen, and heart rate of the sleeping individual in real time, the method does not interfere with the normal sleep state of the patient, the naturalness and reliability of data acquisition are ensured, the quality of the collected data can be determined, accurate and effective data are ensured for subsequent analysis, a solid data basis is provided for subsequent diagnosis and early warning, the data analysis module can timely determine whether the respiratory frequency, blood oxygen, and heart rate are abnormal by comparing and analyzing the collected data with the preset warning value, at the same time, the warning value is dynamically adjusted based on the historical monitoring data of the patient, individual differences and changes in the patient's own condition are fully considered, the early warning is more accurate and personalized, and the possibility of false positives and false negatives is reduced, once an abnormal condition occurs, the abnormal early warning module can grade the abnormality, and set a specific warning sound based on the sleep habits and physiological characteristics of the patient, the personalized warning can more effectively wake up the patient, timely intervene in the respiratory sleep syndrome related dangerous situation, thereby reducing the probability of health risks caused by respiratory pause or abnormal blood oxygen or heart rate, at the same time, the monitoring data and abnormal conditions are synchronously transmitted to a remote medical platform and an account associated with the family members of the patient in real time, so that the family members of the patient can timely understand the condition of the patient, the participation and trust of the patient and the family members in the monitoring process are enhanced, and the patient is more actively cooperated with the monitoring.

[0081] It should be noted that the contactless respiratory sleep syndrome monitoring and early warning device of the present application is not only suitable for professional medical institutions such as hospitals, but also can be used in home environment, thereby expanding the application range of the device, so that the patient can also receive timely and effective respiratory sleep syndrome monitoring and early warning at home.

[0082] The infrared sensing module is configured to emit an infrared light beam to the sleeping individual through the infrared sensor, capture the respiratory condition, blood oxygen information, and heart rate information of the individual during the sleep process, and collect the respiratory data, blood oxygen data, and heart rate data of the patient.

[0083] The data determination module is configured to identify the error data in the respiratory data, blood oxygen data, and heart rate data, extract the data amount information corresponding to the error data in the respiratory data, blood oxygen data, and heart rate data according to the identification result, evaluate the collection quality of the respiratory data, blood oxygen data, and heart rate data, and perform abnormal alarm when the collection quality of the respiratory data, blood oxygen data, and heart rate data is abnormal.

[0084] The intelligent positioning module is configured to use millimeter wave radar technology to locate the position of the patient in the pillow area in real time, and automatically adjust the focusing area of the infrared sensing module when the patient position deviates.

[0085] The technical effects of the above technical solutions are as follows: The infrared sensing module realizes non-contact monitoring, accurately captures the respiratory, blood oxygen information and heart rate information of the sleeping individual through the emission of infrared light beams, continuously and stably collects data, provides real-time and accurate data basis for subsequent analysis, avoids the inconvenience and interference that may be caused by contact monitoring, the data determination module can identify and extract error data in the respiratory data, blood oxygen data and heart rate data, thereby evaluating the data collection quality and ensuring the reliability and accuracy of the monitoring data, and timely alarm when the data quality is abnormal, facilitating the rapid discovery and solution of collection problems, ensuring the continuous stability of the monitoring process, improving the effectiveness of the monitoring results, and determining the quality of the collected data through the data determination module and automatically adjusting when the data quality is abnormal, ensuring the accuracy and reliability of the monitoring data, reducing false alarms and frequent monitoring interruptions caused by data quality problems, improving the trust of patients in the monitoring results, and thereby enhancing the compliance of patients, and the intelligent positioning module uses millimeter wave radar technology to locate the position of the patient in the pillow area in real time, and automatically adjusts the focusing area of the infrared sensing module when the patient position deviates, so as to accurately monitor the patient at all times, avoid monitoring errors caused by position changes, enhance the flexibility and adaptability of monitoring, improve the accuracy and reliability of monitoring, and reduce false alarms and missed alarms caused by monitoring area deviation.

[0086] The data determination module comprises:

[0087] The data identification module is configured to pre-analyze the characteristics of the existing normal respiratory data, blood oxygen data and heart rate data in the numerical range, fluctuation law and change trend, and establish a data identification model, input the collected respiratory data, blood oxygen data and heart rate data into the data identification model for comparison, and regard the data that does not conform to the data identification model as error data.

[0088] The data extraction module is configured to extract first data amount information corresponding to the error data in the respiratory data, extract second data amount information corresponding to the error data in the blood oxygen data, and extract third data amount information corresponding to the error data in the heart rate data.

[0089] The abnormality determination module is configured to compare the first data amount information, the second data amount information and the third data amount information with the preset parameter threshold, specifically:

[0090] If the first data amount information is lower than the preset parameter threshold, it is determined that the collection quality of the respiratory data is abnormal, and an abnormal alarm is performed.

[0091] If the second data quantity information is lower than the preset parameter threshold, it is determined that the collection quality of the blood oxygen data is abnormal, and an abnormal alarm is performed.

[0092] If the third data quantity information is lower than the preset parameter threshold, it is determined that the collection quality of the heart rate data is abnormal, and an abnormal alarm is performed.

[0093] The technical effects of the above technical solutions are: the data recognition module identifies the error data through the established data recognition model, providing a scientific basis and standard for subsequent accurate screening of error data. Based on the identification result of the error data, the data extraction module extracts the data quantity information corresponding to the error data in the respiratory data, blood oxygen data and heart rate data, respectively, providing a quantitative basis for data quality evaluation. By accurately extracting these information, the abnormal situation occurring in the data collection process can be more accurately judged. When the data collection quality is found to be abnormal, the abnormal alarm can be performed in time, which helps to quickly find and solve the problems in data collection, ensuring the continuity and reliability of the monitoring process. Through the above functions, the data recognition module can effectively identify and process error data, ensuring the accuracy and integrity of the monitoring data, and improving the reliability and effectiveness of the entire device monitoring.

[0094] Specifically, the position offset threshold corresponding to the patient position offset for triggering automatic adjustment of the focusing area is set, including:

[0095] Retrieving historical respiratory data and historical blood oxygen data corresponding to the patient;

[0096] Obtaining the respiratory data standard deviation and the respiratory data average value corresponding to the patient according to the historical respiratory data corresponding to the patient;

[0097] Obtaining the blood oxygen data standard deviation and the blood oxygen data average value corresponding to the patient according to the historical blood oxygen data corresponding to the patient;

[0098] Processing the respiratory data standard deviation and the respiratory data average value corresponding to the patient by ratio, to obtain a respiratory ratio coefficient;

[0099] Processing the blood oxygen data standard deviation and the blood oxygen data average value corresponding to the patient by ratio, to obtain a blood oxygen ratio coefficient;

[0100] Comparing the respiratory ratio coefficient and the blood oxygen ratio coefficient to obtain a coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient;

[0101] Comparing the coefficient difference value between the respiratory ratio coefficient and the blood oxygen ratio coefficient with a preset difference value threshold;

[0102] When the coefficient difference between the respiration ratio coefficient and the blood oxygen ratio coefficient does not exceed the preset difference threshold, a preset initial position offset threshold is called as a trigger condition for triggering automatic adjustment of the focusing area.

[0103] When the coefficient difference between the respiration ratio coefficient and the blood oxygen ratio coefficient exceeds the preset difference threshold, the initial position offset threshold is adjusted.

[0104] The technical effects of the above technical solutions are as follows: The historical respiration data and blood oxygen data of a patient are called, and the standard deviation and average value of the respiration data and the standard deviation and average value of the blood oxygen data are calculated. These statistical values reflect the range and reference state of the fluctuations of the patient's respiration and blood oxygen. The standard deviation and average value of the respiration data are processed by ratio to obtain a respiration ratio coefficient, and the standard deviation and average value of the blood oxygen data are processed by ratio to obtain a blood oxygen ratio coefficient. The two ratio coefficients eliminate the influence of the absolute value difference of different patient data and normalize the fluctuation degree of respiration and blood oxygen. The coefficient difference between the respiration ratio coefficient and the blood oxygen ratio coefficient is calculated, and the difference is used to judge the coordination of the fluctuations of respiration and blood oxygen. The coefficient difference is compared with a preset difference threshold. If the difference does not exceed the difference threshold, it indicates that the physiological state of the patient is stable, and a preset initial position offset threshold is called as a trigger condition. If the difference exceeds the difference threshold, it indicates that the patient may have abnormal respiration or blood oxygen, and the initial position offset threshold needs to be adjusted dynamically according to the difference size and the relative size of the two ratio coefficients to adapt to abnormal fluctuations.

[0105] By analyzing the historical respiration and blood oxygen data of a patient, calculating the ratio coefficients and comparing the differences, the position offset threshold can be adjusted according to the individual physiological characteristics of the patient, so that the trigger condition for automatically adjusting the focusing area is more in line with the actual situation, the focusing deviation is reduced, and the accuracy of the adjustment is improved. Combined with the comprehensive judgment of respiration and blood oxygen multidimensional physiological data, the influence of single data noise on threshold setting is reduced, accidental physiological fluctuations are filtered, threshold adjustment is avoided, complex physiological scenarios are adapted to, and the ability of the system to cope with abnormal situations is improved. Based on historical data, the threshold is dynamically generated, human intervention is reduced, threshold setting time is shortened, and the efficiency of the diagnosis and treatment process is improved. The subjective error of manually setting the threshold is reduced, the trigger condition for adjusting the focusing area is ensured to be in line with the physiological regularity of the patient, and the reliability of treatment or examination is enhanced.

[0106] Specifically, when the coefficient difference between the respiration ratio coefficient and the blood oxygen ratio coefficient exceeds the preset difference threshold, the initial position offset threshold is adjusted, including:

[0107] When the coefficient difference between the respiration ratio coefficient and the blood oxygen ratio coefficient exceeds the preset difference threshold, the respiration ratio coefficient and the blood oxygen ratio coefficient are called.

[0108] The historical patient position offset data corresponding to the patient is called.

[0109] acquire a position offset standard deviation and a time interval standard deviation of triggering automatic adjustment of the focus area by using the historical patient position offset data corresponding to the patient;

[0110] normalize the position offset standard deviation and the time interval standard deviation of triggering automatic adjustment of the focus area, to obtain a normalized position offset standard deviation and a normalized time interval standard deviation of triggering automatic adjustment of the focus area;

[0111] adjust an initial position offset threshold by using the normalized position offset standard deviation and the normalized time interval standard deviation of triggering automatic adjustment of the focus area in combination with a respiration ratio coefficient and an oxygen saturation ratio coefficient, to obtain an adjusted position offset threshold.

[0112] The adjusted position offset threshold is obtained by the following formula:

[0113]

[0114] Wherein, S represents the adjusted position offset threshold; S0 represents the position offset threshold before adjustment; B x represents the oxygen saturation ratio coefficient; B h represents the respiration ratio coefficient; f represents a preset adjustment coefficient, and the value range of the adjustment coefficient is 0.24-0.58; L b represents the normalized position offset standard deviation; t b represents the normalized time interval standard deviation of triggering automatic adjustment of the focus area. Specifically, The oxygen saturation ratio coefficient B x and the respiration ratio coefficient B h respectively reflect the degree of fluctuation of oxygen saturation and respiration data relative to the average value. The maximum value of the two is taken in order to determine the dominant factor in physiological data fluctuation. If B x is greater, it means that the oxygen saturation data fluctuation is more significant; if B h is greater, it means that the respiration data fluctuation is more prominent. In this way, the physiological fluctuation factor that has a greater impact on the position offset threshold can be grasped. The normalized position offset standard deviation L b embodies the dispersion degree of the historical position offset amplitude of the patient, and the normalized time interval standard deviation t b reflects the dispersion degree of the position offset adjustment time interval. Multiplying the two and taking the square root comprehensively considers the amplitude and frequency characteristics of the position offset. When the position offset amplitude is large and the adjustment frequency is high, the value will be larger, which means that the patient's position change is complex; otherwise, it means that the position change is relatively stable.

[0115] The preset adjustment coefficient f is used to control the influence degree of physiological fluctuation factors (reflected by max(B x ,B h ) on the adjustment of the position offset threshold. The denominator plays a restraining role. When the position offset amplitude and frequency characteristics (reflected by ) are small, the value of the fraction is mainly determined by , that is, the influence of physiological fluctuation factors on the threshold is emphasized more; when the position offset amplitude and frequency characteristics are large, the denominator increases, which weakens the influence of physiological fluctuation factors, so that the threshold adjustment considers the situation of the position offset itself more. Then, the direction and degree of adjustment are determined. When the value of the fraction is small, 1 minus the value is close to 1, which indicates that the adjusted position offset threshold S is close to the threshold S0 before adjustment, that is, the threshold does not change much; when the value of the fraction is large, 1 minus the value is small, which means that the adjusted threshold S is significantly smaller than the threshold S0 before adjustment, that is, the threshold will be adjusted greatly according to the physiological fluctuation and the position offset situation.

[0116] The technical effects corresponding to the above technical solutions are: by introducing historical position offset data (such as position offset standard deviation, adjustment time interval standard deviation), combined with the difference in the ratio coefficient of respiration and blood oxygen, the threshold adjustment is more in line with the real-time physiological fluctuation mode of the individual patient. For example, if the patient's recent position offset amplitude is large and the adjustment is frequent (i.e., the position offset standard deviation and the time interval standard deviation are large), the system can dynamically amplify or reduce the threshold according to the degree of discordance of respiratory / blood oxygen fluctuation (coefficient difference), avoiding the lag or excessive sensitivity of the focus area adjustment caused by the fixed threshold. Improve the response capability of the system to the dynamic changes of the patient's physiological state, especially suitable for complex scenarios where respiration and blood oxygen fluctuate asynchronously (such as patients with lung disease who have rapid respiration but relatively stable blood oxygen saturation), reducing false triggering or missed triggering caused by single physiological indicator fluctuation. Normalize the position offset standard deviation and the adjustment time interval standard deviation to eliminate the interference of data with different dimensions (such as position offset unit in millimeters, time interval unit in seconds), so that multi-dimensional data can directly participate in threshold calculation. Combined with the respiration / blood oxygen ratio coefficient, a dynamic threshold model is formed by multi-parameter fusion. By quantifying the correlation between physiological fluctuations (respiration / blood oxygen) and position offset, the threshold adjustment step and direction are more accurate, avoiding the roughness of adjusting the threshold based on experience or fixed rules. For example, if the respiration ratio coefficient is significantly greater than the blood oxygen ratio coefficient (respiration fluctuation dominates), and the historical position offset standard deviation is large, the system can specifically reduce the threshold to enhance the sensitivity to respiratory-related displacement. Instead of relying on single physiological fluctuation data, adjusting the threshold based on historical data statistical characteristics (such as standard deviation) can filter accidental noise (such as transient position offset caused by sudden cough of the patient), avoiding frequent threshold oscillation. When the patient's physiological indicators are abnormal for a long time (such as chronic respiratory insufficiency), the threshold is optimized through long-term statistical rules of historical data, reducing system misjudgment caused by short-term data fluctuation, improving the stability of the focus area during treatment or examination, and reducing the medical risks caused by unreasonable threshold (such as radiotherapy target area offset). Without human intervention, the threshold iteration can be automatically completed according to the individual data of the patient (respiration, blood oxygen, position offset history), and through multi-dimensional data cross-validation (physiological fluctuation + position offset behavior), single indicator misjudgment is avoided. While improving the efficiency of automation, the personalized solution of "one person one threshold" is realized, especially suitable for elderly patients, critically ill patients and other groups with large individual differences in physiological indicators, reducing the problem of insufficient adaptability caused by universal threshold.

[0117] The data analysis module comprises:

[0118] The data processing module is configured to receive and preprocess respiration data, blood oxygen data, and heart rate data from the monitoring and sensing module, including removing outliers and filling missing values, converting the respiration data, blood oxygen data, and heart rate data into a unified data format, and temporarily storing the preprocessed respiration data, blood oxygen data, and heart rate data in a cache area;

[0119] a comparison analysis module configured to store preset respiratory alert values, blood oxygen alert values and heart rate alert values, compare the collected respiratory data with the respiratory alert values, compare the collected blood oxygen data with the blood oxygen alert values, and compare the collected heart rate data with the heart rate alert values;

[0120] an abnormality judgment module configured to judge whether the patient's respiration, blood oxygen and heart rate are abnormal according to the comparison analysis result of the comparison analysis module;

[0121] when the respiratory data is higher or lower than the respiratory alert value range, the patient's respiration is abnormal;

[0122] when the blood oxygen data is lower than the blood oxygen alert value range, the patient's blood oxygen is abnormal;

[0123] when the heart rate data is lower or lower than the heart rate alert range, the patient's heart rate is abnormal;

[0124] a dynamic adjustment module configured to collect and store the patient's long-term historical monitoring data, and dynamically adjust the alert values according to the analysis result by analyzing the change trend and rule of the historical monitoring data.

[0125] The technical effects of the above technical solutions are: after the data processing module receives the monitoring data, the abnormal values are removed, the missing values are filled, and the data format is unified and temporarily stored in the cache area, so as to ensure the integrity, accuracy and consistency of the data, lay a solid foundation for subsequent analysis and processing, the comparison analysis module compares the preprocessed data with the preset alert values, can timely find the abnormal situation of the data, so that the abnormality judgment module can judge whether the patient's respiration, blood oxygen and heart rate are abnormal, through timely comparison and judgment, the potential health risk can be quickly identified, early detection and early intervention are achieved, and the dynamic adjustment module collects the patient's long-term historical monitoring data, analyzes the change trend and rule, and dynamically adjusts the alert values, fully considers the individual differences and changes of the patient's own situation, makes the monitoring more targeted and adaptive, and improves the accuracy and effectiveness of the early warning.

[0126] The dynamic adjustment module comprises:

[0127] a data collection module configured to collect the patient's historical monitoring data for a long time, including time sequence records of respiratory frequency, blood oxygen saturation data and heart rate, and corresponding abnormal event records, and store the collected historical monitoring data according to the time sequence;

[0128] The historical data analysis module is configured to determine the normal fluctuation range and change rule of the patient's respiratory, blood oxygen and heart rate data by calculating the mean, standard deviation and variance statistical indicators of the historical monitoring data, and on the other hand, to identify the periodicity, trend and abnormal fluctuation mode in the historical monitoring data by using time series analysis technology, and to determine the dynamic change trend of the patient's health status;

[0129] The warning value adjustment module is configured to dynamically adjust the respiratory warning value, the blood oxygen warning value and the heart rate warning value according to the patient health change trend and rule obtained by the patient feature analysis module, and to feed back the adjusted warning values to the comparative analysis module and the abnormality judgment module to update the judgment basis.

[0130] The technical effects of the above technical solutions are: the data collection module collects the patient's historical monitoring data for a long time, providing detailed data basis for comprehensively understanding the patient's health status, which helps to further analyze the patient's health change trend, the historical data analysis module determines the normal fluctuation range and change rule of the patient's respiratory, blood oxygen and heart rate data by calculating the mean, standard deviation and variance statistical indicators, providing a scientific basis for accurately judging abnormal conditions, and further using time series analysis technology to identify the periodicity, trend and abnormal fluctuation mode in the historical data, which can more accurately grasp the dynamic change trend of the patient's health status, thereby providing a reliable basis for the dynamic adjustment of the warning value, effectively improving the matching degree of the monitoring data and the patient's actual health status, and improving the overall monitoring accuracy, and the warning value adjustment module adjusts the respiratory warning value, the blood oxygen warning value and the heart rate warning value according to the patient health change trend and rule, so that the warning value can be flexibly changed with the change of the patient's health status, better adapting to the individual situation of the patient and the health needs at different stages, avoiding the limitations that may be brought by fixed warning values, improving the timeliness and effectiveness of the warning, providing more accurate early warning for the patient's health, facilitating to take corresponding measures in advance to cope with potential health risks, and feeding back the adjusted warning values to the related modules to update the judgment basis, realizing the dynamic optimization of monitoring and early warning, improving the accuracy and timeliness of the early warning, and reducing the possibility of false positives and false negatives.

[0131] The abnormality early warning module comprises:

[0132] The abnormality grading module is configured to grade the abnormal respiratory data, blood oxygen data and heart rate data according to the comparative analysis results of the data analysis module with the respiratory warning value, the blood oxygen warning value and the heart rate warning value, and to determine different levels according to the severity of the abnormality, including three levels of mild, moderate and severe;

[0133] patient characteristic analysis module configured to collect and analyze the patient's sleep habits and physiological characteristics, and extract the characteristics of the patient's sleep habits and physiological characteristics through statistical analysis, wherein the sleep habits are the patient's sleep time, sleep posture, sleep cycle characteristics and easily disturbed time period, and the physiological characteristics are the patient's age, gender, hearing condition and medical history;

[0134] alert sound setting module configured to set the alert sound parameters, including the type, volume, frequency and duration of the sound, according to the abnormal grading results and the information provided by the patient characteristic analysis module;

[0135] alert playing module configured to play the corresponding alert sound through the speaker on the device to wake up the patient according to the parameter configuration of the alert sound setting module after the data analysis module determines that the patient has a situation;

[0136] synchronization notification module configured to synchronize the monitored respiratory data, blood oxygen data and heart rate data, as well as abnormal data, to a remote medical platform in real time, to ensure that medical personnel can obtain the patient's data in a timely manner, and at the same time, send a notification message to the patient's family members through a mobile application or a short message platform associated with the patient's family account, to inform the patient's current situation.

[0137] The technical effects of the above technical solutions are: the abnormal grading module classifies the abnormal respiratory data, blood oxygen data and heart rate data into mild, moderate and severe levels according to the comparison analysis results, which can accurately identify the severity of the patient's abnormal situation, making the subsequent warning measures more targeted, avoiding the ambiguity of the warning information leading to inaccurate judgment of the abnormal situation by the patient and medical personnel, improving the timeliness and effectiveness of dealing with the abnormality, the patient characteristic analysis module collects the patient's sleep habits and physiological characteristics information, provides personalized basis for the alert sound setting module, so that it can set appropriate sound type, volume, frequency and duration parameters according to the patient's condition, such as considering the patient's sleep depth, hearing condition, etc., so that the alert sound is more easily detected by the patient and reduces the excessive disturbance to the patient, improving the patient's acceptance and cooperation of the warning, and the synchronization notification module synchronizes the monitoring data and abnormal data to the remote medical platform in real time, ensuring that medical personnel can follow up the patient's condition in a timely manner, and at the same time, sends a notification to the patient's family members through a mobile application or a short message platform associated with the patient's family account, so that the family members can understand the patient's condition in a timely manner, and facilitate the family members to assist in care or take the patient to the hospital, etc., realizing the cooperation of the patient, medical personnel and family members in dealing with the patient's health problem, and improving the efficiency and effectiveness of dealing with sudden health conditions.

[0138] abnormal grading module, which is divided into three levels of mild, moderate and severe according to the severity of the abnormality, as follows:

[0139] mild: the respiratory data is higher or lower than the respiratory alert value range by no more than 5%, and the duration is no more than 1 minute;

[0140] The blood oxygen data is less than the blood oxygen warning value range by not more than 5%, and the duration is not more than 2 minutes;

[0141] The heart rate data is higher or lower than the heart rate warning value range by not more than 20%, and the duration is not more than 30 seconds;

[0142] Moderate: The respiratory data is higher or lower than the respiratory warning value range by not more than 10%, and the duration is between 1 and 3 minutes;

[0143] The blood oxygen data is less than the blood oxygen warning value range by not more than 10%, and the duration is between 2 and 5 minutes;

[0144] The heart rate data is higher or lower than the heart rate warning value range by not more than 30%, and the duration is between 30 seconds and 2 minutes;

[0145] Severe: The respiratory data is higher or lower than the respiratory warning value range by more than 10%, and the duration is more than 3 minutes;

[0146] The blood oxygen data is less than the blood oxygen warning value range by more than 30%, and the duration is more than 3 minutes;

[0147] The heart rate data is higher or lower than the heart rate warning value range by more than 20%, and the duration is not more than 30 seconds.

[0148] The technical effects of the above technical solutions are: by refining the abnormal grading standard, the abnormal conditions of respiratory, blood oxygen and heart rate data are accurately graded according to the percentage of deviation from the warning value and the duration, which can more accurately reflect the severity of the patient's abnormal condition, avoid the possible misjudgment or omission of the past single threshold judgment, make the monitoring result closer to the actual health status of the patient, and divide the abnormality into three levels of mild, moderate and severe, so that medical staff or patients themselves can quickly and intuitively understand the emergency degree of the abnormal condition, and take corresponding measures, for example: for mild abnormality, observation first, for moderate abnormality, timely attention, and for severe abnormality, immediate treatment, thereby improving the effectiveness of dealing with abnormality and better protecting the safety of patients.

[0149] The warning sound setting module sets the warning sound parameters individually according to the abnormal grading results and the information provided by the patient characteristic analysis module, specifically as follows:

[0150] Mild: The sound type is a soft prompt tone, the volume is 40-50 decibels, the frequency is 2-3 seconds per prompt, and the duration is 1-2 seconds per prompt;

[0151] Moderate: The sound type is a clear alarm tone, the volume is 60-70 decibels, the frequency is 1-2 seconds per prompt, and the duration is 3-5 seconds per prompt;

[0152] Severe: the sound type is an intense alarm sound, the volume is 80-90 decibels, the frequency is prompted once per second, and the duration is 5-10 seconds per prompt.

[0153] The technical effect of the above technical solution is that the parameters of the warning sound can be configured individually by combining the abnormal grading results with the information provided by the patient characteristic analysis module, so that the warning sound can effectively wake up the patient without excessively affecting the sleep quality of the patient, and especially for patients with poor hearing or deep sleep, more appropriate parameters can be set to ensure that the patient can perceive the warning.

[0154] Working principle: The respiratory frequency, blood oxygen and heart rate of the sleeping individual are monitored in real time by using infrared sensing function without contact, so as to collect the respiratory data, blood oxygen data and heart rate data of the patient, and determine the quality of data collection, so as to ensure the reliability and accuracy of the monitoring data. By comparing and analyzing the collected data with the preset warning value, it can be judged in time whether the respiration, blood oxygen and heart rate are abnormal. At the same time, the warning value is dynamically adjusted based on the historical monitoring data of the patient, which fully considers the individual differences and changes of the patient's own condition, so that the early warning is more accurate and individualized, and the possibility of false alarm and missed alarm is reduced. Once an abnormal situation occurs, the abnormal early warning module can grade the abnormality, and set a specific warning sound according to the patient's sleep habits and physiological characteristics. The individualized warning can more effectively wake up the patient, so that the warning sound is more easily perceived by the patient and the excessive disturbance to the patient is reduced.

[0155] It should be noted that the relationship terms such as first and second in this text are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0156] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application.

Claims

1. A non-contact monitoring and early warning device for sleep apnea syndrome, characterized in that, The device is installed above the pillow at the head of the bed and includes a monitoring and sensing module, a data analysis module, and an anomaly warning module. The monitoring and sensing module is configured to use the infrared sensing function of an infrared sensor to monitor the respiratory rate, blood oxygen, and heart rate of an individual during sleep in real time. It collects the patient's respiratory data, blood oxygen data, and heart rate data, assesses the quality of the data collection, and simultaneously locates the patient's position in the pillow area, automatically adjusting the focusing area of ​​the infrared sensor. Specifically: The patient's historical respiratory and blood oxygen data are retrieved to obtain the standard deviation and mean, the corresponding ratio coefficient is calculated and compared with the preset difference threshold, and the initial position offset threshold is retrieved as the trigger condition or the initial position offset threshold is adjusted based on the comparison result. The position offset threshold used to trigger automatic adjustment of the focusing area is set, including: Retrieve the patient's historical respiratory and blood oxygenation data; Based on the patient's historical respiratory data, obtain the standard deviation and average respiratory data of the patient; Based on the patient's historical blood oxygen data, obtain the standard deviation and average blood oxygen data for the patient. The respiratory ratio coefficient is obtained by comparing the standard deviation of the respiratory data corresponding to the patient with the mean of the respiratory data. The blood oxygenation ratio coefficient is obtained by comparing the standard deviation of the blood oxygenation data corresponding to the patient with the average blood oxygenation data. The respiratory ratio coefficient and the blood oxygen ratio coefficient are compared to obtain the coefficient difference between the respiratory ratio coefficient and the blood oxygen ratio coefficient; The difference between the respiratory ratio coefficient and the blood oxygen ratio coefficient is compared with a preset difference threshold. If the difference between the respiratory ratio coefficient and the blood oxygen ratio coefficient does not exceed the preset difference threshold, then the preset initial position offset threshold is retrieved as the trigger condition for automatically adjusting the focusing area. When the difference between the respiration ratio coefficient and the blood oxygen ratio coefficient exceeds a preset difference threshold, the initial position offset threshold is adjusted. Wherein, when the difference between the respiration ratio coefficient and the blood oxygen ratio coefficient exceeds a preset difference threshold, the initial position offset threshold is adjusted, including: When the difference between the respiratory ratio coefficient and the blood oxygen ratio coefficient exceeds a preset difference threshold, the respiratory ratio coefficient and the blood oxygen ratio coefficient are retrieved. Retrieve the historical patient location offset data corresponding to the patient; The standard deviation of the positional offset and the standard deviation of the time interval for triggering automatic adjustment of the focusing area are obtained using the historical patient positional offset data corresponding to the patient. The standard deviation of the position offset and the standard deviation of the time interval for triggering automatic adjustment of the focus area are normalized to obtain the normalized standard deviation of the position offset and the standard deviation of the time interval for triggering automatic adjustment of the focus area. The initial position offset threshold is adjusted by combining the normalized position offset standard deviation and the standard deviation of the time interval for triggering automatic adjustment of the focusing area with the respiration ratio coefficient and the blood oxygen ratio coefficient, thus obtaining the adjusted position offset threshold. The adjusted position offset threshold is obtained using the following formula: Where S represents the adjusted position offset threshold; S0 represents the original position offset threshold; B x Indicates the blood oxygen ratio coefficient; B h represents the respiratory ratio coefficient; f represents the preset adjustment coefficient, and the value range of the adjustment coefficient is 0.24~0.58; L b t represents the standard deviation of the positional offset after normalization. b This represents the standard deviation of the time interval for triggering automatic focus adjustment after normalization. The data analysis module is configured to compare and analyze the collected respiratory data, blood oxygen data, and heart rate data with preset warning values ​​to determine whether the patient's breathing and blood oxygen are abnormal. At the same time, it dynamically adjusts the warning values ​​based on the patient's historical monitoring data. The abnormality warning module is configured to classify any abnormality in respiratory rate, blood oxygen, or heart rate, combine the abnormality classification results with the patient's sleep habits and physiological characteristics, set the type, volume, frequency, and duration of the warning sound, and then issue a specific warning sound to wake the patient. At the same time, the abnormality classification results, abnormal data, and monitoring data are synchronized to the remote medical platform and the patient's family account associated with the device.

2. The non-contact sleep apnea monitoring and early warning device according to claim 1, characterized in that: The monitoring and sensing module includes: The infrared sensing module is configured to emit an infrared beam to the sleeping individual through an infrared sensor to capture the individual's breathing, blood oxygen information, and heart rate information during sleep, and to collect the patient's respiratory data, blood oxygen data, and heart rate data. The data judgment module is configured to identify erroneous data in respiratory data, blood oxygen data, and heart rate data. Based on the identification results, it extracts the data volume information corresponding to the erroneous data in respiratory data, blood oxygen data, and heart rate data, thereby evaluating the acquisition quality of respiratory data, blood oxygen data, and heart rate data, and issuing an abnormal alarm when the acquisition quality of respiratory data, blood oxygen data, and heart rate data is abnormal. The intelligent positioning module is configured to use millimeter-wave radar technology to locate the patient's position in the pillow area in real time. When the patient's position shifts, the focusing area of ​​the infrared sensing module is automatically adjusted.

3. The non-contact sleep apnea monitoring and early warning device according to claim 2, characterized in that: The data determination module includes: The data recognition module is configured to pre-analyze the characteristics of existing normal respiratory data, blood oxygen data, and heart rate data in terms of numerical range, fluctuation patterns, and trends, and establish a data recognition model. The collected respiratory data, blood oxygen data, and heart rate data are input into the data recognition model for comparison, and data that does not conform to the data recognition model is regarded as erroneous data. The data extraction module is configured to extract the first data volume information corresponding to the erroneous data in the respiratory data, the second data volume information corresponding to the erroneous data in the blood oxygen data, and the third data volume information corresponding to the erroneous data in the heart rate data. The anomaly detection module is configured to compare the first data volume information, the second data volume information, and the third data volume information with preset parameter thresholds, specifically: If the first data volume is lower than the preset parameter threshold, it is determined that the quality of the respiratory data acquisition is abnormal and an abnormal alarm is triggered. If the second data volume is lower than the preset parameter threshold, it is determined that the blood oxygen data acquisition quality is abnormal and an abnormal alarm is triggered. If the third data volume is lower than the preset parameter threshold, it is determined that the heart rate data acquisition quality is abnormal and an abnormal alarm is triggered.

4. The non-contact sleep apnea monitoring and early warning device according to claim 1, characterized in that: The data analysis module includes: The data processing module is configured to receive respiratory data, blood oxygen data, and heart rate data from the monitoring and sensing module and perform preprocessing, including removing outliers and filling missing values. At the same time, it converts the respiratory data, blood oxygen data, and heart rate data into a unified data format and temporarily stores the preprocessed respiratory data, blood oxygen data, and heart rate data in a cache area. The comparison and analysis module is configured to store preset respiratory warning values, blood oxygen warning values, and heart rate warning values, and compare and analyze the collected respiratory data with the respiratory warning values, the collected blood oxygen data with the blood oxygen warning values, and the collected heart rate data with the heart rate warning values. The anomaly detection module is configured to determine whether the patient's respiration, blood oxygenation, and heart rate are abnormal based on the comparison analysis results of the comparison analysis module. When respiratory data are higher or lower than the respiratory warning value range, the patient's breathing is abnormal; When blood oxygen data is below the blood oxygen warning value range, the patient's blood oxygen is abnormal; When the heart rate data is below or below the heart rate warning range, the patient's heart rate is abnormal; The dynamic adjustment module is configured to collect and store long-term historical monitoring data of patients, and dynamically adjust the warning value based on the analysis results by analyzing the changing trends and patterns of historical monitoring data.

5. The non-contact sleep apnea monitoring and early warning device according to claim 4, characterized in that: The dynamic adjustment module includes: The data collection module is configured to collect patients' historical monitoring data over a long period of time, including time-series records of respiratory rate, blood oxygen saturation data and heart rate, as well as corresponding abnormal event records, and store the collected historical monitoring data in time series. The historical data analysis module is configured to determine the normal fluctuation range and change pattern of the patient's respiratory, blood oxygen and heart rate data by calculating the mean, standard deviation and variance statistical indicators of historical monitoring data. On the other hand, it uses time series analysis technology to identify periodic, trend and abnormal fluctuation patterns in historical monitoring data to determine the dynamic change trend of the patient's health status. The warning value adjustment module is configured to dynamically adjust the respiratory warning value, blood oxygen warning value, and heart rate warning value based on the patient health change trends and patterns obtained from the patient characteristic analysis module. At the same time, the adjusted warning values ​​are fed back to the comparison analysis module and the anomaly judgment module to update the judgment criteria.

6. The non-contact sleep apnea monitoring and early warning device according to claim 1, characterized in that: The anomaly warning module includes: The abnormality grading module is configured to grade abnormal respiratory data, blood oxygen data, and heart rate data based on the comparison and analysis results of the data analysis module with respiratory warning values, blood oxygen warning values, and heart rate warning values, and to determine different levels according to the severity of the abnormality, namely mild, moderate, and severe. The patient characteristic analysis module is configured to collect and analyze patients' sleep habits and physiological characteristics. Through statistical analysis methods, it extracts the features of patients' sleep habits and physiological characteristics. Among them, sleep habits include patients' sleep time, sleep posture, sleep cycle characteristics and easily disturbed time periods, and physiological characteristics include patients' age, gender, hearing status and past medical history. The alert sound setting module can be configured to personalize alert sound parameters, including sound type, volume, frequency, and duration, based on information provided by the abnormality grading results and patient characteristic analysis module. The alarm playback module is configured to wake up the patient by playing a corresponding alarm sound through the speaker on the device, based on the parameter configuration of the alarm sound setting module, after the data analysis module determines that the patient has a condition. The synchronous notification module is configured to synchronize the monitored respiratory data, blood oxygen data, heart rate data, and abnormal data to the remote medical platform in real time, ensuring that medical personnel can obtain patient data in a timely manner. At the same time, it can send notification messages to family members through a mobile application or SMS platform associated with the patient's family member's account to inform them of the patient's current condition.

7. The non-contact sleep apnea monitoring and early warning device according to claim 6, characterized in that: The anomaly grading module classifies anomalies into three levels: mild, moderate, and severe, based on their severity, as follows: Mild: Respiratory data is no more than 5% above or below the respiratory warning value, and the duration is no more than 1 minute; Blood oxygen data should be no more than 5% below the blood oxygen warning value, and the duration should not exceed 2 minutes; Heart rate data should be no more than 20% above or below the heart rate warning value, and the duration should not exceed 30 seconds. Moderate: Respiratory data are not more than 10% above or below the respiratory warning value, and the duration is between 1 and 3 minutes; Blood oxygen data is no more than 10% below the blood oxygen warning value, and the duration is between 2 and 5 minutes; Heart rate data should be within 30% of the heart rate warning value, and the duration should be between 30 seconds and 2 minutes. Severe: Respiratory data are more than 10% higher or lower than the respiratory warning value range, and the duration exceeds 3 minutes; Blood oxygen data is more than 30% below the blood oxygen warning value and lasts for more than 3 minutes; Heart rate data that is more than 20% higher or lower than the heart rate warning value, and lasts for no more than 30 seconds.

8. The non-contact sleep apnea monitoring and early warning device according to claim 6, characterized in that: The warning sound setting module, based on the abnormality grading results and information provided by the patient characteristic analysis module, personalizes the warning sound parameters as follows: Mild: The sound type is a gentle prompt tone, the volume is 40~50 decibels, the frequency is once every 2~3 seconds, and the duration is 1~2 seconds for each prompt; Moderate: The sound type is a clear alarm tone, the volume is 60~70 decibels, the frequency is once every 1~2 seconds, and the duration is 3~5 seconds for each alarm. Severe: The sound type is a loud alarm sound, the volume is 80~90 decibels, the frequency is once per second, and the duration is 5~10 seconds per alarm.

Images