A suffocation prevention alarm system

By combining facial features, physiological features and sound information, a detailed assessment and accurate warning of infant suffocation risk is achieved, and the problems of frequent early warnings and misjudgments in the existing technology are solved, which improves the efficiency and accuracy of infant safety monitoring. sex.

CN118570964BActive Publication Date: 2024-12-03PEKING UNIVERSITY THIRD HOSPITAL (THE THIRD CLINICAL MEDICAL SCHOOL OF PEKING UNIVERSITY)
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411053299.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2024-02-08
Filing Date
2024-08-02
Publication Date
2024-12-03
Estimated Expiration
2044-08-02

AI Technical Summary

Technical Problem

The prior art cannot accurately distinguish whether abnormal vital signs of infants are caused by environmental stimulation, resulting in frequent early warning information increasing the anxiety of the guardian, and the inability to distinguish the types of early warning information based on the detection results, which may lead to the guardian's misjudgment and prolong the rescue time for infant suffocation.

Method used

An alarm system for preventing suffocation is adopted. By obtaining facial feature information, physiological feature information and sound information of the target object, combining these data for joint judgment, and outputting different types of early warning instructions to improve the accuracy and early warning information of respiratory status assessment. the degree of precision.

Benefits of technology

The system can provide accurate early warning information based on the specific situation of the baby, reduce the guardian's reaction time, reduce the risk of suffocation caused by misjudgment, and alleviate the guardian's anxiety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118570964B_ABST
    Figure CN118570964B_ABST
Patent Text Reader

Abstract

The present invention relates to an alarm system for preventing suffocation, comprising: a first monitoring module for acquiring facial feature information of a target object; a second monitoring module for acquiring physiological feature information of the target object; a third monitoring module for acquiring sound information, and a processing module, wherein the processing module is respectively connected to the first monitoring module, the second monitoring module and the third monitoring module by signal, and the processing module is configured to: output different types of early warning instructions indicating the dangerous state of the target object according to the mouth and nose exposure conditions displayed by the facial feature information of the target object, the hypoxia conditions displayed by the physiological feature information of the target object, and the sound changes of the target object and / or the environment indicated by the sound information. The present invention can solve the problem of missing the best rescue time caused by the inability of the prior art to distinguish the types of early warning information according to the detection results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and relates to measurement for diagnostic purposes, and specifically to the determination of heart rate, blood oxygen saturation, etc. to identify the state of a target object, and especially to an alarm system for preventing suffocation. Background Art

[0002] Infants are very vulnerable and their safety is of great concern to parents. In daily life, a little carelessness can lead to tragedy, such as: when a baby is sleeping, the mouth and nose are covered by the quilt and cannot breathe; when sleeping, the baby turns to a prone position with the face facing down, causing suffocation; when lying flat, the baby spit milk into the lungs or chokes the nasal cavity, causing nasal obstruction, etc. Infant suffocation is one of the main causes of death. If the baby's sleeping posture is improper or the airway is blocked and not treated in time, hypoxia will occur. Long-term hypoxia can cause brain damage and have a great impact on the baby's intellectual development. In severe cases, it can cause the baby's death. Heart rate is one of the important indicators for evaluating the health status of infants, and heart rate monitoring is one of the important items for infant monitoring. For example, the research of Song Wenshuai et al. revealed that in the infant group, inhalation suffocation and accidental suffocation in bed are the two types with the highest incidence. If they are not properly handled after the occurrence, the baby's brain ischemia and hypoxia or even death will occur within 3 to 5 minutes. (Research Progress on Prevention of Accidental Suffocation of Infants and First Aid Parent Training [J]. Journal of Nursing (2023)). Xu Zhenjie's research revealed that the normal resting heart rate range of newborns is 130-140 beats / min; the normal resting heart rate range of infants aged 1-12 months is 120-130 beats / min; the normal resting heart rate range of infants aged 1-2 years is 100-110 beats / min; the normal resting heart rate range of infants aged 3-4 years is 95-105 beats / min (Research on non-contact adaptive robust real-time monitoring method of infant heart rate [D]. Jiangsu University (2019)). In addition to heart rate, blood oxygen saturation is also an important vital sign. The blood oxygen saturation of infants is usually maintained at 95%-100%. In some cases, such as suffocation or other diseases, blood oxygen saturation will be abnormal. In addition, the influence of the external environment may also cause abnormal fluctuations in the blood oxygen saturation of infants, such as external noisy or stimulating environmental sounds. Ji Nan's research disclosed that in a pig house environment with a relatively complex sound environment, the sound in the pig house is monitored by sound monitoring equipment to obtain the health status of live pigs. Specifically, this study used audio analysis technology to extract sound feature technology to capture the most representative sound features from complex environments in order to accurately identify pig coughs, and then provide auxiliary early warning for mass respiratory diseases in pigs. Although there are various sound interferences in the environment, the sound signals captured by the sound monitoring equipment have a large range of frequency changes and feature differences. Therefore, the external environmental noise and the characteristic sounds of the target user can be identified based on the extraction of sound features (Research on pig cough sound recognition method based on multi-domain feature fusion [D]. Northeast Agricultural University, (2023)).

[0003] It is impossible for a guardian to stay with the baby all the time. Currently, a monitor is usually used to monitor the baby in real time. The monitoring content (video or audio) can be sent to the guardian's mobile phone, so the guardian does not have to stay by the baby all the time, which ensures the baby's safety to a certain extent. However, this method still requires the guardian to check the mobile phone frequently to understand the baby's condition. When the baby chokes on milk and the guardian does not find this situation, the baby may suffocate. If the guardian misses the monitoring content at the moment of choking, and regards the baby's suffocation as sleeping and misses the time for treatment, it will eventually lead to the death of the baby. On the other hand, when the guardian is resting at night, he cannot detect the baby's abnormal condition in time. Therefore, the traditional monitor has a single function and the safety factor for the baby is not high.

[0004] The Chinese invention patent application with publication number CN116110103A provides a method, device, electronic device and storage medium for preventing suffocation of infants and young children. The method for preventing suffocation of infants and young children includes: obtaining a facial image of an infant and young children; locating the eyes of the facial image to determine the position information of both eyes in the facial image; locating the nostril position of the infant and young children based on the position information of both eyes by grayscale integral projection; if it is determined that the nostril position cannot be located, a warning message is issued, wherein the warning message is used to indicate that the nostrils of the infant and young children are covered. Although this technical solution can warn the breathing state of the infant and young children in advance by locating the nostrils of the infant and young children, during the infant's sleep, the mouth and nose are not blocked, but choking or milk spilling may occur. This technical solution cannot warn the suffocation caused by choking or milk spilling during the infant's sleep.

[0005] The Chinese invention patent application with publication number CN113729673A provides a monitoring method for preventing suffocation, including: comprehensive evaluation of breathing and heartbeat, analyzing whether the current breathing and heartbeat frequencies are within the preset normal range, if not within the normal range, analyzing the duration and the trend of increasing or decreasing; according to the data of the comprehensive evaluation, if the duration of the breathing not within the normal range exceeds the preset time value, and the heartbeat exceeds the normal range, it is determined that suffocation occurs, and the determination result is reported for local and remote alarm, that is, by monitoring the physiological characteristics of the baby to determine whether suffocation occurs, this method cannot distinguish whether the abnormality of the detection result is caused by the environment. For example, if there is a sudden sound stimulation or light stimulation in the environment where the baby is located, the baby will be frightened and start crying. When the baby cries, the sympathetic nerves are excited, the heart rate will increase, the breathing will become rapid, and the rapid breathing and shallow breathing will reduce the effective ventilation of the alveoli, thereby causing unstable blood oxygen saturation, or even a decrease in blood oxygen saturation.

[0006] The existing technology cannot distinguish whether the abnormal vital signs of an infant are caused by environmental stimulation. Once the test results are abnormal, a prompt message will be issued. Frequent prompt messages will increase the anxiety or tension of the infant's parents. In addition, the existing technology cannot distinguish the type of warning information based on the test results. The guardian may be confused for the first time after receiving the warning information. In this state, the guardian does not know what to do after receiving the warning information. In some cases, the baby needs to be sent to the hospital immediately, while in other cases, the baby does not need to be sent to the hospital. The guardian's hesitation may cause the baby to miss the best time for rescue.

[0007] In addition, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making the present invention, but due to space limitations, not all details and contents are listed in detail. However, this does not mean that the present invention does not have the characteristics of these prior arts. On the contrary, the present invention already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art to the background technology. Summary of the invention

[0008] The prior art has already seen technical solutions that use a multi-parameter detection protection system to ensure the safety and health of infants and young children. For example, the patent document with publication number CN117100531A discloses a multi-level protection system for a pediatric intensive care unit and a method of using the system, which includes an incubator, a respiratory monitoring module and an early warning module. The respiratory monitoring module includes an image monitoring unit, a sound monitoring unit, a carbon dioxide monitoring unit and a controller. The image monitoring unit includes a camera, the sound monitoring unit includes a respiratory sound sensor, the carbon dioxide monitoring unit includes an end-expiratory carbon dioxide detector, and the incubator is provided with an adjustment device for adjusting the position of the respiratory sound sensor and the end-expiratory carbon dioxide detector. The camera, the respiratory sound sensor and the end-expiratory carbon dioxide detector are all connected to the controller signal, and the early warning module includes a buzzer and an indicator light. This technical solution uses the real-time monitoring and early warning functions of the multi-level protection system to ensure the safety and health of children in an intensive care environment, and provides timely information required by medical staff, providing effective support for the treatment and care of children. However, the monitoring data obtained in this technical solution can only be compared independently, and the monitoring data cannot be judged jointly. As a result, as long as a certain monitoring data exceeds the preset value, an alarm can be triggered. That is, when the controller receives monitoring data that is different from the preset normal value, the system will trigger an alarm, which will reduce the precision of the early warning system. Under this setting, even abnormal data that does not affect the safety of the user can trigger an alarm, resulting in ineffective monitoring attention and reduced efficiency of safety monitoring.

[0009] In view of the deficiencies in the prior art, the present invention provides an alarm system for preventing suffocation, comprising: a first monitoring module for acquiring facial feature information of a target object; a second monitoring module for acquiring physiological feature information of the target object; a third monitoring module for acquiring sound information, and a processing module, wherein the processing module is signal-connected to the first monitoring module, the second monitoring module and the third monitoring module, respectively, and the processing module is configured to output different types of early warning instructions indicating a dangerous state of the target object according to the mouth and nose exposure conditions displayed by the facial feature information of the target object, the hypoxia conditions displayed by the physiological feature information of the target object, and the sound changes of the target object and / or the environment indicated by the sound information.

[0010] Compared with the above-mentioned prior art, the processing module of the present invention can accurately judge the respiratory state of the target object by combining multiple monitoring parameters, and output different types of early warning instructions according to different judgment results. Based on the above-mentioned distinguishing technical features, the problems to be solved by the present invention may include: how to improve the accuracy of the evaluation of the actual respiratory state of the target object, and improve the precision of the early warning information type. Specifically, the beneficial effects of the present technical solution: the present application obtains the facial feature information of the target object by setting a first monitoring module to help judge the exposure of the mouth and nose, and whether the mouth and nose are exposed directly affects the breathing smoothness of the target object. In particular, the exposure of the mouth and nose can reflect the lying posture of the target object and the degree to which the quilt covers the face. This information helps the face of the target object to be discovered in time before it is covered. The present application obtains the physiological feature information of the target object by setting a second monitoring module to help judge whether the target object is hypoxic. When the human body is hypoxic, physiological characteristics such as rapid breathing, decreased blood oxygen saturation, and compensatory increase in heart rate will occur. The present application obtains sound information by setting a third monitoring module, wherein the sound information includes the sound information of the target object and the environmental sound information. In traditional monitoring methods, when the target object cries, it is impossible to determine whether it is caused by environmental sounds. Usually, when there is a sudden sound stimulus in the environment, the target object is easily frightened and cries non-stop, and the crying is more intense. When crying is more intense, the target object's sympathetic nerves are excited, the heart rate will increase, and the breathing will become rapid. If only the monitoring results of physiological characteristics are relied upon, the warning information of hypoxia will be fed back. In fact, the hypoxia is a temporary phenomenon caused by the target object's crying and will not affect their physical health.

[0011] The present application can analyze various situations that may occur when the target object is unattended by combining the target object's oral and nasal exposure, physiological characteristic information, environmental sounds and the target object's sound, and remind the guardian with different warning information indicating the dangerous state that the target object may be in. When the guardian receives the warning information, he or she can respond quickly according to the type of warning information. In an emergency, this setting is conducive to reducing the guardian's response time.

[0012] During the initialization process of the alarm system according to the present invention, the processing module is further configured to:

[0013] Based on the age information of the corresponding infant as the monitored object input from the mobile terminal paired with the anti-suffocation alarm system, the processing module initializes and determines the initial reference threshold of the physiological characteristic information (physiological indicators), wherein the initial reference threshold is corrected based on the data collected by the second monitoring module during the initialization stage and the age information of the monitored object and the change of the age information.

[0014] When alarm monitoring is enabled, the initial reference threshold of the physiological index corrected in the initialization phase is loaded as the benchmark reference threshold of the alarm system, so as to analyze and output different types of early warning instructions indicating the dangerous state of the target object; and after a preset time period has passed since the alarm monitoring is enabled, the benchmark reference threshold is divided into a first reference threshold representing a resting state, a second reference threshold representing a moving state, a third reference threshold representing a post-eating state, a fourth reference threshold representing a sleeping state, and a fifth reference threshold representing a crying state according to the behavior of the corresponding infant as the monitored object, wherein the initialization of the first, second, third, fourth and fifth reference thresholds can be determined or modified by the second monitoring module collecting physiological indicators accordingly at multiple collection time points (for example, by manual menu selection) determined by the mobile terminal paired with the alarm system when the mobile terminal is at rest, kicking legs, drinking water and feeding, during sleep and crying. Positive, wherein, after the preset time period has passed since the alarm monitoring was enabled, the processing unit dynamically corrects the first, second, third, fourth and fifth reference thresholds according to the data collected by the second monitoring module at multiple collection time points determined by the mobile terminal paired with the alarm system, wherein the dynamic correction is performed continuously by the processing module according to the initial reference threshold determined and corrected in the initialization phase, and in combination with the state determined by the second monitoring module and / or the third monitoring module, so as to obtain a dynamic baseline reference threshold corresponding to the infant as the monitoring object after the preset time period, wherein the dynamic baseline reference threshold is continuously iteratively corrected, and the iteration is, for example, periodic, especially for infants under one month old, with the first beat every 2 days, for infants over one month old but under two months old, with the time beat every one week, and for infants whose birth date exceeds two months old, with the time beat every two weeks. In particular, the time beat of the periodic correction can also be dynamically adaptively set according to the growth state of the corresponding object, in particular, the time interval is manually set according to the weight of the infant by means of the mobile terminal paired with the alarm system.

[0015] The benchmark reference threshold value set according to the present invention is to form a first set of initialization data for the initialization stage, and the initialization data is set based on the evidence-based medical data and the measurement of the second monitoring module and / or the third monitoring module as the basis for correction, thereby being able to exclude erroneous initialization caused by unexpected measurement data while satisfying conventional monitoring alarms. However, since the various data of newborns vary greatly in the initial stage, and the dynamic range of physiological data caused by various behaviors also varies greatly, monitoring that is not corrected and not associated with behavior will inevitably lead to an extremely high false alarm rate, which is also the main reason why related monitoring products cannot play an effective role. Therefore, dynamic corrections related to nonlinear changes in growth age and targeted corrections associated with behavior are of great significance for suppressing extremely high false alarm rates. The inventor discovered during the course of his work that excessive information analysis, excessive collection and correction would lead to unnecessary security redundancy. Excessive correction would also cause the system data to be so huge that the system analysis capability would be insufficient to cope with the situation. Therefore, the present invention only selects the five most representative items from the massive behavioral data for correction. In actual use, the user can also use a mobile terminal paired with the alarm system to select two or three of the first, second, third, fourth and fifth reference thresholds for configuration and correction according to the specific situation of the monitored object, so as to improve the accuracy of the alarm.

[0016] In order to improve the accuracy of infant status detection, the prior art has already developed a technical solution for realizing infant status analysis through machine vision. For example, the patent document with publication number CN109598204A discloses an AI mother-infant care system, including a detection module, a device module, a cloud connection server, an AI server, a management server, a cloud data server, and a cloud analysis server. The detection module collects images through a front-end camera, and then communicates and connects with the microphone, speaker, storage SD card, night infrared fill light device and cloud server through the front-end camera detection algorithm. The mouth and nose occlusion detection module in the detection module of this technical solution takes the front-end camera detection algorithm as the core, adds the judgment of the application layer logic to strengthen the prediction of occlusion, and gives a suggested sound and light alarm and alarm push to reduce the risk of suffocation and avoid suffocation. It can realize functions such as infant emotion recording, infant behavior analysis, family control, infant status detection, and infant activity range monitoring. However, the suffocation state judgment in the technical solution is still judged only by the single image information collected by the camera. The occlusion of the mouth and nose state on the image does not necessarily cause the baby to suffocate. If only relying on the image detected by the mouth and nose occlusion detection module to push the corresponding warning, it will obviously increase the number of responses of the guardian, resulting in excessive warning of the baby's monitoring state. Compared with the above-mentioned prior art, the processing module of the present invention can monitor the suffocation state of the target object according to the mouth and nose exposure situation displayed by the facial feature information of the target object and in combination with the hypoxia situation and sound information of the target object. Based on the above-mentioned distinguishing technical features, the problems to be solved by the present invention may include: how to improve the accuracy of the actual respiratory state assessment referred to by the mouth and nose exposure situation. Furthermore, although the physiological feature information can directly reflect whether the target object is hypoxic, the target object has various hypoxia situations, such as hypoxia caused by occlusion of the mouth and nose, hypoxia caused by normal crying, hypoxia caused by choking, etc. If all hypoxia situations are reminded by the same type of warning information, the danger level of different hypoxia situations will be blurred. In some cases, the target object suffers from hypoxia due to normal crying, which is less dangerous. In other cases, the target object suffers from hypoxia due to blocked mouth and nose, which is very dangerous and can cause brain hypoxia in just 3 minutes. When receiving different types of warning information, guardians can respond quickly according to the type of warning information, reducing the situation where the time for taking rescue measures for the target object is prolonged due to nervousness.

[0017] On the other hand, the crying of the target object is also divided into different situations, such as crying caused by sudden choking while sleeping, crying caused by external environmental sound stimulation while sleeping, etc. When crying causes hypoxia, the guardian will receive an early warning message. Usually, the target object cries more frequently every day, that is, the guardian will receive multiple early warning messages of hypoxia of the target object every day, which increases the tension and anxiety of the guardian. At the same time, since the early warning information is not distinguished, the guardian will be confused after receiving the early warning information and will not know what to do next. In the case of emergency medical treatment, this operation will obviously delay the rescue time of the target object. In the case of no need for medical treatment, some guardians will send the target object to the hospital because of excessive worry. This operation not only increases medical costs, but also increases the pressure on the hospital to visit.

[0018] The present application can distinguish situations in which the target object suffers from hypoxia, so that the guardian can not only receive warning information in a timely manner, but also obtain the degree of danger of the target object through the type of warning information, which is conducive to the guardian to quickly take corresponding rescue measures for the target object, reduce the guardian's hesitation time, and relieve his anxiety.

[0019] According to a preferred embodiment, the alarm system for preventing suffocation also includes a mobile terminal for receiving warning instructions and issuing warning prompts, and the processing module is configured to: when the facial feature information of the target object shows that the mouth is blocked, control the mobile terminal to issue a first type of warning information indicating that the target object's breathing is about to be obstructed.

[0020] Beneficial effects of the technical solution: The mobile terminal is a device that the guardian carries with him, which is conducive to the guardian receiving early warning information in time. Since the mouth and nose are the respiratory passages of the human body, the present application chooses to directly monitor the exposure of the mouth and nose of the target object. In particular, when the target object is sleeping, the quilt or other coverings may move toward the face due to the movement of the target object when sleeping. When the quilt or other coverings have covered the mouth of the target object, it means that the quilt has a tendency to continue to move toward the nose. At this time, the nose of the target object can still breathe normally. The present technical solution sends out the first type of early warning information when the mouth of the target object is covered, so as to remind the guardian before the mouth and nose are covered, and then the guardian can deal with the covering of the mouth and nose before the target object suffers from hypoxia, thereby reducing the probability of accidents. The defect of the prior art is that the early warning information is usually issued only when the mouth and nose of the target object are covered and hypoxia occurs. Although the target object can struggle, it will not raise its head. At this time, the target object has experienced hypoxia. If the guardian is delayed in arriving at the location of the target object, it is very likely to cause the target object to suffocate.

[0021] When the target is an infant, and its mouth and nose are blocked and hypoxia occurs, there is very little time left for rescue. Mild hypoxia may occur when the infant's mouth and nose are blocked for tens of seconds, and brain hypoxia may occur when the blockage time exceeds 3 minutes. Damage caused by brain hypoxia may lead to symptoms such as mental retardation, impaired speech function, and cerebral palsy. If the hypoxia time continues to prolong, it will affect the normal functions of multiple organs such as the heart, lungs, and liver, and may even lead to the death of the target.

[0022] This technical solution sends out the first type of warning information indicating that breathing is about to be obstructed when the mouth of the target object is blocked. At this time, the target object can still breathe normally and there is no hypoxia. The guardian can respond quickly after receiving the warning information, which not only leaves time for the guardian to process, but also reminds the guardian of the next processing operation, reducing their reaction time and effectively reducing the risk of suffocation of the target object.

[0023] According to a preferred embodiment, the processing module is configured to: when the facial feature information of the target object shows that the mouth and nose are blocked, control the mobile terminal to issue a second type of warning information indicating that the target object's face is blocked and breathing is obstructed.

[0024] The beneficial effect of the technical solution is that when the target object is sleeping without anyone watching, its sleeping posture may change from supine or side-lying to prone posture. Prone sleeping posture can cause the respiratory tract to be squeezed and hypoxia, which leads to suffocation. Therefore, the technical solution is particularly suitable for the case where the target object has a prone sleeping posture. When its sleeping posture changes to a prone posture, the first monitoring module cannot obtain the facial features of the target object, that is, it cannot identify the mouth and nose of the target object. When this happens, the processing module directly controls the mobile terminal to issue a second type of warning information indicating that the face of the target object is blocked and the breathing is obstructed. When the guardian receives the second type of warning information, it can know in advance that the mouth and nose of the target object are blocked. Therefore, it can distinguish the importance of things and give priority to arriving at the location of the target object to deal with the target object. In daily situations, some guardians lack medical knowledge. Even if they receive warning information, they may choose to finish what they are doing first, and then go to see the situation of the target object. This operation is likely to cause the target object to suffocate due to prolonged hypoxia.

[0025] The second type of warning information of the present technical solution has indicated that the target object's breathing is obstructed. Therefore, the guardian can give priority to dealing with the target object's situation. Moreover, when the guardian receives the second type of warning information, he already knows the next operating steps. When he arrives at the location of the target object, he can take measures immediately to shorten the time when the target object's breathing is obstructed to the greatest extent.

[0026] According to a preferred embodiment, the alarm system for preventing suffocation also includes a timing module connected to the processing module signal, and the processing module is configured as follows: when the facial feature information of the target object shows that the mouth and nose are blocked, the timing module starts timing, and stops timing after the mouth and nose of the target object are exposed, so as to obtain the time when the mouth and nose of the target object are blocked, wherein when the mouth and nose of the target object are blocked for a period of time exceeding a first time length that can cause brain hypoxia, the mobile terminal is controlled to issue a second type of warning information indicating that the target object's face is blocked and breathing is obstructed, and the mobile terminal is controlled to automatically call an emergency number.

[0027] Beneficial effects of this technical solution: Since the target object belongs to a very vulnerable group of people, when their mouth and nose are blocked, their breathing is hindered, and the target object will experience hypoxia in a very short time. When the hypoxia time increases to a certain time, brain hypoxia will occur. This technical solution obtains the length of time the target object's breathing is suppressed by setting a timing module. When the target object's breathing is suppressed for more than the first time length that can cause brain hypoxia, the processing module can control the mobile terminal to automatically call an emergency number. Since brain hypoxia can affect the development of intelligence, language, movement, etc., therefore, when the brain is hypoxic, the target object needs to be sent to a doctor in time and examined under the guidance of a doctor.

[0028] In the existing technology, in an emergency, when the guardian receives the warning information, he may forget the mobile terminal he carries at the original location, or put the mobile terminal somewhere casually. When he arrives at the location of the target object, even if he finds that the target object is suffering from hypoxia or suffocation, he cannot immediately find the mobile device to ask for help. In addition, due to the limited cognitive level of the guardian, when he arrives at the location of the target object, he will first choose to observe the state of the target object or observe the physiological state of the target object through the second monitoring module. He will not choose to send the target object to the doctor or call the emergency number until the target object has shown very obvious danger signals. In this case, the guardian may also spend a certain amount of time looking for the mobile device. From the time the target object's breathing is suppressed to the time the emergency number is called, a long time has been wasted, and it is likely to miss the best time for emergency treatment.

[0029] This technical solution can well overcome the defects of the prior art. Once it is detected that the target object's breathing is suppressed for a period of time exceeding the first time length, an emergency call will be automatically made. If the breathing is suppressed for a period of time exceeding the first time length, it indicates that brain hypoxia may occur. Brain hypoxia has adverse effects on various central functions of the target object. Therefore, a professional doctor is required to examine the target object to assess the degree of damage caused by hypoxia in various body systems.

[0030] According to a preferred embodiment, the sound information acquired by the third monitoring module includes sound information from the target object itself and sound information from the external environment, wherein the sound information refers to the sound intensity and the duration at the corresponding sound intensity.

[0031] Beneficial effects of the present technical solution: The sound information acquired by the present application includes the sound information of the target object and the sound information of the external environment. This setting can help distinguish the hypoxia caused by the crying of the target object caused by the sound stimulation of the external environment. When the sound intensity of the external environment is high, it is easy to cause the target object to cry. When crying, the breathing is faster and shallower, causing the effective ventilation volume of the alveoli to decrease, which in turn leads to a decrease in blood oxygen saturation. The existing technology is unable to distinguish the cause of hypoxia displayed by physiological characteristic information. Once a decrease in blood oxygen saturation and / or an increase in heart rate is detected, an early warning message will be sent. After receiving the early warning message of hypoxia, the guardian will feel nervous and worried. When a large number of early warning messages of hypoxia are received daily, it will significantly affect the guardian's mental state.

[0032] According to a preferred embodiment, the processing module is configured to: when the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object is suffering from hypoxia, and the sound information of the external environment shows that the sound intensity exceeds a first intensity that can cause a crying response of the target object and the sound intensity of the target object continues to maintain or continues to exceed a second intensity indicating that the airway is unobstructed, control the mobile terminal to issue a first prompt message that the target object is frightened.

[0033] Beneficial effects of this technical solution: When the mouth and nose of the target object are not blocked and the physiological characteristic information shows that it is hypoxic, this technical solution further obtains the environmental sound and the sound of the target object to analyze the specific cause of the hypoxia of the target object. When the sound intensity of the external environment exceeds the first intensity that can cause the target object to cry and the sound intensity of the target object continues to maintain or continues to exceed the second intensity indicating that the respiratory tract is unobstructed, it means that the hypoxia of the target object is caused by the external environmental sound stimulation. The target object may cry due to the stimulation of the external environmental sound, which leads to an increase in heart rate and / or a decrease in blood oxygen saturation. In this technical solution, although the target object is monitored to be hypoxic, its sound intensity exceeds the second intensity, indicating that the respiratory tract is smooth, so the hypoxia is only caused by environmental stimulation. When the guardian receives the prompt information, it can be known that the target object is frightened and needs to be comforted. This setting can prevent the guardian from being in a tense state of mind at all times, and in this case, there is no need to send the target object to the hospital for examination, and the guardian does not have to send the target object to the hospital due to excessive worry, reducing the number of ineffective medical treatment.

[0034] According to a preferred embodiment, the processing module is configured to: when the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object is suffering from hypoxia, the target object makes choking and crying sounds, and the sound intensity continues to maintain or continues to exceed the second intensity indicating that the airway is unobstructed, control the mobile terminal to issue a second prompt information indicating that the target object is suffering from slight choking.

[0035] Beneficial effects of this technical solution: When the mouth and nose of the target object are not blocked and the physiological characteristic information shows that there is hypoxia, this technical solution further monitors the sound of the target object. When the target object chokes and cries, and the sound intensity continues to maintain or exceeds the second intensity, it means that the target object may have a slight choking. Since the target object can still cry loudly, it means that its respiratory tract remains smooth, and the choking or crying causes temporary hypoxia. Therefore, the processing module controls the mobile terminal to issue a second prompt information indicating that the target object has a slight choking. When the guardian receives the prompt information, he can quickly take measures to relieve the discomfort caused by the choking of the target object as soon as possible.

[0036] According to a preferred embodiment, the processing module is configured to: when the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object has hypoxia, and the sound intensity of the target object is continuously lower than the second intensity, control the mobile terminal to issue a third type of warning information indicating that the target object may have hypoxia caused by choking.

[0037] The beneficial effects of the present technical solution are as follows: when the mouth and nose of the target object are not blocked and the physiological characteristic information shows the presence of hypoxia, and the sound intensity of the target object is continuously lower than the second intensity, it indicates that the target object is unable to make a greater response due to hypoxia, and this situation may be caused by the target object choking on a large amount of milk. Choking on a large amount of milk may cause the overflowing milk to enter the lungs or nasal cavity, thereby blocking the respiratory tract, and the target object's breathing is obstructed, and it is unable to make a greater response or make a sound. In this case, the processing module can control the mobile terminal to issue a third type of warning information indicating that the target object may have hypoxia caused by choking on milk. When the guardian receives the third type of warning information, he or she can take emergency measures as soon as possible, such as: clearing foreign objects from the target object's mouth, picking up the target object and patting its back to reduce the obstruction of the respiratory tract, and avoiding the guardian from increasing the judgment time due to the inability to determine the cause of the target object's hypoxia, thereby prolonging the length of time the target object's respiratory tract is obstructed.

[0038] According to a preferred embodiment, the third monitoring module includes a sound receiving unit and a sound processing unit. The sound processing unit distinguishes the sound of the target object itself and the sound of the external environment through characteristic quantities of the sound, and then obtains the sound intensity of the target object and / or the external environment and the duration at the corresponding sound intensity.

[0039] Preferably, the characteristic quantity of the sound is, for example, the signal waveform, loudness, pitch and timbre of the sound.

[0040] According to a preferred embodiment, the second monitoring module is a monitoring device wearable on the wrist of the target object. Preferably, the second monitoring module can be configured as a heart rate and blood oxygen sensor.

[0041] According to a preferred embodiment, the second monitoring module and the third monitoring module are integrated into one body.

[0042] Beneficial effects of this technical solution: The second monitoring module and the third monitoring module are set as one unit to better obtain sound information. When the target object is sleeping, the hands are usually raised on both sides of the head, that is, close to the ears. When the second monitoring module and the third monitoring module are close to the ears, the third monitoring module can better receive the sound made by the baby. At the same time, the sound intensity of the external environment received by the third monitoring module is also closer to the sound that the baby actually hears. On the other hand, the second monitoring module and the third monitoring module are set as one unit to simplify the structure and avoid wearing multiple devices on the target object's body to affect the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a simplified schematic diagram of module connection relationships of a suffocation prevention alarm system according to a preferred embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the working process of a suffocation prevention alarm system according to a preferred embodiment of the present invention;

[0045] Figure 3 is a structural schematic diagram of a second monitoring module of a preferred embodiment provided by the present invention;

[0046] Figure 4 It is a schematic diagram of the working principle of a suffocation prevention alarm system according to a preferred embodiment of the present invention;

[0047] Figure 5 It is a schematic diagram of the working principle of an alarm system for preventing suffocation according to another preferred embodiment of the present invention;

[0048] Figure 6 It is a preferred embodiment of the first monitoring module according to the present invention.

[0049] Reference numerals list

[0050] 100: first monitoring module; 200: second monitoring module; 210: first ring member; 211: first part; 212: second part; 213: fixing part; 220: intermediate connecting member; 230: second ring member; 240: monitoring unit; 300: third monitoring module; 310: sound receiving unit; 320: sound processing unit; 400: processing module; 500: mobile terminal; 600: timing module. DETAILED DESCRIPTION

[0051] The following is a detailed description with reference to the accompanying drawings.

[0052] Facial feature information refers to the characteristic information of facial organs, mainly referring to the exposure of the mouth and nose.

[0053] Physiological characteristic information refers to information used to determine the vital signs of the target object, such as heart rate, body temperature, blood oxygen saturation, respiratory rate, etc.

[0054] Sound information refers to the interfering sound information in the external environment that the target object can receive and the sound from the target object, such as the target object's crying, choking, and interfering sounds in the external environment.

[0055] The suffocation prevention alarm system provided in the present application is particularly suitable for monitoring infants, and the target objects are mainly infants. Example 1

[0056] This embodiment provides an alarm system for preventing suffocation, such as Figure 1 and Figure 4As shown, it includes a first monitoring module 100, a second monitoring module 200, a third monitoring module 300 and a processing module 400. The first monitoring module 100 is used to obtain facial feature information of the baby. Preferably, the first monitoring module 100 can be set at a position corresponding to the baby's head rather than the baby's feet. For example, the first monitoring module 100 is set at a position on the roof corresponding to the baby's head; or, it is set at a position on the bedside table corresponding to the baby's head. This setting can reduce the erroneous data caused by the angle when the first monitoring module 100 collects facial feature information. If the first monitoring module 100 is set at a position corresponding to the baby's feet, when collecting facial feature information, even if the baby's mouth and nose are not covered, the facial feature information may be displayed as both the mouth and nose are covered, thereby reducing the monitoring accuracy of the system. Specifically, the first monitoring module 100 can be a video monitor, which is usually composed of a camera, an electric lens, a pan / tilt, a multi-function decoder, etc. Facial feature information mainly refers to the exposure of the baby's mouth and nose. The second monitoring module 200 is used to obtain physiological feature information of the baby, such as heart rate and blood oxygen saturation. Specifically, the second monitoring module 200 obtains physiological characteristic information through the PPG heart rate and blood oxygen sensor. The third monitoring module 300 is used to obtain sound information. The sound information mainly includes the crying and choking sounds of the baby and the interference sounds in the environment. By utilizing the information collected by the first monitoring module, the second monitoring module and / or the third monitoring module, the alarm system for preventing suffocation of this embodiment is suitable for babies with difficulty breathing due to their mouths and noses being covered when there is no family member to accompany them; on the other hand, it is suitable for suffocation caused by blockage of the respiratory tract after the baby chokes on milk; the third aspect is also suitable for the judgment of abnormal physiological parameters when the baby is stimulated by the external environment.

[0057] According to a preferred embodiment, the third monitoring module 300 includes a sound receiving unit 310 and a sound processing unit 320. The sound processing unit 320 distinguishes the sound of the target object itself and the sound of the external environment through the characteristic amount of the sound, and then obtains the sound intensity of the target object and / or the external environment and the duration under the corresponding sound intensity. Preferably, the characteristic amount of the sound is such as the signal waveform, loudness, pitch and timbre of the sound. Preferably, the sound receiving unit 310 can be a sound sensor.

[0058] According to a preferred embodiment, the processing module 400 is set as a processor, which serves as an execution module for information processing and program operation. The processor mainly includes two parts, namely a controller and an operator, which also includes a cache memory and a bus for realizing data and control of the connection between them. The processor can also be a mobile phone, a tablet computer, a computer, etc. with a built-in microprocessor chip or a dedicated integrated circuit, which can receive information from the first monitoring module 100, the second monitoring module 200 and / or the third monitoring module 300, and quickly process and analyze the received information. The processor has a certain storage ability and storage capacity, which receives temporary data generated during the monitoring process of the first monitoring module 100, the second monitoring module 200 and the third monitoring module 300, and filters out the data actually used for analysis for subsequent processing. Since each monitoring module converts the received signal into data information at a different rate, and not all acquired monitoring data are used to determine the infant's condition, according to a preferred embodiment, the first monitoring module 100, the second monitoring module 200, and the third monitoring module 300 collect different amounts of data, have different processing times, and have different transmission times to the processing module 400. When different monitoring modules send the collected information to the processing module 400, the processing module 400 filters out the data at the same time on the timeline, and then performs a joint analysis. After the processor marks and filters the data acquired by the first monitoring module 100, the second monitoring module 200, and the third monitoring module 300 based on time changes, it uses the filtered data for analysis to obtain results that can reflect the actual condition of the infant. For example: the first monitoring module 100 obtains the exposure information of the infant's mouth and nose at 8:10, 8:12, 8:14, 8:16, 8:18 and 8:20 on March 10, 2023, and sends the oral and nasal exposure information to the processor based on the timeline; the second monitoring module 200 obtains the infant's physiological characteristic information at 8:10, 8:11, 8:12, 8:13, 8:14, 8:15, 8:16, 8:17, 8:18, 8:19 and 8:20 on March 10, 2023, and sends the physiological characteristic information to the processor based on the timeline; the third monitoring module 300 obtains the sound information at 8:10, 8:12, 8:14, 8:16, 8:18 and 8:20 on March 10, 2023, and sends the sound information to the processor based on the timeline. After receiving the information from each monitoring module, the processor screens it, selects the information generated at the same time, assigns an identification tag, and then analyzes the screened information. For example, the processor screens the facial information, physiological feature information, and sound information obtained at 8:10, 8:12, 8:14, 8:16, 8:18, and 8:20, analyzes the information corresponding to the screened time points, and triggers subsequent operations.

[0059] The processing module 400 controls the mobile terminal 500 to issue warning information and prompt information indicating the dangerous state of the target object based on the exposure of the mouth and nose displayed by the facial feature information of the target object, the hypoxia displayed by the physiological feature information of the target object, and the sound changes of the target object and / or the environment indicated by the sound information. The dangerous states of the target object represented by the warning information and the prompt information are different. According to a preferred embodiment, the processing module 400 can switch the type of information issued by the mobile terminal 500 based on the analysis results of the data collected by the first monitoring module 100, the second monitoring module 200 and / or the third monitoring module 300.

[0060] According to a preferred embodiment, the processing module 400 can switch the working modes of the second monitoring module 200 and the third monitoring module 300 based on the oral and nasal exposure conditions monitored by the first monitoring module 100. Specifically, the processing module 400 can control the opening or closing of the second monitoring module 200 and the third monitoring module 300 based on the oral and nasal exposure conditions monitored by the first monitoring module 100.

[0061] According to a preferred embodiment, the processing module 400 can control the collection cycle of the second monitoring module 200 and the third monitoring module 300 based on the exposure of the mouth and nose monitored by the first monitoring module 100. For example: when the first monitoring module 100 collects that the mouth or nose of the target object is blocked, the processing module 400 controls the second monitoring module 200 and the timing module 600 to be continuously turned on to monitor the physiological data of the target object, and obtains the duration of the specific state of the target object (such as the duration of the mouth and nose being blocked), and controls the third monitoring module 300 to be closed. Since the target object is in danger of suffocation, the processing module 400 can reduce the analysis of the data of the third processing module 300 and can quickly issue early warning information when the amount of data is reduced. Continuously monitoring the changes in the physiological data of the specific state of the target object is conducive to the doctor's understanding of the condition during subsequent medical treatment, so as to quickly carry out rescue measures.

[0062] When the first monitoring module 100 collects that the mouth and nose of the target object are not blocked, the processing module 400 controls to increase the collection cycle of the second monitoring module 200, the third monitoring module 300 and the timing module 600. The mouth and nose of the target object are not blocked, and the possibility of suffocation is relatively lower. In this case, the mobile terminal 500 sends a prompt message, so the collection cycle of the second monitoring module 200, the third monitoring module 300 and the timing module 600 can be increased to reduce the data processing volume and increase the monitoring accuracy.

[0063] According to a preferred embodiment, the processing module 400 can also control the collection cycle of the third monitoring module 300 and the timing module 600 based on the oral and nasal exposure conditions monitored by the first monitoring module 100 and the second monitoring module 200 .

[0064] According to a preferred embodiment, the processing module 400 is further configured to: according to the age information of the corresponding infant as the target object input by the mobile terminal 500 paired with the suffocation prevention alarm system, the processing module 400 initializes and determines the initial reference threshold of the physiological characteristic information, wherein the initial reference threshold is corrected according to the data collected by the second monitoring module 200 in the initialization stage and the age information of the target object and the change of the age information, such as Figure 5 shown.

[0065] According to a preferred embodiment, the step of sound recognition includes: the sound receiving unit 310 obtains sound information and transmits it to the sound processing unit 320. The sound processing unit 320 recognizes the received sound information. Preferably, the sound processing unit 320 distinguishes the sound frequency, loudness, pitch, and timbre through the signal waveforms of different sound sources, and then distinguishes the received sound information. After the sound processing unit 320 divides the received sound information into different sound intervals, when the sound intervals are obtained, the processing module 400 can extract the duration of different sound intervals, so that the sound intensity of the target object and / or the external environment and the duration at the corresponding sound intensity can be obtained.

[0066] According to a preferred embodiment, the processing module 400 is respectively connected to the first monitoring module 100, the second monitoring module 200 and the third monitoring module 300 by signals. Preferably, the signal connection mode can be a wireless connection, such as WLAN, Bluetooth connection.

[0067] According to a preferred embodiment, the processing module 400 is configured to output warning instructions indicating different types of dangerous states of the target object according to the exposure of the mouth and nose displayed by the facial feature information of the target object, the hypoxia displayed by the physiological feature information of the target object, and the sound changes of the target object and / or the environment indicated by the sound information. The specific situations include: ① The facial feature information shows that the mouth is blocked; ② The facial feature information shows that the mouth and nose are blocked; ③ The facial feature information shows that the mouth and nose are not blocked, and the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object has hypoxia. In the third situation, the facial feature information shows that the mouth and nose are not blocked, and the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object has hypoxia. It can also include: A. The sound information of the external environment shows that the sound intensity exceeds the first intensity that can cause the target object to cry and the sound intensity of the target object continues to maintain or continue to exceed the second intensity indicating that the airway is unobstructed; B. The sound of choking and crying is emitted, and the sound intensity continues to maintain or continue to exceed the second intensity indicating that the airway is unobstructed; C. The sound intensity of the target object is continuously lower than the second intensity.

[0068] The first intensity can be set to 70, 80, or 90 decibels, for example. When the sound of the external environment (such as horns on the road, fireworks, and other stimulating sounds) exceeds the first intensity, the target object will become anxious and cry.

[0069] The second intensity can be set to 100, 110, or 120 decibels, for example. When the target object has a smooth respiratory tract, the audio frequency of the crying sound is high, indicating that the target object can breathe enough oxygen self-sufficiently.

[0070] Different situations correspond to different degrees of danger. This embodiment divides the situations in which the infant may suffer from hypoxia and the causes of hypoxia so that the guardian can quickly understand the specific situation of the infant and take targeted measures in a timely manner.

[0071] According to a preferred embodiment, the facial recognition step includes: obtaining real-time dynamic video of the baby's face and sending the video to the processing module 400; the processing module 400 extracts the facial image after receiving the dynamic video; identifies the features of the mouth and nose of the face after performing fill light, sharpening and other processing on the facial image; and compares with the pre-stored facial feature image of the baby. When the mouth and nose in the subsequently extracted facial image are successfully compared with the mouth and nose in the pre-stored facial feature image of the baby, it is confirmed that the mouth and nose of the baby are exposed to the external environment.

[0072] Figure 2 Schematic diagram of the work flow of the suffocation prevention alarm system provided in this embodiment.

[0073] According to a preferred embodiment, the alarm system for preventing suffocation further includes a mobile terminal 500 for receiving an early warning instruction and issuing an early warning prompt, and the processing module 400 is configured to: when the facial feature information of the target object shows that the mouth is blocked, control the mobile terminal 500 to issue a first type of early warning information indicating that the breathing of the target object is about to be obstructed. Specifically, the alarm system for preventing suffocation includes a mobile terminal 500 for receiving an early warning instruction and issuing an early warning prompt, and a processing module 400. The processing module 400 is configured to perform the following operations:

[0074] S1. Facial feature judgment

[0075] When the target object’s facial feature information meets the following conditions: ,

[0076] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0077] S2. Warning information output

[0078] When the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a first type of warning information, which indicates that the target object's breathing is about to be obstructed.

[0079] Preferably, the first type of warning information can be set as a voice warning information, for example, the mobile terminal 500 issues a voice prompt: the mouth of the target object is blocked, and breathing is about to be obstructed, please check and deal with it in time.

[0080] Preferably, the mobile terminal 500 can be a smart phone, a tablet, a smart watch, or a smart bracelet.

[0081] According to a preferred embodiment, the processing module 400 is configured to: when the facial feature information of the target object shows that the mouth and nose are blocked, control the mobile terminal 500 to issue a second type of warning information indicating that the target object's face is blocked and breathing is obstructed. Specifically, the suffocation prevention processing module 400 is configured to perform the following operations: when the facial feature information of the target object meets the following conditions:

[0082] S1. Facial feature judgment

[0083] When the target object’s facial feature information meets the following conditions: ,

[0084] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0085] S2. Warning information output

[0086] When the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a second type of warning information, which indicates that the target object's face is blocked and its breathing is obstructed.

[0087] Preferably, the second type of warning information can be set as sound and light warning information, for example, the mobile terminal 500 issues a voice and a red light flashing prompt: the mouth and nose of the target object are blocked, breathing is obstructed, please go to the corresponding place immediately for processing.

[0088] According to a preferred embodiment, the alarm system for preventing suffocation also includes a timing module 600 connected to the signal of the processing module 400, and the processing module 400 is configured as follows: when the facial feature information of the target object shows that the mouth and nose are blocked, the timing module 600 starts timing, and stops timing after the mouth and nose of the target object are exposed, so as to obtain the time when the mouth and nose of the target object are blocked, and when the time when the mouth and nose of the target object are blocked exceeds the first time length that can cause brain hypoxia, the mobile terminal 500 is controlled to issue a second type of warning information indicating that the face of the target object is blocked and the breathing is obstructed, and the mobile terminal 500 is controlled to automatically dial an emergency number. Preferably, the timing module 600 can be a timer. Specifically, the alarm system for preventing suffocation includes the processing module 400, the mobile terminal 500 and the timing module 600. The processing module 400 is configured to perform the following operations:

[0089] S1. Facial feature judgment

[0090] When the target object’s facial feature information meets the following conditions: ,

[0091] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0092] S2. Start timing

[0093] The timing module 600 starts timing and stops timing when the mouth and nose of the target object are exposed to obtain the time when the mouth and nose of the target object are blocked, which is recorded as T 遮挡 ;

[0094] S3. Determine the length of time the mouth and nose are blocked

[0095] When T 遮挡 Exceeding the first time length that can cause brain hypoxia (denoted as T 阈值 ), that is: T 遮挡> T 阈值 , showing that the target object’s mouth and nose were blocked, which could cause brain hypoxia;

[0096] S4. Warning information output

[0097] When all the above conditions are met, the processing module 400 performs the following operations:

[0098] Controlling the mobile terminal 500 to issue a second type of warning information, which indicates that the target object's face is blocked and breathing is obstructed;

[0099] Control the mobile terminal 500 to automatically dial an emergency number.

[0100] According to a preferred embodiment, the sound information acquired by the third monitoring module 300 includes sound information from the target object itself and sound information from the external environment, wherein the sound information refers to the sound intensity and the duration at the corresponding sound intensity.

[0101] According to a preferred embodiment, the processing module 400 is configured to: when the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object has hypoxia, and the sound information of the external environment shows that the sound intensity exceeds the first intensity that can cause the target object to cry and the sound intensity of the target object continues to maintain or continues to exceed the second intensity indicating that the airway is unobstructed, control the mobile terminal 500 to issue a first prompt information that the target object is frightened. Preferably, when the heart rate exceeds the set heart rate threshold or the blood oxygen saturation is lower than the blood oxygen saturation threshold, it is displayed that the target object has hypoxia. Preferably, when the heart rate exceeds the set heart rate threshold and the blood oxygen saturation is lower than the blood oxygen saturation threshold, it is displayed that the target object has hypoxia.

[0102] Specifically, the processing module 400 is configured to perform the following operations:

[0103] S1. Facial feature judgment

[0104] When the target object’s facial feature information meets the following conditions: ,

[0105] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0106] S2. Physiological characteristics judgment

[0107] When the target object's physiological characteristics meet the following conditions:

[0108] The heart rate (HR) exceeds the set heart rate threshold (HR 阈值 ), or the blood oxygen saturation (recorded as SpO2) is lower than the blood oxygen saturation threshold (recorded as SpO 2阈值 ), that is: HR>HR 阈值 or SpO 2< SpO 2阈值 , showing that the target object has hypoxia;

[0109] S3. Sound intensity judgment

[0110] When the sound information in the external environment meets the following conditions:

[0111] Sound intensity (denoted as I 环境 ) exceeds the first intensity (denoted as I 阈值1 ), and the sound intensity of the target object (denoted as I 对象 ) continuously maintains or continuously exceeds the second intensity (denoted as I 阈值2 ), that is: I 环境> I 阈值1 And I 对象 ≥I 阈值2 , showing the target object being frightened;

[0112] S4. Warning information output

[0113] When all the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a first prompt message indicating that the target object is frightened.

[0114] In another operation scenario, the processing module 400 is configured to perform the following operations:

[0115] S1. Facial feature judgment

[0116] When the target object’s facial feature information meets the following conditions: ,

[0117] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0118] S2. Physiological characteristics judgment

[0119] When the target object's physiological characteristics meet the following conditions:

[0120] The heart rate (HR) exceeds the set heart rate threshold (HR 阈值 ), and the blood oxygen saturation (denoted as SpO2) is lower than the blood oxygen saturation threshold (denoted as SpO 2阈值 ), that is: HR>HR 阈值 And SpO 2< SpO 2阈值 , showing that the target object has hypoxia;

[0121] S3. Sound intensity judgment

[0122] When the sound information in the external environment meets the following conditions:

[0123] Sound intensity (denoted as I环境 ) exceeds the first intensity (denoted as I 阈值1 ), and the sound intensity of the target object (denoted as I 对象 ) continuously maintains or continuously exceeds the second intensity (denoted as I 阈值2 ), that is: I 环境> I 阈值1 And I 对象 ≥I 阈值2 , showing the target object being frightened;

[0124] S4. Warning information output

[0125] When all the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a first prompt message indicating that the target object is frightened.

[0126] In the physiological characteristic information, when the heart rate exceeds 120 beats / minute, it indicates that the target object may be hypoxic. When the blood oxygen saturation is lower than 90%, it indicates that the target object may be hypoxic. When the heart rate exceeds 120 beats / minute and the blood oxygen saturation is lower than 90%, it indicates that the target object may be hypoxic.

[0127] Preferably, the first type of prompt information can be set as vibration prompt information, for example, the mobile terminal 500 vibrates to prompt the target object to be frightened.

[0128] According to a preferred embodiment, the processing module 400 is configured to: when the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object is suffering from hypoxia, the target object makes choking and crying sounds, and the sound intensity continues to maintain or continues to exceed the second intensity indicating that the airway is unobstructed, control the mobile terminal 500 to send a second prompt information indicating that the target object is suffering from slight choking.

[0129] Specifically, the processing module 400 is configured to perform the following operations:

[0130] S1. Facial feature judgment

[0131] When the target object’s facial feature information meets the following conditions: ,

[0132] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object.

[0133] S2. Physiological characteristics judgment

[0134] When the target object's physiological characteristics meet the following conditions:

[0135] The heart rate (HR) exceeds the set heart rate threshold (HR 阈值 ), or the blood oxygen saturation (recorded as SpO2) is lower than the blood oxygen saturation threshold (recorded as SpO 2阈值 ), that is: HR>HR 阈值 or SpO 2< SpO 2阈值 , showing that the target object has hypoxia;

[0136] S3. Sound intensity judgment

[0137] When the target object's sound information meets the following conditions: the target object makes choking and crying sounds, and the sound intensity (denoted as I 对象 ) continuously maintains or continuously exceeds the second intensity (denoted as I 阈值2 ), that is: I 对象 ≥I 阈值2 , showing that the target object choked slightly;

[0138] S4. Warning information output

[0139] When all the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a second prompt message indicating that the target object has slightly choked on milk.

[0140] In another operation scenario, the processing module 400 is configured to perform the following operations:

[0141] S1. Facial feature judgment

[0142] When the target object’s facial feature information meets the following conditions: ,

[0143] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0144] S2. Physiological characteristics judgment

[0145] When the target object's physiological characteristics meet the following conditions:

[0146] The heart rate (HR) exceeds the set heart rate threshold (HR 阈值 ), and the blood oxygen saturation (denoted as SpO2) is lower than the blood oxygen saturation threshold (denoted as SpO 2阈值 ), that is: HR>HR 阈值 And SpO 2< SpO 2阈值 , showing that the target object has hypoxia;

[0147] S3. Sound intensity judgment

[0148] When the target object's sound information meets the following conditions:

[0149] The target object makes choking and crying sounds, and the sound intensity (denoted as I 对象 ) continuously maintains or continuously exceeds the second intensity (denoted as I 阈值2 ), that is: I 对象 ≥I 阈值2 , showing that the target object choked slightly;

[0150] S4. Warning information output

[0151] When all the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a second prompt message indicating that the target object has slightly choked on milk.

[0152] Preferably, the second type of prompt information can be set as vibration plus flash prompt information, for example, the mobile terminal 500 vibrates and flashes yellow light to prompt the target object to choke slightly on milk.

[0153] According to a preferred embodiment, the processing module 400 is configured to: when the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object has hypoxia, and the sound intensity of the target object is continuously lower than the second intensity, control the mobile terminal 500 to issue a third type of warning information indicating that the target object may have hypoxia caused by choking.

[0154] Specifically, the processing module 400 is configured to perform the following operations:

[0155] S1. Facial feature judgment

[0156] When the target object’s facial feature information meets the following conditions: ,

[0157] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0158] S2. Physiological characteristics judgment

[0159] The heart rate (HR) exceeds the set heart rate threshold (HR 阈值 ), or the blood oxygen saturation (recorded as SpO2) is lower than the blood oxygen saturation threshold (recorded as SpO 2阈值 ), that is, HR>HR 阈值 or SpO 2< SpO 2阈值 , among which, HR 阈值 and SpO 2阈值 Respectively represent the set heart rate threshold and blood oxygen saturation threshold;

[0160] S3. Sound intensity judgment

[0161] The sound intensity of the target object (denoted as I 对象 ) is continuously lower than the second intensity (denoted as I 阈值2 ), that is: I 对象 <I 阈值2 , indicating that the target object may be suffering from hypoxia due to choking;

[0162] S4. Warning information output

[0163] When all the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a third type of warning information indicating that the target object may suffer from hypoxia caused by choking on milk.

[0164] In another operation scenario, the processing module 400 is configured to perform the following operations:

[0165] S1. Facial feature judgment

[0166] When the target object’s facial feature information meets the following conditions: ,

[0167] Wherein, F(x) represents the facial feature information function of the target object, and x represents the facial feature data of the target object;

[0168] S2. Physiological characteristics judgment:

[0169] The heart rate (HR) exceeds the set heart rate threshold (HR 阈值 ), and the blood oxygen saturation (denoted as SpO2) is lower than the blood oxygen saturation threshold (denoted as SpO 2阈值 ), that is: HR>HR 阈值 And SpO 2< SpO 2阈值 , among which, HR 阈值 and SpO 2阈值 Respectively represent the set heart rate threshold and blood oxygen saturation threshold;

[0170] S3. Sound intensity judgment

[0171] The sound intensity of the target object (denoted as I 对象 ) is continuously lower than the second intensity (denoted as I 阈值2 ), that is: I 对象 <I 阈值2 , indicating that the target object may be suffering from hypoxia due to choking;

[0172] S4. Warning information output

[0173] When all the above conditions are met, the processing module 400 controls the mobile terminal 500 to issue a third type of warning information indicating that the target object may suffer from hypoxia caused by choking on milk.

[0174] Preferably, the third type of warning information can be set as voice, vibration and flashing prompt information, for example, the mobile terminal 500 issues a voice prompt: the target object may choke and suffer from hypoxia, and simultaneously issues a warning information of vibration and red flashing light.

[0175] According to a preferred embodiment, the second monitoring module 200 is a monitoring device wearable on the wrist of the target object.

[0176] According to a preferred embodiment (such as Figure 6 As shown in FIG. 1 ), the first monitoring module 100 and the third monitoring module 300 are integrally arranged. Preferably, the first monitoring module 100 is integrated with the third monitoring module 300 having a sound receiving unit, wherein the sound receiving unit is an omnidirectional or directional microphone, or an array unit of a combination of omnidirectional and directional microphones.

[0177] According to a preferred embodiment, the second monitoring module 200 and the third monitoring module 300 are integrated.

[0178] According to a preferred embodiment, Figure 4 As shown, the processing module 400 can also be equipped with a standardized communication protocol so that it can be connected to the Hospital Information System (HIS). Through this connection, the processing module 400 can call the relevant physiological indicators of the target object (infant) stored in the HIS. HIS stores various information and historical records of hospitalized infants, such as information, medical advice, examinations, and surgeries, and infants who have undergone relevant examinations or diagnoses in outpatient clinics also have archived physiological indicator examination results. The early warning system of the present invention can also rationalize and personalize the relevant parameters required for early warning (first intensity, second intensity, first time length, etc.) according to the more detailed physiological indicators of the infant, so as to formulate an early warning plan that is more in line with the actual situation of the infant. Specifically, since different infants may have different physiological indicators due to age and development, there will be slight differences in the threshold values ​​of the relevant parameters required for early warning for different infants. For example, some infants have more fully developed vocal cords, so they may be more likely to cry at a higher decibel. For such infants, the threshold of the second intensity set by the processing module 400 should be appropriately increased; some infants have relatively poor cardiopulmonary function, and they may experience suffocation symptoms in a shorter time under the same conditions. For such infants, the first time length set by the processing module 400 should be appropriately shortened. By obtaining the infant's physiological indicators from the HIS system, the processing module 400 can achieve more accurate early warning system settings. Example 2

[0179] This embodiment provides a second monitoring module 200 for monitoring physiological characteristic information of an infant, such as Figure 3 As shown. When the second monitoring module 200 is worn on the baby's wrist, it is limited to a specific range to avoid sliding up to the thicker part of the arm and getting stuck there, thereby causing blood circulation obstruction in the baby's arm. Preferably, the second monitoring module 200 is configured as a monitoring bracelet. The monitoring bracelet includes a first ring 210 that surrounds the wrist, an intermediate connecting member 220, a second ring 230 that is sleeved on the finger, and a monitoring unit 240 that monitors physiological characteristic information. The first ring 210 is connected to the second ring 230 through the intermediate connecting member 220. Preferably, the intermediate connecting member 220 has a specific length to control the moving distance of the second ring 230 on the arm within a smaller range.

[0180] According to a preferred embodiment, the first ring member 210 is opened and closed by a locking portion. The first ring member 210 includes a first portion 211 provided with a snap-in nail and a second portion 212 provided with a snap-in hole, wherein the shape of the snap-in nail is adapted to the snap-in hole, and when the snap-in nail of the first portion 211 is buckled into the snap-in hole of the second portion 212, the first portion 211 is connected with the second portion 212 to form a closed state of the first ring member 210, and when the snap-in nail of the first portion 211 is separated from the snap-in hole of the second portion 212, the first portion 211 is separated from the second portion 212 to form an open state of the first ring member 210. Preferably, the snap-in nail and the snap-in hole are respectively provided at both ends of the first ring member 210.

[0181] Preferably, the monitoring unit 240 is fixed on the first ring member 210. Preferably, a fixing portion 213 is provided in the middle of the first ring member 210. The fixing portion 213 is used to install the monitoring unit 240. Preferably, the fixing portion 213 can be set as a hollow square shell or a circular shell, and the shape here is only an example, as long as the structure can be used to install the monitoring unit 240.

[0182] The monitoring unit 240 can be configured as a heart rate and blood oxygen sensor. Preferably, the third monitoring module 300 is disposed in the second monitoring module 200, so that the second monitoring module 200 and the third monitoring module 300 are integrally formed. Preferably, the sound receiving unit 310 and the sound processing unit 320 of the third monitoring module 300 are disposed in the fixing portion 213. Preferably, the third monitoring module 300 can also be independent of the second monitoring module 200 and placed at the head of the baby's bed in an external form (such as a teddy bear, a soft ball, a toy box, etc.) that can soothe and / or attract children. Preferably, the sound receiving unit is equivalent to the sound receiving unit 310, which can be disposed on the surface of the teddy bear. Preferably, the sound processing unit 320 is disposed in the teddy bear.

[0183] Preferably, the second ring 230 is configured as a finger ring that can be put on the base of a baby's finger. The intermediate connecting member 220 can be configured as a flexible line segment.

[0184] It should be noted that the above-mentioned specific embodiments are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the disclosure scope of the present invention and fall within the protection scope of the present invention. Those skilled in the art should understand that the present invention specification and its drawings are illustrative and do not constitute a limitation of the claims. The scope of protection of the present invention is defined by the claims and their equivalents. The present invention specification contains multiple inventive concepts, such as "preferably" and "according to a preferred embodiment", both of which indicate that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application based on each inventive concept. Throughout the text, the features guided by "preferably" are only an optional method and should not be understood as being required. Therefore, the applicant reserves the right to abandon or delete the relevant preferred features at any time.

Claims

1. An alarm system to prevent suffocation, include: A first monitoring module (100) for acquiring facial feature information of a target object; A second monitoring module (200) for acquiring physiological characteristic information of the target object; A third monitoring module (300) for acquiring sound information, and a processing module (400), wherein the processing module (400) is respectively connected to the first monitoring module (100), the second monitoring module (200) and the third monitoring module (300) by signals, wherein the processing module (400) is configured as follows: Output different types of warning instructions indicating the dangerous state of the target object according to the exposure of the mouth and nose shown by the facial feature information of the target object, the hypoxia shown by the physiological feature information of the target object, and the sound changes of the target object and / or the environment indicated by the sound information, The processing module (400) is configured to: When the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object is suffering from hypoxia, and the sound information of the external environment shows that the sound intensity exceeds a first intensity that can cause the target object to cry, and the sound intensity of the target object continues to maintain or continues to exceed a second intensity indicating that the airway is unobstructed, the mobile terminal (500) is controlled to issue a first prompt message indicating that the target object is frightened, When the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object is suffering from hypoxia, the target object makes choking and crying sounds, and the sound intensity continues to maintain or continues to exceed the second intensity indicating that the airway is unobstructed, the mobile terminal (500) is controlled to issue a second prompt information indicating that the target object is suffering from slight choking, When the facial feature information of the target object shows that the mouth and nose are not blocked, the heart rate and / or blood oxygen saturation of the physiological feature information show that the target object is suffering from hypoxia, and the sound intensity of the target object is continuously lower than the second intensity, the mobile terminal (500) is controlled to issue a third type of warning information indicating that the target object may suffer from hypoxia caused by choking.

2. The system according to claim 1, It is characterized in that The suffocation prevention alarm system further comprises a mobile terminal (500) for receiving an early warning instruction and issuing an early warning prompt, and the processing module (400) is configured to: When the facial feature information of the target object shows that the mouth is blocked, the mobile terminal (500) is controlled to issue a first type of warning information indicating that the breathing of the target object is about to be obstructed.

3. The system according to claim 2, It is characterized in that The processing module (400) is configured to: When the facial feature information of the target object shows that the mouth and nose are blocked, the mobile terminal (500) is controlled to issue a second type of warning information indicating that the target object's face is blocked and breathing is obstructed.

4. The system according to claim 3, It is characterized in that The suffocation prevention alarm system further comprises a timing module (600) connected to the processing module (400) by signal, wherein the processing module (400) is configured to: When the facial feature information of the target object shows that the mouth and nose are blocked, the timing module (600) starts timing, and stops timing after the mouth and nose of the target object are exposed, so as to obtain the time when the mouth and nose of the target object are blocked. in, When the target object's mouth and nose are blocked for a period of time exceeding a first time length that can cause brain hypoxia, the mobile terminal (500) is controlled to issue a second type of warning message indicating that the target object's face is blocked and breathing is obstructed, and the mobile terminal (500) is controlled to automatically dial an emergency number.

5. The system according to claim 1, It is characterized in that The sound information acquired by the third monitoring module (300) includes sound information from the target object itself and sound information from the external environment, wherein the sound information refers to the sound intensity and the duration at the corresponding sound intensity.

6. The system according to claim 5, It is characterized in that The third monitoring module (300) comprises a sound receiving unit (310) and a sound processing unit (320), wherein the sound processing unit (320) distinguishes the sound of the target object itself and the sound of the external environment through a characteristic amount of the sound, thereby obtaining the sound intensity of the target object and / or the external environment and the duration at the corresponding sound intensity.

7. The system according to claim 1, It is characterized in that The processing module (400) is also configured to: Based on age information of a corresponding infant as a target object input from a mobile terminal (500) paired with the anti-asphyxia alarm system, the processing module (400) initializes and determines an initial reference threshold value of physiological characteristic information, wherein the initial reference threshold value is corrected based on data collected by the second monitoring module (200) during the initialization phase and the age information of the target object and changes in the age information.

Citation Information

Patent Citations

  • An AI mother and infant nursing system

    CN109598204A

  • Anti-suffocation monitoring method and device

    CN113729673A

  • Infant anti-suffocation method and device, electronic equipment and storage medium

    CN116110103A

  • Children intensive care unit multi-stage protection system and use method thereof

    CN117100531A

  • Alarming method and device, wearable equipment and storage medium

    CN107595259A