Unattended SOS calling method and system
Through the combination of smart wearable devices and home monitoring, identifying falls in the elderly and conducting accurate rescue inquiries and health analysis, the problems of missed help and waste of resources in the existing technology are solved, and more accurate emergency rescue services are achieved.
Patent Information
- Application Number
- CN202410937677.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-07-12
AI Technical Summary
The existing unmanned monitoring technology may lead to missed help and waste of first aid resources after the elderly fall, and the elderly are unwilling to turn on the rescue function.
Obtain physiological and behavioral information through smart wearable devices, conduct scene recognition and health analysis. If a fall is suspected, initiate an audio or graphic rescue inquiry, and determine whether to send a call request to the designated unit based on feedback and health analysis results, or Contact your guardian to ensure accurate emergency rescue if necessary.
Reduce rescue false alarms and missed reports, and provide more accurate emergency rescue services to ensure that the elderly can receive timely attention and help after falling.
Smart Images

Figure CN118524175B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of unattended technology, and more particularly to an unattended SOS calling method and system. Background Art
[0002] As people age, their body functions gradually decline, and the chances of falling and sudden illness increase. The frequency and probability of these happening to the elderly are relatively higher, so they need to be cared for. However, children have their own lives and jobs, and the caregivers in nursing homes and convalescent homes are always not around the elderly, so it is necessary to provide rescue plans when there is no one to supervise them.
[0003] At present, one solution is that smart watches and smart bracelets actively send a rescue request signal to 120 or related medical units after the elderly fall, so that relevant personnel can go to the rescue in time; however, the above method is often due to the rough rescue mechanism, which may lead to problems such as wrong requests for help and waste of emergency resources. Some elderly people are even unwilling to turn on this function. Therefore, this application proposes a new technical solution. Summary of the invention
[0004] In order to provide more accurate emergency rescue services for the elderly and reduce the waste of emergency resources, the present application provides an unattended SOS calling method and system.
[0005] In the first aspect, the present application provides an unattended SOS calling method, which adopts the following technical solution:
[0006] An unattended SOS calling method, comprising:
[0007] Obtain physiological and behavioral information obtained by monitoring the user's physical signs and behaviors through smart wearable devices;
[0008] Perform scene recognition based on physiological information and behavioral information to obtain the current scene;
[0009] If the current scene is a suspected fall, a post-fall health analysis is performed based on physiological information, and rescue inquiries in the form of audio, text and images are initiated through the pre-established interactive unit;
[0010] If the response to the rescue inquiry is that rescue is needed, a call request is sent to the terminal of the designated unit / personnel;
[0011] If the response to the rescue inquiry is no response or no rescue is needed, determine whether to call the designated guardian autonomously based on the health analysis results after the fall, and obtain and record the communication content after communication;
[0012] If the communication content and / or the health analysis results after the fall meet the preset rescue conditions, a call request is sent to the terminal of the designated unit / personnel.
[0013] Optionally, also include:
[0014] Obtain the location information of smart wearable devices and the user's home monitoring installation information;
[0015] If the current scene is suspected of falling, and the positioning information meets the preset home conditions, and the home monitoring installation information meets the preset image cross-verification conditions, the current monitoring image is called and portrait recognition is performed;
[0016] If the user is not recognized, monitoring rotation tracking control information is sent;
[0017] If the user is identified, fall recognition is performed;
[0018] If the recognition result is a fall, a safety timer is performed;
[0019] If the timing exceeds the recovery time threshold and the recognition result remains a fall, a call request is sent to the terminal of the designated unit / personnel.
[0020] Optionally, also include:
[0021] If the home surveillance installation information meets the preset image cross-verification conditions and the preset active inquiry conditions, then:
[0022] Use home monitoring as an interactive unit to initiate rescue inquiries in audio, graphic, and text formats;
[0023] Get the power of smart wearable devices;
[0024] If the power level of the smart wearable device meets the preset high power threshold and the parameter information of the smart wearable device meets the preset interaction conditions, the smart wearable device is also used as an interaction unit to initiate a rescue inquiry in audio, graphic and text form.
[0025] Optionally, also include:
[0026] If the positioning information meets the preset outing conditions, the battery life analysis after at least one standard interaction is performed based on the power of the smart wearable device;
[0027] If the battery life meets the preset charging reminder conditions, a charging reminder will be given.
[0028] Optionally, it also includes: if the user is not identified within T1 time after the current suspected fall scene occurs, and the feedback of the rescue inquiry is no response, it is determined to autonomously call the designated guardian and send physiological information to the designated guardian; wherein the T1 time is a preset value.
[0029] Optionally, it also includes: if the power level of the smart wearable device meets the preset low power threshold, a charging reminder is output, and a device offline risk warning message is sent to the designated guardian; if a power-saving operation control plan is pre-stored, the power-saving operation control plan is output when the charging reminder is output.
[0030] Optionally, also include:
[0031] Obtain the user's medical history or physiological abnormalities;
[0032] The post-fall health analysis criteria were adjusted according to medical history or physiological abnormalities.
[0033] Optionally, also include:
[0034] If safety timing is performed, the physiological information within the preset tracking and monitoring time is tracked and analyzed;
[0035] If the tracking and analysis results meet the risk conditions, a call request is sent to the terminal of the designated unit / personnel;
[0036] In addition to the preset tracking and monitoring time before and after the safety timing, the preset classification rules are used for classification processing and alarm conditions are allocated respectively.
[0037] Optionally, also include:
[0038] If the current physiological information and behavioral information meet the preset dirty data conditions, the data will be cleaned or the response will be stopped.
[0039] In the second aspect, the present application provides an unattended SOS calling system, which adopts the following technical solution:
[0040] An unattended SOS calling system includes a smart wearable device, a cloud platform and a home monitoring camera, wherein the cloud platform is wirelessly connected to the smart wearable device and the home monitoring camera, and the cloud platform is configured to store a computer program that can be loaded by a processor and execute any of the unattended SOS calling methods described above.
[0041] To summarize, the present application includes the following beneficial technical effects: after a user (mainly the elderly) falls, an inquiry can be actively initiated, and based on the feedback obtained from the inquiry, it can be determined whether to contact a hospital, etc.; at the same time, if the inquiry is not answered or indicates that no rescue is needed, the health condition will be analyzed based on physiological information, and the guardian will be contacted in time when necessary to let him pay attention to the user and communicate, and further analyze the content of the communication, and call for rescue for the user in time when the conditions are met, so that this method can reduce false alarms, missed alarms, etc., and provide more accurate emergency rescue services for the elderly. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 This is a schematic diagram of the process of the method of this application Figure 1 ;
[0043] Figure 2 This is a schematic diagram of the process of the method of this application Figure 2
[0044] Figure 3 It is a schematic diagram of the structure of the system of this application. DETAILED DESCRIPTION
[0045] The following is combined with Figure 1-3 This application is described in further detail.
[0046] The embodiment of the present application discloses an unattended SOS calling method.
[0047] Reference Figure 1 , unattended SOS calling methods include:
[0048] S11. Acquire physiological information and behavioral information obtained by monitoring the user's physical signs and behaviors through the smart wearable device.
[0049] Among them, in this embodiment, the smart wearable device is one of a smart ring and a smart bracelet; in this embodiment, a smart ring is used as an example, and the smart ring should be able to monitor the user's physiological information, such as blood oxygen, heart rate, body temperature, and exercise, and derive information such as heart rate variability, sleep quality, and exercise volume. At the same time, the smart wearable device also includes one of a smart watch and a smart necklace, which should integrate remote call function and positioning function to meet the use requirements.
[0050] S12. Perform scene recognition based on physiological information and behavioral information to obtain the current scene.
[0051] Example: Assuming that the smart wearable device is integrated with an accelerometer, the parameters fed back by the accelerometer match the acceleration change conditions during a fall, and it is identified as a suspected fall scene;
[0052] Assuming that the current heart rate is consistent with the heart rate range during sleep, it is identified as a suspected sleep scene.
[0053] S13. If the current scene is a suspected fall, a post-fall health analysis is performed based on the physiological information, and a rescue inquiry in the form of audio, text and graphics is initiated through an interactive unit with a pre-established connection.
[0054] Among them, post-fall health analysis includes: comparing parameters such as heart rate and blood oxygen with standard range thresholds or with reference range values during stroke and cerebral infarction.
[0055] The above-mentioned interactive unit can be a smart watch or smart necklace if the smart watch or smart necklace in the smart wearable device can perform audio and text and picture interactions; this embodiment gives priority to the audio inquiry method, and the text and picture forms are mostly in the form of pop-up windows combined with vibration, which is relatively poor for the elderly, especially those who have fallen; the content of the rescue inquiry can be: audio "Do you need rescue?"
[0056] S14. If the feedback of the rescue inquiry is that rescue is needed, a call request is sent to the terminal of the designated unit / personnel (e.g., through the Web, APP, using a computer, or a mobile phone); the designated unit is such as a nursing home, hospital, or sanatorium; the designated person is such as a caregiver or the customer service of a medical institution. The request is implemented in the form of a phone call.
[0057] If the feedback to the rescue inquiry is no response or no rescue is needed, the system will determine whether to call the designated guardian autonomously based on the health analysis results after the fall. For example, if the analysis results indicate that the heart rate is within the abnormal heart rate range, the system will start to contact the guardian by phone for verification. The guardian can be a caregiver, family member, etc., and the content of the communication will be obtained and recorded after the communication.
[0058] It should be noted that because the lack of response to inquiries may be caused by damage to the smart wearable device, when actively calling the guardian, the communication may not necessarily be through the smart wearable device, but may also be through a mobile phone carried with the person.
[0059] If the communication content and / or the health analysis results after the fall meet the preset rescue conditions, for example: the user (i.e., the elderly) clearly states in the communication content that rescue is needed; the heart rate is consistent with the heart rate changes during ischemic stroke, then a call request is sent to the terminal of the designated unit / personnel.
[0060] According to the above content, this method can actively initiate inquiries after the user (mainly the elderly) falls, and determine whether to contact the hospital based on the feedback obtained from the inquiry; at the same time, if the inquiry is not answered or it indicates that no rescue is needed, the health condition will be analyzed based on physiological information, and the guardian will be contacted in time when necessary to let him pay attention to the user and communicate, and further analyze the communication content, and call for rescue for the user in time when the conditions are met, so that this method can reduce false alarms and missed alarms of rescue, and provide more accurate emergency rescue services for the elderly.
[0061] Reference Figure 2 In one embodiment of the method, if the user's home is equipped with a home surveillance system, the method is configured as follows:
[0062] S21. Obtain positioning information of the smart wearable device (such as GPS positioning, Beidou positioning) and the user's home monitoring installation information (such as the number and location of installed monitoring).
[0063] S22: If the current scene is suspected of falling, and the positioning information meets the preset home conditions, and the home monitoring installation information meets the preset image cross-verification conditions, the current monitoring image is called and portrait recognition is performed.
[0064] Among them, it meets the preset home conditions, such as: the real-time positioning information is less than 20m from the pre-stored home address. It should be noted that the premise of this item is that the physiological information of the smart wearable device indicates that the user wears it. It meets the preset image cross-validation conditions, such as: the number of monitoring is greater than 1, and the monitoring location is in the living room and kitchen. It is understandable that identifying people based on monitoring is an existing technology, so it will not be repeated.
[0065] S23. If the user is not recognized, monitoring rotation tracking control information is sent, that is, the cloud platform corresponding to the monitoring camera is used to control its rotation and adjust the viewing angle to avoid the user being outside the current monitoring viewing angle.
[0066] S24. If the user is recognized, fall recognition is performed; it should be noted that if there are only elderly people at home, portrait recognition can be converted into human body recognition to reduce the difficulty of recognition.
[0067] S25. If the recognition result is a fall, a safety timer is performed.
[0068] S26. If the timing exceeds the recovery time threshold and the recognition result remains a fall, for example, if the person has not stood up for 5 minutes after the fall, a call request is sent to the terminal of the designated unit / personnel.
[0069] According to the above content, this method also uses the monitoring installed in the user's home to conduct further fall analysis. On the basis of the above, rescue can be provided in time when the elderly fall and lose the ability to actively seek help.
[0070] Furthermore, the method further comprises:
[0071] If the home monitoring installation information meets the preset image cross-verification conditions and meets the preset active inquiry conditions, the home monitoring is used as an interactive unit to initiate rescue inquiries in audio and text forms.
[0072] Among them, the preset active inquiry conditions are met, such as: the surveillance camera has a voice call function, and the user is currently at home.
[0073] According to the above settings, the power of the smart wearable device can be relatively saved, so that the power can be used when it is more needed; at the same time, it can also prevent the smart wearable device from being low on power and affecting the effect of the method.
[0074] The method also includes: obtaining the power of the smart wearable device;
[0075] If the power level of the smart wearable device meets the preset high power threshold (e.g. >50%), and the parameter information of the smart wearable device meets the preset interaction conditions (i.e., it has audio interaction function), the smart wearable device is also used as an interaction unit to initiate rescue inquiries in audio, graphic and text forms.
[0076] According to the above settings, the smart wearable device can be used to supplement its functions to prevent the elderly from speaking too softly or the surveillance camera's audio collection function from being defective, which affects the effectiveness of the method.
[0077] In another embodiment of the method, the method further comprises:
[0078] If the positioning information meets the preset outing conditions, the battery life analysis after at least one standard interaction is performed based on the power of the smart wearable device.
[0079] Among them, the preset outing conditions are met, for example: the distance between the real-time positioning information and the pre-stored home address is greater than 80m. The above-mentioned battery life analysis after at least one standard interaction includes: assuming a standard interaction, the duration is 5 minutes, and the power consumption is 20%, then the battery life is calculated according to the power consumption set by the manufacturer after the current power is reduced by 20%.
[0080] If the battery life meets the preset charging reminder conditions, for example: the battery life is less than 30 minutes, a charging reminder will be given.
[0081] According to the above settings, it can prevent the battery of the smart wearable device from being relatively insufficient when the elderly go out, resulting in the possibility of power outage during the call for help, and the safety of device use can be improved.
[0082] Device charging prompts do not only occur in the above situations. In this method, if the power of the smart wearable device meets the preset low power threshold, such as: <30%, a charging prompt is output; at the same time, this method will also send device offline risk prompt information to the designated guardian to prevent the elderly from accidentally falling and being unable to ask for help when the device is turned off, and the guardian thinks that the device is working normally and pays less attention to it, causing safety hazards. If a power-saving operation control plan is pre-stored, the power-saving operation control plan is output when the charging prompt is output; among them, the power-saving operation control plan is preset by the staff, which can be to stop some relatively unimportant sensors (temperature sensors) or reduce the sampling frequency of sensors.
[0083] Because the human characteristics monitored by smart rings and smart bracelets have certain limitations, in the case of some people falling and losing consciousness or suddenly falling ill and losing the ability to call for help, it is impossible to identify them only by physiological information. Therefore, in another embodiment of the present method, the method also includes: if the user is not identified within T1 time after the current suspected fall scene occurs, and the feedback of the rescue inquiry is no response, it is determined to autonomously call the designated guardian and send physiological information to the designated guardian.
[0084] Among them, T1 duration is a preset value; according to the above settings, if the monitoring cannot find the user and the inquiry is not correct, this method believes that the user may have accidentally fallen in an area such as a bathroom and lost consciousness and the ability to seek help, and calls the guardian in time, etc., so as to provide the user with adequate health care effects.
[0085] In another embodiment of the method, considering that different users have different physical qualities, whether they are ill, and different types of illnesses, the impact of falling is different, so the method is set as follows:
[0086] Obtain the user's medical history or physiological abnormalities; both information can be uploaded by the user or by medical units / staff.
[0087] Correct the post-fall health analysis standards based on medical history or physiological abnormalities; for example, adjust the upper and lower limits of the evaluation standard range for users with a history of heart disease when their heart rate is abnormal, and relax the standards for heart rate fluctuations.
[0088] According to the above settings, this method can better match each user for use.
[0089] In one embodiment of the method, the method further comprises:
[0090] If safety timing is performed, the physiological information within the preset tracking and monitoring time is tracked and analyzed; wherein, the tracking and monitoring time is longer than the recovery time threshold, and the purpose is to continue tracking and measuring to prevent "unnoticed" secondary injuries, such as internal bleeding, etc.; the tracking and analysis may be analysis of heart rate, etc.;
[0091] If the tracking and analysis results meet the risk conditions, a call request is sent to the terminal of the designated unit / personnel;
[0092] In addition to the preset tracking and monitoring time before and after the safety timing, the preset classification rules are used for classification processing, and alarm conditions are assigned respectively; that is, in normal times, body temperature, heart rate, etc. are divided into levels according to their importance, and alarm conditions are assigned to avoid late or late alarms.
[0093] In this embodiment, the method further includes:
[0094] If the current physiological information and behavioral information meet the preset dirty data conditions, for example, if the scene is recognized as bathing and the body temperature data is dirty data, the data will be cleaned or the response will be stopped.
[0095] According to the above settings, on the one hand, analysis interference is reduced, and on the other hand, user privacy can be protected.
[0096] The embodiment of the present application also discloses an unattended SOS calling system.
[0097] Reference Figure 3 , an unattended SOS calling system, comprising a smart wearable device, a cloud platform and a home monitoring camera, wherein the cloud platform is wirelessly connected to the smart wearable device and the home monitoring camera, and the cloud platform is configured to store a computer program that can be loaded by a processor and execute any of the above-mentioned unattended SOS calling methods.
[0098] The unattended SOS calling method has been described in the above embodiments, so it will not be described again.
[0099] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.
Claims
1. An unattended SOS calling method, characterized in that: include: Obtain physiological and behavioral information obtained by monitoring the user's physical signs and behaviors through smart wearable devices; Perform scene recognition based on physiological information and behavioral information to obtain the current scene; If the current scene is a suspected fall, a post-fall health analysis is performed based on physiological information, and rescue inquiries in the form of audio, text and images are initiated through the pre-established interactive unit; If the response to the rescue inquiry is that rescue is needed, a call request is sent to the terminal of the designated unit / personnel; If the response to the rescue inquiry is no response or no rescue is needed, determine whether to call the designated guardian autonomously based on the health analysis results after the fall, and obtain and record the communication content after communication; If the communication content and / or the post-fall health analysis results meet the preset rescue conditions, a call request is sent to the terminal of the designated unit / personnel; Also includes: Obtain the location information of smart wearable devices and the user's home monitoring installation information; If the current scene is suspected of falling, and the positioning information meets the preset home conditions, and the home monitoring installation information meets the preset image cross-verification conditions, the current monitoring image is called and portrait recognition is performed; If the user is not recognized, monitoring rotation tracking control information is sent; If the user is identified, fall recognition is performed; If the recognition result is a fall, a safety timer is performed; If the timing exceeds the recovery time threshold and the identification result remains a fall, a call request is sent to the terminal of the designated unit / personnel; If the home surveillance installation information meets the preset image cross-verification conditions and the preset active inquiry conditions, then: Use home monitoring as an interactive unit to initiate rescue inquiries in audio, graphic, and text formats; Get the power of smart wearable devices; If the power level of the smart wearable device meets the preset high power threshold, and the parameter information of the smart wearable device meets the preset interaction conditions, the smart wearable device is also used as an interaction unit to initiate a rescue inquiry in the form of audio, text and images; Also includes: If the positioning information meets the preset outing conditions, the battery life analysis after at least one standard interaction is performed based on the power of the smart wearable device; If the battery life meets the preset charging reminder conditions, a charging reminder will be given.
2. The unattended SOS calling method according to claim 1, characterized in that: Also includes: If the user is not identified within T1 duration after the current suspected fall scene occurs, and the feedback of the rescue inquiry is no response, it is determined to autonomously call the designated guardian and send physiological information to the designated guardian; wherein, the T1 duration is a preset value.
3. The unattended SOS calling method according to claim 2, characterized in that: Also includes: If the battery level of the smart wearable device meets the preset low battery threshold, a charging reminder will be output, and a device offline risk warning message will be sent to the designated guardian; If a power-saving operation control scheme is pre-stored, the power-saving operation control scheme is output when the charging prompt is output.
4. The unattended SOS calling method according to claim 1, characterized in that: Also includes: Obtain the user's medical history or physiological abnormalities; The post-fall health analysis criteria were adjusted according to medical history or physiological abnormalities.
5. The unattended SOS calling method according to claim 1, characterized in that: Also includes: If safety timing is performed, the physiological information within the preset tracking and monitoring time is tracked and analyzed; If the tracking and analysis results meet the risk conditions, a call request is sent to the terminal of the designated unit / personnel; In addition to the preset tracking and monitoring time before and after the safety timing, the preset classification rules are used for classification processing and alarm conditions are allocated respectively.
6. The unattended SOS calling method according to claim 3, characterized in that: Also includes: If the current physiological information and behavioral information meet the preset dirty data conditions, the data will be cleaned or the response will be stopped.
7. An unattended SOS calling system, characterized in that: The invention comprises a smart wearable device, a cloud platform and a home monitoring camera, wherein the cloud platform is wirelessly connected to the smart wearable device and the home monitoring camera, and the cloud platform is configured to store a computer program that can be loaded by a processor and execute any one of claims 1 to 6 of the unattended SOS calling method.
Citation Information
Patent Citations
Abnormal behavior video monitoring method and device for smart home old-age care
CN115631605A
Wearable emergency call device and system
CN115880857A
Control system based on motion sensing game controller
CN118001713A
