Fall detection device and method

Abstract

A fall detection device and method are provided. In at least some embodiments, the system can utilize sensing devices such as smartwatches or smartphones (e.g., using iOS, Android, or Pebble operating systems) to read a user's vital signs and apply algorithms to interpret these signs. If the patient is interpreted as having a fall, tremor, or seizure, a notification is sent to a designated caregiver via a reporting process. In at least some embodiments, a doctor or other party can log into a protected dashboard and review patient data in real time. Also in at least some preferred forms, the doctor or other party can analyze the patient's medical history. In at least some embodiments, the user / patient can also use the data to track the onset and progression of falls, tremors, or seizures. Embodiments of the invention can be applied, for example, to situations where the patient / user experiences a condition such as epilepsy that makes the patient / user susceptible to falls and related events.

Description

[0001] This invention is a divisional application of the international patent application filed on October 5, 2017, with international application number PCT / AU2017 / 000209 and national application number CN201780062016.7, entitled "Alarm System". This international application claims priority to Australian patent application AU2016904045, filed on October 5, 2016. Technical Field

[0002] The present invention relates to an alarm system, and more specifically, though not exclusively, to a system suitable for (though not exclusively) assisting in the management of people prone to falls (whether due to medical condition, age or other circumstances). Background Technology

[0003] To date, systems for monitoring people have not been specifically designed to detect selected conditions, including one or more specific conditions (i.e., falls, seizures, or sleepwalking events or related events), to systematically analyze events and transmit them locally and to remote locations.

[0004] US 9689887, transferred to Amazon Technologies, describes a method for detecting fall events associated with packages, etc.

[0005] However, due to the complexity and variability of how humans can fall to the ground, different methods are needed to detect human falls.

[0006] In this particular form, the primary sensing will be performed by sensors worn on the body (more specifically, limb-mounted sensors, and more specifically, sensors mounted on the wrist). Again, there are complexities associated with using limbs to sense movements relevant to the entire human body.

[0007] One object of the present invention is to solve or at least improve some of the disadvantages mentioned above.

[0008] It is also advantageous if the alarm system can be applied to sensing, analyzing, and transmitting situations other than the fall conditions mentioned above, or other situations besides the fall conditions mentioned above, thus providing a multi-functional alarm system.

[0009] Notice

[0010] The term “comprising” (and its grammatical variations) is used in this specification in the inclusive sense of “having” or “including”, rather than in the exclusive sense of “consisting of only”.

[0011] The above discussion of the prior art in the background of this invention is not an admission that any information discussed herein is part of the prior art that can be cited or common knowledge to those skilled in the art in any country. Summary of the Invention

[0012] definition:

[0013] In this specification, a body-worn sensor or wearable device sensor is a sensor mechanically associated with a user's body, such that the sensor can at least sense the acceleration of the body relative to a frame of reference. In a particular form, the primary sensing in embodiments of the invention will be performed by a body-worn sensor (more specifically, a limb-mounted sensor, and more specifically, a sensor mounted on the wrist).

[0014] In this specification, the frame of reference is the frame of reference related to the acceleration of the sensed body. In a preferred embodiment, the frame of reference will be the surface supporting the user. In most cases, the frame of reference will be the ground. If the user has already moved relative to the ground (e.g., they are in an elevator, airplane, or other moving vehicle), then the frame of reference will be that elevator, airplane, or vehicle, and more specifically, the surface inside the vehicle, elevator, or airplane supporting the user.

[0015] Therefore, in a broader form of the invention, an alarm system is provided for transmitting events sensed by sensors worn on the body.

[0016] Preferably, the wearable sensors are mechanically associated with the body.

[0017] Preferably, the event is a fall event.

[0018] Preferably, the sensor includes a processor that communicates with a memory for airborne processing of at least one signal.

[0019] Preferably, the sensor includes a timer.

[0020] Preferably, the sensor includes a GPS device.

[0021] Preferably, the sensor includes a communication device.

[0022] Preferably, the communication equipment includes broadband network interconnectivity for connecting to the Internet.

[0023] Preferably, the communication device includes cellular telephone network interconnectivity for connecting the device to a local cellular telephone network.

[0024] Preferably, the sensor includes an accelerometer.

[0025] Preferably, at least one signal is an acceleration signal.

[0026] Preferably, at least one signal is a timing signal.

[0027] Preferably, the signal is an acceleration signal derived from the accelerometer.

[0028] Preferably, the signal is a timing signal derived from a timer.

[0029] Preferably, the signal is a GPS signal derived from a GPS device.

[0030] Preferably, the event is a fall event.

[0031] Preferably, the event is an epileptic seizure event.

[0032] Preferably, the event is a sleepwalking event.

[0033] In a preferred embodiment, the system also includes additional monitoring or sensing devices.

[0034] Preferably, the additional monitoring or sensing device includes at least a speaker and a microphone, and communicates with a web-enabled server.

[0035] Preferably, a web-enabled server executes the application, thereby supplementing the functionality of the wearable sensors with the functionality of additional monitoring or sensing devices.

[0036] Preferably, the wearable sensor is mounted on the user's wrist.

[0037] Preferably, the artificial intelligence (AI) capability is programmed into the sensor's memory 18 for execution by the sensor's processor 1 worn on the body.

[0038] Preferably, the AI ​​program is executed on a processor associated with a server located remotely from sensor 14.

[0039] Preferably, the AI ​​capability learns from the determination of false positive and false negative events in order to statistically improve the reliability of event detection over time, with particular reference to the learned attributes of data associated with any given user 12.

[0040] In another broader form of the invention, a fall detection device is provided, comprising:

[0041] The accelerometer transmits acceleration signals to the processor.

[0042] The acceleration signal is essentially continuous relative to the reference frame, quantizing the acceleration.

[0043] The timer transmits a time reference signal to the processor.

[0044] The processor essentially monitors the acceleration signal continuously;

[0045] The processor essentially monitors timing signals continuously;

[0046] And thus, if the acceleration signal is within a first low acceleration range during a predetermined time period and is followed by a second high acceleration signal during a second predetermined time period, the processor determines the fall condition.

[0047] Preferably, the processor monitors the timing signal and the acceleration signal during a third predetermined time period after the second predetermined time period. Thus, if the acceleration signal remains within a predetermined very low range during the third predetermined time period, it is determined that the user is not moving and a fall detection event is confirmed.

[0048] Preferably, when the processor determines a fall, a fall signal is sent to a remote location.

[0049] Preferably, when the processor determines a fall, a fall signal is transmitted locally.

[0050] Preferably, the acceleration signal is referenced relative to a reference frame.

[0051] Preferably, the reference frame is the surface of the user supporting the fall detection device.

[0052] Preferably, the fall detection device is a fall detection device installed on the wrist.

[0053] In another broader form of the invention, a detection and communication system is provided that uses sensing devices to read a user's vital signs and applies algorithms to interpret those signs. If the user is interpreted as having fallen, trembling, or having an epileptic seizure, a notification is sent to a designated caregiver through a reporting process.

[0054] Preferably, the device is a smartwatch or smartphone (e.g., using the iOS, Android or Pebble operating system).

[0055] Preferably, doctors or other parties can log in to the protected dashboard and check user data in real time.

[0056] Preferably, doctors or other parties can analyze the user's medical history.

[0057] Preferably, users / patients can also use user data exported by the system to track falls, tremors, or epileptic seizures and monitor their progression.

[0058] In another broader form of the invention, an epileptic seizure detection device is provided, comprising:

[0059] The accelerometer transmits acceleration signals to the processor.

[0060] The acceleration signal is essentially continuous relative to the reference frame, quantizing the acceleration.

[0061] The timer transmits a time reference signal to the processor.

[0062] The processor essentially monitors the acceleration signal continuously;

[0063] The processor essentially monitors timing signals continuously;

[0064] And thus, if the acceleration signal oscillates within a predetermined range over a predetermined time period, a concurrent signal is determined to notify of an epileptic seizure event.

[0065] Preferably, the epileptic seizure detection device is an epileptic seizure detection device installed on the wrist.

[0066] In another broader form of the invention, a sleepwalking detection device is provided, comprising:

[0067] The accelerometer transmits acceleration signals to the processor.

[0068] The acceleration signal is essentially continuous relative to the reference frame, quantizing the acceleration.

[0069] The timer transmits a time reference signal to the processor.

[0070] The processor essentially monitors the acceleration signal continuously;

[0071] The processor essentially monitors timing signals continuously;

[0072] And thus, if the acceleration signal indicates walking movement during a predetermined period exceeding the minimum walking time and determined to be the user's bedtime, a concurrent signal is determined to notify of a sleepwalking event.

[0073] Preferably, the sleepwalking detection device is a sleepwalking detection device installed on the wrist.

[0074] In another broader form of the invention, a method for detecting a fall event is provided, comprising:

[0075] It provides an accelerometer that transmits acceleration signals to the processor;

[0076] The acceleration signal is essentially continuous relative to the reference frame, quantizing the acceleration.

[0077] Provides a timer that transmits a time reference signal to the processor;

[0078] The processor essentially monitors the acceleration signal continuously;

[0079] The processor essentially monitors timing signals continuously;

[0080] And thus, if the acceleration signal is within a first low acceleration range during a predetermined time period and is followed by a second high acceleration signal during a second predetermined time period, the processor determines the fall condition.

[0081] In another broader form of the invention, a method for detecting epileptic seizures is provided, comprising:

[0082] An accelerometer is provided that transmits acceleration signals to a processor;

[0083] The acceleration signal is essentially continuous relative to the reference frame, quantizing the acceleration.

[0084] Provides a timer that transmits a time reference signal to the processor;

[0085] The processor essentially monitors the acceleration signal continuously;

[0086] The processor essentially monitors timing signals continuously;

[0087] And thus, if the acceleration signal oscillates within a predetermined range over a predetermined time period, a concurrent signal is determined to notify of an epileptic seizure event.

[0088] In another, broader form of the invention, a method for detecting sleepwalking events is provided, comprising:

[0089] It provides an accelerometer that transmits acceleration signals to the processor;

[0090] The acceleration signal is essentially continuous relative to the reference frame, quantizing the acceleration.

[0091] Provides a timer that transmits a time reference signal to the processor;

[0092] The processor essentially monitors the acceleration signal continuously;

[0093] The processor essentially monitors timing signals continuously;

[0094] And thus, if the acceleration signal indicates walking movement during a predetermined period exceeding the minimum walking time and determined to be the user's bedtime, a concurrent signal is determined to notify of a sleepwalking event. Attached Figure Description

[0095] Embodiments of the invention will now be described with reference to the accompanying drawings, in which:

[0096] Figure 1 This is a logic flowchart of an alarm system according to an embodiment of the present invention;

[0097] Figure 2 It can be applied to Figure 1The flowchart of the fall detection algorithm of the system;

[0098] Figure 3 It can be applied to Figure 1 The flowchart of the system's epileptic seizure detection algorithm;

[0099] Figure 4 It can be applied to Figure 1 The flowchart of the system's sleepwalking detection algorithm;

[0100] Figure 5 yes Figure 1 Electronic block diagram of the system implementation;

[0101] Figure 6 yes Figure 1 Electronic block diagram of another implementation of the system;

[0102] Figure 7 yes Figure 1 Another implementation of the system is shown in the electronic block diagram. Detailed Implementation

[0103] In summary, the disclosed invention relates to an apparatus, method, and system in at least some embodiments that can utilize sensing devices such as smartwatches or smartphones (e.g., using iOS, Android, or Pebble or Tizen operating systems) to read a user's vital signs, apply algorithms to interpret these signs, and then, if the patient is interpreted as having fallen, trembling, or experiencing an epileptic seizure, send a notification to a designated caregiver via an escalation process. In at least some embodiments, a doctor or other party can log into a protected dashboard and review patient data in real time. Similarly, in at least some preferred forms, a doctor or other party can analyze the patient's medical history.

[0104] In at least some embodiments, users / patients can also use the data to track falls, tremors, or the onset of epileptic seizures and monitor their progression.

[0105] Embodiments of the present invention can be applied to situations, for example, where a patient / user suffers from a condition such as epilepsy that makes the patient / user prone to falls and related events.

[0106] refer to Figure 1 and Figure 5 An alarm system 10 according to a first embodiment of the present invention is shown.

[0107] In this configuration, the alarm system 10 monitors and analyzes the data derived from the sensor 11. In a preferred embodiment, the sensor 11 is a wearable sensor. In a particular embodiment, it may be strapped to the wrist of the user 12. In other embodiments, it may be mounted on the chest, ankle, or otherwise, creating a mechanical link between the sensor 11 and the user 12's body sufficient to allow the sensor to detect parameters associated with the user 12's body.

[0108] These parameters can include the movement of the body relative to a frame of reference. In a preferred embodiment, the frame of reference will be the surface supporting the user 12.

[0109] Other parameters may include physiological parameters such as heart rate, ECG waveform, EEG waveform, blood pressure, blood sugar, sweat, body temperature, etc.

[0110] Other parameters may include geographic location information and data such as that exported from a GPS module. Figure 6 An embodiment of a device incorporating GPS capability is shown in the figure. Figure 6 The same components are numbered the same as in the first embodiment, except that they are in the 100 series. In this case, in addition to the timer module 119, the acceleration sensing module 120, and the communication module 121, a GPS module 34 that communicates with satellite 35 is also included, and optionally, a Wi-Fi signal that can be provided by Wi-Fi router 124 is also included.

[0111] In some cases, user 12 will be referred to as a patient, although there are also contexts where using alarm system 10 would make user 12 an object monitored by system 10, but describing him as a "patient" may not be appropriate.

[0112] In summary, System 10 includes networked components, and in most cases, these components will be geographically separated from each other.

[0113] In a particular form, system 10 includes sensors 11 mechanically associated with user 12, which communicate with server 14. In many cases, the sensors and / or server 14 will also communicate with caregiver digital communication device 15, and additionally, separately with call center digital communication device 16.

[0114] In a particularly preferred form, the sensor 11 is in the form of a wearable device, preferably attached to the wrist of the user 12.

[0115] Sensor 11 includes a processor 17, a memory 18, a timer module 19, an acceleration sensing module 20, and a communication module 21, or communicates locally with them. In a preferred embodiment, components 17, 18, 19, 20, and 21 communicate with each other via bus 22.

[0116] In another distribution configuration, at least the acceleration detection module and the communication module can communicate with other components forming sensor 11 via Bluetooth or other short-range radio or electromagnetic transmission capabilities.

[0117] In a preferred embodiment, the acceleration sensing module 20 is implemented as at least a three-axis accelerometer, which allows acceleration to be decomposed on three orthogonal axes.

[0118] The communication module 21 can communicate with the Internet 23 or other wide area networks via Wi-Fi router 24 or via cellular telephone network 25, thereby enabling sensor 11 to communicate with server 14, caregiver digital communication device 15 and call center digital communication device 16.

[0119] System 10 also includes a scheduler 36, which, in a preferred embodiment, executes as an application on server 14. The primary function of scheduler 36 is to start and stop the monitoring implemented by sensor 11.

[0120] In a specific form, the function is to automatically start monitoring the application on sensor 11 in the morning and turn it off at night for the detection of fall and seizure events. For sleepwalking event detection, it will start at bedtime and turn off in the morning.

[0121] In use

[0122] Initially in Figure 1 To see it best, Figure 5 The arrangement is designed to at least monitor accelerometer data and apply a reference algorithm to at least the timing data derived from timer module 19 in order to determine whether a fall condition / event (such as...) has occurred. Figure 2 As shown in the flowchart), whether an epileptic seizure event has been detected (according to...) Figure 3 (Flowchart) or whether a sleepwalking event has been detected (see reference) Figure 4 (Flowchart).

[0123] Then, according to Figure 1 The flowchart shows how events are transmitted to one or more of the following: server 14, caregiver digital communication device 15, and call center digital communication device 16.

[0124] In a particular form, the event is also transmitted locally to user 12. In a preferred form, the event is transmitted locally via display 26 associated with sensor 11.

[0125] In a preferred embodiment, display 26 may be a touch-sensitive display (or voice-activated, Apple Siri or OK Google Assistance), thereby allowing the user to communicate with one or more of server 14, caregiver digital communication device 15, or call center digital communication device 16.

[0126] Integrated sensors and communication devices

[0127] In a particularly preferred form, sensors 11, 111, and 211 can be implemented as a smartwatch application running on a standalone smartwatch (e.g., an Apple Watch Series 3 or an LG Urbane LTE smartwatch) with an integrated SIM or eSIM card.

[0128] Machine learning adaptation

[0129] In a particularly preferred form, artificial intelligence (AI) capabilities can be programmed into memory 18 for execution by processor 17. Alternatively, or additionally, the AI ​​program can execute on a processor associated with server 14. A particular application of the AI ​​capabilities is learning from the identification of false positive and false negative events to statistically improve the reliability of event detection over time, with particular reference to the learned attributes of data associated with any given user 12.

[0130] Sleepwalking detection

[0131] Combination Figure 4 refer to Figure 1 The instructions for the algorithm can be stored in memory 18 and executed by processor 17, which, according to... Figure 4 The flowchart operation is used to detect, transmit, and warn of sleepwalking events as appropriate.

[0132] Heart rate monitoring event detection

[0133] In a preferred embodiment, sensor 11 may include ECG monitoring capabilities, thereby enabling heart rate monitoring to provide alerts to patients and caregivers when abnormal heart rates / pulses are recorded.

[0134] Audio function

[0135] An audio alert is issued to people in the vicinity and emergency services when an event such as a fall, seizure, or sleepwalking is detected. In a preferred form, this is achieved by a sensor that emits an audible sound. In a particularly preferred form, the sound is loud enough for people in the vicinity to hear it.

[0136] Sensor condition monitoring and communication

[0137] The app's add-on capability is used to send notifications to caregivers about the app's monitoring status (ensuring the app is monitoring) and the watch's battery level, so caregivers can contact the patient if there are any issues with app monitoring.

[0138] Integration with other systems—Remote health

[0139] In a special form and reference Figure 7 The same components are numbered the same as in the first embodiment, except that they are from the 200 series. The additional monitoring or sensor device 27 can be located in association with the user 12. In a preferred embodiment, the additional monitoring or sensor device can be located in the user's home, office, or other locations where the user may spend a predetermined period of time.

[0140] The additional monitoring or sensor device 27 includes functions and communication capabilities similar to those of sensor 11, but more specifically includes at least microphone 28, and in a preferred form, also includes speaker 29 communicating with bus 30, which also communicates with processor 31 and memory 32 and thus with Wi-Fi router 224, Internet 223 and subsequently Web-enabled database 33.

[0141] In a particular form, the additional monitoring or sensor device 27 may take the form of a smart microphone and speaker device currently sold as Amazon Echo or Google Home devices or Apple's HomePod.

[0142] These devices typically allow audio pickup from the entire room and also allow audio playback throughout the room. Third-party applications can run on a web-enabled server to provide specific functionalities to supplement the basic functionality, which may include speech recognition and enabling voice commands by communicating with other nearby devices.

[0143] In this example, this arrangement facilitates remote health functionality, enabling users at home to communicate with caregivers and emergency workers, at least using the voice recognition system built into the attached monitoring or sensing device 27. In a preferred form, the application would be loaded onto a web-enabled server 33, which, when executed, integrates the functionality of the attached monitoring or sensing device 27 with that of the sensor 211.

[0144] In a particular form, this combination of features offers the integration of powerful wearable sensors with local indoor sensors, which at least have audio pickup and playback capabilities.

[0145] Industrial applicability

[0146] Embodiments of the present invention are applicable whenever it is desired to monitor and transmit conditions or events associated with a user.

[0147] In a particular form, the system has applications for detecting falls and transmitting the data to a remote location for assistance or at least monitoring.

[0148] In at least some embodiments, the system can be advantageously applied to utilize sensing devices such as smartwatches or smartphones (e.g., using iOS, Android, or Pebble operating systems) to read a user's vital signs, and to apply algorithms to interpret these signs. If the patient is interpreted as having fallen, trembling, or experiencing an epileptic seizure, a notification is sent to a designated caregiver via a reporting process. In at least some embodiments, a doctor or other party can log into a protected dashboard and review patient data in real time. Also in at least some preferred forms, the doctor or other party can analyze the patient's medical history.

[0149] In at least some embodiments, users / patients can also use the data to track falls, tremors, or the onset of epileptic seizures and monitor their progression.

[0150] Embodiments of the present invention can be applied to situations, for example, where a patient / user suffers from a condition such as epilepsy that makes the patient / user prone to falls and related events.

[0151] The above description only describes some embodiments of the present invention, and modifications that are obvious to those skilled in the art can be made to the embodiments without departing from the scope of the present invention.

Claims

1. A fall detection device, comprising: The processor is configured to monitor acceleration and timing events; An accelerometer is configured to detect acceleration events and transmit information related to the acceleration events to the processor; as well as A timer is configured to transmit information related to timing events to the processor; in: If the processor receives information about an acceleration event that is within a first low acceleration range during a first predetermined time period and is followed by a high acceleration event during a second predetermined time period, the processor determines a fall condition; and If, throughout the entire third predetermined time period, the processor receives information on one or more acceleration events that fall within the second low acceleration range, then the processor confirms the fall condition; and Specifically, the acceleration events are monitored substantially continuously between the first predetermined time period, the second predetermined time period, and the third predetermined time period, so that there are no time periods in which the processor does not monitor acceleration events.

2. The fall detection device according to claim 1 further includes a communication module; wherein, When the fall is confirmed, the fall signal is sent to a remote location via the communication module.

3. The fall detection device according to claim 1, wherein, The acceleration events are determined based on a reference frame.

4. The fall detection device according to claim 3, wherein, The reference frame is the surface of the user supporting the fall detection device.

5. The fall detection device according to claim 1, wherein, The fall detection device is a fall detection device installed on the wrist.

6. The fall detection device according to claim 1 further includes a storage module configured to communicate with the processor.

7. The fall detection device according to claim 6, wherein, Artificial intelligence (AI) functions are programmed into the storage module and configured to execute instructions via the processor.

8. The fall detection device according to claim 7, wherein, The AI ​​function learns from the determination of false positive and false negative events of the fall event in order to statistically improve the reliability of the detection of the fall event over time, with reference to the learned attributes of data associated with any given user.

9. The fall detection device of claim 1 further includes a Global Positioning System (GPS) device configured to communicate with the processor.

10. The fall detection device according to claim 1, wherein, The processor is configured to receive and determine an acceleration event immediately following the second predetermined time period, and the second low acceleration range is a customizable range.

11. A method for detecting fall events, comprising: Use accelerometers to identify acceleration events; Use a timer to identify the time period; as well as The acceleration event and the timing period are transmitted to the processor; in: The processor is configured to determine a fall condition if it receives an acceleration event that is in a first low acceleration range during a first predetermined time period and is followed by a high acceleration event during a second predetermined time period. The processor is configured to confirm the fall condition if, during the entire third predetermined time period, the processor receives information indicating that one or more acceleration events are below a second low acceleration threshold during the third predetermined time period; and The processor is configured to continuously receive acceleration events from the first predetermined time period to the third predetermined time period.

12. The method for detecting a fall event according to claim 11, further comprising: When the fall is confirmed, the fall event is sent to a remote location.

13. The method for detecting fall events according to claim 11, wherein, The acceleration events are determined based on a reference frame.

14. The method for detecting fall events according to claim 13, wherein, The reference frame is the surface of the user supporting the fall detection device.

15. The method for detecting fall events according to claim 1, wherein, The method is performed in part using a fall detection device attached to the user's wrist.

16. The method for detecting fall events according to claim 11, wherein, The storage module is communicatively coupled to the processor.

17. The method for detecting a fall event according to claim 16, further comprising: The processor executes the instructions corresponding to the artificial intelligence (AI) functions stored in the storage module.

18. The method for detecting a fall event according to claim 17, wherein, The AI ​​function learns from the determination of false positive and false negative events of the fall event in order to statistically improve the reliability of the detection of the fall event over time, with reference to the learned attributes of data associated with any given user.

19. The method for detecting fall events according to claim 1, wherein, The processor is communicatively coupled to a Global Positioning System (GPS) device.

20. The method for detecting fall events according to claim 1, wherein: The processor is configured to receive and determine the acceleration event immediately following the second predetermined time period; and The second low acceleration range is a customizable range.

Images