Fall detection system using a combination of accelerometer, audio input and magnetometer

• US 8,843,101 B2
• Filed: 09/20/2011
• Issued: 09/23/2014
• Est. Priority Date: 10/04/2010
• Status: Active Grant

• Current Assignee: NICE North America LLC
• Original Assignee: Numera, Inc. (Innova Pure Water Incorporated)
• Inventors: Fish, Ram David Adva, Messenger, Henry, Baryudin, Leonid, Dardashti, Soroush Salehian, Goldshtein, Evgenia
• Primary Examiner(s): Prez-Gutirrez, Rafael
• Assistant Examiner(s): DONADO, FRANK E
• Application Number:
  US13/237,857
• Publication Number:
  US 20120083237A1
• Time in Patent Office: 1,099 Days
• Field of Search: 455/404.1, 455/67.13
• US Class Current: 455/404.1
• CPC Class Codes: G01P 1/07 Indicating devices, e.g. for remote indication indicating working conditions of vehicles G07C5/00
• CPC Class Codes: H04M 1/72421 with automatic activation of emergency service functions, e.g. upon sensing an alarm
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2250/02 including a Bluetooth interface
• CPC Class Codes: H04M 2250/12 including a sensor for measuring a physical value, e.g. temperature or motion
• CPC Class Codes: H04W 4/90 Services for handling of emergency or hazardous situations, e.g. earthquake and tsunami warning systems [ETWS]

Abstract

A wearable device for detecting a user state is disclosed. The wearable device includes an accelerometer for measuring an acceleration of a user, a magnetometer for measuring a magnetic field associated with the user'"'"'s change of orientation, a microphone for receiving audio, a memory for storing the audio, and at least one processor communicatively connected to the accelerometer, the magnetometer, the microphone, and the memory. The processor is identified to declare a measured acceleration as a suspected user state, and to categorize the suspected user state based on the stored audio as one of an activity of daily life (ADL), a confirmed user state, or an inconclusive event.

Claims

1. An apparatus wearable by a user for detecting a user state, comprising:

an accelerometer for measuring an acceleration of the user;

a magnetometer for measuring a magnetic field associated with the user and calculation of an orientation of the user based on a measured magnetic field;

a microphone for receiving audio;

a memory for storing the audio; and

at least one processor communicatively connected to the accelerometer, the magnetometer, the microphone, and the memory, and configured to;

identify a measured acceleration and a change of orientation of the user as a suspected user state;

extract at least one feature from the stored audio and at least one of the measured acceleration and the change of orientation, where the at least one feature is an inter-signal dynamic property that is based on relationships between audio energy and physical movement, and the inter-signal dynamic property is one of elapsed time between acceleration and audio peaks and acceleration and rotation rate peaks;

categorize the suspected user state based on the at least one feature as one of an activity of daily life (ADL), a confirmed user state indicating that a fall has occurred, and an inconclusive event;

transmit the stored audio and at least one of the measured acceleration and the change of orientation to the cloud computing system; and

receive a re-confirmation or change of classification from the cloud computing system based on the stored audio and at least one of the measured acceleration and the change of orientation.

2. The apparatus of claim 1, wherein categorizing the suspected user state based on the stored audio comprises an opt-out mechanism wherein specific sound patterns are used to assess the suspected user state as an ADL.

3. The apparatus of claim 1, wherein categorizing the suspected user state based on the stored audio comprises an opt-in mechanism wherein specific sound are detected and employed to confirm that the fall has occurred.

4. The apparatus of claim 1, further comprising a gyroscope communicatively connected to the at least one processor, wherein the processor is configured to calculate the change of orientation of the user from the gyroscope, the magnetometer, and accelerometer that is more accurate than a change of orientation calculated from the magnetometer and accelerometer alone.

5. The apparatus of claim 4, wherein the at least one processor is further configured to extract at least one feature from the stored audio and at least one of the measured acceleration or the measured magnetic field.

6. The apparatus of claim 5, wherein the at least one feature is an inter-signal dynamic property.

7. The apparatus of claim 6, wherein the inter-signal dynamic property is based on relationships between audio energy and physical movement.

8. The apparatus of claim 7, wherein the inter-signal dynamic property is one of elapsed time between acceleration and audio peaks and acceleration and rotation rate peaks.

9. The apparatus of claim 1, further comprising:

a speaker and a cellular transceiver each communicatively connected to the at least one processor, wherein the at least one processor is configured to employ the speaker, the microphone, and the cellular transceiver to receive a notification and an optional further confirmation from a voice conversation with a call center or the user.

10. A computer implemented method performed by an apparatus wearable by a user for detecting a fall of a user, comprising:

measuring an acceleration of the user;

measuring a magnetic field associated with the user and calculating of an orientation of the user based on a measured magnetic field;

receiving and storing audio;

identifying, by a processor, a measurement of acceleration and a change of orientation of the user as a suspected fall;

extracting at least one feature from the stored audio and at least one of the measured acceleration and the change of orientation, where the at least one feature is an inter-signal dynamic property that is based on relationships between audio energy and physical movement, and the inter-signal dynamic property is one of elapsed time between acceleration and audio peaks and acceleration and rotation rate peaks;

categorizing, by the processor, the suspected fall based on the at least one feature as one of an activity of daily life (ADL), a confirmed fall, and an inconclusive event;

transmitting the stored audio and at least one of the measured acceleration and the change of orientation to a cloud computing system; and

receiving a re-confirmation or change of classification from the cloud computing system based on the stored audio and at least one of the measured acceleration and the change of orientation.

11. The method of claim 10, wherein categorizing the suspected fall based on the stored audio comprises an opt-out mechanism wherein specific sound patterns are used to assess the suspected fall as an ADL.

12. The method of claim 10, wherein categorizing the suspected fall based on the stored audio comprises an opt-in mechanism wherein specific sounds characteristics are detected and employed to confirm that a fall has happened.

13. The method of claim 10, further comprising calculating the change of orientation of the user from the gyroscope, the magnetometer, and accelerometer that is more accurate than a change of orientation calculated from the magnetometer and accelerometer alone.

14. The method of claim 13, further comprising extracting at least one feature from the stored audio and at least one of the measured acceleration and the calculated change of orientation.

15. The method of claim 14, wherein:

the at least one feature is an inter-signal dynamic property is based on relationships between audio energy and physical movement; and

the inter-signal dynamic property is one of elapsed time between acceleration and audio peaks and acceleration and rotation rate peaks.

16. The method of claim 10, further comprising receiving a notification and an optional further confirmation from a voice conversation with a call center or the user.

17. A non-transitory computer readable storage medium including instructions that, when executed by a processor included in an apparatus wearable by a user, cause the processor to perform operations comprising:

measuring an acceleration of the user;

measuring a magnetic field associated with the user and calculating of an orientation of the user based on a measured magnetic field;

receiving and storing audio;

identifying, by the processor, a measurement of acceleration and a change of orientation of the user as a suspected fall;

extracting at least one feature from the stored audio and at least one of the measured acceleration and the change of orientation, where the at least one feature is an inter-signal dynamic property that is based on relationships between audio energy and physical movement, and the inter-signal dynamic property is one of elapsed time between acceleration and audio peaks and acceleration and rotation rate peaks;

categorizing, by the processor, the suspected fall based on the at least one feature as one of an activity of daily life (ADL), a confirmed fall, and an inconclusive event;

transmitting the stored audio and at least one of the measured acceleration and the change of orientation to a cloud computing system; and

receiving a re-confirmation or change of classification from the cloud computing system based on the stored audio and at least one of the measured acceleration and the change of orientation.

18. The non-transitory computer-readable medium of claim 17, wherein the method further comprises:

calculating the change of orientation of the user from the gyroscope, the magnetometer, and accelerometer that is more accurate than a change of orientation calculated from the magnetometer and accelerometer alone.