WO2008129452A1 - Multi-sensory fall detection system - Google Patents

Multi-sensory fall detection system Download PDF

Info

Publication number
WO2008129452A1
WO2008129452A1 PCT/IB2008/051425 IB2008051425W WO2008129452A1 WO 2008129452 A1 WO2008129452 A1 WO 2008129452A1 IB 2008051425 W IB2008051425 W IB 2008051425W WO 2008129452 A1 WO2008129452 A1 WO 2008129452A1
Authority
WO
WIPO (PCT)
Prior art keywords
sensors
logic
fall detection
detection system
data
Prior art date
Application number
PCT/IB2008/051425
Other languages
French (fr)
Inventor
Ningjiang Chen
Yang Peng
Original Assignee
Koninklijke Philips Electronics N.V.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips Electronics N.V. filed Critical Koninklijke Philips Electronics N.V.
Publication of WO2008129452A1 publication Critical patent/WO2008129452A1/en

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/1116—Determining posture transitions
    • A61B5/1117—Fall detection
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
    • G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall

Definitions

  • the present invention relates to a fall detection system for automatically detecting a fall accident of a user.
  • a medical emergency aiding system is important for seniors when an accident occurs. As is well known, fall accidents are important causes why a senior is unable to attend to himself or herself. In some regions, fall accidents even rank third in causes of accidental death. In addition, seniors may reduce their social activities due to fear of falling, thus becoming depressed and stiff. This decreases their body coordinating ability, which in turn increases the likelihood of falling.
  • Some portable manual self-aiding apparatus designed for seniors are commercially available. In case of an accident, they can press a help button for requesting help. However, this cannot solve all problems. For example, if a senior becomes delirious or too tense (?) after having a fall, he may lose the ability to press the button. Therefore, the automatic fall detection is necessary in a medical emergency aiding system.
  • the present invention provides a multi- sensory fall detection system in which sensors are in complementary and cooperating relationship.
  • the multi-sensory system when one or more sensors encounter a malfunction such as damage, a power break or abnormal communication among the sensors, the multi-sensory fall detection system can still work properly to accurately detect a fall.
  • the present invention provides a multi-sensory fall detection system capable of automatically detecting the number of sensors in normal operation (sensors operating as a function of a predetermined object, without any damage, power break or communication failure) and of automatically adjusting the fall detection system accordingly.
  • Fig. 1 is an architecture diagram showing a multi-sensory fall detection system
  • Fig. 2 is a flow chart showing the determination of a fall detection result from a certain sensor in various cases
  • Fig. 3 shows the progress of the fall detection flow chart for a decision-making sensor
  • Fig. 4 shows the progress of the fall detection flow chart for a non-decision-making sensor.
  • the advantage of the multi-sensory fall detection system over the single-sensory fall detection system resides in that most occurrences of error warning signals are avoided by comparing and analyzing associated signals from different sensors.
  • the multi-sensory fall detection scheme may become unreliable when one or more sensors fail.
  • the entire (fall detection) system must adjust the manner of operation thereof by analyzing the number of fault-free sensors.
  • Multiple sensors should be arranged at specific locations of a human body to establish a complementary relationship, so as to recognize true and false fall occurrences and prevent error warnings. For example, according to a specific calculating method:
  • a faultless sensor refers to a sensor that can operate according to a designed principle (?) without any damage, power break or failure.
  • data from this sensor should match, or not match, that from other sensors.
  • the first is real-time detection of the number of faultless sensors.
  • the second is adjustment of the detection algorithm according to the number of faultless sensors.
  • Each sensor is placed in a specified location of the body, such as waist, wrist or ankle.
  • the simplest method is to design each sensor in a form (e.g. package) such that it can be fixed in a convenient manner at a certain part of the body.
  • Each sensor has a corresponding identifier number for recognizing its fixed position. For each fixed position, the degree of matching with sensors at other positions, either for a fall or for daily activities, should be calculated in advance.
  • the basic fall detection logic is essentially a modularized framework.
  • Each sensor node includes all necessary logic modules (single-sensory detection logic, decision-making logic, matching degree analyzing logic and control logic). All the modules have an active mode and an inactive mode. Only one specified sensor operates as a core detection module (i.e. a decision-making logic node). For this sensor, all logic modules are in the active mode.
  • the main portion of fall detection is run at the decision-making logic node, except for the single- sensory detection algorithm. However, the matching degree data of other sensors are also considered.
  • the single- sensory detection logic and the control logic are activated while the matching analyzing logic and the decision-making logic are not activated.
  • Their detection logic is established on the basis of the decision-making node. They jointly constitute a detection framework for accurately detecting fall occurrences and avoiding error warning signals. All the sensors may become the decision-making node as long as the decision-making module and the matching degree analyzing module are activated.
  • the decision-making node sends the detection result to the external world through for example gateway@home (e.g. a home communicator) and is connected to a service center, as shown in Fig. 1.
  • gateway@home e.g. a home communicator
  • Fig. 1 a service center
  • the nodes (?) are subject to routine inspection to determine whether the sensor nodes can work properly (e.g. power break, malfunction, damage, abnormal matching value or large variation) . If a sensor node cannot work properly, it will notify other nodes and then exit the detection logic automatically.
  • Fig. 2 is a flow chart showing the determination of a fall detection result of a certain sensor in various cases, wherein the sensor is powered. Regardless of the number of faultless sensors, the result of fall detection can be determined by this flow chart. If any sensor is damaged, power is off or it cannot be operated, it will exit the ad-hoc network. Once the sensor is repaired, it can join the network again.
  • Fig. 3 and Fig. 4 show respectively the progress of the fall detection flow chart for decision-making and non-decision-making sensors.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Pathology (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Social Psychology (AREA)
  • General Physics & Mathematics (AREA)
  • Physiology (AREA)
  • Dentistry (AREA)
  • Psychology (AREA)
  • Psychiatry (AREA)
  • Biophysics (AREA)
  • Gerontology & Geriatric Medicine (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Alarm Systems (AREA)

Abstract

The present invention relates to a fall detection system for automatically detecting a fall accident of a user. The detection system comprises: at least two sensors to be worn by a user, each capable of acquiring motion data of the user's body independently; and a data analyzing system for analyzing the data acquired by the sensors. The data analyzing system comprises a modularized framework in which there is one node at each position where a sensor is located, the node comprising 4 logic modules including a single- sensory detection logic module, a decision-making logic module, a matching degree analyzing logic module and a control logic module. A specified sensor operates in the manner of the core detection module, the node at which the specified sensor is located is a decision-making logic node and the nodes at which other sensors are located are member nodes. All logic modules of the decision-making logic node are in the active mode. For member nodes, the single-sensory detection logic and the control logic module thereof are activated while the matching degree analyzing logic and the decision-making logic thereof are not activated.

Description

MULTI-SENSORY FALL DETECTION SYSTEM
Technical Field
The present invention relates to a fall detection system for automatically detecting a fall accident of a user.
Background Art
A medical emergency aiding system is important for seniors when an accident occurs. As is well known, fall accidents are important causes why a senior is unable to attend to himself or herself. In some regions, fall accidents even rank third in causes of accidental death. In addition, seniors may reduce their social activities due to fear of falling, thus becoming depressed and stiff. This decreases their body coordinating ability, which in turn increases the likelihood of falling. Some portable manual self-aiding apparatus designed for seniors are commercially available. In case of an accident, they can press a help button for requesting help. However, this cannot solve all problems. For example, if a senior becomes delirious or too tense (?) after having a fall, he may lose the ability to press the button. Therefore, the automatic fall detection is necessary in a medical emergency aiding system.
Existing fall detection systems in the market, such as accelerometers and vibration sensors, basically use a single sensor. These products can detect a fall when a person actually falls. There is the problem that these products often issue false warning signals when a person is not falling. This on the one hand annoys a user, and on the other hand keeps service providers busy handling error warnings while missing true warning signals.
Summary of the Invention A significant drawback of the existing single- sensory fall detection system is that error warning signals may still be generated in many cases. In order to improve this situation, the present invention provides a multi- sensory fall detection system in which sensors are in complementary and cooperating relationship. In the multi-sensory system, when one or more sensors encounter a malfunction such as damage, a power break or abnormal communication among the sensors, the multi-sensory fall detection system can still work properly to accurately detect a fall.
The present invention provides a multi-sensory fall detection system capable of automatically detecting the number of sensors in normal operation (sensors operating as a function of a predetermined object, without any damage, power break or communication failure) and of automatically adjusting the fall detection system accordingly.
Description of the Drawings For explaining the invention, the exemplary embodiments of the invention will be described in connection with the accompanying drawings, in which:
Fig. 1 is an architecture diagram showing a multi-sensory fall detection system; Fig. 2 is a flow chart showing the determination of a fall detection result from a certain sensor in various cases; Fig. 3 shows the progress of the fall detection flow chart for a decision-making sensor; and
Fig. 4 shows the progress of the fall detection flow chart for a non-decision-making sensor.
Detailed description of the Invention
The advantage of the multi-sensory fall detection system over the single-sensory fall detection system resides in that most occurrences of error warning signals are avoided by comparing and analyzing associated signals from different sensors. However, the multi-sensory fall detection scheme may become unreliable when one or more sensors fail. Theoretically, the entire (fall detection) system must adjust the manner of operation thereof by analyzing the number of fault-free sensors.
Multiple sensors should be arranged at specific locations of a human body to establish a complementary relationship, so as to recognize true and false fall occurrences and prevent error warnings. For example, according to a specific calculating method:
1. when a user falls, data from sensors at ankles should match that from sensors at the waist.
2. in most motion activities such as jumping, dust collecting and squatting, data from the sensors at ankles will not match that from the sensors at the waist.
3. in motionless activities such as sitting in a chair or standing, data from the sensors at ankles will match that from the sensors at the waist.
Therefore, if data from the sensors at ankles matches that from the sensors at the waist well, it is most likely that the user is either falling or holding still. Further, motion and rest can be distinguished from the variation of data from sensors. Therefore, the degree to which two sensors' data match each other can be used to determine whether the user is falling. The use of the degree of data matching needs to be adjusted accordingly if the sensors are placed at other locations.
Hereinafter, a faultless sensor refers to a sensor that can operate according to a designed principle (?) without any damage, power break or failure. In addition, while the user is falling, data from this sensor should match, or not match, that from other sensors.
At least two inventive steps are involved in the present invention. The first is real-time detection of the number of faultless sensors. The second is adjustment of the detection algorithm according to the number of faultless sensors.
Each sensor is placed in a specified location of the body, such as waist, wrist or ankle. Under this premise, the simplest method is to design each sensor in a form (e.g. package) such that it can be fixed in a convenient manner at a certain part of the body. Each sensor has a corresponding identifier number for recognizing its fixed position. For each fixed position, the degree of matching with sensors at other positions, either for a fall or for daily activities, should be calculated in advance.
As shown in Fig.l, the basic fall detection logic is essentially a modularized framework. Each sensor node includes all necessary logic modules (single-sensory detection logic, decision-making logic, matching degree analyzing logic and control logic). All the modules have an active mode and an inactive mode. Only one specified sensor operates as a core detection module (i.e. a decision-making logic node). For this sensor, all logic modules are in the active mode. The main portion of fall detection is run at the decision-making logic node, except for the single- sensory detection algorithm. However, the matching degree data of other sensors are also considered. As for other sensor nodes (member nodes), the single- sensory detection logic and the control logic are activated while the matching analyzing logic and the decision-making logic are not activated. Their detection logic is established on the basis of the decision-making node. They jointly constitute a detection framework for accurately detecting fall occurrences and avoiding error warning signals. All the sensors may become the decision-making node as long as the decision-making module and the matching degree analyzing module are activated.
Similar to a sensor network, in a multi-sensory fall detection system, all the sensors are connected to an ad-hoc network in a self-organizing manner. The entire system operates as a self-consistent system.
The decision-making node sends the detection result to the external world through for example gateway@home (e.g. a home communicator) and is connected to a service center, as shown in Fig. 1. For each sensor that is currently activated in the work configuration, the nodes (?) are subject to routine inspection to determine whether the sensor nodes can work properly ( e.g. power break, malfunction, damage, abnormal matching value or large variation) . If a sensor node cannot work properly, it will notify other nodes and then exit the detection logic automatically.
Fig. 2 is a flow chart showing the determination of a fall detection result of a certain sensor in various cases, wherein the sensor is powered. Regardless of the number of faultless sensors, the result of fall detection can be determined by this flow chart. If any sensor is damaged, power is off or it cannot be operated, it will exit the ad-hoc network. Once the sensor is repaired, it can join the network again. Fig. 3 and Fig. 4 show respectively the progress of the fall detection flow chart for decision-making and non-decision-making sensors.
The present invention is in no sense limited to the exemplary embodiments illustrated in the description and the drawings. It will be understood that all combinations of the illustrated and described (parts of) embodiments are intended to be within the description and also fall within the scope of protection of the present invention. Also, many variants are within the scope of the present invention, as defined by the appended claims.

Claims

Claims
1. A fall detection system comprising: at least two sensors to be worn by a user, each configured to acquire motion data of the user's body independently; and a data analyzing system, configured to analyze the data acquired by the sensors.
2. The fall detection system of claim 1, wherein the data analyzing system has a function: detecting whether each of the sensors is in faultless operation by analyzing the data.
3. The fall detection system of claim 1, wherein the data analyzing system comprises a program by means of which the data analyzing system calculates whether or not the user is falling.
4. The fall detection system of claim 2, wherein the data analyzing system comprises a program by means of which the data analyzing system calculates whether or not the user is falling.
5. The fall detection system of claim 4, wherein the data analyzing system adjusts an algorithm of the program according to the data of whether each of the sensors is in proper operation.
6. The fall detection system of claim 1, wherein each of the sensors is placed in a predetermined location of the body.
7. The fall detection system of claim 4, wherein the run-time program calculates whether the user is falling by calculating a relative position change of each of the sensors.
8. The fall detection system of claim 4, wherein the data analyzing system adjusts an algorithm of the program according to the number of faultless sensors.
9. The fall detection system of claim 6, wherein for each predetermined body part, the system calculates in advance a matching degree with sensors at other predetermined parts for at least one of a state where the user is falling and a state where the user is in daily activities.
10. The fall detection system of claim 1, wherein the data analyzing system comprises a modularized framework in which there is one node at each position where a sensor is located, and the node comprising 4 logic modules including a single- sensory detection logic module, a decision-making logic module, a matching degree analyzing logic module and a control logic module.
11. The fall detection system of claim 10, wherein each of the logic modules has an active mode and an inactive mode.
12. The fall detection system of claim 11, wherein a specified sensor operates in a manner of the core detection module, the node at which the specified sensor is located is a decision-making logic node and the nodes at which other sensors are located are member nodes.
13. The fall detection system of claim 12, wherein all logic modules of the decision-making logic node are in the active mode.
14. The fall detection system of claim 12, wherein for the member nodes, the single- sensory detection logic and the control logic module thereof are activated while the matching degree analyzing logic and the decision-making logic thereof are not activated.
PCT/IB2008/051425 2007-04-19 2008-04-15 Multi-sensory fall detection system WO2008129452A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200710096644.5 2007-04-19
CN200710096644 2007-04-19

Publications (1)

Publication Number Publication Date
WO2008129452A1 true WO2008129452A1 (en) 2008-10-30

Family

ID=39615764

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2008/051425 WO2008129452A1 (en) 2007-04-19 2008-04-15 Multi-sensory fall detection system

Country Status (2)

Country Link
TW (1) TW200842767A (en)
WO (1) WO2008129452A1 (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010113092A1 (en) * 2009-04-03 2010-10-07 Koninklijke Philips Electronics N.V. Method and system for detecting a fall of a user
WO2011012166A1 (en) * 2009-07-31 2011-02-03 Nec Europe Ltd. System and a method for employing swarms of electronic devices to detect and locate fall victims in an indoor environment
WO2011018168A1 (en) * 2009-08-11 2011-02-17 Schuberth Gmbh System for detecting an accident situation and emergency call activation and method for same
CN102458248A (en) * 2009-06-23 2012-05-16 皇家飞利浦电子股份有限公司 Methods and apparatus for detecting fall of user
WO2018194523A1 (en) 2017-04-19 2018-10-25 National Science And Technology Development Agency System for recording, analyzing risk(s) of accident(s) or need of assistance and providing real-time warning(s) based on continuous sensor signals
WO2020236091A2 (en) 2019-05-17 2020-11-26 National Science And Technology Development Agency Method for detecting falls by using relative barometric pressure signals

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050131736A1 (en) * 2003-12-16 2005-06-16 Adventium Labs And Red Wing Technologies, Inc. Activity monitoring
WO2006101587A2 (en) * 2005-03-22 2006-09-28 Freescale Semiconductor, Inc. System and method for human body fall detection

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050131736A1 (en) * 2003-12-16 2005-06-16 Adventium Labs And Red Wing Technologies, Inc. Activity monitoring
WO2006101587A2 (en) * 2005-03-22 2006-09-28 Freescale Semiconductor, Inc. System and method for human body fall detection

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010113092A1 (en) * 2009-04-03 2010-10-07 Koninklijke Philips Electronics N.V. Method and system for detecting a fall of a user
CN102368948A (en) * 2009-04-03 2012-03-07 皇家飞利浦电子股份有限公司 Method and system for detecting a fall of a user
CN102458248A (en) * 2009-06-23 2012-05-16 皇家飞利浦电子股份有限公司 Methods and apparatus for detecting fall of user
WO2011012166A1 (en) * 2009-07-31 2011-02-03 Nec Europe Ltd. System and a method for employing swarms of electronic devices to detect and locate fall victims in an indoor environment
WO2011018168A1 (en) * 2009-08-11 2011-02-17 Schuberth Gmbh System for detecting an accident situation and emergency call activation and method for same
WO2018194523A1 (en) 2017-04-19 2018-10-25 National Science And Technology Development Agency System for recording, analyzing risk(s) of accident(s) or need of assistance and providing real-time warning(s) based on continuous sensor signals
WO2020236091A2 (en) 2019-05-17 2020-11-26 National Science And Technology Development Agency Method for detecting falls by using relative barometric pressure signals

Also Published As

Publication number Publication date
TW200842767A (en) 2008-11-01

Similar Documents

Publication Publication Date Title
US10485452B2 (en) Fall detection systems and methods
WO2008129452A1 (en) Multi-sensory fall detection system
US9940810B2 (en) Person support apparatuses with patient mobility monitoring
US20160183847A1 (en) Apparatus and method for detecting a fall
CN109035699A (en) The elderly house overall view monitoring nursing system and method based on artificial intelligence
WO2007147012A2 (en) Motion sensing in a wireless rf network
KR20120108335A (en) Emergency monitoring system based on newly developed fall detection algorithm
US9728062B2 (en) Wearable device and system capable of automatically notifying personal emergency, and method thereof
KR20170011631A (en) Wearable device having detection function of fall-down position and system for fall-down position detection using the same
US20230252884A1 (en) Devices, systems and methods for fall detection and preventing false alarms
KR20180057368A (en) Remote management system using living patterns of the elderly people living alone
US10986199B2 (en) Method for tracking persons and state of tracked person and device employing method
Gjoreski et al. Using smartwatch as telecare and fall detection device
KR101754576B1 (en) Biological signal analysis system and method
TWI721576B (en) Nursing system and nursing method
WO2016184189A1 (en) Warning method, wearable device, and computer storage medium
CN109362030A (en) A kind of movable-type intelligent custodial care facility for apartment for elderly people
EP3350786A1 (en) System to detect movement of a user
KR101993599B1 (en) Apparatus for handling emergency situation based on internet of things smartware
JP5722659B2 (en) Bed leaving detector
CN206610414U (en) A kind of the elderly falls down monitoring system
KR20190043737A (en) A sos system with wearable device
KR102232700B1 (en) falldown detection method of patients and senior citizen
JP2004070842A (en) Anomaly notifying system
WO2014098765A1 (en) Medical monitoring systems and methods for using a medical monitoring system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08737849

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08737849

Country of ref document: EP

Kind code of ref document: A1