U.S. patent application number 14/114546 was filed with the patent office on 2014-05-22 for method for detecting potential falls and a fall detector.
This patent application is currently assigned to KONINKLIJKE PHILIPS N.V.. The applicant listed for this patent is Constant Paul Marie Jozef Baggen, Heribert Baldus, Martin Ouwerkerk, Anthony Robert Andre Schoofs, Warner Rudolph Theophile Ten Kate, Wei Zhang. Invention is credited to Constant Paul Marie Jozef Baggen, Heribert Baldus, Martin Ouwerkerk, Anthony Robert Andre Schoofs, Warner Rudolph Theophile Ten Kate, Wei Zhang.
Application Number | 20140142460 14/114546 |
Document ID | / |
Family ID | 44681507 |
Filed Date | 2014-05-22 |
United States Patent
Application |
20140142460 |
Kind Code |
A1 |
Zhang; Wei ; et al. |
May 22, 2014 |
METHOD FOR DETECTING POTENTIAL FALLS AND A FALL DETECTOR
Abstract
There is provided a fall detector for detecting a potential fall
by a user, the fall detector comprising a sensor for measuring the
skin conductivity of the user; and a processor for analyzing the
skin conductivity measurements to determine whether the user has
had a potential fall, wherein the processor is configured to
analyze the skin conductivity measurements by matching the skin
conductivity measurements to a predetermined skin conductivity
response corresponding to a stressful event.
Inventors: |
Zhang; Wei; (Eindhoven,
NL) ; Baggen; Constant Paul Marie Jozef; (Blerick,
NL) ; Baldus; Heribert; (Aachen, DE) ; Ten
Kate; Warner Rudolph Theophile; (Waalre, NL) ;
Ouwerkerk; Martin; (Culemborg, NL) ; Schoofs; Anthony
Robert Andre; (Dublin, IE) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Zhang; Wei
Baggen; Constant Paul Marie Jozef
Baldus; Heribert
Ten Kate; Warner Rudolph Theophile
Ouwerkerk; Martin
Schoofs; Anthony Robert Andre |
Eindhoven
Blerick
Aachen
Waalre
Culemborg
Dublin |
|
NL
NL
DE
NL
NL
IE |
|
|
Assignee: |
KONINKLIJKE PHILIPS N.V.
EINDHOVEN
NL
|
Family ID: |
44681507 |
Appl. No.: |
14/114546 |
Filed: |
April 25, 2012 |
PCT Filed: |
April 25, 2012 |
PCT NO: |
PCT/IB2012/052067 |
371 Date: |
February 4, 2014 |
Current U.S.
Class: |
600/547 |
Current CPC
Class: |
A61B 2562/0219 20130101;
A61B 5/1117 20130101; G08B 21/02 20130101; A61B 5/021 20130101;
A61B 5/02438 20130101; A61B 5/1116 20130101; A61B 5/7203 20130101;
A61B 5/7275 20130101; A61B 5/0531 20130101; A61B 5/1126 20130101;
A61B 5/747 20130101; A61B 5/02055 20130101; A61B 5/1102 20130101;
H04M 2250/12 20130101; H04M 3/5116 20130101; A61B 5/14551 20130101;
G08B 21/0453 20130101; A61B 5/725 20130101; A61B 5/7282 20130101;
G08B 21/0446 20130101 |
Class at
Publication: |
600/547 |
International
Class: |
A61B 5/11 20060101
A61B005/11; A61B 5/00 20060101 A61B005/00 |
Foreign Application Data
Date |
Code |
Application Number |
Apr 29, 2011 |
EP |
11164240.1 |
Claims
1. A fall detector for detecting a potential fall by a user, the
fall detector comprising: a sensor for measuring the skin
conductivity of the user; and a processor for analyzing the skin
conductivity measurements to determine whether the user has had a
potential fall, wherein the processor is configured to use a
matched filter to match the skin conductivity measurements to a
predetermined skin conductivity response corresponding to a
stressful event to generate skin conductivity coefficients, the
value of each coefficient indicating the match of a respective set
of consecutive skin conductivity measurements to the predetermined
skin conductivity response; and to determine whether the user has
had a potential fall by analyzing the skin conductivity
coefficients.
2. (canceled)
3. A fall detector as claimed in claim 1, wherein the processor is
configured to analyze the skin conductivity coefficients to
determine whether the skin conductivity measurements are consistent
with a user having had a fall by determining one or more of (i) the
maximum value of the coefficients, (ii) the mean of the
coefficients, and/or (iii) the median of the coefficients.
4. A fall detector as claimed in claim 1, wherein the processor is
configured to filter the skin conductivity measurements to remove
noise and/or interference prior to matching the skin conductivity
measurements to the predetermined skin conductivity response.
5. A fall detector as claimed in claim 4, wherein the processor is
configured to use a median filter to remove the noise and/or
interference from the skin conductivity measurements.
6. A fall detector as claimed in claim 1, further comprising at
least one movement sensor for sensing the movement of the user, the
processor being further configured to analyze measurements from the
at least one movement sensor to determine if the user may have
fallen.
7. A fall detector as claimed in claim 6, wherein the processor is
configured to analyze the skin conductivity measurements only in
the event that the measurements from the at least one movement
sensor indicate that the user may have fallen.
8. A fall detector as claimed in claim 6, wherein the processor is
configured to use the skin conductivity sensor to measure the skin
conductivity in the event that the measurements from the at least
one movement sensor indicate that the user may have fallen.
9. A fall detector as claimed in claim 6, wherein the processor is
configured to analyze the skin conductivity measurements obtained
during a time window following a time at which the measurements
from the at least one sensor indicate that the user may have
fallen.
10. A method of detecting a potential fall by a user, the method
comprising: measuring the skin conductivity of the user; and
analyzing the skin conductivity measurements to determine whether
the user has had a potential fall by using a matched filter to
match the skin conductivity measurements to a predetermined skin
conductivity response corresponding to a stressful event to
generate skin conductivity coefficients, the value of each
coefficient indicating the match of a respective set of consecutive
skin conductivity measurements to the predetermined skin
conductivity response, and determining whether the user has had a
potential fall by analyzing the skin conductivity coefficients.
11. (canceled)
12. A method as claimed in claim 11, wherein analyzing the skin
conductivity coefficients to determine whether the response is
consistent with a user having had a fall comprises determining one
or more of (i) the maximum value of the coefficients, (ii) the mean
of the coefficients, and/or (iii) the median of the
coefficients.
13. A method as claimed in claim 10, further comprising the steps
of: sensing the movement of the user; and analyzing the movement of
the user to determine if the user may have fallen.
14. A method as claimed in claim 13, wherein the step of analyzing
the skin conductivity measurements is performed only in the event
that the step of analyzing the movement of the user indicates that
the user may have fallen.
15. A computer program product comprising computer readable code
embodied therein, the computer readable code being configured such
that, upon execution by a suitable computer or processor, the
computer or processor performs the method claimed of claim 10.
Description
TECHNICAL FIELD OF THE INVENTION
[0001] The invention relates to a method for detecting potential
falls by a user and a fall detector, and in particular relates to a
method for detecting a potential fall and a fall detector that
provides increased fall detection reliability.
BACKGROUND TO THE INVENTION
[0002] Falls affect millions of people each year and result in
significant injuries, particularly among the elderly. In fact, it
has been estimated that falls are one of the top three causes of
death in elderly people. A fall is defined as a sudden,
uncontrolled and unintentional downward displacement of the body to
the ground, followed by an impact, after which the body stays down
on the ground.
[0003] Personal Help Buttons (PHBs) are available that require the
user to push the button to summon help in an emergency. However, if
the user suffers a severe fall (for example if they are knocked
unconscious), the user might be unable to push the button, which
might mean that help doesn't arrive for a significant period of
time, particularly if the user lives alone.
[0004] Fall detectors are also available that process the output of
one or more movement sensors to determine if the user has suffered
a fall. Most existing body-worn fall detectors make use of an
accelerometer (usually an accelerometer that measures acceleration
in three dimensions) and they try to infer the occurrence of a fall
by processing the time series generated by the accelerometer. Some
fall detectors can also include an air pressure sensor, for example
as described in WO 2004/114245. On detecting a fall, an alarm is
triggered by the fall detector.
[0005] Some fall detectors are designed to be worn as a pendant
around the neck of the user, whereas others are designed to be worn
on the torso or limbs of the user, for example at the wrist.
However, the wrist is capable of complex movement patterns and has
a large range of movement, which means that existing fall detection
methods based on analyzing measurements from an accelerometer do
not provide a sufficiently high detection rate while minimizing the
number of false alarms for this type of fall detector.
[0006] Therefore, there is a need for an improved method for
detecting a potential fall and a fall detector implementing the
same.
SUMMARY OF THE INVENTION
[0007] According to a first aspect of the invention, there is
provided a fall detector for detecting a potential fall by a user,
the fall detector comprising a sensor for measuring the skin
conductivity of the user; and a processor for analyzing the skin
conductivity measurements to determine whether the user has had a
potential fall, wherein the processor is configured to analyze the
skin conductivity measurements by matching the skin conductivity
measurements to a predetermined skin conductivity response
corresponding to a stressful event.
[0008] In a preferred embodiment, the processor is configured to
generate skin conductivity coefficients by matching the skin
conductivity measurements to a predetermined skin conductivity
response corresponding to a stressful event, the value of each
coefficient indicating the match to the predetermined skin
conductivity response, and wherein the processor is configured to
determine whether the response is consistent with a user having had
a fall by analyzing the skin conductivity coefficients.
[0009] In an embodiment, the processor is configured to analyze the
skin conductivity coefficients to determine whether the response is
consistent with a user having had a fall by determining one or more
of (i) the maximum value of the coefficients, (ii) the mean of the
coefficients, and/or (iii) the median of the coefficients.
[0010] In some embodiments, the processor is configured to filter
the skin conductivity measurements to remove noise and/or
interference prior to matching the skin conductivity measurements
to the predetermined skin conductivity response.
[0011] In those embodiments, the processor is preferably configured
to use a median filter to remove the noise and/or interference from
the skin conductivity measurements.
[0012] In preferred embodiments, the fall detector further
comprises at least one movement sensor for sensing the movement of
the user, the processor being further configured to analyze
measurements from the at least one movement sensor to determine if
the user may have fallen.
[0013] In a preferred embodiment, the processor is configured to
analyze the skin conductivity measurements only in the event that
the measurements from the at least one movement sensor indicate
that the user may have fallen.
[0014] Preferably, the processor is configured to use the skin
conductivity sensor to measure the skin conductivity in the event
that the measurements from the at least one movement sensor
indicate that the user may have fallen.
[0015] In embodiments of the invention, the processor is configured
to analyze the measurements from the at least one sensor to
identify whether an impact and/or a change in altitude greater than
respective threshold values have occurred.
[0016] Preferably, the processor is configured to analyze the skin
conductivity measurements obtained during a time window following a
time at which the measurements from the at least one sensor
indicate that the user may have fallen.
[0017] According to a second aspect of the invention, there is
provided a method of detecting a potential fall by a user, the
method comprising measuring the skin conductivity of the user and
analyzing the skin conductivity measurements to determine whether
the user has had a potential fall by matching the skin conductivity
measurements to a predetermined skin conductivity response
corresponding to a stressful event.
[0018] In a preferred embodiment, the step of analyzing comprises
analyzing skin conductivity coefficients generated from matching
the skin conductivity measurements to the predetermined skin
conductivity response, the value of each coefficient indicating the
match to the predetermined skin conductivity response.
[0019] In an embodiment, analyzing the skin conductivity
coefficients to determine whether the response is consistent with a
user having had a fall comprises determining one or more of (i) the
maximum value of the coefficients, (ii) the mean of the
coefficients, and/or (iii) the median of the coefficients.
[0020] In some embodiments, the method further comprises the step
of filtering the skin conductivity measurements to remove noise
and/or interference prior to matching the skin conductivity
measurements to the predetermined skin conductivity response.
[0021] In those embodiments, the method preferably comprises using
a median filter to filter the skin conductivity measurements to
remove the noise and/or interference from the skin conductivity
measurements.
[0022] In preferred embodiments, the method further comprises the
steps of sensing the movement of the user and analyzing the
movement of the user to determine if the user may have fallen.
[0023] Preferably, the step of analyzing the skin conductivity
measurements is performed only in the event that the step of
analyzing the movement of the user indicates that the user may have
fallen.
[0024] In preferred embodiments, the step of measuring the skin
conductivity is performed only in the event that the step of
analyzing the movement of the user indicates that the user may have
fallen.
[0025] In embodiments of the invention, the step of analyzing the
movement of the user comprises determining whether an impact and/or
a change in altitude greater than respective threshold values have
occurred.
[0026] Preferably, the step of analyzing skin conductivity
measurements comprises analyzing skin conductivity measurements
obtained during a time window following a time at which the
movements of the user indicate that the user may have fallen.
[0027] According to a third aspect of the invention, there is
provided a computer program product comprising computer readable
code embodied therein, the computer readable code being configured
such that, upon execution by a suitable computer or processor, the
computer or processor performs the method described above.
BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Exemplary embodiments of the invention will now be
described, by way of example only, with reference to the following
drawings, in which:
[0029] FIG. 1 is a block diagram of a fall detector in accordance
with the invention;
[0030] FIG. 2 is a flow chart illustrating a method of detecting
falls in accordance with an embodiment of the invention;
[0031] FIG. 3 is a flow chart illustrating the analysis of skin
conductivity measurements according to an embodiment of the
invention;
[0032] FIGS. 4(a)-(d) are a series of graphs illustrating the skin
conductivity measurements and the results of the processing steps
shown in FIG. 3;
[0033] FIG. 5 is a graph illustrating the filter used in step 1095
of the method shown in FIG. 3; and
[0034] FIG. 6 is a graph illustrating an analysis window for the
skin conductivity coefficients determined from the method shown in
FIG. 3.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0035] A fall detector 2 according to an embodiment of the
invention is shown in FIG. 1. In a preferred embodiment of the
invention, the fall detector 2 is designed to be worn by a user on
their wrist, although it will be appreciated that the invention is
not limited to this use, and the fall detector 2 could instead be
designed to be worn at the user's waist, on their chest or back or
as a pendant around their neck or carried in their pocket.
[0036] In this exemplary embodiment, the fall detector 2 comprises
two movement sensors--an accelerometer 4 and pressure sensor
6--which are connected to a processor 8. The processor 8 receives
measurements from the movement sensors 4, 6, and processes the
measurements to determine if a user of the fall detector 2 has
suffered a fall. Although two movement sensors are shown in this
embodiment, it will be appreciated that fall detectors according to
alternative embodiments may comprise only one movement sensor (for
example just the accelerometer 4).
[0037] The fall detector 2 further comprises an audible alarm unit
10 that can be activated by the processor 8 in the event that the
processor 8 determines that the user has suffered a fall. The fall
detector 2 may also be provided with a button (not shown in FIG. 1)
that allows the user to manually activate the audible alarm unit 10
if they require assistance (or deactivate the alarm if assistance
is not required).
[0038] The fall detector 2 also comprises a transmitter unit 12
that allows the fall detector 2 to transmit an alarm signal to a
base station associated with the fall detector 2 (which can then
issue an alarm or summon help from a healthcare provider or the
emergency services) or directly to a remote station (for example
located in call centre of a healthcare provider) if a fall is
detected, so that assistance can be summoned for the user. In some
embodiments, the processor 8 in the fall detector 2 may not execute
the algorithm on the data from the sensors 4, 6 to determine if the
user has fallen; instead the processor 8 and transmitter unit 12
may provide the raw data from the sensors 4, 6 to the base station
and a processor in the base station can execute the algorithm on
the data from the sensors 4, 6 to determine if the user has
fallen.
[0039] As existing fall detection methods based on analyzing
measurements from an accelerometer in a wrist-worn fall detector do
not provide a sufficiently high detection rate while limiting the
number of false alarms, the fall detector 2 according to some
aspects of the invention further comprises a sensor 14 for
measuring the conductivity of the user's skin and the processor 8
is configured to analyze the skin conductivity measurements in
conjunction with the measurements from the movement sensors 4, 6 to
provide a more reliable indication of whether the user has suffered
a fall. In a wrist-worn fall detector 2, the skin conductivity
sensor 14 is preferably arranged to contact the skin of the user on
the volar side of their wrist. In some embodiments, the fall
detector 2 comprises multiple skin conductivity sensors 14 that are
to be placed at different positions on the user's body. In this
case, at least one of those skin conductivity sensors 14 can be
integrated into a separate housing to the rest of the components of
the fall detector 2.
[0040] In the illustrated embodiment, all of the components of the
fall detector 2 are integrated into a single housing that is to be
placed in contact with the user's skin. In alternative embodiments,
for example where part of the fall detector is in the form of a
pendant to be worn around the user's neck (and so might not be in
contact with the user's skin at all times), the skin conductivity
sensor 14 can be provided in a housing that is separate from the
pendant (the pendant including the accelerometer 4 and pressure
sensor 6), so that the skin conductivity sensor 14 can be located
in contact with the user's skin during use.
[0041] The operation of the fall detector 2 will now be described
in more detail with reference to the flow chart in FIG. 2.
[0042] In order for the processor 8 in the fall detector 2 (or, in
the alternative embodiment described above, for the processor in
the base station) to determine if the user has suffered a fall, it
is necessary to extract values for various features that are
associated with a fall from the movement sensor measurements. Thus,
in step 101, the accelerometer 4 and air pressure sensor 6 measure
the accelerations and air pressure changes experienced by the fall
detector 2 and these measurements are provided to the processor 8
by the sensors 4, 6, and the processor 8 analyses them to determine
whether the user might have suffered a fall (step 103).
[0043] A fall can be broadly characterized by, for example, a
change in altitude of around 0.5 to 1.5 meters (the range may be
different depending on the part of the body that the fall detector
2 is to be worn), culminating in a significant impact, followed by
a period in which the user does not move very much. Thus,
conventionally, in order to determine if a fall has taken place,
the processor 8 needs to process the sensor measurements to extract
values for features including a change in altitude (which is
usually derived from the measurements from the pressure sensor 6,
but can also or alternatively be derived from the measurements from
the accelerometer 4), a maximum activity level (i.e. an impact)
around the time that the change in altitude occurs (typically
derived from the measurements from the accelerometer 4) and a
period in which the user is relatively inactive following the
impact (again typically derived from the measurements from the
accelerometer 4). It will be appreciated that other features can
further improve the detection algorithm. For example, the detection
of a change in orientation upon falling can improve the likelihood
that the signal is due to a fall.
[0044] The analysis performed by the processor 8 in step 103 will
not be described in further detail herein, but those skilled in the
art will be aware of various algorithms and techniques that can be
applied to determine whether a user may have suffered a fall from
accelerometer and/or pressure sensor measurements.
[0045] If the processor 8 determines that a user may have suffered
a fall (step 105), i.e. if the analysis of the accelerometer 4
and/or pressure sensor 6 measurements indicates that a fall has
taken place, then measurements of the skin conductivity of the user
are taken by the skin conductivity sensor 14 (step 107). If the
processor 8 has not detected a possible fall, the process returns
to step 101 and further measurements are taken and analyzed (step
103).
[0046] In some embodiments, the skin conductivity sensor 14 is only
activated once a possible fall (or simply a change in altitude of
at least 0.5 m detected in the measurements from the pressure
sensor 6) has been detected from the analysis of the accelerometer
4 and/or pressure sensor 6 measurements, thus reducing the power
consumption of the fall detector 2. As the analysis in step 103 is
performed by the processor 8 substantially in real time or with
only a small delay, the skin conductivity sensor 14 will be
activated shortly after a fall event has occurred.
[0047] In alternative embodiments, the skin conductivity sensor 14
may measure skin conductivity constantly or frequently whenever the
fall detector 2 is in use (i.e. even when a possible fall has not
yet been detected). This way, skin conductivity measurements will
be available to the processor 8 as soon as a possible fall is
detected.
[0048] In step 109, the processor 8 analyses the measurements from
the skin conductivity sensor 14 for characteristics associated with
a fall. In particular, the processor 8 analyses the measurements to
identify whether there has been a measurable skin conductivity
response that is consistent with the user having suffered a
stressful event, such as an accidental fall. In the alternative
embodiment described above where the skin conductivity sensor 14 is
measuring skin conductivity constantly or frequently, the processor
8 may analyze the skin conductivity measurements only when a
possible fall or change in altitude greater than a predetermined
amount, for example 0.5 m, has been detected from the analysis of
the measurements from the movement sensor(s) 4, 6. Alternatively,
the processor 8 may analyze the skin conductivity measurements for
a response consistent with the user experiencing a stressful event,
such as suffering a fall, regardless of whether a possible fall or
change in altitude greater than 0.5 m has been detected yet.
[0049] When a person accidentally falls, the process of falling,
from losing balance to the impact with the ground and the
(possible) inability to get up after the fall, causes a measureable
change in the skin conductivity of the user as a result of the
stress suffered by the user. However, a movement, such as sitting
down or intentionally `falling` onto a chair, which might appear to
be a fall from an analysis of the accelerometer 4 and pressure
sensor 6 measurements alone, would not cause this stress response,
and the skin conductivity measurements can be used to qualify
whether an accidental fall has taken place.
[0050] Thus, the analysis by the processor 8 in step 109 aims to
identify a skin conductivity response associated with a stressful
event (i.e. an accidental fall) from the measurements obtained
after (and in some embodiments, during) the possible fall event.
The skin conductivity measurements analyzed in step 109 may relate
to a period of around 15 seconds after the possible fall event has
occurred, although those skilled in the art will appreciate that
time windows having different lengths can be used.
[0051] In step 111, the processor 8 uses the result of the analysis
of the skin conductivity measurements and the result of the
analysis of the acceleration and air pressure measurements to
determine if the user has potentially suffered a fall. If the user
is determined to have fallen, the processor 8 can trigger an alarm
to obtain help for the user (step 113), as described above. After
triggering the alarm, the process can return to step 101 to
continue the monitoring of the user. If it is determined in step
111 that the user has not fallen, the process also returns to step
101 to continue the monitoring of the user.
[0052] It will be appreciated that the results of the analysis
steps (steps 103 and 109) can be combined in a number of different
ways. For example, where the skin conductivity sensor 14 is only
activated once a possible fall has been detected, a fall may be
determined in step 111 if the skin conductivity measurements are
consistent with a fall having taken place. Alternatively, features
(such as an impact, altitude change, skin conductivity response)
can be extracted from the sensor measurements, and the features
combined (possibly using weightings for each extracted feature) to
determine whether a fall has taken place.
[0053] In the event that the measurements from the skin
conductivity sensor 14 indicate that the sensor 14 is not in
contact with the user's skin, i.e. the measurements are very low or
zero (for example if the fall detector 2 is not being worn properly
or if the sensor 14 has been knocked out of position by the fall),
the processor 8 can take a decision on whether a person has fallen
just based on the measurements from the accelerometer 4 and/or
pressure sensor 6.
[0054] The analysis of the skin conductivity measurements by the
processor 8 in step 109 of FIG. 2, will now be described in more
detail with reference to the flow chart in FIG. 3, and the graphs
shown in FIGS. 4, 5 and 6.
[0055] In step 1091, the processor 8 receives the skin conductivity
measurements. An exemplary set of measurements is shown in FIG.
4(a). The skin conductivity is measured in micro Siemens,
.mu.LS.
[0056] Firstly, the processor 8 filters the skin conductivity
measurements to remove noise and interference (step 1093). The
interference may result from electrical interference in the
circuitry of the fall detector 2, poor contact between the skin
conductivity sensor 14 and the user's skin, etc.). Preferably step
1093 comprises applying a median filter to the sensor measurements.
The result of the filtering step is illustrated in FIG. 4(b), and
it can be seen that spike in the measurements at around time=550
seconds has been removed.
[0057] Then, in step 1095, the filtered skin conductivity
measurements are matched to a predetermined pattern corresponding
to a skin conductivity response associated with a (any) stressful
event, such as an accidental fall, using a matched filter. The
output is a set of coefficients that indicate the match of the
measurements to the pattern. Each coefficient represents the match
of a number of consecutive measurement samples (covering a time
period of the same length as the predetermined pattern) to the
predetermined pattern. The higher the coefficient, the better the
match of the measurements to the pattern (and therefore the greater
the chance that a stressful event, including a fall, has
occurred).
[0058] In one embodiment, this step is implemented using a matched
filter that is DC-free. This means that the DC component in the
measurements from the skin conductivity sensor 14 will not
adversely affect the filtering, which will produce a `0` skin
conductivity coefficient for non-variant (or very low variant) skin
conductance.
[0059] An exemplary predetermined pattern, i.e. impulse response of
the matched filter, that represents the skin conductivity response
that would be measured at the wrist of a person who suffers a fall
or other sudden stress-inducing event can be given by the
function
match.sub.-- filter=A(t.sup.n=B)e.sup.-t/.tau.-DC.sub.remove
(1)
where t is time (0.ltoreq.t.ltoreq.30), A is a scaling factor that
determines the height of the initial part of the pattern (i.e. how
strong the skin conductivity response will be), n and B are further
scaling factors that determine the gradient of the initial part of
the pattern (i.e. how quickly the skin conductivity will increase),
.tau. is a scaling factor that determines the gradient of the part
of the pattern after the initial peak in the pattern (i.e. how
quickly the skin conductivity returns to the initial level) and
DC_remove is the DC component excluded from the pattern.
[0060] This function is illustrated in FIG. 5 and shows that there
is a large steep increase in the skin conductivity measurement
following a stressful event, such as a fall, followed by a slow
recovery, with A, n, B, .tau. and DC_remove taking the values,
1/150, 300, 3, 3 and 2 respectively. It has been found that the
recovery of skin conductivity happens relatively quickly after
normal emotional arousal, whereas recovery is much longer after an
event in which the person is scared or stressed, such as a fall.
The recovery or decay in skin conductivity after such an event can
occur in two stages after the skin conductivity maximum, with the
decay coefficients differing by at least a factor of 10. The y-axis
in FIG. 5 has no unit, but it's scale is the same as the skin
conductance measurement in FIG. 4(a) (i.e. a difference of 1 unit
in the y-axis direction of FIG. 5 corresponds to a difference of 1
Siemens difference in the y-axis direction in FIG. 4(a). An
exemplary output of the matched filtering process is shown in FIG.
4(c). It will be appreciated that the values of A, n, B, .tau. and
DC_remove can be varied from the exemplary values given above
depending on available data (perhaps user-specific data) relating
to the response of skin conductivity to a fall or other
stress-inducing event. An optimization technique, such as higher
order degree polynomial regression, can be used to fit the curve of
equation (1) to the available data.
[0061] It will also be appreciated by those skilled in the art that
the predetermined pattern can be represented by alternative
functions to that shown in equation (1). For example, any function
that defines a curve having characteristics similar to that shown
in FIG. 5 can be used (i.e. a sharp increase at the start, followed
by a gradual recovery). Also, different curves can be used, with
the processor 8 in the fall detector 2 selecting one for use based
on the estimated severity of the fall. For example, where there is
a high impact value deduced from the accelerometer data, a response
curve (predetermined pattern) corresponding to high tension can be
used. In addition or alternatively, the predetermined pattern can
be personalized to the user, since different people exhibit
different skin conductivity levels and responses to different
events (although generally the differences concern the shape,
rather than the magnitude).
[0062] The skin conductivity coefficients used in subsequent
processing are determined by reversing the magnitude of the
coefficients output by the matched filter (see FIG. 4(d)).
[0063] The processor 8 can then analyze the skin conductivity
coefficients to provide an indication of whether the user has
potentially suffered a fall (step 1097). In particular, the
processor 8 can analyze the coefficients within an analysis window
that spans the skin conductivity measurements obtained during a
predetermined period after an altitude drop of more than a
predetermined value has been observed. The predetermined period can
be 15 seconds, although any suitable time period can be used. The
predetermined value for the altitude drop may be 0.5 meters,
although this value can be set based on user-specific parameters,
such as the height of the user. An exemplary 15-second analysis
window is shown in FIG. 6. It will be appreciated that in the
embodiment where skin conductivity measurements are only collected
following the detection of a change in altitude of at least 0.5 m
or a possible impact, the analysis window will start effectively
when the skin conductivity sensor starts to measure the skin
conductivity (i.e. there will not be any data to the left of the
analysis window in FIG. 6).
[0064] In some embodiments, the processor 8 analyses the skin
conductivity coefficients in the analysis window by determining any
one or more of the following: (i) the maximum value of the
coefficients in the analysis window (ii) the mean of the
coefficients in the analysis window, and/or (iii) the median of the
coefficients in the analysis window. The processor 8 can analyze
individual ones of the features listed above to determine if there
has been a skin conductivity response consistent with a stressful
event, such as a fall, having taken place, or multiple ones of the
features listed above, for example by comparison of the features
with respective thresholds. In that case, the processor 8 can
determine whether there has been a skin conductivity response based
on a combination or a weighted combination of the derived
features.
[0065] It will be appreciated that the final decision by the
processor 8 on whether the user has suffered a fall can be based on
a combination or weighted combination of the results of the
analysis of the measurements from the movement sensor(s) 4, 6 and
the analysis of the measurements from the skin conductivity sensor
14.
[0066] In an alternative embodiment of the invention, a fall
detector is provided that determines if the user has potentially
suffered a fall based on measurements from a single sensor
(specifically a skin conductivity sensor 14). In this embodiment,
the measurements from the skin conductivity sensor 14 can be
analyzed as described above to provide an indication of whether the
user has potentially suffered a fall. Thus, the measurements of the
skin conductivity are matched to a predetermined skin conductivity
response corresponding to a stressful event to determine if the
user has experienced such an event. Although this fall detector may
not necessarily be able to distinguish between a fall and other
types of stressful event that might be experienced by the user, the
analysis of the skin conductivity measurements described above
would still allow stressful events (including falls) to be
identified from other types of event (including normal emotional
arousal) that lead to a or no significant change in skin
conductivity. Although a fall detector according to this embodiment
of the invention might provide some false positive indications
(because non-fall stressful events are also identified by the
analysis of the skin conductivity measurements) this is still a
useful embodiment since the rate of false negatives (i.e. where
there would be no indication of a fall when in fact the user has
fallen) will be quite low.
[0067] There is therefore provided a method for detecting a
potential fall and a fall detector that provides increased fall
detection reliability compared to conventional techniques.
[0068] While the invention has been illustrated and described in
detail in the drawings and foregoing description, such illustration
and description are to be considered illustrative or exemplary and
not restrictive; the invention is not limited to the disclosed
embodiments.
[0069] Variations to the disclosed embodiments can be understood
and effected by those skilled in the art in practicing the claimed
invention, from a study of the drawings, the disclosure, and the
appended claims. In the claims, the word "comprising" does not
exclude other elements or steps, and the indefinite article "a" or
"an" does not exclude a plurality. A single processor or other unit
may fulfill the functions of several items recited in the claims.
The mere fact that certain measures are recited in mutually
different dependent claims does not indicate that a combination of
these measures cannot be used to advantage. A computer program may
be stored/distributed on a suitable medium, such as an optical
storage medium or a solid-state medium supplied together with or as
part of other hardware, but may also be distributed in other forms,
such as via the Internet or other wired or wireless
telecommunication systems. Any reference signs in the claims should
not be construed as limiting the scope.
* * * * *