Method for estimating core body temperature
View Patent ↗The invention in at least one embodiment includes a method for determining the core body temperature of a person by determining an initial core body temperature and heart rate of said person; providing the initial core body temperature and heart rate of the person to a processor; and calculating a predicted core body temperature by the processor using an extended Kalman filter based on the heart rate and the initial core body temperature. In another embodiment, a system for performing the method.
1. A method for indirectly predicting a core body temperature of a person in substantially real-time to allow the person to act on the core body temperature to avoid a heat illness, the method comprising:
setting an initial core body temperature with a mobile device comprising a heart rate monitor, a processor, and a display;
receiving at least two heart rates of the person with the processor, where the heart rates are detected by the heart rate monitor attached to the person, wherein the at least two heart rates include a first heart rate and a second heart rate;
calculating the core body temperature with the processor using an extended Kalman filter based on a sigmoid model using solely the second heart rate and the initial core body temperature;
providing the core body temperature to the display and/or an external device; and,
using the provided core body temperature by the person to adjust an activity level in real-time when the core body temperature indicates there is an elevated risk of suffering from heat illness, and
wherein the initial core body temperature is set by the processor
based solely on the first heart rate of the person or
on a predetermined resting core body temperature for that person; and
the sigmoid model uses an error between the measured heart rate and an estimated heart rate based on the estimated core body temperature to adjust the sigmoid model.
2. The method according to claim 1 , further comprising:
receiving an additional heart rate of the person with the processor, where the additional heart rate is detected by the heart rate monitor;
calculating a new predicted core body temperature with the processor using the extended Kalman filter based on the sigmoid model using solely the additional heart rate and the last predicted core body temperature from the last calculation of the core body temperature; and
repeating the receiving additional heart rate and calculating a new predicted core body temperature at predetermined intervals.
3. The method according to claim 2 , further comprising:
when the new predicted core body temperature exceeds a predetermined threshold, sending an alarm signal from said processor prior to proceeding to repeating, wherein the alarm signal triggers an alarm; and
when the new predicted core body temperature does not exceed a predetermined threshold, proceeding to repeating.
4. The method according to claim 2 , wherein the method repeats on predetermined intervals of 1 minute intervals, 2 minute intervals, or 5 minute intervals.
5. The method according to claim 1 , further comprising recording the predicted core body temperature in memory in the mobile device; and
wherein the extended Kalman filter includes a time update model and an observation update model.
6. The method according to claim 1 , wherein providing the predicted core body temperature includes transmitting the predicted core body temperature with a transmitter in the mobile device to an external device.
7. The method according to claim 1 , wherein calculating the predicted core body temperature includes
computing a preliminary core body temperature estimate;
computing a preliminary estimate of the variance of the preliminary core body temperature estimate;
computing an extended Kalman filter mapping function variance coefficient;
computing a Kalman gain weighting factor based on the preliminary estimate of variance and the extended Kalman filter mapping function variance coefficient; and
computing the predicted core body temperature (CT t ) using a preliminary core body temperature estimate (ĈT t ), and an error between the new heart rate (HR t ) and an expected heart rate given the preliminary estimate of the core body temperature with the following relationship
C
T
t
=
C
ˆ
T
t
+
k
t
{
HR
t
-
(
41
+
1
5
2
(
1
+
0
.
0
6
e
-
0
8
9
(
C
^
T
t
-
37.84
)
)
1
/
0.07
)
}
,
and
wherein the expected heart rate (expected HR) is determined based on
expected
HR
t
=
A
+
K
-
A
(
1
+
Qe
-
B
(
C
^
T
t
-
M
)
)
1
/
v
where A=lower asymptote, K=upper asymptote, Q=is the value when estimated core temperature (ĈT t )=M, B=the growth rate, M=sets the starting ĈT t value, and v=affects the asymptote where maximum growth occurs, or
the expected heart rate (expected HR) is determined based on
expected
H
R
=
4
1
+
1
5
2
(
1
+
0.06
e
-
0.89
(
C
^
T
t
-
37.84
)
)
1
/
0.07
.
8. The method according to claim 1 , wherein a relationship between the core body temperature and the heart rate is a sigmoid relationship that varies over a range of heart rate measurements.
9. The method according to claim 1 , further comprising:
receiving from an accelerometer activity data;
determining based on the activity data whether the person is exercising or resting; and
changing the extended Kalman filter based on the activity data to affect a rate of core temperature rise or fall.
10. A mobile device for indirectly determining a core body temperature of a person to allow the person to act on the core body temperature to avoid a heat illness, the mobile device comprising:
a heart rate monitor capable of attachment to the person whose core body temperature is being determined;
a memory to store core body temperature data;
a processor in communication with said heart rate monitor, said processor calculates a core body temperature using an extended Kalman filter having a plurality of constants and the extended Kalman filter is based on a sigmoid model having as the only inputs a heart rate from said heart rate monitor and prior core body temperature data stored in said memory; and
a display in communication with said processor, and
wherein the sigmoid model uses an error between the measured heart rate and an estimated heart rate based on the estimated core body temperature to adjust the sigmoid model; and
wherein said processor provides the calculated core body temperature in real-time to said display, the person acts on the displayed calculated core body temperature by adjusting an activity level when the calculated core body temperature indicates the person is at an elevated risk of suffering from heat illness.
11. The mobile device according to claim 10 , further comprising a transmitter for communication with an external system, said transmitter in communication with said processor.
12. The mobile device according to claim 10 , wherein the processor is configured to calculate the core body temperature by
computing a preliminary core body temperature estimate;
computing a preliminary estimate of the variance of the preliminary core body temperature estimate;
computing an extended Kalman filter mapping function variance coefficient;
computing a Kalman gain weighting factor based on the preliminary estimate of variance and the extended Kalman filter mapping function variance coefficient; and
computing the predicted core body temperature (CT t ) using a preliminary core body temperature estimate (ĈT t ), and an error between the new heart rate (HR t ) and an expected heart rate given the preliminary estimate of the core body temperature with the following relationship
C
T
t
=
C
ˆ
T
t
+
k
t
{
HR
t
-
(
41
+
1
5
2
(
1
+
0
.
0
6
e
-
0
.
8
9
(
C
^
T
t
-
3
7
.
8
4
)
)
1
/
0.07
)
}
,
and
wherein the expected heart rate (expected HR) is determined based on
expected
HR
t
=
A
+
K
-
A
(
1
+
Qe
-
B
(
C
^
T
t
-
M
)
)
1
/
v
where A=lower asymptote, K=upper asymptote, Q=is the value when estimated core temperature (ĈT t )=M, B=the growth rate, M=sets the starting ĈT t value, and v=affects the asymptote where maximum growth occurs.
13. The mobile device according to claim 10 , wherein said processor includes
an initial core body temperature module for setting an initial core body temperature,
a heart rate module for obtaining the heart rate based on a signal from said heart rate monitor, and
a core body temperature calculator module for calculating the core body temperature, said core body temperature calculator module in communication with said initial core body temperature module and said heart rate module.
14. The mobile device according to claim 13 , where said processor further includes a timer module in communication with said heart rate module and said core body temperature calculator module.
15. The mobile device according to claim 14 , where said processor further includes an alarm module for triggering an alarm when the core body temperature exceeds a threshold, said alarm module in communication with said core body temperature calculator module.
16. The mobile device according to claim 15 , further comprising:
a display in communication with said core body temperature calculator module;
a transmitter in communication with said core body temperature calculator module; and
a temperature sensor in communication with at least one of said timer module and said alarm module.
17. The mobile device according to claim 10 , wherein a relationship between the core body temperature and the heart rate is a sigmoid relationship that varies over a range of heart rate measurements.
18. A mobile device for indirectly determining a core body temperature of a person based on a heart rate to allow the person to act on the core body temperature to avoid a heat illness, the mobile device comprising:
a heart rate monitor attached to a heart rate monitor processor configured to convert signals from said heart rate monitor into the heart rate, said heart rate monitor configured to contact the person during use; and
a processor in communication with said heart rate monitor, said processor having program code embodied therewith, the program code executable by the processor in real time to
set an initial core body temperature;
receive the heart rate of the person from said heart rate monitor;
calculate the core body temperature using an extended Kalman filter based on a sigmoid model having as its only inputs the heart rate and at least one of the initial core body temperature and the prior core body temperature; and
provide the core body temperature of the person, and
wherein the person reduces the activity level in response to the provided core body temperature when the core body temperature indicates an elevated risk of suffering from heat illness,
wherein the sigmoid model uses an error between the measured heart rate and an estimated heart rate based on the estimated core body temperature to adjust the sigmoid model.
19. The mobile device according to claim 18 , further comprising:
a display in communication with said processor; and
a transmitter in communication with said processor, and
wherein the initial core body temperature is set based on the initial heart rate of the person.
20. The mobile device according to claim 18 , wherein said processor having further program code embodied therewith, the further program code executable by the processor to calculate the predicted core body temperature by
computing a preliminary core body temperature estimate;
computing a preliminary estimate of the variance of the preliminary core body temperature estimate;
computing an extended Kalman filter mapping function variance coefficient;
computing a Kalman gain weighting factor based on the preliminary estimate of variance and the extended Kalman filter mapping function variance coefficient; and
computing the predicted core body temperature (CT) using a preliminary core body temperature estimate (ĈT t ), an error between the new heart rate (HR t ) and an expected heart rate given the preliminary estimate of the core body temperature with the following relationship
C
T
t
=
C
ˆ
T
t
+
k
t
{
HR
t
-
(
41
+
1
5
2
(
1
+
0
.
0
6
e
-
0
.
8
9
(
C
^
T
t
-
3
7
.
8
4
)
)
1
/
0.07
)
}
,
and
wherein the expected heart rate (expected HR) is determined based on
expected
HR
t
=
A
+
K
-
A
(
1
+
Qe
-
B
(
C
^
T
t
-
M
)
)
1
/
v
where A=lower asymptote, K=upper asymptote, Q=is the value when estimated core temperature (ĈT t )=M, B=the growth rate, M=sets the starting ĈT t value, and v=affects the asymptote where maximum growth occurs.