WO2025078720A1

WO2025078720A1 - Wearable health measurement device with heat flow measurement

Info

Publication number: WO2025078720A1
Application number: PCT/FI2024/050520
Authority: WO
Inventors: Antti Immonen; Saku LEVIKARI; Antti Jokela; Tomi Harjunmaa
Original assignee: Vire Health Oy
Current assignee: Vire Health Oy
Priority date: 2023-10-12
Filing date: 2024-10-04
Publication date: 2025-04-17
Anticipated expiration: 2026-04-12
Also published as: FI20236135A1

Abstract

The embodiments relate to a device (100) for measuring health-related parameters from a user. The device (100) comprises a measurement part (101) comprising a first side (101a) and a second side (101b), the first side (101a) being in contact with user's skin. The first side (101a) comprises at least a first temperature sensor (202) to measure a first temperature, the first temperature being a skin temperature. The measurement part (101) comprises at least a second temperature sensor to measure a second temperature. The first temperature sensor (202) and the second temperature sensor comprise a heat flow channel therein between. The measurement part comprises means for determining when the first side (101a) is disconnected from the skin and when the first side (101a) is reconnected to the skin. A body core temperature value is determined based on the heat flow channel and taking into account the measured signal on the disconnection or reconnection of the measurement part (101).

Description

WEARABLE HEALTH MEASUREMENT DEVICE WITH HEAT FLOW MEASUREMENT

Technical Field

The present solution generally relates to an apparatus for measuring health- related parameters on a user.

Background

Over decades, athletes have been interested in their heart rates (HR) during exercise, whereupon devices for personal HR monitoring have rapidly evolved. Upon the development of the sensor technology, devices with more intelligent measurement and analysis capabilities have become available for any consumer, who is interested in their health and well-being.

Sports-related measurement devices typically comprise a belt or a band to receive physical signals directly from user’s chest, an arm or a forehead, which signals are then transmitted wirelessly to user’s wristwatch or other device for further analysis or reporting. So called activity watches (or smartwatches) are capable of measuring the heart rate and other body signals directly from the wrist and making the analysis on user’s well-being and activity by themselves. More wearable health measurement devices are entering the market. One example is a ring that measures body signals from user’s finger, and transmits the measured data to an analysis device, such as a mobile phone.

These devices are limited to their possibilities to make health-related analysis, whereupon broader perspective on the matter is needed.

Summary

The present invention improves the prior art by providing a health measurement device that is not limited to the data obtained from fingers, chest, forehead, but can provide more extensive analysis on well-being of the user. The scope of protection sought for various embodiments of the invention is set out by the independent claims. The embodiments and features, if any, described in this specification that do not fall under the scope of the independent claims are to be interpreted as examples useful for understanding various embodiments of the invention.

Various aspects include a method and an apparatus, which are characterized by what is stated in the independent claims. Various embodiments are disclosed in the dependent claims.

According to a first aspect, there is provided a wearable device for measuring health-related parameters from a user, the device comprising a measurement part and a clip part, wherein the measurement part comprises a first side and a second side, the first side being in contact with user’s skin, and wherein the clip part is connected to the measurement part; wherein the first side of the measurement part comprises at least a first temperature sensor to measure a first temperature, the first temperature being a skin temperature, and wherein the measurement part comprises at least a second temperature sensor to measure a second temperature, wherein the first temperature sensor and the second temperature sensor comprises a heat flow channel therein between, and wherein each of the temperature sensors has a time constant; the measurement part comprising means for measuring signal from the skin for determining when the first side of the measurement part is disconnected from the skin and when the first side of the measurement part is reconnected to the skin; and means for determining a body core temperature value based on the heat flow channel and taking into account the measured signal on the disconnection or reconnection of the measurement part.

According to a second aspect, there is provided a computer-implemented method for determining health-related parameters, the method comprising receiving a first temperature measured by a first temperature sensor of wearable health measurement device, the first temperature being a skin temperature; receiving a second temperature measured by a second temperature sensor of the wearable health measurement device; wherein the first temperature sensor and the second temperature sensor has a heat flow channel therein between, and wherein each of the temperature sensors has a time constant; measuring signal from the skin for determining when the first side of the measurement part is disconnected from the skin and when the first side of the measurement part is reconnected to the skin; and determining a body core temperature value based on the heat flow channel and taking into account the measured signal on the disconnection or reconnection of the measurement part.

According to an embodiment, said means for measuring signal is a noncontact temperature sensor configured to measure a third temperature close to the first temperature sensor, wherein the wearable device further comprises means for detecting whether the first temperature is different from the third temperature.

According to an embodiment, said means for measuring signal is a skin conductance sensor using two skin electrodes on the first side of the measurement part.

According to an embodiment, said means for measuring signal is an optical reflectance sensor on the first side of the measurement part.

According to an embodiment, when it is determined that the device is disconnected from the skin, the data is discarded, and when it is determined that device is reconnected to the skin, the first and second temperature values are corrected for receiving the core temperature of the body of the user.

According to an embodiment, the correction is performed by defining a time constant coefficient (/3) to each of the first and second temperature sensors.

According to an embodiment, the time constant coefficient (0) is defined numerically from the data measured by using any combination of the following: forward and backward differences, lowpass filtering of the derivative signal, or curve fitting methods.

According to an embodiment, the device comprises two heat flux channels, both having their own first temperature sensors and second temperature sensors. According to an embodiment, the device comprises a humidity sensor configured to measure humidity of air close to the device.

According to an embodiment, the device comprises means for measuring a skin impedance between electrodes or electrode areas.

According to an embodiment, the device comprises means for selecting a temperature calculation model.

According to an embodiment, the device comprises a third temperature sensor, and means for determining a difference between the third temperature sensor and at least one of the first temperature sensors.

According to an embodiment, the device comprises at least one processor, memory including computer program code.

According to an embodiment, the computer program product is embodied on a non-transitory computer readable medium.

Description of the Drawings

In the following, various embodiments will be described in more detail with reference to the appended drawings, in which

Fig. 1 shows a simplified example of a wearable health measurement device according to an embodiment;

Fig. 2 shows an example of the measurement part of the wearable health measurement device according to an embodiment;

Fig. 3a shows a principle of two channel heat flux sensor as an example;

Fig 3b shows an example of a contact model for a skin temperature sensor;

Fig. 3c shows an example of a signal indicating when the device has a skin contact or when it has not; Fig. 4a shows an example of the wearable health measurement device being attached to a clothing;

Fig. 4b shows an intersection of the wearable health measurement device shown in Fig. 4a;

Fig. 5 shows another example of the wearable health measurement device being attached to a clothing;

Fig. 6 is an example of a system according to an embodiment; and

Fig. 7 is a flowchart illustrating a method according to an embodiment.

Description of Example Embodiments

The following description and drawings are illustrative and are not to be construed as unnecessarily limiting. The specific details are provided for a thorough understanding of the disclosure. However, in certain instances, well- known or conventional details are not described in order to avoid obscuring the description. Reference in this specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosure.

The present embodiments relate to a wearable health measurement device. Despite the term, the device can be considered as an activity measurement device, well-being device, health device, or any device that is able to measure health-related parameters or signals from a user’s body and is able to derive analysis therefrom relating to user’s well-being. Shortly, the wearable health measurement device is referred interchangeably to as “a device” or “an apparatus”. Term “health-related parameters” refers to electrical and nonelectrical signals that can be measured and monitored from a user’s body. In addition, for the purposes of the present embodiments, “health-related parameters” also cover other measurable data from a user or user’s environment. Such data may comprise any one or more of the following: temperature, humidity, altitude, air condition, air quality, carbon dioxide (CO2) or carbon monoxide (CO) or other gases in the air. It is appreciated that in addition to these examples, or alternatively, other environmental data can be measured as well.

Figure 1 illustrates a structure of the wearable health measurement device

100, according to an embodiment. It is to be noticed that the illustration in Figure 1 is a simplified example of a possible configuration of the device, and the design can vary from what is shown in Figure 1. The device may be consisted of two parts, wherein one of the parts is a measurement part that houses electronics and sensors. Another part, i.e., clip part, acts as fastening mechanism. The parts of the device can be mechanically connected, or the parts of the device can be integrated.

Figure 1 shows the device 100 comprising a measurement part 101 and a clip part 103. The measurement part 101 comprises sensing elements (not shown in Figure 1 ) and purpose of the clip part 103 is to attach the device 100 to a place (e.g., a clothing, a garment) where the measurement is to be carried out from the user’s body.

The measurement part comprises a first side 101 a and a second side 101 b, wherein the first side 101a is (at least partly) in contact with user’s skin when the device 100 is used, and the second side 101 b is facing the clip part 103. The first side 101 a of the device 100 comprises the sensors/electrodes for measuring data from the user. The second side 101 b of the device 100 is outwards from user’s skin and body. The clip part 103 is connected to the measurement part 101 with a hinge 104 at one end of the measurement part

101. The device 100 comprises a groove part 105 between the measurement part 101 and the clip part 103 to receive the edge of the clothing or waistband. The clip part 103 has a design, where a locking has an offset distance between a closing end 106 and the hinge 104 axis. When the clip part 103 is opened, the closing end 106 opens and allows a clothing or garment to be set to the groove part 105.

Figure 2 illustrates an example of the measurement part 101 according to an embodiment. As said, the measurement part 101 comprises a plurality of sensors for measuring data from a user. The measurement part 101 may comprise first sensor(s) a located on (or in primary transducing contact with) the first side facing the skin, and second sensor(s) located on (or in primary transducing contact with) second side of the measurement part 101. The second sensors may benefit from the mechanical properties of the clip mechanism. The first sensors may comprise at least one heart sensor, such as a photoplethysmogram (PPG) sensor, and skin temperature sensor. The PPG sensor may comprise at least one light-emitting diode (LED) and one photodetector. In addition, the first side may have one or more light emitters, and one or more light detectors for optical measurements, e.g., plethysmography. In addition, the first sensors may comprise at least one skin electrode.

The sides (e.g., second side) not facing the skin may include, e.g., one or more instances of a contact temperature measurement point (temperature sensor), and infrared (IR) sensor, ambient light sensor, or a relative humidity measurement (RH) sensor.

In the example of Figure 2, there are thermal contacts including temperature sensors 202 to electrode areas 201. In addition, in the example of Figure 2, there is an infrared (IR) temperature sensor 203, which is does not have to be in contact with the skin. In addition, or alternatively, the IR temperature sensor can also be arranged on the second side of the measurement part, as described above. In the example of Figure 2, there are also receivers, such as photodetectors 204 (“photosensors”), and light sources, such as light-emitting diodes (LEDs) 205 for IR, red light, and green light. In the example of Figure 2, there are also electrical contacts 206 to electrode areas 201 . Finally, the electrode areas 200 are provided with metal or other electrical conducting material (not shown in Figure 2) on the top of the electrode area grooves. These elements 201 , 202, 203, 204, 205, 206 are provided on the first side 101 a of the measurement part 101 .

The device may also comprise at least one movement sensor (e.g., linear acceleration)(not shown in Figure 2) situated inside the measurement part 101 .

The device may have electrodes or contact points for electrodes on all sides of the device. On the first side of the measurement part, they directly interface the skin impedance or biopotential measurements. On the other sides, the electrodes may constitute an impedance-based humidity/evaporation measurement or interface the clip to external circuitry on clothing. Thus, as shown above, the measurement part comprises a plurality of sensors being configured to measure health-related data from user’s body, which health-related data is used to make analysis on user’s health and well-being. Also, the measurement part may comprise sensors being configured to measure data from the environment. In addition to, or instead of, the sensors as discussed above, the device may comprise one or more of the following sensors:

- optical heart rate sensor(s) (OH RM), which can operate in a transmissive and/or reflective mode;

- electrodes and electronics for electrocardiography (ECG);

- electrodes and electronics for measuring electrical impedances, such as galvanic skin resistance (GSR) and/or bioimpedance analysis (BIA);

- contact or contactless temperature sensor(s), such as one or more of the following: therm istor(s), infrared temperature sensor(s), thermocouple(s);

- heat flow sensors, which are also interchangeably referred to as heat flux sensor(s);

- humidity sensor(s);

- inertial sensor(s), such as one or more of the following: linear accelerometer(s), magnetometer(s), gyroscope(s);

- barometer(s);

- mechanical force sensor(s), such as pressure sensor(s) and/or strain gauge(s);

- acoustic sensor(s), such as contact microphone(s) and/or contactless microphone(s);

- optical sensor(s), such as charge-coupled device (CCD) imaging sensor(s) and/or complementary metal-oxide-semiconductor (CMOS) imaging sensor(s).

The optical heart rate sensor may be placed close to the point, where the wearable device experiences the greatest mechanical support from the user’s clothing, such as near the joint of the clip mechanism.

The humidity sensor(s) may be placed such that at least one of the sensors is located away from the user’s skin, but underneath the user’s clothing, such as at one end of the wearable device. The measurement data being measured by the sensors may comprise one or more of the following: PPG, HR, skin conductance/impedance, skin temperature, electrocardiography (ECG), tissue impedance for determining tissue content (fat, muscle, bone), electromyography (EMG), infrared sensor for skin temperature, humidity, accelerometer.

Based on the measured data, a core temperature can be defined. Core temperature refers to temperature within the body where vital organs are located. The core temperature is different (typically at least one Celsius degree higher) from a skin temperature. The core temperature may be defined based on a heat flow channel. A heat flow channel is formed between at least a first temperature sensor being located on the first side of the measurement part and a second temperature sensor being located on the second side of the measurement part, which both measure different temperatures. For more accurate results, two heat flow channels may be used, which two heat flow channels are formed between two first temperature sensors and two second temperature sensors. When the temperatures of these two heat flow channels are measured, it is possible to calculate the body core temperature generating the heat flux.

The principle of two heat flow channels (i.e., two channel heat flux sensor) is presented in Figure 3a. Ti and T2 refers to temperatures measured by the first temperature sensors from the skin 310. T3 and T4 refers to temperatures measured by the second temperature sensors located on the second side of the measurement part. Tcore is the temperature of the core 320 of the body. Rskin is heat resistance of the skin 310, and R1 and R2 are heat resistance of the two heat paths (between T1 and T3, and between T2 and T4) inside the measurement device. Assuming that the heat flow from the deep body into the measurement device remains the same, Fourier’s law applies as follows:

(Equation 1 )

and

(Equation 2)

Now, core temperature Tcore can be determined as: ( \Equation 3) /

where k=(Ri/R2) and can be represented as the ratio of the heights or distances or related parameters of the thermal paths (such as cylinders) used in the measurement construction structure. The ratio k can be defined, for example, empirically from the material parameters and dimensions of the heat flow structure of the device.

The thermal paths may be arranged so that temperatures Ts and

are measured by the second temperature sensors located close to another side of the measurement part of the device. It is also possible to arrange the thermal pathways so that the one or the both second temperature sensors are placed inside the measurement path.

More detailed analysis can be found in the research paper Structural Optimization of a Wearable Deep Body Thermometer: From Theoretical Simulation to Experimental Verification, M. Huang et al., Journal of Sensors, vol 2016, Article 4828093.

The above formula can be assumed to work under steady-state conditions. However, in practice, the device may become momentarily disconnected from the skin. For example, when the ambient temperature is lower than that of the skin, the disconnection causes a rapid drop in the observed temperatures T1...T4, followed by longer settling times as the device becomes in contact with the skin again. For any of the temperatures 77 (/ = 1... 4), the event of the device becoming disconnected and then reconnected with the user’s skin can be visualized as in Figures 3b and 3c. The time constants a (i = 1...4) of each temperature sensor 77 (/ = 1...4) shows how fast or slow each temperature sensor changes to a new temperature value. The events of the device becoming disconnected with skin can occur up to dozens to times per day, and the settling of thermal measurements back to their steady-state values can take approximately from ten minutes to over an hour. This can add up to a significant fraction of the data being unreliable and unusable.

When a humidity of the skin is changed or sensor’s contact from the skin comes off, the measurement of the core temperature may suffer. Heat flow sensors have low time constants. However, these time constants may be estimated and determined, whereupon needed corrections may be done. The present embodiment provides a solution to detect if the skin contact is loosened. Therefore, the present embodiments provide various models for a skin temperature Tskin according to which it is able to detect how the skin temperature behaves.

Therefore, according to an embodiment, body core temperature is determined by heat flow sensors (i.e., one or more pairs of temperature sensors). Heat flow describes how much heat energy is going through the structure being measured. The heat flow can define the amount of energy as well as properties of a tissue being in contact with the heat flow sensor. Each time the heat flow values are measured, they are stored as history and reference data to storage means, e.g., to a memory of a smart device or to cloud. Then newly measured values may be compared to the reference values or earlier history data. The skin temperature values measured by conventional temperature sensors may complement the heat flow data. In addition, the heat flow data can be combined with HR, skin conductance or skin electrical impedance, PPG signals, whereupon the heat flow values can be used, for example, for recognizing rhythms and activity in stomach/waist area for detecting eating.

As discussed above, the heat flow sensor is formed of at least one pair of temperature sensors. According to another embodiment, the heat flow sensor can be formed of a non-contact (i.e., not having a contact with a skin) temperature sensor 203, at least one pair of temperature sensors and a humidity sensor. However, according to yet another embodiment, the humidity sensor is not required. The non-contact temperature sensor 203 can be an IR sensor being located on a first side 101a of the measurement part 101. IR sensor is able to determine object’s skin temperature based on infrared radiation in a known manner.

The pair of temperature sensors can be formed of a first temperature sensor 202 and second temperature sensor, wherein the first temperature sensor 202 is in contact with the skin and is located on a first side 101 a of the measurement part 101 , and wherein the second temperature sensor is located inside the measurement part 101 close to the second side 101 b of the measurement part 101 or located on the second side 101 b of the measurement part 101. The non-contact temperature sensor 203 may measure a skin temperature close to the first temperature sensor 202.

In such embodiment, the non-contact temperature sensor 203 can be utilized for quick and non-contact measurement for a skin temperature, according to which the measurement of the first temperature sensor 202 can be verified. In a normal situation, the measurement of the first temperature sensor 202 and the measurement of the non-contact temperature sensor 203 are aligned. For example, the skin temperature Tskin and the non-contacted skin temperature TNC may be monitored and measured and compared to each other. Tskin can be temperature Ti from the first temperature sensor 202, or temperature T2 from the second temperature sensor. When their difference (in percents or degrees) exceeds a predetermined limit (as shown in Figure 3c, T, (1,2, 3, or 4) - TNC ), it is determined that the measurement Tskin is different. This may indicate that the contact of the temperature sensor to the skin is not good, whereupon the measurement of the temperature sensor can be corrected. Correspondingly, when the skin contact is restored, i.e., the device contacts the skin again, the temperature difference T, (1,2, 3, or 4) - TNC drops first. This indicates that the non-contact temperature sensor 203 having a fast temperature response is measuring almost the real skin temperature, and the skin temperature sensor starts to measure the skin temperature and its value is approaching the skin temperature by its time constant a. Tskin is used here as an indication of any used temperature sensor T1-T2. For implementing this, a contact model can be used for heat flow sensor or a first temperature sensor 202 that is against the skin. The contact model refers to behavior of each temperature sensor (T1 -T4), how their temperature value is changed if the sensor comes off from the skin or comes on to the skin. The device can use the contact model to correct the heat flow value, until the first temperature sensor 202 is restored and the measured temperature is returned to the value it had when the contact was good.

Fig 3b shows an example of a situation where the one or both of the first temperature sensors 202 comes off the skin at time point ti. Temperatures Ti and T2 measured by the first temperature sensors 202 change from the skin temperature Tskin towards temperature of the environment TENVIRONMENT. The speed of the change is determined by their time constants, ci and 2 which are determined by the heat mass of the sensors and construction. When the first temperature sensors are back to the skin contact (time points t4 and ts), the temperatures Ti and T2 will start approaching the temperature of the skin, TSKIN, reaching it at time points te and t , respectively.

The time constants can be determined by measuring the temperature change at determined temperatures or during normal use between contacting first temperature sensors and non-contacting first temperature sensors from the skin. When knowing the time constant, temperatures T1 and T2 can be modelled to correct the heat flow measurement to get more accurate core temperature Tcore and also to get correct core temperature Tcore faster after the change of the skin contact of the measurement device and first temperature sensors.

It is important to define time points when the device becomes disconnected, and then reconnected with the skin.

As described above, means for measuring a signal from the skin for determining a skin contact can be a non-contact IR temperature sensor that is used as a fast response temperature sensor whose value can be compared to the slower response skin temperature sensor, which is first sensor of the heat flow sensor. Alternatively, a skin conductance or electrical impedance measured between two electrodes on the skin can be also a skin contact monitor, i.e., means for measuring a signal from the skin for determining a skin contact. Also, optical reflectance value of optical heart rate sensor (PPG) can be used as an indication of the skin contact. A signal related threshold value can be set to make decision when the measured signal is indicating that device is having a skin contact (SC-yes) when it is not (SC-no). This has been illustrated in Figure 3c. Figure 3c also shows some examples of signals (noncontact temperature sensor such as IR temperature sensor, optical reflectance sensor, and electrical impedance sensor), and corresponding thresholds when they can form the output (skin contact) indicating if the device is having skin contact or not.

As mentioned above, the events of the device becoming disconnected with skin can occur up to dozens to times per day, and the settling of thermal measurements back to their steady-state values can take from approximately ten minutes to over an hour. This can add up to a significant fraction of the data being unreliable and unusable. Thus, it is valuable to try to minimize time when the data is not available or it is not reliable.

By means of the present embodiments, data coverage in daily use is improved by estimating the value which the observed temperatures are converging when the device is still thermally settling. For any of the temperatures T1...T4 (in Equation 3), this can be achieved as follows: i. detecting the event of the device becoming unattached to skin. This can be done by e.g., observing rapid negative changes in the observed temperature, or by other means, such as the loss of signal quality in PPG measurement; ii. discarding the temperature observations when the device is not in contact with the skin; ill. after the skin contact is re-established, estimating the steady-state value which the temperature observations converge.

Step (iii) involves two assumptions. At first, it is assumed that the observed temperatures indeed converge towards a steady-state value. In the time scale of dozens of minutes or 1 -2 hours, this is a reasonable assumption, slight changes in ambient or skin temperature can be treated as noise. Secondly, it is assumed that each of the temperatures 77 can be treated separately as a first-order system, the time constant of which can be determined by the thermally connected masses close to temperature Ti, such as the mechanical structure of the device, tissue and/or fabric surrounding or close to temperature sensor measuring the temperature Ti. Under these assumptions, the process of thermal settling for temperature Ti is treated as the step response of a first- order system. When the skin contact is reacquired, it is assumed that the steady-state value ,_ss, which is converging to, can be represented as a sum of the current observed temperature and some residual term ,res. Thus, the observation model for the temperature becomes

Ti,ss = T + f_{i res} (Equation 4)

Because the thermal settling can be treated as a step response of a 1 st order system, the residual term , _res is of the form T_{i res} ( * Ae^at (Equation

The time derivative of the residual term becomes

^^res = aAe^at (Equation 6) from which coefficient s can be obtained as

(Equation 7)

Substituting Equation 7 to Equation 6 yields

(Equation 8)

Noting that dT_{i res} —dTi dt dt time constant correcting coefficient can be denoted p := -a^-1, obtaining

Ti.res = P ^ (Equation 9) which, substituting to Equation 4, finally yields

Ti,ss = Ti + P ^ (Equation 10)

For calculating the convergence value Ti,_ss during thermal settling, the coefficient fi as well as the time derivative of 77 may be determined. For discrete-time sampled observations Ti[t], the latter can be calculated e.g., as the forward difference between two temperature observations. However, the coefficient fi requires either empirical testing or numerical optimization.

Depending on the location of the temperature sensor, the coefficient /? can be determined either by the thermal time constant of the device alone, or by the combined time constant of the device and the tissue the device is in contact with. As the latter is affected by several factor, such as the body composition of the user, the coefficient /3 is optimized here using numerical methods. To numerically calculate the settling-corrected values of Ti, estimates of the temperature derivatives (denoted here as T for brevity) may be needed. For discrete-time signals Ti[t], the simplest way is to utilize the backward difference, i.e.

T t] = T t] - T t - 1] (Equation 11 )

However, derivative estimates obtained by simple backward or forward difference methods are prone to noise, which can occur due to many reasons during thermal settling (such as contact variations, humidity build-up, air currents etc.). Therefore, more robust methods can be utilized for obtaining the derivatives, such as the combination of forward and backward differences, lowpass filtering of the derivative signal, or curve fitting methods such as the Savitsky-Golay filter. After calculating the derivative estimates T_t , the full Tcore model becomes C_0re(/c, Pl> P2> P > P ) ⁼

(Equation 12) where k is a predefined constant, and fii...p>4 can be obtained e.g., by means of numerical optimization.

As the model (Equation 12) relies on estimation of the derivative terms, the temperatures may be let to settle for some period after reacquiring skin contact. For example, a wait time of 10 - 180 seconds, preferably 60-120 seconds, can allow the temperature signals to recover a large part of the temperature drop, after which the estimation of the /3-dependent correction terms in Equation 10 becomes more reliable, as the derivative estimates become more accurate. If the device and skin reach thermal equilibrium, the thermal readings become constant, reducing the derivative terms T_t to zero. In such a case, the equation 12 will reduce to its original, uncorrected form (Equation 3). To optimize the coefficient /3 values, a loss function L can be defined. An example of such a function is to apply a rolling variance operator to the time series data obtained by evaluating Equation 12, with the length of the rolling function window 1/1/ set to two hours, and taking the grand mean of the result: Equation 13

Instead of applying an arithmetic mean as the outer aggregate function in Equation 13, also other methods such as a mean of some high percentile (e.g., 80%) can be calculated, to place a greater emphasis on the settling events. The optimization task can be carried out by any suitable numerical optimization algorithm, such as the Nelder-Mead method.

In addition to the time constant coefficients /?, the settling-corrected 7^ model (Equation 12) (as well as the original, uncorrected model (Equation 3)) may require the thermal resistance ratio to be specified. While this can be done empirically by placing the device in contact with a heat source with a known temperature, and numerically estimating the value for to match the output of the Tcore model with the known temperature, the ratio k can also be estimated without such a calibration setup. For example, assuming that the thermal resistance of both of the heat paths (Ti... T₃) and (T₂. . . T₄) is high compared to that of the skin and subcutaneous tissue, the ratio can be estimated as the ratio of temperature differences. Because the temperature difference in both of the heat paths is proportional to the thermal resistance of the path, it can be assumed that (Equation 14)

Where the temperature values may be recorded when placing the device in contact with a heat source, such as the user, and calculating the k estimate using the data collected. Similarly, the ratio may be estimated as the ratio of slopes between the heat channels (Equation 15)

where the measurement may be performed as above but collecting the measurements at several different skin temperature values T?and T₄.

Skin contact of the heat flow sensor can also be checked by an impedance measurement. This can be performed especially if the sensor areas are the same or close to each other or thermally connected.

The heat flow sensor, according to another embodiment, can contain two or more pairs of temperature sensors located in such way, that one sensor of a pair of sensors is in skin contact, and another sensor of the pair of sensors is not in contact with the skin, but e.g., inside the measurement part. By this configuration, a heat flow channel is formed between the pair of sensors. The different heat flow channels have different heat flow capabilities and have different heat flow resistivities. On the other hand, the heat flux between the different heat flow channels, i.e., between the different pairs of sensors, is different. This is because the thermal resistance or conductance is different in different channels, because they have different amount of thermal insulation.

The temperature sensors according to embodiments may be able to measure temperature at a resolution of at least 0.01 degrees Celsius relative to one another, and 0.1 degrees Celsius absolute temperature accuracy. It is appreciated that in some embodiments, the temperature scale can be other than Celsius scale.

Thus, a body core temperature is defined by using a heat flow arrangement as discussed above. The wearable health measurement device (as shown in Figures 1 and 2) comprising the heat flow arrangement may be attached to the band or edge of underwear or pants or other clothing. The first side 101 a of the measurement device 100 comprises a metal area (i.e., an electrode area) 201 that is connected thermally to a first temperature sensor 202 and the second temperature sensor is inside the device 100 close to the second side 101 b. The method comprises forming a heat flow channel along the electrode area 201 , the first temperature sensor 201 , a medium between the first and the second temperature sensors, the second temperature sensor, and the second side 101 b of the device, and possible the material of the clothing. A time constant of each temperature sensor is known and can be expressed by a formula f= (T, t), wherein temperature is denoted with T and time is denoted with t. As described above, the device may comprise a non-contact temperature sensor 203, e.g., an IR sensor, which can measure the skin temperature close to the first temperature sensor 202. If the temperature measured by the first temperature sensor 202 drops and it is different to the non-contact sensor 203, then it is determined that the first temperature sensor 202 does not contact the skin. Instead of non-contact temperature sensor, other means to detect disconnection of the measurement part from the skin. When it is determined that the first temperature sensor 202 is not in contact with the skin, the correction is made to the first temperature sensor value and to the second temperature sensor value based on the formula f. The body core temperature value can be calculated based on the corrected values of the first temperature sensor 202 and the second temperature sensor. When the first temperature sensor value is close to the non-contact temperature sensor value (and the second temperature value is stabilized or after a certain time period), the correction of the first and the second temperature values are not used.

According to an embodiment, the temperature drop of the first temperature sensor 202 is detected as a drop measured as a difference to the temperature of the non-contact temperature sensor.

According to an embodiment, the wearable health measurement device can comprise two heat flux channels (i.e., a first heat flux channel and a second heat flux channel), both having their own first temperature sensors 202 and second temperature sensors.

According to an embodiment, the heat flux through the first heat flux channel and the second heat flux channel are different.

According to an embodiment, the wearable health measurement device may have an additional temperature sensor inside the device, which additional temperature sensor comprises a heat flux medium (on the PCB). The additional temperature sensor is used for correcting the time constant and the formula f.

In addition to the two heat flux channels, with skin temperatures Ti and T2, the device may be equipped with a third skin temperature sensor T5, located between temperature sensors Ti and T2. One possible location for such a sensor is the housing of the PPG sensor unit (Fig. 2, 207). Difference between T5 and T1 and/or T2 can be used to estimate the divergence of the heat flow through the device, by examining the lateral temperature differences in the skin side of the device.

This information can, for example, be utilized to estimate the temperature distribution underneath the sensor, which in turn can be utilized for inferring the state of thermal settling. For example, when the differences between T5, T1 and T2 decreases below a certain threshold, e.g., 0.1 degrees Celsius, the device can be treated as fully thermally settled, which indicates that the thermal readings are reliable.

Previously defined thermal resistance ratio k and time constant correction coefficients (/?/) can be defined once, or they can be defined every time when there is new reconnecting of the measurement part to the skin. Once defined, a group of time constant correction coefficient (/?/) and group of thermal resistance ratio k can be named as a one calculation model. It is possible to define different calculation models for different use cases or use conditions. It is also possible to define thermal resistance ratio k and time constant correction coefficients (/?/) independently, so that for example one group of time constant correction coefficient (/?/) are used with one thermal resistance ratio k. Different models can be defined for different humidity conditions or wet and dry skin conditions, for example. Also, different calculation models can be defined for different skin conductance or skin resistance or skin impedance representing different skin or tissue conditions.

According to an embodiment, the wearable health measurement device comprises a humidity sensor configured to measure humidity of air close to the device. The humidity sensor may be located on a second side 101 b of the measurement part. The humidity value may be used to correct the non-contact temperature sensor (i.e., IR sensor) value. The humidity value may be used for determining whether the skin is wet (e.g., due to sweating). When the skin is wet, the device is able to determine that the heat flow measurement is not reliable. The humidity value may be used for selecting a temperature calculation model to be used for defining core temperature. According to an embodiment, the wearable health measurement device comprises means for measuring a skin impedance between the metal electrodes. The impedance value may be used for detecting if the first sensor is in contact to the skin. The impedance value may further be used for correcting the time constant of the temperature sensors on the first side and on the second side of the device part.

As mentioned above, the humidity value can be used for selecting a temperature calculation model. Alternatively, or in addition, also time of the day (i.e., day or night) can be used for selecting the temperature calculation model.

Figure 4a illustrates an example of the wearable health measurement device being attached to a waistline 310 of a clothing 300. As show in Figure 4a, the clip part 103 of the device is located outside the clothing 300, and the measurement part 101 is located inside the clothing 300 so that the measurement part 101 is at least mostly in contact with user’s skin. Figure 3b is an exploded view of the wearable health measurement device of the intersection A-A of Figure 4a. In Figure 4b, elements the measurement part 101 with hinge 104 and the clip part 103 are shown. The first side 101a of the measurement part 101 is towards user’s skin. The waistline 310 fabric of the clothing 300 is shown to be located between the measurement part 101 and the clip part 103. In addition, Figure 4b shows a textile contact area 315 on the clothing 310 that is in contact with the second side 101 b of the measurement part 101 .

The purpose of the clip part 103 is to attach the device to the user’s clothes such as the waistband of pants or underwear or other clothing. The clip part 103 may include a base part and a cam actor level mounted rotatably to the base part. The clip part 103 is attached to the measurement part on a side other than the skin-contacting side. Thus, the device is mechanically attached into user’s clothing in such a way that the device is in contact with the user’s skin. The clip mechanism allows attaching and detaching of the device easily without damaging the clothing.

Figure 5 shows the wearable health measurement device according to another embodiment being attached to a clothing 300. In this embodiment, the device is additionally electrically connected to a piece of clothing, such as underwear, such that the piece of clothing can be used to provide e.g., external skin contact electrodes for ECG or BIA measurements. Thus, in the embodiment of Figure 5, the clothing comprises a textile conductor 420 within the waistband 310 and an additional electrode area 425, wherein the conductor 420 electrically connects the additional electrode area 425 to the device. The additional electrode area 425 may comprise silver conductive fabric and an evaporative guard (e.g., nitrite). The additional electrode area 425 may be placed on the waistband 310 of the clothing 300. In this embodiment, the clothing may be a specific clothing to be used with the wearable health measurement device according to the embodiments. The clothing 300 of this example may be sold as a device accessory, and in addition to the electrode area the clothing may have further measurement units.

In addition to above elements, the wearable health measurement device according to an embodiment comprises other electronics as follows:

- a microprocessor;

- memory for storing data measured by the sensors;

- a radio frequency communication device for relaying the data measured by the sensors;

- a battery;

- optionally one or more of the following: wireless charging means; a display and/or indicator lights and/or switches in the enclosure of the device; a loudspeaker; a mechanical vibrator.

The physical structure and enclosure of the wearable health measurement device according to embodiments have been designed such that the device has few or no sharp edges for user comfort. In addition, the wearable health measurement device according to embodiments has a flat shape. The thickness may be less than 5mm. However, in some embodiments the thickness may be 5mm or more. Other dimensions (width and/or height and/or diameter) may be less than 50mm. However, in some embodiments, one or more of the dimensions may be 50mm or more. It is appreciated that these dimensions may have the same size or each of them may have their own size differing from the others. According to an embodiment, the measurement part has larger dimension in thickness than the clip part. This enables a contact pressure against the skin. The wearable health measurement device according to embodiments is able to perform measurements from the user and the environment continuously at predetermined intervals. Alternatively, the device can reduce, halt, or continue its measurements and operation under an external influence, such as movement, in order to increase the battery’s life of the device. The device can store all of its measurement data from a time span of 24 hours. Alternatively, the device can store the measurement data over a smaller or larger time period. The measurement data from the wearable health measurement device can be transferred to another device, such as a mobile phone, for processing, data visualization and user feedback. Alternatively, or in addition, the measurement data can be delivered from the device to an external server (e.g., in a cloud network), where the processing and long-term storage of the measurement data takes place. The data can be downloaded from the cloud to another device, such as a mobile phone, for data visualization and user feedback.

Also, the wearable health measurement device may comprise means for giving feedback to a user. Such means may comprise a LED of piezo or vibrating element to provide optical and/or haptic feedback and/or alarms and/or indication triggered by measured data and/or time or other feedback defined by the device or communicated from a mobile device through a wireless connection to the device,

Some of the embodiments of the wearable health measurement device have been discussed above. Some of the further embodiments are considered next.

According to an embodiment, the thickness of measurement part 101 may be two times bigger that the thickness of the clip part 103. The thickness of measurement part may be 3-10mm and the thickness of the clip part may be 1 - 4 mm measured from the thickest point.

According to an embodiment, a PPG sensor may comprise at least one light source (e.g., a LED) and one receiver (e.g., a photodetector). According to another embodiment, the PPG sensor comprises at least two light sources and two receivers. The line of light sources and the receivers may be perpendicular to the axis of the hinge. According to an embodiment, the first side 101 a may comprise two skin electrodes (against the skin) for measuring the impedance between the electrodes.

According to an embodiment, the electrode area may have larger dimension in the perpendicular direction to the axis of the hinge.

According to an embodiment, the second side of the measurement part may have at least one temperature sensor to form a temperature heat channel with one temperature sensor on the first side to form a heat flux sensor.

According to an embodiment, the second side of the measurement part may have at least one electrical connecting area for contacting the second side to the external electrode area supported by the inner side (towards the skin) of the upper edge of the underwear or panties.

According to an embodiment, the clip locking mechanism may tighten the upper edge of the garment between the clip and the second side of the device preventing the garment to move after locking the clip.

According to an embodiment, the edge of the garment may be positioned at least over the middle point the measurement part in vertical (y- axis) dimension.

According to an embodiment, open space may be formed in the part of the wearable device, between the measurement part and clip part when they are closed, and the garment is set to its designed position.

According to an embodiment, one electrode of the first side can be used as first ECG electrode and the electrode connected to the connecting area of the second side can be used as the second ECG electrode. The ECG signal can be measured between the first and second ECG electrodes.

According to an embodiment, one electrode area may be used for impedance or ECG measurement is connected thermally to the skin temperature sensor, so that metal of the electrode area conduits the skin temperature to the skin temperature sensor connected thermally to the electrode in one point of the electrode area.

The measurements made by the wearable health measurement device generate an output, such as HR, heart rate variability (HRV), skin temperature, body temperature, perfusion index, ECG, EMG, oxygen saturation (SpO2), activity, body position (sitting, standing, laying), sleep stature, respiration. These outputs will further be formed and communicated with features such as overall vitality; sleep and recovery, activeness, and strain, HRV and respiration, circadian alignment, hormonal alignment, biological rhythms, and habits identification, feeding or meal detection, readiness, forward-looking recommendations, etc.

The wearable health measurement device can transmit the collected data to a client application that can locate on a mobile device. Figure 6 illustrates such an example. The implementation may be done in a similar way as the data is transmitted from a smartwatch or smart ring to a mobile device and further to a (e.g., cloud) server. In the example of Figure 6, the wearable health measurement device 100 forms a data transfer network with a mobile device 610 for data delivery 1. The mobile device can be a smartphone, a tablet device, a laptop, a general computer etc. The data transfer network may be a Bluetooth connection, a Near Field Connection (NFC) or any other current or future wireless or wired data transfer network. For that respect, both the wearable health measurement device 100 and the mobile device 610 have corresponding radio interface and communication means. The mobile device 610 stores the client application 620 that may have been downloaded from a server to the mobile device 610. Alternatively, the mobile device 610 may only comprise a browser that is able to retrieve a view to the client application being located in a (e.g., cloud) server. However, despite the location of the client application, the client application is part of the computing system of the wearable health measurement device.

The purpose of the client application 620 is at least to visualize the data received from the wearable health measurement device 100 and generate feedback or analysis on the well-being of the user. The mobile device 610 may be in connection with a (e.g., cloud) server 650 to deliver data from the client application 620 to the server 650. The server 650 may gather data from multiple users, to make broader analysis on a general wellbeing of plurality of users and the generate statistics. According to some embodiments, the user-related analysis may also be done at the server 650 instead of the mobile device 610, whereupon such analysis may be returned back to the mobile device 610 for visualization. The data transfer network 2 between the mobile device 610 and the server 650 may be created by any known wireless or wired communication technology. Examples of such are mobile networks of different generations (3G, 4G, 5G, etc.) and local and/or wide area networks.

The mobile device 610 comprises at least one processor and at least one memory, wherein the memory stores computer-implemented instructions according to which visualization of data, processing of data, analysis of data, etc. can be implemented. The memory may also store measurement data. In addition, the mobile device 610 comprises a user interface, e.g., in the form of graphical user interface (GUI). In addition to the GUI, the mobile device 610 may contain other user interaction means, e.g., a microphone and/or loudspeakers.

The measurement data received from the wearable health measurement device may be at least partly stored in the mobile device 610 and at least partly in the cloud 650. For example, the cloud 650 may store history data, whereas the mobile device 610 may store measurement data obtained within a certain shorter time period (e.g., a week, a month).

It is appreciated that a client application may be interpreted to be an element belonging to the wearable health measurement device, despite it has a physical location at the mobile device. Therefore, the wearable health measurement device may utilize capabilities (e.g., processing power) of the mobile device through the client application.

The method according to an embodiment is shown in Figure 7. The computer- implemented method for determining health-related parameters generally comprises receiving 710 a first temperature measured by a first temperature sensor of wearable health measurement device, the first temperature being a skin temperature; receiving 720 a second temperature measured by a second temperature sensor of the wearable health measurement device, wherein the first temperature sensor and the second temperature sensor has a heat flow channel therein between; measuring 730 signal from the skin for determining when the first side of the measurement part is disconnected from the skin and when the first side of the measurement part is reconnected to the skin; and determining 740 a body core temperature value based on the heat flow channel and taking into account the measured signal on the disconnection or reconnection of the measurement part. Each of the steps can be implemented by a respective module of a computer system.

If desired, the different functions discussed herein may be performed in a different order and/or concurrently with other. Furthermore, if desired, one or more of the above-described functions and embodiments may be optional or may be combined.

Although various aspects of the embodiments are set out in the independent claims, other aspects comprise other combinations of features from the described embodiments and/or the dependent claims with the features of the independent claims, and not solely the combinations explicitly set out in the claims.

It is also noted herein that while the above describes example embodiments, these descriptions should not be viewed in a limiting sense. Rather, there are several variations and modifications, which may be made without departing from the scope of the present disclosure as, defined in the appended claims.

Claims

Claims:

1. A wearable device (100) for measuring health-related parameters from a user, the device (100) comprising

- a measurement part (101 ) and a clip part (103), wherein the measurement part (101 ) comprises a first side (101 a) and a second side (101 b), the first side (101 a) being in contact with user’s skin, and wherein the clip part (103) is connected to the measurement part (101 );

- wherein the first side (101 a) of the measurement part (101 ) comprises at least a first temperature sensor (202) to measure a first temperature, the first temperature being a skin temperature, and wherein the measurement part (101 ) comprises at least a second temperature sensor to measure a second temperature, wherein the first temperature sensor (202) and the second temperature sensor comprises a heat flow channel therein between, and wherein each of the temperature sensors has a time constant;

- the measurement part (101 ) comprising means for measuring signal from the skin for determining when the first side (101 a) of the measurement part (100) is disconnected from the skin and when the first side (101a) of the measurement part (100) is reconnected to the skin; and

- means for determining a body core temperature value based on the heat flow channel and taking into account the measured signal on the disconnection or reconnection of the measurement part (100).

2. The wearable device (100) according to claim 1 , wherein said means for measuring signal is a non-contact temperature sensor (203) configured to measure a third temperature close to the first temperature sensor (202), wherein the wearable device (100) further comprises means for detecting whether the first temperature is different from the third temperature.

3. The wearable device (100) according to claim 1 , wherein said means for measuring signal is a skin conductance sensor using two skin electrodes on the first side of the measurement part.

4. The wearable device (100) according to claim 1 , wherein said means for measuring signal is an optical reflectance sensor on the first side of the measurement part.

5. The wearable device (100) according to any of the preceding claims 1 to 4, wherein when it is determined that the device (100) is disconnected from the skin, the data is discarded, and when it is determined that device (100) is reconnected to the skin, the first and second temperature values are corrected for receiving the core temperature of the body of the user.

6. The wearable device (100) according to claim 5, further comprising means for performing the correction by defining a time constant coefficient (/3) to each of the first (202) and second temperature sensors.

7. The wearable device (100) according to claim 6, wherein the time constant coefficient (0) is defined numerically from the data measured by using any combination of the following: forward and backward differences, lowpass filtering of the derivative signal, or curve fitting methods.

8. The wearable device (100) according to any of the claims 1 to 7, further comprising two heat flux channels, both having their own first temperature sensors (202) and second temperature sensors.

9. The wearable device (100) according to any of the claims 1 to 8, further comprising a humidity sensor configured to measure humidity of air close to the device.

10. The wearable device (100) according to any of the claims 1 to 9, further comprising means for measuring a skin impedance between electrodes or electrode areas (201 ).

11 .The wearable device (100) according to any of the claims 1 to 10, further comprising means for selecting a temperature calculation model.

12. The wearable device (100) according to any of the claims 1 to 11 , further comprising a third temperature sensor, and means for determining a difference between the third temperature sensor and at least one of the first temperature sensors.

13. The wearable device (100) according to any of the claims 1 to 12, further comprising at least one processor, and a memory including computer program code as means for determining the body core temperature value based on the heat flow channel.

14. A computer-implemented method for determining health-related parameters, the method comprising

- receiving a first temperature measured by a first temperature sensor (202) of wearable health measurement device (100), the first temperature being a skin temperature;

- receiving a second temperature measured by a second temperature sensor of the wearable health measurement device (100); wherein the first temperature sensor (202) and the second temperature sensor have a heat flow channel therein between, and wherein each of the temperature sensors has a time constant;

- measuring signal from the skin for determining when a first side (101 a) of a measurement part (100) of the health measurement device (100) is disconnected from the skin and when the first side (101 a) of the measurement part (100) is reconnected to the skin; and

- determining a body core temperature value based on the heat flow channel and taking into account the measured signal on the disconnection or reconnection of the measurement part (100).