Method and system for estimating whole-body cutaneous ionic loss

Abstract

A computer-implemented method is described for estimating whole-body cutaneous loss of a target ion from a subject over a time period. The method comprises estimating a local ionic concentration within sweat at a localised sub-region of the subject's body over the time period. The method further comprises estimating the whole-body cutaneous loss of the target ion over the time period based on: a temperature at or near the localised sub-region during the time period; an energy expenditure of the subject during the time period; a physiological parameter for the subject; and a partitioning function applied to the local ionic concentration. The partitioning function is selected based on the target ion and a location of the localised sub-region. A processor assembly for performing the method is also described, as well as a sweat collection device configured to collect a sweat sample from the localised sub-region of the subject's body.

Description

[0001] Method and system for estimating whole-body cutaneous ionic loss

[0002] This application claims priority from US provisional application number 63532441 filed 14 August 2023, and GB application number 2402391.3 filed 20 February 2024, the contents and elements of which are herein incorporated by reference for all purposes.

[0003] Field of the Invention

[0004] The present invention relates to a method and devices for estimating the whole-body cutaneous loss of one or more ions from a subject’s body. The ionic loss may be estimated based on a measurement of sweat from a localised region of the subject’s body, in combination with user metrics.

[0005] Background

[0006] It is desirable to provide a system and method for measuring whole-body electrolyte / ionic losses of a subject, e.g. during athletic activity. Such a measurement can be used to establish a hydration strategy and thereby help limit the impacts of dehydration and electrolyte loss. The thermoregulatory response of the human body uses sweat to cool itself, relying on the evaporation of sweat from the skin to rid excess heat. Sweat removes water and electrolytes from the body. This can be a particular concern, for example, when exercise is performed in warm environments or for prolonged periods of time, resulting in a greater loss of sweat (and thus a greater loss of water and electrolytes). Electrolytes / ions (e.g. chloride, sodium, potassium, calcium, magnesium) are essential for motor and neuron function. However, whole-body ionic losses remain challenging to determine accurately.

[0007] Traditionally, ionic losses have been determined by a whole body washdown. Here, a subject is washed with deionised water pre-workout to remove electrolytes already present on the skin. Next, the subject is dried and weighed. Later, the subject is weighed again post-workout to determine an amount of fluid lost during the workout (with 1 kg of weight loss corresponding to 1 litre of fluid loss). Any fluid intake (drinking) or loss (urination, respiration) is taken into account at this stage, as well as any additional mass changes due to food intake. After the workout, the subject is washed with deionised water containing a specific, non-sweat, marker such as ammonium-sulfate. The mixed volume of sweat and deionised water is collected and analysed. All equipment must be washed thoroughly prior to the workout to prevent contamination with electrolytes already present on the skin and equipment, and all water used in the washdown must be collected. In practice, this means that whole-body washdowns are performed in the nude and in well controlled chambers. The final mixture is analysed using lab-grade quantification tools such as flame-photometry (accuracy ±3%). It should be apparent that although the whole-body washdown technique is highly accurate, it is impractical, having complicated procedures and long set-up times. Additionally, the resulting measurement is only a spot measurement of the electrolytes lost, rather than providing continuous and / or real-time data, which severely limits its ability to provide useful physiological insights during athletic activity.

[0008] A more practical technique has been developed using disposable absorbent patches which can be attached to one or more sites on the body!1 2f The patches absorb cutaneous sweat until they become saturated, at which point they cease to function and should therefore be replaced. After use, sweat is extracted from each patch using centrifugation or compression, and the volume is measured. The collected sweat is analysed using flame-photometry or a hand-held reader to measure the concentration of a specific electrolyte in the collected sample. An algorithmic scale- up, developed by Baker et al, can be used to estimate whole-body loss of the specific electrolyte that has been measured.!1 2] By using a number of patches (e.g. at 12 different locations), which are continually attached and replaced throughout an exercise session, the accuracy of these estimates of whole-body ionic losses can be improved compared to using a single patch. However, this is still less accurate than the whole-body washdown technique. Additionally, contamination of the sweat samples is common, affecting measurements. Further, since the sweat can only be measured after removal of the disposable patches, the absorbent patches still cannot provide continuous and / or real-time monitoring. Instead, this method can generally only provide one or two samples per hour (assuming normal operating conditions and average sweat rates, e.g. 1 Ltr1).

[0009] Although a number of techniques have been developed to provide real-time measurements of local sweat losses, further developments are still desirable to improve accuracy, practicability, and the level of insights into electrolyte losses. Known techniques require means for capturing a volume of sweat, measuring the volume of the captured sweat, and measuring a specific electrolyte concentration within the captured sweat. The volume measurements can be used, in combination with the measured concentration, to determine a rate or amount of electrolyte loss from the local area of the body on which the device is applied. In some devices, a chamber / channel (rather than an absorbent patch) is used to collect sweat for measurement. However, accuracy is limited by the volume measurements within this chamber, which are prone to errors due to the small amounts of fluid being collected. Further, it is not practical to reuse such devices once the sweat-collection areas have been saturated / filled with sweat (e.g. because they would be extremely difficult to empty to the precision required for subsequent accurate volume measurements), and thus these devices have relatively short lifespans and are unable to provide true continuous and / or real-time monitoring.

[0010] It should be apparent that determining local sweat rate requires a dedicated sensor device or sensor disposed within the sweat patch. The additional sensor increases the overall footprint of the device and significantly increases the complexity of the manufacture and electronic read-out. In addition, the measurement is prone to error as both sweat-rate and ionic concentration must be measured directly and accurately. In particular, the capture of sweat is complicated; altered sweating due to coverage of the skin by the wearable device or the addition of an adhesive to attach the device to the skin as is commonly employed results in an over capture of sweat, causing inaccuracies in sweat rate measurements and thus errors in whole body estimates.

[0011] Examples of prior art devices and methods for collecting sweat at a local level are described in US10463273B, US1 1523775B, US11337681 B, US1 12301 1 B, and US11219410B.

[0012] US10463273B discloses a simple hydration monitor that uses skin impedance measurements to determine the fluid (hydration) balance within a user’s body through analysis of the bioimpedance signal. This type of system does not specifically target ions lost through sweat.

[0013] US11523775B uses a conductivity measurement in combination with a processor unit to determine hydration status. The measurement system uses the cutaneous sweat collecting on the skin to determine hydration status. In such a design the resolution of the ion concentrations within sweat are limited, inter-user variability is not accounted for, and whole body ionic loss is not determined. Further, the cutaneous sweat on the skin can be pooled from adjacent areas of the skin near the sensor, limiting the scaling accuracy. Accuracy can also be affected by the placement of the sensor in contact with the skin, since the measured conductivity can include conductivity values of the skin itself.

[0014] US11337681 B uses a fluidic capture system to transport sweat from the skin to a colour indicator for a specific ionic species. The resulting, colorimetric read-out is representative of the ionic concentration within the sweat sample, as well as a variant for sweat rate. A scaling principle is then applied to obtain a whole body concentration. However, the device has no readout capabilities, and still exhibits a maximum operating time (affected by the wearer's sweat rate) after which saturation occurs and no further sweat rate information can be inferred.

[0015] US11123011 B provides a fluidic sweat collection device in combination with an impedimetric measurement of the sweat sample. The impedimetric measurement electrodes are used to determine the sweat rate and electrolyte contents (as an osmolality) of the sweat sample. However this design relies on local sweat rate measurement, which requires an extremely accurate sweat capture geometry, as discussed above.

[0016] Alternative methods have been proposed for whole body fluid loss as described in US11219410B. A system to determine whole body fluid losses using air / ambient temperature in combination with energy, associated with a particular form of exercise is described and gives higher accuracy than conventional scale-up methods. However, such a system, specifically applied to ionic losses is still lacking and the need for the accurate estimation of whole-body cutaneous ionic losses remains unmet.

[0017] Therefore, further developments are desirable to provide systems and methods for estimating whole-body cutaneous ionic losses, e.g. in a manner that is accurate, convenient, and / or can provide continuous measurements.

[0018] The present invention has been devised in light of the above considerations. Summary of the Invention

[0019] At its most general, the present invention provides a method, device, and system for estimating whole-body electrolyte (i.e. ionic) losses.

[0020] According to a first aspect of the invention, there is provided a computer-implemented method for estimating whole-body cutaneous loss of a target ion from a subject over a time period, the method comprising: estimating a local ionic concentration within sweat at a localised sub-region of the subject’s body over the time period; and estimating the whole-body cutaneous loss of the target ion over the time period based on: a temperature (e.g. skin temperature, core temperature, and / or ambient temperature) at or near the localised sub-region during the time period; an energy expenditure of the subject during the time period; a physiological parameter for the subject; and a partitioning function applied to the local ionic concentration; wherein the partitioning function is selected based on (e.g. determined as a function of) the target ion, and a location (body location) of the localised sub-region.

[0021] This method allows the whole-body cutaneous loss of the target ion to be determined in a convenient, accurate, and / or repeated / reusable manner.

[0022] Advantageously, the claimed method can estimate whole-body cutaneous ionic losses without relying on a measurement of local sweat rate (e.g. by collecting sweat within a closed chamber or channel), because the amount of sweat lost is instead accounted for based on contributing factors of the localised sub-region of the body (via the partitioning function), temperature, energy expenditure, and physiological information for the subject. Combining these factors with the estimated local ionic concentration allows for whole-body cutaneous electrolyte / ionic loss to be estimated accurately in a relatively simple manner, e.g. without requiring a direct measurement of sweat rate to estimate the user’s total fluid loss. This helps to reduce system complexity, since a volume-based sensor or fixed-volume sweat collection chamber is not required. In turn, this can also help improve accuracy, by eliminating potential sources of error that could otherwise arise from relying on localised volumetric measurements of sweat (e.g. since such volumes can be difficult to measure precisely).

[0023] Even further, since the claimed method utilises an interrelationship between factors such as temperature, energy expenditure, physiological data, and ionic loss, it can provide valuable insights into the extent to which a specific ion is affected by these criteria. These insights can be incorporated, for example, into a machine learning algorithm and / or to generate specific profiles (e.g. a user-specific profile, or population-specific profile), improving accuracy of future estimations. Ultimately this may even allow ionic losses to be estimated based on correlations with the other factors, e.g. even in circumstances where the sweat itself is not directly measured.

[0024] As used herein, the time period over which the ionic loss is estimated may also be referred to as a “predetermined time period”, “measurement time period”, “continuous time period”, “evaluated time period”, “monitored time period”. The time period may correspond to a total exercise period, or may be one of a plurality of time periods (i.e. a “first time period”) within a total exercise period (e.g. to provide continual and / or substantially real-time monitoring throughout the exercise period).

[0025] As used herein, the term “target ion” refers to an ion produced by an electrolyte within the body and excreted cutaneously (e.g. through sweat / perspiration). Examples of suitable ions include chloride, sodium, potassium, calcium, magnesium, and iron.

[0026] As used herein, the term “localised sub-region” (or simply, “sub-region”) may refer to any portion of the subject’s body (e.g. human body) through which the target ion may be excreted cutaneously. The localised sub-region may have sweat glands. For example, the localised subregion may be a particular appendage (e.g. arm, leg) or part thereof (e.g. upper arm, forearm, wrist, quad, thigh). The localised sub-region may be located e.g. on the chest or back (e.g. upper, lower, or central chest / back). The localised sub-region may be on the front or rear of the body. Sweat monitoring devices described herein may be mountable at the localised sub-region, to measure sweat excreted therefrom.

[0027] As used herein, the term “whole-body cutaneous loss” may refer to a total loss of the target ion from the subject’s whole body, as excreted through the skin (e.g. via sweat / perspiration). This may be extrapolated based on measurements from the localised sub-region, using the partitioning function. The partitioning function may be selected based on the type of target ion (e.g. chloride, sodium, potassium, calcium, magnesium, iron, etc) and an identified location of the localised sub-region (e.g. upper arm, upper back, etc), e.g. by using a look-up table (stored in a memory) that provides values for the partitioning function in dependence of the target ion and identified body location.

[0028] As used herein, the “partitioning function” may also be referred to as an “ion-specific scaling function”. The partitioning function helps to identify the location-dependent and / or ion-dependent contributions of local ionic losses to whole-body losses. The partitioning function may differ based on (i.e. may be dependent on) the location and / or ion being measured, and so may be determined e.g. using a table of look-up values for a particular ion and body location.

[0029] The partitioning function may also be dependent on a measurement technique used to estimate local ionic concentration (e.g. ion selective, electrochemical, conductivity, etc). Accordingly, optionally, the partitioning function may be selected (e.g. using a look-up table) based on this measurement technique. For example, in embodiments where the estimated local ionic concentration measures a plurality of ions including the target ion (e.g. by measuring the total ionic concentration, e.g. using a conductivity sensor), the partitioning function may convert the estimated concentration (e.g. total ionic concentration) to a concentration for the target ion, i.e. by effectively separating out a portion of the estimated concentration that is attributable (e.g. solely attributable) to the target ion. In embodiments where the estimated local ionic concentration measures one or more ions that do not include the target ion, the partitioning function may convert the concentration to a concentration for the target ion, by providing an expected correlation between the measured ion(s) and the target ion. The partitioning function may therefore effectively calibrate a local ionic concentration that is not specific to the target ion into a concentration that is specific to the target ion. Of course, it will be apparent that, in embodiments where the estimated local ionic concentration already measures the target ion specifically (e.g. using an optical sensor), no conversion is required in this regard by the partitioning function.

[0030] The whole-body cutaneous loss of the target ion may be estimated using the following equation:

[0031] I = at ■ k(a) ■ c ■ EE ■ f(T~)

[0032] (1) where I is the whole-body cutaneous loss of the target ion, is the local ionic concentration, fc(cz) is the partitioning function, c is the physiological parameter for the subject, EE is the energy expenditure by the subject during the time period, and f(T~) is a heat transfer function based on the temperature at the localised sub-region during the time period.

[0033] The local ionic concentration may be a concentration of the total ions within the sweat, a concentration of a combination (but sub-total) of ions within the sweat (e.g. sodium chloride), or a concentration of a single ion within the sweat (e.g. sodium), e.g. depending on the type of sensor used. Local ionic concentration can be measured for example, using one or more of an optical sensor, biosensor, and / or electrochemical (e.g. ion selective, conductivity) sensor.

[0034] Optionally, the local ionic concentration is derived from a total concentration of ions within the sweat from the localised sub-region. This provides multiple advantages. Firstly, by measuring total concentration, the method can be used to evaluate whole-body losses for any potential ion and / or a plurality of ions (by selecting suitable partitioning function(s)). Secondly, total ionic concentration is relatively simple to measure, and allows for a relatively simple device structure. For example, the total ionic concentration may be measured as a conductivity value of the sweat from the localised sub-region, using a conductivity sensor at the localised sub-region. A conductivity sensor does not provide measurements that are specific to a particular ion, but rather provides a value that is indicative of the total concentration of all ions within the sample, which can then be converted to be representative of the target ion by using a suitable scaling factor (partitioning function).

[0035] Accordingly, optionally, the method further includes: receiving a user input selecting the target ion from a plurality of potential (target) ions; and determining the partitioning function based on the selected target ion. The user may therefore monitor any desired ion of interest from the plurality of potential target ions.

[0036] In variant embodiments, the processing unit / assembly implementing the method may be preprogrammed to measure one or more particular target ions, e.g. without providing means for customisation by the user. In variant embodiments, the local ionic concentration may be estimated differently, e.g. as a concentration of a sub-total of ions (e.g. only the target ion) within the sweat. This may be measured using an optical sensor, e.g. using fluorescence or absorbance-based methods. Such arrangements may improve accuracy, but typically require more complex sensor arrangements than those which measure a total ionic concentration.

[0037] Optionally, the computer-implemented method comprises identifying a measurement technique used to estimate the local ionic concentration. The partitioning function may be selected based on the identified measurement technique. This allows the measurement technique to be customisable, thereby allowing ionic losses to be estimated with a variety of different sensor arrangements.

[0038] The measurement technique may be identified manually, e.g. by receiving a user input that identifies the measurement technique. Alternatively, the measurement technique may be identified automatically, e.g. based on data received from a sensor that is indicative of the measurement technique (e.g. conductivity data).

[0039] In variant embodiments, the processing unit / assembly implementing the method may be preprogrammed for use with a particular measurement technique (which may be stored in a memory), e.g. without providing means for customisation by the user.

[0040] Optionally, the computer-implemented method further comprises: receiving positional data from a positional sensor located at or near the localised sub-region; and identifying the location of the localised sub-region using the received positional data. The partitioning function may then be determined based on the identified location of the localised sub-region. This allows the location of the sub-region (and hence the relevant partitioning function) to be evaluated in a user- convenient and autonomous manner, e.g. without requiring input from the user to specify the location from which the sweat is monitored. The resulting whole-body cutaneous loss can therefore be estimated without requiring user input to identify the location at which the local ionic concentration is estimated. For example, the positional sensor may comprise an accelerometer.

[0041] Alternatively, optionally, the partitioning function is determined based on a received user input that identifies the location of the localised sub-region. Accordingly, the method may include: receiving a user input identifying the location of the localised sub-region. The partitioning function may then be determined (selected) based on the identified location. This allows for a simplified device design, by omitting the need for a positional sensor. This also provides for more simplified processing. For example, the method may be implemented (at least in part) in an app, e.g. on the user’s mobile phone, and the user may identify the location using the app.

[0042] In variant embodiments, the location of the localised sub-region may not need to be identified as part of the method. Rather, the location may be predetermined. That is, the method may be configured to apply a predetermined partitioning function to the measured concentration, said partitioning function being specific to a predetermined localised sub-region of the body from which the local concentration must be estimated.

[0043] As used herein, the “physiological parameter” may also be referred to as a “physiological factor” or “subject-specific physiological parameter (or factor)”. This parameter / factor accounts for subject-specific physiological information (physiological data) that affects sweat rate, such as physical metrics and / or data history. For example, the physiological information may include one or more of the following: sex, age, health status (e.g. illness, medication), diet, exercise history, physique (e.g. weight, BMI, height, muscle mass), and hydration history. Methods and devices described herein may be configured to collect any of this physiological information via one or more of: a user input, sensor, memory, and / or remote server. The physiological parameter may then be estimated based on such physiological information, e.g. using a look-up table of values stored in a memory. The look-up table may be formed based on the subject’s individual data and / or the data from a population of subjects.

[0044] As used herein, the term “energy expenditure” may refer to a factor that accounts for an amount of energy expended by the subject over the time period. Energy expenditure affects the user’s fluid loss (through sweat) and hence the whole-body ionic loss. The energy expenditure may be provided e.g. as a number of calories burned during the time period, or may be estimated based on one or more additional user metrics (e.g. heart rate) and / or session metrics (e.g. speed, distance, power, weight, height, time). The energy expenditure may be estimated using one or more sensors, e.g. implemented in a wearable device (e.g. as part of the device that estimates the local ionic concentration, or a separate device) or a third-party application (e.g. on the subject’s mobile phone). For example, suitable means for measuring energy expenditure may include one or more of: a calorie-counter, smart-watch, heart-rate monitor, pedometer, or global positioning system (GPS) for measuring movement.

[0045] Accordingly, optionally, the method may include a step of estimating the energy expenditure based on activity data received from an activity sensor (e.g. a heart rate sensor, pedometer, or accelerometer). The activity data may indicate a heart rate and / or movement of the subject over the subject over the time period. The method may utilise a stored value (e.g. utilising a stored profile comprising a look-up table or graph) to determine the energy expenditure factor based on the activity data. The stored profile may comprise historical data from the individual subject and / or from a population of subjects, as will be discussed further herein.

[0046] As used herein, the term “heat transfer function” (T) may refer to a function that accounts for a temperature difference (e.g. thermodynamic gradient) between the subject (e.g. between their skin and / or core) and their surrounding environment (e.g. an ambient temperature). Without wishing to be bound by theory, the temperature difference between a subject and their environment is believed to impact sweat rate, by governing the amount of heat that the body can transfer away through its sweat response. The heat transfer function (T) may provide a relationship between two or more of: the ambient temperature Tambient, core body temperature TCore , and skin temperature Tskin. The relationship between these temperatures may be substantially linear within a physiologically relevant range (e.g. 17°C < Tskin< 40°C; 36°C < Tcore< 42°C). Each may be used to determine the other.

[0047] When evaluated at a particular temperature (or combination of temperatures), the heat transfer function f (T) can provide a temperature dependent factor that helps account for the impact of temperature on fluid losses, in turn contributing to the accurate estimation of total losses of the target ion. The temperature factor may be evaluated based on a measured temperature (e.g. skin temperature Tskinand / or core temperature Tcore), at the same localised sub-region from which the local ionic concentration is estimated (e.g. at or near a sensor that measures the local ionic concentration, e.g. on the same device), for greatest accuracy.

[0048] For example, Tcoremay be measured using an internal temperature sensor (e.g. telemetry pill) or may be estimated e.g. using heart rate data (using estimations found in literature). Alternatively, skin and / or ambient temperatures may be measured using temperature sensors at an exterior of the body, which may be particularly convenient.

[0049] As an example, based on a measured skin temperature, within a physiological range (e.g. 17°C < Tskin<40°C; 36°C < Tcore, 42°C ) the heat transfer function may be represented as: ff'T' — r skin-ambientrr >rr skin-ambient

[0050] J Jskin-ambientQ 1skin10

[0051] (2) f T > r skin-core - irrskin-core

[0052] J Jskin-coreQ 1skin '10

[0053] (3) Here Tbkin~ambientand Tokin~coreare offset factors that vary in magnitude depending on whether the Tamblentor Tcoreis being calculated, in which Tbkin~core> T^kin~ambient. G represents the gradient scaling factor, with Gskin~ambient< 1, and Gskin~core> 1. The offset factors and gradient scaling factors may be determined empirically, through the measurement of skin and ambient temperatures for a specific body location and / or exercise modality, and may be stored in a memory.

[0054] Optionally, the computer-implemented method can be used as part of a machine learning process.

[0055] Optionally, the computer-implemented method includes storing, in a memory, a dataset comprising two or more of: the local ionic concentration, temperature, energy expenditure, physiological parameter, and whole-body cutaneous loss of the target ion. The stored data may advantageously be used to identify correlations and / or to improve accuracy of subsequent measurements.

[0056] The dataset may be stored as part of a profile (e.g. graph, look-up table). The profile may include a plurality of said datasets (i.e. a plurality of datasets based on the same two or more factors of: local ionic concentration, temperature, energy expenditure, physiological parameter, and wholebody cutaneous loss of the target ion). The plurality of datasets may be based on data collected from a plurality of time periods and / or a plurality of subjects (e.g. by performing the computer- implemented method at a plurality of time periods and / or for a plurality of subjects). The profile may therefore be used to identify a correlation between the two or more factors, which can in turn be used to improve accuracy and / or convenience of subsequent measurements. Optionally, the profile may relate to a particular localised sub-region of the body. Accordingly, optionally, the memory may include a plurality of location-specific profiles, each profile providing a correlation for a different region of the body.

[0057] The profile may be used to estimate values for relevant factors (e.g. ionic concentration, temperature, etc) based on its correlation with another factor, thereby enabling whole-body ionic loss to be estimated in a simplified manner without requiring one or more relevant sensors. For example, in some embodiments, whole-body cutaneous loss of the target ion may be estimated without requiring a sweat sensor to measure local ionic concentration, e.g. by utilising a stored correlation between the local ionic concentration at a particular body location and one or more other factors such as energy expenditure, temperature, and physiological parameter, and inputting this estimated value for local ionic concentration into the estimate for whole-body ionic loss (e.g. using equation (1)). Accordingly, optionally, the method includes estimating the wholebody cutaneous loss of the target ion over a subsequent time period using the profile. The step of estimating the whole-body cutaneous loss of the target ion may include utilising the correlation (in the profile) to estimate a value for one more factors within the profile. In other words, the profile may be used to estimate any one or more of these variables in dependence of another variable in the profile. This provides for simpler monitoring, by omitting a need to measure each of the relevant variables, since they can instead be inferred from the stored profile. The method may include steps that correspond to any of those discussed above, i.e. estimating a local ionic concentration within sweat at the localised sub-region of the subject’s body over the subsequent time period, and estimating the whole-body cutaneous loss of the target ion over the time period based on the local ionic concentration within the sweat at the localised sub-region; a temperature at the localised sub-region during the time period; an energy expenditure of the subject during the time period; a physiological parameter for the subject; and a partitioning function, wherein the partitioning function is based on the target ion and a location of the localised sub-region. These steps may be repeated, and the stored profile can be further updated in a machine learning process, as new data is collected.

[0058] The method may therefore contribute to a machine learning process that can be used to improve accuracy of subsequent measurements and / or provide insights into the relationships between different variables e.g. for particular users or exercise modalities.

[0059] Optionally, the profile includes data for a plurality of subjects. This enables accuracy to be even further improved, by allowing measurements from a large population of users to be applied for ionic estimations. In such embodiments, each dataset preferably includes the physiological parameter for the respective subject. Optionally, the method includes a step of filtering data in the profile based on the subject’s physiological parameter before estimating the whole-body cutaneous loss of the target ion over the subsequent time period using the profile (i.e. using the filtered data from the profile). The step of filtering the datasets effectively allows the method to disregard certain datasets that have been obtained from subjects having physiological parameters that are very different from the physiological parameter of the present subject. This ensures that the estimations will be based on the data that is most relevant to the subject, i.e. as determined based on the shared physiological attributes with the subjects whose data is in the filtered profile.

[0060] Optionally, the profile includes user-specific data collected for the subject (i.e. the subject whose whole-body cutaneous losses are being estimated) over a plurality of time periods (e.g. historical data stored following previous measurements). This can improve accuracy for a subject’s particular body features. For example, some user-dependent physiological attributes such as scar tissue or tattoos may affect sweat rate. Although the physiological parameter helps account for this, accuracy may be even further improved by providing a user-specific profile (or by filtering a population-based profile based on the individual user’s data), so that any estimates have improved accuracy according to the individual. The user-specific data may include, for example, any of local ionic concentration, temperature, and / or physiological parameter, in correlation with expended energy. These may each provide a substantially linear relationship. Correspondingly, population-specific data may also provide such linear relationships, once filtered by physiological parameter.

[0061] According to another aspect of the invention, there is provided a method for estimating wholebody cutaneous loss of a target ion from a subject, the method comprising: collecting a sweat sample from a localised sub-region of the subject’s body over a time period (e.g. using a sweat collection device); and using a processing unit to perform the computer-implemented method of any preceding claim; wherein the step of estimating the local ionic concentration is performed based on the collected sweat sample.

[0062] Optionally, the step of estimating the local concentration may comprise measuring a conductivity of the sweat sample.

[0063] Optionally, the method includes monitoring the local ionic concentration over a plurality of time periods, by repeating (e.g. periodically repeating) the steps for collecting a sweat sample and performing using the processing unit to perform the computer-implemented method. This can advantageously provide substantially continuous and / or real-time monitoring.

[0064] As will be evident from the above discussion regarding stored datasets, in variant embodiments, it may not be necessary to perform a sweat collection step, e.g. by instead estimating the local ionic concentration based on its correlation with a measured value as inferred using a stored profile. According to another aspect of the invention, there is provided a processor assembly having at least one processor and at least one memory including computer program code, wherein the computer program code is configured to, with the at least one processor, cause the processing unit to perform the computer-implemented method described herein.

[0065] The processor assembly can comprise a single processor (processing unit) or a combination of distributed processors (processing units) that collect and communicate the required information. All or a subset of the relevant data may be visible to all or a subset of the processors at any time. The processor assembly may be configured to determine ionic losses in real-time and provide an indication of the ionic loss to a user, e.g. through a user interface such as a display or audio interface. This may be implemented, for example, via a mobile phone application or computer program. An example of a processor (processing unit) can be a microcontroller embedded in a wearable device. An example of distributed processors (processing units) could comprise any two or more of: a microcontroller on a wearable device, a processor on a smartphone, and a processor on the server; where each processor has access to a memory and communication link to transfer data from one processor to another.

[0066] If more than one processor is employed, it should be apparent that the data collected by each processor may be available to any other processing unit at the same time, for further aggregated analysis. The processors do not need to be of the same type nor positioned in the same location. Examples include but are not limited to ASIC, microcontroller, microprocessors, server, cloud servers.

[0067] Embodiment methods can be performed using discrete / point-based sweat collection devices (e.g. an absorbent patch), or using real-time sweat collection.

[0068] According to another aspect of the invention, there is provided a sweat collection device configured to collect a sweat sample from a localised sub-region of the subject’s body over a time period. Optionally, the sweat collection device may be provided as part of a system for estimating whole-body cutaneous loss of a target ion from a subject, the system comprising the sweat collection device and the processor assembly. The sweat collection device may comprise a sensor coupled to the processor assembly, for providing the estimations of local ionic concentration.

[0069] The sweat collection device may comprise a substrate; a fluid pathway extending through the substrate, the fluid pathway having: an inlet port; an outlet port; and a sensor (ionic sensor, e.g. conductivity or optical sensor) configured to measure the local ionic concentration of sweat within the fluid pathway. This allows sweat to flow in and out of the device in use, meaning the device can be continuously used in real-time and can be conveniently cleaned if desired.

[0070] This can be contrasted from known devices which are limited in the amount of sweat that they can capture before becoming full or saturated, and which rely on accurate volume measurements of sweat within a chamber in order to determine total ionic fluid loss. Volumetric sweat measurements are prone to error, e.g. based on the amount of skin covered or the addition of an adhesive to attach the device to the skin, which can result in an over-capture of sweat, causing inaccuracies. The present arrangement overcomes these challenges by providing a continuous fluid pathway through the sweat collection device, and by providing a method for determining the whole body cutaneous ionic losses without reliance on a sensor having a sweat collection chamber for measuring a volume of sweat.

[0071] Optionally, the sweat collection device comprises a capture area configured (e.g. sized, shaped, dimensioned) to funnel sweat (from the subject’s skin) into the inlet port. This helps to efficiently capture sweat, thereby helping provide a quick capture rate. The capture area may also be referred to as a “wicking region”, which may comprise one or more grooves within the substrate. The capture area may be location on a surface of the sweat collection device that is configured to face the subject’s skin in use, e.g. on a planar surface of the substrate, which may be located on an opposite surface of the substrate from the sensor. Optionally, the capture area (e.g. wicking region) may be shaped substantially like a star, web, asterix, snow-flake or cross mark (e.g. by defining grooves arranged in such a shape). The capture area may comprise grooves (e.g. capi Hi ary grooves) having a width and / or depth of at least 70 m up to 500 gm (e.g. at least 100 gtm up to 400 gtm, e.g. at least 200 gtm up to 300 gtm). These dimensions are small enough to facilitate wicking of sweat, whilst being large enough to help avoid the risk of clogging or a backpressure within the device preventing sweat from flowing through the fluid pathway.

[0072] Optionally, the substrate may comprise (e.g. consist of) any one or more of: silicone, textile, adhesive tape, polyurethane, silicone, epoxy, metal, acrylic, PEEK, PET, teflon-based materials, ABS, PC, ABS / PC, glass. These materials may be particularly useful for a reusable sweat collection device.

[0073] Optionally, the outlet port may have a wider opening than the inlet port. This facilitates cleaning by allowing the run-through of fluid (e.g. water and / or air) to expel liquid (sweat) from the fluidic channel. For example, cleaning may be performed by blowing with breath, pumping or wicking of the fluid or using a hydrophilic material such as a sponge or paper. This is therefore particularly useful for providing a reusable sweat collection device. As a further advantage, a reusable sweat collection device may be implemented with more sophisticated and / or robust sensor equipment (e.g. more accurate ionic sensor) than disposable devices, while remaining cost-effective due to the reusability of the device.

[0074] Optionally, the sweat collection device is modular. For example, optionally, the sweat collection device comprises a first portion and a second portion that are removably detachable from each other to provide access to an interior of the fluid pathway. This may facilitate reusability and cleaning of the fluid pathway, by allowing access into the fluid pathway via separation of the first and second portions. The first portion may comprise the inlet port and / or outlet port, and the second portion comprise the ionic concentration sensor (e.g. conductivity sensor, optical sensor, etc). The sensor may comprise an electrochemical sensor (also referred to herein as a conductivity sensor or impedimetric sensor). The electrochemical sensor which may comprise a plurality of conductive metal electrodes not limited to gold, platinum, silver. The conductive metal electrodes may have a relatively high surface area, or may be made from a material with an intrinsic high surface area not limited to carbon, iridium-oxide, ruthenium-oxide, platinum black, silver-silver chloride; in order to reduce a double layer capacitance of the sensor. The processor assembly may be electrically connected to the sensor to apply an alternating voltage potential across the conductive metal electrodes, e.g. of 0.01 -0.90 Vpp with a frequency of 1 ,000 - 500,000 Hz.

[0075] An impedimetric (also referred to as conductivity), measurement is a simple electrochemical technique for determining the ionic contents of a liquid. Ions possess an intrinsic charge (negative or positive) which determines the ability to conduct electrons through an aqueous sample. By applying an alternating excitation voltage or current at a given frequency to the sample via a pair of electrodes and by measuring the current or voltage, an estimate of the resistivity is obtained. A phase component that represents the capacitive elements of the system and sample is also measured. The phase is representative of capacitive charge-layer distributions, henceforth referred to as the doublelayer, at the electrode-sample interface. The characteristics of the double-layer are dependent on the electrode material, ionic concentration, electrode geometry and frequency of the applied potential. A phase shift close to zero is desirable to obtain the most accurate measurement of the sample resistance. However, larger phase shifts (e.g. up to -14°) can be used as long as the resistive component can be substantially isolated from the measurement result. By increasing the frequency, the double-layers effects can be omitted. Likewise, increasing the geometric or effective electrode surface increases the double-layer capacitance, lowering the frequency that is required to obtain accurate measurements.

[0076] Typically, a conductivity sensor may comprise a minimum of two electrodes. The sensor may be driven by a controller unit which applies an alternating potential or current across the sensor electrodes. The potential or current across or between the same (2-electrode) or an additional set of electrodes (4-electrode) can be measured. The latter configuration can significantly reduce capacitive effects by bypassing excluding the double-layer from the resistive measurement. When the configuration consists of multiple electrode pairs for current and voltage excitation and recordings (up to 9 electrodes) the accuracy of the measurement can be enhanced further. The sample resistivity is converted into an ionic conductivity through a geometric constant of the electrodes. Here the distance between the electrodes, the area of the electrodes and the surface quality determines a cell constant k. By dividing the value of k by the measured resistivity (in Q) the conductivity of the sample may be determined. Since conductivity is highly dependent on temperature (2% C°1), temperature compensation may be performed to improve accuracy. The final result is a measure of the total ionic concentration in the sample which can be used by the disclosed system to determine whole body ionic losses. Optionally, the sensor may comprise an optical sensor. Optical measurements of the local ions can be done through fluorescence or absorbance based methods. The concentration may be measured as a change in the emitted, refracted or reflected wavelength. Flame Photometry is also a valid method when using point-based analysis. Optical methods are highly specific to the ion of interest but likewise can provide a measure of the combined ionic constituents within a sample. Similarly, the ion detection can be done through the use of mass, thermal conductance, impedimetric, potentiometric or amperometric methods using biorecognition elements specific to the ionic species of interest. One example is a two-electrode potentiometric ion selective electrode (ISE). An ISE utilises an ion selective compound (ionophore) immobilised in a polymer matrix. When immersed in or put in contact with a sample, ions are bound to the ionophore establishing an electromotive force at the liquid-membrane and membrane-electrode interface. The binding event is transduced into a potential difference (electrical). The same binding principles can be transduced into other physical readouts, (including optical) where the binding of the ion results in an increase in weight (sensed either directly or through a piezoelectric or vibrational element); reduced thermal conductivity (measurable by a change in heat-transfer at the sensor and liquid interface), electrical conductivity (impedance spectroscopy).

[0077] It should be apparent that the method of quantifying ions locally can vary and can encompass any of (or combination of) the aforementioned sensing methods, and is not limited to the aforementioned sensing methods.

[0078] Optionally, the sweat collection device may include one or more additional sensors, e.g. a temperature sensor.

[0079] In variant embodiments, the sweat collection device may be provided independently without the processor assembly.

[0080] The invention includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

[0081] Summary of the Figures

[0082] Embodiments and experiments illustrating the principles of the invention will now be discussed with reference to the accompanying figures in which:

[0083] Figure 1 is a flow diagram of a computer-implemented method according to an embodiment of the invention.

[0084] Figure 2 is a schematic diagram that depicts the various data inputs provided to the processing assembly in order to estimate whole-body ionic loss.

[0085] Figure 3 is a graph showing the relationship between ambient temperature, core temperature, and skin temperature. Figure 4A is a schematic diagram highlighting different localised sub-regions suitable for local sweat measurements on the front and rear of a subject’s body.

[0086] Figure 4B is a chart illustrating a partitioning function for different body locations from Figure 4A and different target ions of Na+and K+.

[0087] Figure 5 is a schematic illustrating how correlations in datasets from a user and / or population of individuals can be used to facilitate estimations of whole body ionic loss.

[0088] Figure 6 is an isometric view of a sweat collection device, showing a surface of the device that is configured to face the user’s skin in use.

[0089] Figure 7 is an isometric view of an interior of the sweat collection device of Figure 6, showing an interior surface of the device that faces away from the user’s skin in use.

[0090] Figure 8 is a cross-sectional view of a fluid-capture area and inlet port of the sweat collection device of Figure 6.

[0091] Figure 9 is an isometric view of an ionic conductivity sensor that may be incorporated in the sweat collection device of Figure 6.

[0092] Figure 10 is a side view of a modular sweat collection device.

[0093] Figure 11 is a schematic drawing of a processing assembly for implementing a computer- implemented method according to an embodiment of the invention.

[0094] Detailed Description of the Invention

[0095] Aspects and embodiments of the present invention will now be discussed with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

[0096] Embodiment methods and devices allow the collection of a sample of sweat from the skin through a fluidic sweat collection device. The sweat conductivity may be determined with an ionic sensor (e.g. conductivity sensor) that can measure a combination of or the total amount of ions. The sampling location may be known or determined through a positioning system in order to apply the partitioning function, specific to the ion of interest, this may be combined with the users physiological metrics including previous exercise history and a measure of effort which may be represented by energy expenditure. A skin-temperature sensor may be placed in close proximity to the sample site to measure skin- or near-skin temperature. The aforementioned parameters, measurements and factors may be combined by a processor unit to determine the whole body cutaneous ionic loss.

[0097] Figure 1 is a flow diagram of a computer-implemented method 100 according to an embodiment of the invention. The method provides an estimation of whole-body cutaneous loss of a target ion from a subject over a time period. The method may be implemented by a processor assembly. In step 102, the processor assembly estimates a local ionic concentration within sweat at a localised sub-region of the subject’s body. This estimate may be obtained based on a sensor measurement (e.g. using an ionic sensor such as an optical sensor or conductivity sensor), or may be inferred based on other measured data (e.g. using a stored profile as will be discussed further herein).

[0098] In step 104, the processor assembly identifies the location of the localised sub-region on the subject’s body. For example, the processor assembly may identify that the localised sub-region is on a front or back of the subject’s body, and is on a particular appendage such as an arm or leg (or region thereof). The location may be identified e.g. based on positional sensor data, or based on a user input.

[0099] In step 106, the processor assembly estimates suitable values for a number of other sweatinfluencing factors including: temperature, energy expenditure, a physiological parameter (specific to the subject), and a partitioning function. The partitioning function is determined based on an identification of the target ion, and the identified location of the localised sub-region. The partitioning function may also be determined based on a measurement technique that was used to estimate the local ionic concentration in step 102.

[0100] In step 108, the collected data is combined in order to estimate whole-body cutaneous loss of the target ion. For example, this may be performed using equation (1) discussed further above.

[0101] It will be recognised that the method steps in Figure 1 may alternatively be performed in other orders, e.g. by changing the order of steps 102 to 106 and / or by performing any of these steps simultaneously.

[0102] Figure 2 shows a schematic diagram that depicts the various data inputs provided to a processor 200, from which the processor assembly can estimate the whole-body ionic loss, according to an embodiment of the invention. The processor 200 may be configured to communicate with a memory 202 that stores user data. The processor 200 may be configured to communicate with the memory to retrieve a value for the physiological parameter c from the memory.

[0103] The processor 200 may further be configured to communicate with a sweat sensing device 204 (also referred to herein as a “sweat collection device”), which may detect local ionic constituents within sweat.

[0104] The constituents can be measured as a combination of ions, which can be specific (i.e. total ionic concentration, sodium chloride), or a chosen ion of interest (e.g., sodium), which can be a concentration. The measurement can be performed using optical, biosensor and electrochemical measurement methods. The sample can be obtained from a point-based sample collection (i.e. absorbent patch) or a real-time measurement such as a wearable sweat collection device. In both cases, the target ion must be present within the sample (e.g. sodium, chloride, potassium, magnesium, calcium and iron). The sweat sensing device 204 may provide the processor 200 with a value for the local ionic concentration In this embodiment, the sweat sensing device 204 comprises an electrochemical sensor, and thus the local ionic concentration is a conductivity value. In variant embodiments, the sweat sensing device may be configured in different manners, e.g. using an optical sensor.

[0105] The processor 200 is also in communication with a temperature sensor 206, which provides an input in the form of a temperature of the subject, e.g. skin temperature Tskin. The temperature sensor 206 is placed in contact with the skin in close proximity to the sample collection area in order to obtain a measurement of the skin temperature, Tskin. The temperature sensor 206 should be in contact with or in close proximity to the sweat collection site, specifically the body position from where the local ionic loss is determined. The temperature sensor 206 is used to measure or estimate the near and skin temperature of the user. The thermodynamic gradient between the (near-) skin temperature and the ambient temperature governs the amount of heat that can be transferred away from the skin (and thus the body) through the innate sweat response. The value for Tskinis incorporated in the method for obtaining whole body ionic losses through a specific temperature function, f(T).

[0106] One or more physiological sensors 208A-B are also in communication with the processor 200, to provide physiological data used for the energy expenditure factor EE. For improved accuracy, one or more physiological parameters associated with the user, c, should be obtained. The exercise metrics that constitute effort (expended energy, EE) and govern the heat that is dissipated during exercise are collected. This affects the user’s fluid loss and hence whole body ionic loss, through sweat. The data are available to one or more processor units. If more than one processor is employed it should be apparent that the data collected by each processor is available to any other processing unit at the same time, for further aggregated analysis. The processors do not need to be of the same type nor positioned in the same location. Examples include but are not limited to ASIC, microcontroller, microprocessors, server, cloud servers. The value for EE can be measured directly through a wearable (e.g. calorie-counter, smart-watch, heart-rate monitor) or determined from other exercise and session metrics not limited to speed, distance, power, weight, height, time. These can be available through third party applications that monitor exercise such as a pedometer or global positioning system on the users mobile phone.

[0107] The processor 200 may also be in communication with a user interface (not shown) at which the user may provide data such as an identification of the body location at which the sweat sensor 204 is located, and an identification of the target ion. Based on this data, and with reference to a look-up table of values (e.g. stored in the memory 202 or another memory, e.g. a remote server), the processor 200 may estimate a value for the partitioning function fc(cz). Combining all these values together, the processor 200 may then estimate the whole-body cutaneous loss of the target ion I, e.g. according to equation (1).

[0108] In some embodiments, the sensors 204, 206, and 208A-B may be provided in the same or different units. For example, optionally, the temperature sensor 206 can be incorporated in the physiological sensor 208 or incorporated in the wearable or sweat collection device 204. As discussed above, the temperature function f(T) may be a heat transfer function relating the subject’s skin, ambient, and / or core temperatures, e.g. according to equations (2) or (3). Figure 3 shows a graph that helps to illustrate the linear relationships between the core temperature (represented by line 210), skin temperature (represented by line 212), and ambient temperature (represented by line 214). The x-axis in Figure 3 corresponds to the ambient temperature.

[0109] Figure 4A shows a schematic diagram of a plurality of localised sub-regions of a subject’s body, suitable for local ionic measurements according to embodiments of the invention.

[0110] Figure 4B shows a graph of partitioning functions (partitioning function) fc(cz) for different body locations (e.g. upper back (UB), lower back (LB), upper arm (UA), etc), based on different target ions of Na+and K+respectively. Suitable values can be determined empirically for any target ion, body location, and / or measurement technique to provide a repository of look-up values for the partitioning function to be used in embodiment methods and systems.

[0111] The partitioning function in combination with the local ionic concentration measurement is needed to determine the total ionic losses specific to the target ion (e.g. sodium or potassium). There are differences in ionic concentration for different body locations, for example on the front and back of a person. For example, a significantly higher sodium concentration is present in samples collected from the back compared to the upper arms. In addition, certain ions collect within sweat via the active transport (Na+) mechanism. In healthy individuals, certain ions are partially reabsorbed by the skin overtime. While partitioning functions vary across different body locations, it is understood that side-to- side variability in partitioning functions (between different people) is minimal. The partitioning function therefore allows for accurate estimations across populations of people.

[0112] Optionally, partitioning functions may also be determined and / or updated on an individual level to further improve the accuracy. For example, if a given subject has scar tissue and / or a tattoo on a particular region of their body (e.g. upper arm), the processor may update their partitioning function for that region for the given individual, e.g. using stored profiles as discussed further herein.

[0113] In Figure 4B, the partitioning function and how it varies across body locations is presented as a relative value. For sodium, most body locations apart from the wrist are higher in concentration compared to a reference value. For potassium the upper arm, upper back and lower back are all lower compared to the reference value.

[0114] To determine which partitioning function is to be used the body location from which the concentration was estimated (e.g. from which a sweat sample was required) must be determined. This can be done simply through the user’s manual input or in an automatic fashion through more complex positioning systems disposed within the sweat sensing wearable which can be an accelerometer or heart-rate monitor. The partitioning function is obtained by the processor. Once applied, the partitioning function outputs the final local ionic loss, i.e. the amount (concentration) the target ion excreted from the localised subregion. The function can be used to determine the local ionic concentration of different ionic species, not limited to sodium, potassium, calcium, magnesium, iron, or chloride. The temperature, energy expenditure, and physiological parameter can then be used to effectively “scale-up” the local ionic loss to provide a whole-body cutaneous loss for the target ion, e.g. via equation (1).

[0115] The whole body ionic loss for an individual can be collected and used to further establish a user profile by correcting and optimising the value of c, or to create a separate function all together taking into account session history, fitness level changes, and training methods over a longer time period. Similarly, an individual's metrics can be part of data from a community of users that can be segmented (filtered) by key factors such as modality, age, sex and fitness level. Over-time such a system, as depicted in FIGURE 5 could be used to replace the need for a physical input of a\, instead relying on predictive learning algorithms to determine local ionic losses given a set of session parameters and the user profile.

[0116] As shown in Figure 5, based on data collected from a plurality of time periods and / or a plurality of subjects (within a population of users), one or more profiles may be generated and stored within a memory. This may include a user-specific profile 216 and a population-specific profile 218.

[0117] The user-specific profile may include a plurality of datasets from a given subject, as collected over a plurality of time periods. For example, the datasets may provide a correlation between any one or more of local ionic concentration, temperature (e.g. skin temperature Tsfcira), or physiological parameter c in dependence of energy expenditure EE. These may each provide substantially linear correlations, as shown in Figure 5, which can thus be exploited to infer values such as local ionic concentration or temperature in dependence of a given energy expenditure, or vice versa. The physiological parameter c may likewise be used to compare results to users of similar physiological parameters (e.g. as filtered from the population-specific profile), to further improve either profile 216 or 218.

[0118] The population-specific profile may have a plurality of datasets from a plurality of subjects, e.g. over a plurality of different time periods. The plurality of subjects can be distinguished and filtered based on their physiological parameters c. As shown in Figure 5, plotting the whole-body cutaneous loss I, concentration ai tor temperature (e.g. Tskin) does not provide a strong (e.g. linear) correlation against the physiological parameter c. However, by filtering the data to select the data points with a particular physiological parameter c (e.g. corresponding to that of the current subject), and then evaluating a relationship against energy expenditure EE for the filtered data (e.g. in a similar manner to the user profile 216), a correlation may be identified to provide useful insights, in a similar manner to the user-specific profile 216.

[0119] Figures 6 to 9 show a sweat collection device 300 according to an embodiment of the invention. The sweat collection device is configured to collect a sweat sample from a localised sub-region of a subject’s body over a time period, e.g. from one of the localised sub-regions shown in Figure 4A. The sweat collection device may be implemented e.g. as the device 204 discussed in relation to Figure 2. The sweat collection device 300 includes a fluidic component and an ionic sensor 314. The fluidic component comprises a substrate 302 and a fluid pathway 304 extending through the substrate 302. The fluid pathway 304 has an inlet port 306 and an outlet port 308, i.e. for conveying sweat from the inlet port to the outlet port. The sensor 314 is configured to measure the local ionic concentration of sweat within the fluid pathway.

[0120] Figure 7 shows an internal view of the sweat collection device 300, from which it can be seen that the outlet port 308 has a wider opening than the inlet port 306.

[0121] In an embodiment, the sweat collection device 300 capable of determining the local ionic conductivity within a sample may be placed on a body position of interest on the front or back of the body (e.g. a location shown in Figure 4A). The ionic sensor 314 may be a reusable or singleuse sensor. Figure 9 shows the ionic sensor 314 in detail, having a sensor substrate 316 and sensor electrodes 318. The sensing electrodes 318 may be formed in the sensor system by a conducting material disposed on the sensor substrate 316 which can be glass, silicon or polymeric. The conducting material may be an inert metal including but not limited to gold, platinum, titanium, stainless-steel, Electroplated Nickel-Gold (ENIG) or a material of high surface area including ruthenium-oxide, carbon, carbon-nanotubes, iridium oxide, platinum black; as to lower the frequency of the excitation signal and improve the overall accuracy. The high surface materials can be a coating on top of a base electrode. The frequency and amplitude of the waveform is important and is selected such that the wide variety of sweat conductivities within subjects (0.5-18 mS cm1) and across body locations can be measured. The potential may lie in the range of 10 - 900 mVpp and the frequency between 1 ,000-500,000 Hz.

[0122] The device 300 (e.g. the fluidic component thereof) further includes a fluid capture area 312 configured to funnel sweat into the inlet port 306. The fluid capture area 312 comprises a series of grooves that extend radially outward from the inlet port 306, across a surface of the substrate 302. In this embodiment, the grooves are arranged in a shape substantially like an asterix. The grooves each have a width and / or depth of at least 70 m up to 500 gm.

[0123] The fluidic pathway 304 and fluid capture area 312 are configured to convey cutaneous sweat excreted by the sweat glands across electrodes of a sensor. Figure 8 shows a cross-sectional view of the fluid capture area 312 formed within the substrate 302. The fluid capture area 312 is fabricated on the skin side of the sweat collection device. As sweat rates can vary significantly between users, sexes, body-locations and environments the volume of the capture area 312 should ideally be kept small and the shape and material should facilitate the quick and efficient transport of sweat. This shape can be akin to a star, snowflake, serpentine, swirl or maze which may have a depth D and / or width l / l / of at least 70 up to 500 gtm. The capture area 312 conveys sweat to the one or more inlet ports 306 that extend through the substrate 302 to an internal channel in which the sensing electrodes 318 are positioned. The end of the channel, beyond the sensing electrodes 318, serves as the outlet 308 and is used to expel any sweat during measurement creating a continuous flow of sweat from skin to outlet 308. The outlet 308 also facilitates the expulsion of sweat samples from the internal channel through a user interaction such as blowing with breath, pumping by hand or wicking of the fluid using a hydrophilic material, which can be a sponge or textile, to allow the device to be reused. The accuracy of the sensor system can be enhanced through the geometric design of the sensing electrodes 318 and the fluidic component by controlling the distance between the electrodes 318 and the area they cover in the internal channel (fluid pathway).

[0124] A controller (not shown) may be configured to apply an alternating current between the sensing electrodes 318 and to measure the potential across the inner electrodes. The measured potential is directly proportional to the current and resistivity of the sample and inversely proportional to the conductance of the sample. By multiplying the conductance with the cell constant, k, a value for the ionic conductivity, at of the sample may be obtained. This type of system can be used to continuously capture and measure sweat from the user's skin during exercise providing the value of ai required for the determination of whole body ionic losses.

[0125] Figure 10 shows an embodiment sweat collection device 300 which is modular, having a first portion 320 that is removably detachable from a second portion 322, e.g. via snap-fit features 324 or other attachment means. In this arrangement, the second portion 322 includes the ionic sensor 314, which can thus be entirely separated from the first portion 320 (having the inlet) for cleaning. When the first portion 320 is attached to the second portion 322, the two portions define the fluid pathway at least partially therebetween. Thus, detachment of the first portion 320 from the second portion 322 facilitates access to an interior of the fluid pathway.

[0126] Due to this modular arrangement, the second portion 322 (housing the conductivity sensor and electronics (processor unit)) may be separated from the microfluidic channel, and exchanged for other ion-or biosensors (e.g. glucose, cortisol etc). Additionally or alternatively, the first portion 320 can be exchanged to provide a modified channel geometry and capture area (star, snowflake), for accommodating different sweating rates.

[0127] Figure 11 is a schematic drawing of a processing assembly 3000 according to an embodiment of the invention. The example processing assembly 3000 includes a processor 3004 for executing software routines. Although a single processor 3004 is shown for the sake of clarity, the processing assembly 3000 may also include a multi-processor system. The processor 3004 is connected to a communication infrastructure 3006 for communication with other components of the processing assembly 3000. The communication infrastructure 3006 may include, for example, a communications bus, cross-bar, or network.

[0128] The processing unit 3000 further includes a main memory 3008, such as a random-access memory (RAM), and a secondary memory 3010. The secondary memory 3010 may include, for example, a hard disk drive 3012 and / or a removable storage drive 3014, which may include an optical disk drive, solid state storage or the like. The removable storage drive 3014 reads from and / or writes to a removable storage unit 3018 in a well-known manner. The removable storage unit 3018 may include an optical disk, removable solid-state storage (e.g. SD card) or the like, which is read by and written to by removable storage drive 3014. As will be appreciated by persons skilled in the relevant art(s), the removable storage unit 3018 includes a computer readable storage medium having stored therein computer executable program code instructions and / or data.

[0129] In an alternative implementation, the secondary memory 3010 may additionally or alternatively include other similar means for allowing computer programs or other instructions to be loaded into the processing assembly 3000. Such means can include, for example, a removable storage unit 3022 and an interface 3020. Examples of a removable storage unit 3022 and interface 3020 include a program cartridge and cartridge interface (such as that found in video game console devices), a removable memory chip (such as an EPROM or PROM) and associated socket, and other removable storage units 3022 and interfaces 3020 which allow software and data to be transferred from the removable storage unit 3022 to the processing assembly 3000.

[0130] The processing assembly 3000 also includes at least one communication interface 3024. The communication interface 3024 allows software and data to be transferred between processing assembly 3000 and external devices (e.g. one or more sensors) via a communication path 3026. In various embodiments, the communication interface 3024 permits data to be transferred between the processing assembly 3000 and a data communication network, such as a public data or private data communication network. The communication interface 3024 may be used to exchange data between a plurality of different processing assembly 3000 that together form an interconnected computer network. Examples of a communication interface 3024 can include a modem, a network interface (such as an Ethernet card), a communication port, an antenna with associated circuitry and the like. The communication interface 3024 may be wired or may be wireless. Software and data transferred via the communication interface 3024 are in the form of signals which can be electronic, electromagnetic, optical, or other signals capable of being received by communication interface 3024. These signals are provided to the communication interface via the communication path 3026.

[0131] The processing assembly 3000 further includes a display interface 3002 which performs operations for rendering images to an associated display 3030 and an audio interface 3032 for performing operations for playing audio content via associated speaker(s) 3034.

[0132] As used herein, the term “computer program product” may refer, in part, to removable storage unit 3018, removable storage unit 3022, a hard disk installed in hard disk drive 3012, or a carrier wave carrying software over communication path 3026 (wireless link or cable) to communication interface 3024. These computer program products are devices for providing software to the processing assembly 3000. A computer readable medium can include magnetic media, optical media, or other recordable media, or media that transmits a carrier wave or other signal.

[0133] The computer programs (also called computer program code) are stored in main memory 3008 and / or secondary memory 3010. Computer programs can also be received via the communication interface 3024. Such computer programs, when executed, enable the processing assembly 3000 to perform one or more features of embodiments discussed herein. In various embodiments, the computer programs, when executed, enable the processor 3004 to perform features of the embodiments discussed herein. Accordingly, such computer programs represent controllers of the processing assembly 3000.

[0134] Software may be stored in a computer program product and loaded into the processing assembly 3000 using the removable storage drive 3014, the hard disk drive 3012, or the interface 3020. Alternatively, the computer program product may be downloaded to the processing assembly 3000 over the communications path 3026. The software, when executed by the processor 3004, causes the processing assembly 3000 to perform functions of embodiments described herein.

[0135] It is to be understood that the embodiment of Figure 11 is presented merely by way of example. Therefore, in some embodiments one or more features of the processing assembly 3000 may be omitted. Also, in some embodiments, one or more features of the processing assembly 3000 may be combined. Additionally, in some embodiments, one or more features of the processing assembly 3000 may be split into one or more component parts.

[0136] It will be appreciated that the elements illustrated in Figure 11 function to provide means for performing the various functions and operations of the computer-implemented methods described herein.

[0137] Embodiment methods described herein may be performed in combination with a variety of different types of sweat collection devices, e.g. as explained in the following examples.

[0138] Example 1

[0139] In one example, a new user decides to perform an exercise, cycling for 40 minutes. They are interested in their whole body ionic loss for Na+. They create a profile and input user metrics comprising their weight, height, age and sex, which is incorporated into the physiological parameter c. They choose an absorbent patch with a hand-held conductivity meter to determine the local ionic conductivity. They position the absorbent patch on their upper arm and attach a calorie counter as part of the additional metrics to monitor energy expenditure EE. The calorie counter also measures skin temperature Tskin. At the end of the session the user removes the absorbent patch, extracts the sweat sample and measures the total ionic conductivity as a,. They use a user interface coupled to a processor to input the value for conductivity 07 and the location a from which the sample was obtained. The processor is connected to the calorie counter and draws in the value for EE and the average for Tskin over the time period. The user selects Na+and conductivity measurement to be the output so that the processor selects the appropriate partitioning function k(a). The processor combines the partitioning function, the function f(T) for Tskin, and EE to determine the whole body ionic loss. As the subject is new, no previous history is available, hence no individual value for c 116 can be retrieved besides their inputs. Instead, the processing unit uses the available population data 218 for the given sex, weight, height and duration to finally compute the whole body sodium loss. This value is returned and displayed to the user. Example 2

[0140] In a second example, a regular user decides to perform an exercise: running for 30 min. They are interested in their whole body ionic loss for K+. They choose a wearable, continuous, sweat collection device (similar to device 300) for the determination of the local ionic concentration 07. in real-time and attach it to their upper back. The sweat patch contains a skin-temperature sensor 206 (not shown on device 300 in Figures 6 to 10). The user has no direct access to energy expenditure EE (i.e. no calorie counter) but is able to use their sports watch to record speed, heart-rate and distance in real-time as part of the additional session metrics for an indirectly indication of energy expenditure EE. The user connects the sweat patch and sports watch to the processor unit 200 available on their phone. Throughout the session, the processor unit 200 pulls in the values for 07 and Tskin. It also retrieves the speed, distance and heart-rate to calculate the value for EE in real-time. The value for a relates to the upper-back and is determined by an onboard accelerometer in the sweat collection device. As the user has a session history, the processor loads in the value for c from their user profile 216 and determines the whole body ionic loss in real-time, throughout the session. The final results are added to the individual’s dataset 216 as well as the population database 218.

[0141] The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

[0142] While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

[0143] For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

[0144] Any section headings used herein are for organisational purposes only and are not to be construed as limiting the subject matter described.

[0145] Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and “include”, and variations such as “comprises”, “comprising”, and “including” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.lt must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment. The term “about” in relation to a numerical value is optional and means for example + / - 10%.

[0146] References

[0147] A number of publications are cited above in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Full citations for these references are provided below. The entirety of each of these references is incorporated herein.

[0148] [1] Baker LB, Nuccio RP, Reimel AJ, Brown SD, Ungaro CT, De Chavez PJD, Barnes KA. Cross-validation of equations to predict whole-body sweat sodium concentration from regional measures during exercise. Physiol Rep. 2020. [2] Baker LB, Ungaro CT, Sopena BC, Nuccio RP, Reimel AJ, Carter JM, Stofan JR, Barnes

[0149] KA. Body map of regional vs. whole body sweating rate and sweat electrolyte concentrations in men and women during moderate exercise-heat stress. J Appl Physiol (1985). 2018.

Claims

Claims:1 . A computer-implemented method for estimating whole-body cutaneous loss of a target ion from a subject over a time period, the method comprising: estimating a local ionic concentration within sweat at a localised sub-region of the subject’s body over the time period; and estimating the whole-body cutaneous loss of the target ion over the time period based on: a temperature at or near the localised sub-region during the time period; an energy expenditure of the subject during the time period; a physiological parameter for the subject; and a partitioning function applied to the local ionic concentration; wherein the partitioning function is selected based on the target ion and a location of the localised sub-region.

2. The computer-implemented method of claim 1 , wherein the local ionic concentration is a total concentration of ions within the sweat from the localised sub-region.

3. The computer-implemented method of claim 2, further comprising: receiving a user input selecting the target ion from a plurality of potential ions; and determining the partitioning function based on the selected target ion.

4. The computer-implemented method of any preceding claim, further comprising: receiving positional data from a positional sensor located at or near the localised sub-region; identifying the location of the localised sub-region using the received positional data; and determining the partitioning function based on the identified location of the localised sub-region.

5. The computer-implemented method of any one of claims 1 to 3, further comprising: receiving a user input identifying the location of the localised sub-region; and determining the partitioning function based on the identified location of the localised sub-region.

6. The computer-implemented method of any preceding claim, further comprising: estimating the energy expenditure based on activity data received from an activity sensor such as a heart rate sensor, pedometer, or accelerometer.

7. The computer-implemented method of any preceding claim, further comprising:storing, in a memory, a dataset comprising two or more of: the local ionic concentration, temperature, energy expenditure, physiological parameter, and wholebody cutaneous loss of the target ion.

8. The computer-implemented method of claim 7, wherein the dataset is stored as part of a profile that includes a plurality of said datasets based on data collected from a plurality of time periods and / or plurality of subjects.

9. The computer-implemented method of claim 8, further comprising: estimating the whole-body cutaneous loss of the target ion over a subsequent time period using the profile.

10. The computer-implemented method of claim 9, wherein the profile includes data for a plurality of subjects; and wherein the method includes a step of filtering data in the profile based on the subject’s physiological parameter before estimating the whole-body cutaneous loss of the target ion over the subsequent time period using the profile.11 . The computer-implemented method of any one of claims 8 to 10, wherein the profile includes user-specific data collected for the subject over a plurality of time periods.

12. A method for estimating whole-body cutaneous loss of a target ion from a subject, the method comprising: collecting a sweat sample from a localised sub-region of the subject’s body over a time period; using a processing unit to perform the computer-implemented method of any preceding claim; wherein the local ionic concentration is estimated based on the collected sweat sample.

13. The method of claim 12, wherein the method comprises monitoring the local ionic concentration over a plurality of time periods, by repeating the steps for collecting a sweat sample and performing using the processing unit to perform the computer- implemented method.

14. A processor assembly having at least one processor and at least one memory including computer program code, wherein the computer program code is configured to, with the at least one processor, cause the processor assembly to perform the computer- implemented method of any one of claims 1 to 11 .

15. A system for estimating whole-body cutaneous loss of a target ion from a subject, the system comprising: a sweat collection device configured to collect a sweat sample from a localised sub-region of the subject’s body over a time period; and the processor assembly of claim 14.

16. The system of claim 15, wherein the sweat collection device comprises: a substrate; a fluid pathway extending through the substrate, the fluid pathway having: an inlet port; an outlet port; a sensor configured to measure the local ionic concentration of sweat within the fluid pathway.

17. The system of claim 16, wherein the sweat collection device comprising a fluid capture area configured to funnel sweat into the inlet port.

18. The system of claim 17, wherein the fluid capture area is shaped substantially like a star, web, asterix, snow-flake or cross mark.

19. The system of claim 17 or 18, wherein the fluid capture area comprises grooves having a width and / or depth of at least 70 m up to 500 gm.

20. The system of any one of claims 16 to 19, wherein the substrate comprises any one or more of: silicone, textile, adhesive tape, polyurethane, silicone, epoxy, metal, acrylic, PEEK, PET, teflon-based materials, ABS, PC, ABS / PC, glass.21 . The system of any one of claims 16 to 20, wherein the outlet port has a wider opening than the inlet port.

22. The system of any one of claims 16 to 21 , wherein the sweat collection device comprises a first portion and a second portion which are removably detachable from each other to provide access to an interior of the fluid pathway.

23. The system of any one of claims 16 to 22, wherein the sensor comprises a conductivity sensor.

24. The system of any one of claims 16 to 23, wherein the sensor comprises an optical sensor.

Images