Wearable sweat sensor

Abstract

A wearable sweat sensor for detecting one or more analytes in human sweat includes an optical module including at least one light source and at least one photodetector, at least one sensor layer optically coupled to the optical module and having optical absorbance characteristics dependent on a concentration of a target analyte among the one or more analytes, and one or more processors in communication with the optical module, wherein the one or more processors are configured to transmit light from the at least one light source toward and / or through the at least one sensor layer, obtain one or more optical signals reflected from and / or transmitted from the at least one sensor layer via the at least one photodetector, and determine the target analyte concentration from at least one wavelength component of the one or more optical signals.

Description

[Technical Field]

[0001] The present disclosure relates to a wearable sweat sensor. [Background technology]

[0002] While recent fitness trackers and smartwatches measure important health indicators such as heart rate, SpO2 concentration, sleep cycles, etc., they do not track health indicators at the molecular level. Therefore, various attempts have been made to develop chemical sensors that can measure analytes non-invasively and can be utilized in live biofluids such as sweat, tears, and urine. Of all the available live biofluids, sweat is the easiest and most non-invasively obtained and therefore the most appropriate choice for continuous real-time monitoring of indicators at the molecular level. Furthermore, sweat is a valuable source of sodium (Na + ), chloride (Cl - ), potassium (K + ), calcium (Ca 2+ ), pH, glucose, lactate, and more.

[0003] Among the biomarkers detectable in sweat, pH plays a vital role in the diagnosis of many important health conditions. Changes in skin pH can help diagnose skin conditions such as dermatitis, acne, and other skin infections. Also, sweat from dehydrated individuals contains high levels of Na. + This means that the pH of sweat increases with increasing levels of sodium (Na+). + It's also worth noting that an increase in pH is a proxy for increased concentrations of ATP. Increased sweating rates also trigger increases in sweat pH. Sweat pH therefore provides information about hydration levels, which is important for tracking fitness as well as diagnosing certain medical conditions.

[0004] Wearable devices provide a natural platform for real-time, continuous sweat sensing because they are in constant contact with human skin. As with existing fitness trackers that repeatedly reuse the same low-cost wearable to continuously sense a person's heart rate, sweat sensing on a wearable can also be achieved on a low-cost, reusable wearable platform.

[0005] Conventional wearable devices for continuous, real-time monitoring of sweat pH include those incorporating colorimetric sensors that measure color changes in textured fabrics or colorimetric reagents contained in microfluidic channels. However, these devices are not reusable due to their short lifespan or the need to replace the reagents. Furthermore, conventional wearable devices are unable to track time-dependent changes in sweat biomarker concentrations, making them unsuitable for real-time, continuous perception of sweat pH.

[0006] Other devices incorporate electrochemical sensors, but these devices either have a short stability period (on the order of 10 days) or are extremely complex to fabricate and require expensive instrumentation.

[0007] Other proposed sweat sensors use a pH-sensitive dye along with a paired emitter-detector red LED to continuously sense pH, but the pH-sensitive dye cannot be reused and the sensor requires a complex pumping method to pump sweat to the pH sensor.

[0008] It would therefore be desirable to provide a sweat sensor that addresses one or more of the above difficulties, or at least provides a useful alternative. Summary of the Invention

[0009] Disclosed herein is a wearable sweat sensor for detecting one or more analytes in human sweat, the wearable sweat sensor comprising: an optical module comprising at least one light source and at least one light detector mounted on a support; at least one sensor layer optically coupled to the optical module, the at least one sensor layer having an optical absorbance characteristic that depends on a concentration of a target analyte among the one or more analytes; one or more processors in communication with the optical module, transmitting light from at least one light source towards and / or through at least one sensor layer; acquiring one or more optical signals reflected and / or transmitted from the at least one sensor layer from at least one photodetector; Determining a target analyte concentration from at least one wavelength component of the one or more optical signals one or more processors configured to Equipped with.

[0010] Some embodiments of wearable sweat sensors according to the present teachings will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a schematic cross-sectional view of a first embodiment of a wearable sweat sensor. [Figure 2] FIG. 10 is a schematic cross-sectional view of a second embodiment of a wearable sweat sensor. [Figure 3] FIG. 10 is a schematic cross-sectional view of a third embodiment of a wearable sweat sensor. [Figure 4] FIG. 1 is a block diagram of a prior art pulse oximeter suitable for use as part of certain embodiments. [Figure 5] FIG. 1 illustrates a system-level block diagram of a wearable sweat sensor in accordance with certain embodiments. [Figure 6] FIG. 1 is a schematic diagram illustrating the operating principle of a wearable sweat sensor according to certain embodiments. [Figure 7] FIG. 1 is a block diagram illustrating data flow and process steps implemented by a wearable sensor in accordance with certain embodiments. [Figure 8] Figures 8(a), 8(b), 8(c), and 8(d) are a series of graphs illustrating the performance of polyaniline films suitable for use with certain embodiments. [Figure 9] 9(a) and 9(b) are graphs showing the permeability characteristics of polyaniline films. [Figure 10] 10(a) and 10(b) are graphs showing the infrared / red signal ratio as a function of pH. [Figure 11] 10 is a graph showing pH versus infrared / red ratio for an on-body trial of a wearable sensor, according to certain embodiments. [Figure 12] Graphs from real-time continuous monitoring of pH, heart rate, and SpO2 in four participants in a study conducted using wearable sensors. [Figure 13] 1 is a graph showing the cumulative distribution function of pH error for four study participants. [Figure 14] 1 is a graph showing pH error for study participants with two different skin types. [Figure 15] FIG. 1 is a diagram illustrating pairing of a wearable sweat sensor with a smartphone. DETAILED DESCRIPTION OF THE INVENTION

[0012] An embodiment of the present invention relates to a wearable sweat sensor incorporating at least one sensor layer for detecting one or more analytes in sweat. The at least one sensor layer is optically coupled to at least one light source and at least one photodetector. The at least one sensor layer has optical absorbance characteristics that depend on the concentration of a target analyte among the one or more analytes. Light transmitted by the at least one light source and reflected from the at least one sensor layer can be detected by the at least one photodetector, and different wavelength components of the detected signal (e.g., an infrared component and a red component) can be measured. When the material of the sensor layer comes into contact with sweat on the surface of a user's skin and the optical properties of the material of the sensor layer change as a result of a change in the concentration of the target analyte (e.g., hydrogen ions, cortisol, glucose), the change in the ratio of the wavelength components can be used to estimate the target analyte concentration, for example, from a calibration curve relating one or more ratios to pH value, cortisol concentration, or glucose concentration.

[0013] 1 , in a first example, a wearable sweat sensor 100 includes an optical module 110 that includes a support, such as a PCB 112, on which are mounted a number of light sources 116, 118, and 120. The light sources emit at different wavelengths. For example, the light sources may include an infrared LED 116, a red LED 118, and a green LED 120. It will be appreciated that fewer or more light sources may be provided as part of the optical module 110. For example, the optical module 110 may include only the infrared LED 116 and the red LED 118.

[0014] Also mounted on PCB 112 is a photodetector 114, such as a photodiode. It will be appreciated that additional detectors can be provided. For example, each light source can have a detector associated with it, and each detector can be configured to detect only the light emitted by its respective light source, for example, using an appropriate bandpass filter (e.g., an interference filter).

[0015] In some embodiments, a single broadband light source can be provided as part of the optical module 110, which can include multiple detectors configured to selectively detect different wavelengths, or can include a single detector, such as an optical spectrometer.

[0016] The light sources 116, 118 and 120 and the photodetector 114 are protected by a transparent protective layer 122, which may comprise a packaging layer and optionally other protection in the form of a glass layer.

[0017] The PCB 112 includes other components not shown in FIG. 1, such as signal acquisition and processing circuitry, data storage, and interfaces for connecting the PCB 112 to external devices. In this regard, the wearable sensor 100 may include, for example, an I 2 It may also include a processing module 130 that connects to the circuitry of the optical module 110 via a C interface.

[0018] The processing module 130 2 The optical module 110 may include drivers for sending control signals to the light sources 116, 118, and 120 via a C interface, one or more additional communication interfaces (such as a Bluetooth interface ("Bluetooth" is a registered trademark)) for communicating with external devices, and data acquisition components for retrieving data from storage components of the optical module 110 for analysis by one or more processors, as described in more detail below.

[0019] The optical module 110 and the processing module 130 may be partially or entirely contained within the external housing 102. The housing 102 may facilitate attachment of the sensor 100 to a user. For example, the housing 102 may be attached to a band, strap, or clip (not shown) for attachment to the user, or may be integrated as part of the housing 102.

[0020] The sensor layer 106 is disposed on the protective layer 122 so as to be optically coupled to the light source and the detector. For example, the sensor layer may be a layer of a pH-sensitive polymer. In one example, the pH-sensitive polymer may be polyaniline (PANI). PANI has been found to be particularly suitable for sensing pH, as will be shown below with reference to experimental results obtained using embodiments of the wearable sensor. PANI is also biocompatible.

[0021] In other embodiments, the sensor layer 106 can include another biocompatible material (e.g., a polymer) whose absorbance characteristics change in response to changes in the concentration of a particular target analyte, such as a glucose-responsive hydrogel. In yet other embodiments, the sensor layer 106 can include a substrate, such as a polymer (e.g., PDMS) substrate, to which aptamer-conjugated gold nanoparticles are tethered, where the aptamer can specifically bind to the target analyte (e.g., cortisol).

[0022] The sensor layer 106 may be supported on another layer 104, in particular a flexible layer such as a PDMS layer. This may have several advantages, including facilitating manufacturing of the layer 106 and / or attachment of the layer 106 to the protective layer 122, and improving skin contact and flexible fit when the sensor 100 is worn by a user. The sensor layer 106 or support layer 104 may be attached to the protective layer 122 of the optical module 110 in any suitable manner, for example, using a transparent adhesive (not shown).

[0023] In some embodiments, the optical module 110 may be or include a pulse oximeter of the type commonly found in wearable devices such as fitness trackers and smartwatches. Thus, in such embodiments, an existing device adapted for a specific type of measurement, such as heart rate and SpO2, can be repurposed to act as a sweat sensor (in addition to the existing measurement functions of the existing device). This repurposing can be achieved without modifying any of the internal hardware components of the existing device by attaching the sensor layer 106 to the existing device and augmenting or replacing software components (e.g., stored in the processing module 130) for calculating (for example) pH values. To do this, the properties of the sensor layer 106 can be tailored to the emission wavelength of the light source of the optical module 110 and / or the emission wavelength of the target analyte. For example, the sensor layer 106 can be doped (e.g., with various dopant acids) to adjust the peak position of the absorption spectrum of the layer material to more closely match the emission peak of the light source.

[0024] Another example of a wearable sweat sensor 200 is shown in FIG. 2. Components in FIG. 2 that are identical to those in FIG. 1 are assigned the same reference numerals as those in FIG. 1, and detailed descriptions thereof will be omitted. As can be seen from FIG. 2, the sensor 200 includes an optical module 210, in which the light sources 116, 118, and 120 and the photodetector 114 are encapsulated in a protective layer 206. In this example, the protective layer 206 is itself formed by a sensor layer, such as a pH-sensitive polymer. The sensor layer 206 thus functions as both a protective layer and a sensory layer. Therefore, the sensor 200 can have a thinner form factor. Furthermore, the outer surface 207 of the sensor layer 206 can have a structure that enhances the optical properties of the sensor layer 206. For example, the outer surface 207 can be curved so that the polymer layer 206 acts as a lens. In other embodiments, the outer surface 207 can have a surface structure, such as a diffractive structure, so that the polymer layer 206 can function as a diffractive optical element (DOE). In either case, the structure and / or shape of the exterior surface 207 may be such that the light is focused towards, for example, the photodetector 114 .

[0025] In the example of FIG. 1, optical module 210 may include more or fewer light sources and / or photodetectors than those shown in FIG. 2, and may optionally include bandpass filters, etc., to detect different wavelength components of light reflected from and / or transmitted from polymer layer 206.

[0026] Yet another example of a wearable sweat sensor 300 is shown in Figure 3. Components in Figure 3 that are identical to those in Figure 1 are assigned the same reference numerals as those in Figure 1, and detailed descriptions thereof will be omitted.

[0027] Sensor 300 may be identical in all respects to sensor 100 of Figure 1, except that the sensor layer includes multiple regions 106, 302, 304, in Figure 3, that are responsive to different respective analytes. Regions 106 are formed from a pH-sensitive polymer such as polyaniline, which may be doped to adjust the absorbance peak(s) of region 106, as discussed above.

[0028] Region 302 is formed from different materials that are sensitive to different target analytes, such as glucose. For example, a glucose-responsive hydrogel can be incorporated as the material of layer 302. The hydrogel can be based on polyacrylamide, N,N'-methylenebisacrylamide polymerized with phenylboronic acid, or 3-(acrylamido)phenylboronic acid. Gold nanoparticles with an absorbance peak at 600 nm can be introduced inside the hydrogel. Glucose binding to phenylboronic acid physically swells the hydrogel, which leads to different aggregation stages of the AuNPs. Therefore, the presence of different glucose concentrations results in different optical absorbances.

[0029] Region 304 can be formed from yet another different material that is sensitive to yet another target analyte, such as cortisol (a steroid hormone closely associated with stress and low blood-glucose concentrations). Instead of the hydrogel strategy used for glucose, a nucleic acid probe can be immobilized on a PDMS substrate and then tethered with an AuNP-modified aptamer complementary to the probe. Region 304 can thus comprise PDMS to which aptamer-conjugated gold nanoparticles have been tethered. Sweat containing cortisol leads to a conformational change in the aptamer, releasing the AuNPs, resulting in a change in the local concentration of AuNPs and, therefore, in the absorbance intensity. The absorbance intensity will have a direct relationship to the cortisol concentration in the sweat solution. Similar principles apply to other analytes, and aptamers can be designed (by methods known to those skilled in the art) so that other analytes can be detected by a sensor layer (or region thereof) with appropriate aptamer-conjugated nanoparticles, such as aptamer-conjugated gold nanoparticles.

[0030] As will be appreciated, if it is desired to detect more than one analyte, the optical module 110 can be modified to have additional light sources emitting at different wavelengths, and the sensor regions 106, 302, 304 of the sensor layer can be tailored so that their absorbance peaks coincide or closely match the emission peaks of the various light sources (e.g., using different types or concentrations of dopants such as acids, nanoparticles, etc.) to facilitate detection of the different analytes.

[0031] As shown in the cross-sectional view of Figure 3, the regions 106, 302, 304 targeted for different analytes are positioned next to each other. It will be appreciated that the regions targeted for different analytes may be arranged in any suitable manner across the plane of the sensor layer. For example, patches or stripes of different analyte-specific materials may be interleaved with one another and tiled across the plane of the sensor layer.

[0032] The regions 106, 302, 304 shown in Figure 3 are separated by air gaps. It will be appreciated that other methods of separating the regions are possible, such as providing optically transparent barriers between the regions that do not react with any biomolecular components of human sweat. These barriers can be formed, for example, from PDMS.

[0033] As mentioned above, the optical module 110 may be (or may comprise) a standard off-the-shelf pulse oximeter, such as the MAX30101 High-Sensitivity Pulse Oximeter 400 from Maxim Integrated Products, Inc., whose block architecture is shown in FIG. 4. The Maxim MAX30101 chip 400 is a reflective LED-based sensor that eliminates the need for a probe for transmitted light sensing, thereby enabling a small footprint, ultra-low power operation, and robust motion artifact resilience. The Maxim MAX30101 chip 400 has been adopted in various wearable hardware, such as the OpenHAK kit and Hexiwear platform. The chip has two main blocks, 402 and 404: one optical block and the other electrical block. The optical portion 402 integrates a cover glass 406 for optimal and robust performance. The electrical subsystem 404 integrates a red (peak at 660 nm) LED 116 and an infrared (peak at 880 nm) LED 118 for emitting light, and an LED driver 408 modulates the LED pulses to measure SpO2. The photodiode 114 senses reflected visible and non-visible light and converts it into an electrical signal proportional to the light intensity. A subsequent 18-bit current ADC (analog-to-digital converter) 410 samples and converts the signal into a digitized code using an ambient light cancellation (ALC) function 412. The signal contains pulse rate periodicity information, known as a photoplethysmogram (PPG). The ALC 412 cancels ambient light noise from reflected light and has an internal track / hold circuit to improve dynamic range. The Maxim 30101 system 400 also includes an on-chip temperature sensor (not shown) for calibrating for temperature changes.

[0034] A specific example of such a wearable sensor 510 with pulse oximetry is shown in FIG. 5. In the sensor 510 of FIG. 5, the processing module 530 may be or comprise an off-the-shelf module, such as a Texas Instruments Inc. CC2650STK Sensor Tag. Thus, not only can the processing module 530 include a microcontroller and application software for driving the optical module 400 and receiving and analyzing data from the optical module 400, but the processing module 530 itself can also include additional sensors, such as an accelerometer, a gyroscope, temperature and humidity sensors, etc. For example, as shown in FIG. 5, the processing module 530 can include a microcontroller 534 (which can integrate with or communicate with sensors such as an inertial measurement unit or GPS unit), as well as a debugger module 532 and a display 536 for displaying heart rate, SpO2, and pH values.

[0035] As shown in FIG. 5 , during use, the sensor 510 is attached to a user, for example, by a band or clip, so that the pH-sensitive polymer layer 106 is in contact with the user's skin 500. As a layer of sweat 502 forms on the user's skin 500, the pH of the sweat 502 changes the optical properties of the polymer (PANI) layer 106. Specifically, as shown schematically in FIG. 6 , when the PANI layer 106 comes into contact with an acidic (low pH) solution, the polymer becomes protonated by conversion of emeraldine base (EB) to emeraldine salt (ES). This protonation changes the optical properties of the polymer, resulting in strong transmission of light at 680 nm and strong absorbance of light at 880 nm. When the PANI layer 106 comes into contact with an alkaline (high pH) solution, the polymer becomes deprotonated. In this state, PANI 106 strongly absorbs light at 660 nm and transmits most light at 880 nm. By measuring the relative transmittance of these two wavelengths of light, the sensor 510 can be used to measure the pH of sweat in real time.

[0036] Each of the sensors 100, 200, 510 includes one or more processors in communication with the optical module 110, 210, or 400. At least one of these processors is part of a processing module 130 or 530 and is configured to transmit light from the light source 116, 118, or 120 toward the pH-sensitive polymer layer 106. For example, the processing module 130 or 530 may be, for example, the I shown in FIG. 2 The optical module 110, 210, or 400 may include a microcontroller (MCU) that is used to drive the light source via a C interface. The MCU may be configured to send a sequence of control signals to pulse the light source in a particular sequence at a desired sampling rate for a desired duration. For example, the MCU may be configured to cause the optical module 110, 210, or 400 to sequentially emit pulses of light from the red LED 116 and the infrared LED 118 to measure SpO2. The signal detected by the photodiode 114 in response to these pulses may also be used to determine pH, as described below.

[0037] During each sampling period, the MCU 2 The processing module 130 / 530 also receives data indicative of the optical signal detected by the photodetector 114 via the C interface. This data includes signals corresponding to different wavelengths (e.g., red and infrared). Once sampling has been performed for a desired duration, another processor in the processing module 130 / 530, or even a processor in an external device with which the processing module 130 / 530 is in communication (e.g., via a Bluetooth connection), can determine the user's pH level from the two different wavelength components (red and infrared) of the optical signal.

[0038] An exemplary software architecture implemented by the wearable sweat sensor is shown in Figure 7. This software architecture and the processes performed by the wearable sweat sensor are described below with reference to the MAX30101 pulse oximeter and CC2650 sensor tag embodiment of Figure 5, although it will be recognized that it can be readily adapted for other optical and / or processing modules.

[0039] In sensor 510, all software algorithms for pH, heart rate, and SpO2 prediction are implemented on the CC2650 sensor tag 534 for each 5 second time window. The number of samples to process is M=F due to the small SRAM of the CC2650. s The software is limited to T / 4 = 100 x 5 / 4 = 125 samples. However, it is also possible to use a faster CPU and increase the processing time window T to process more samples. The software flow for the sensor 510 shown in Figure 7 is as follows:

[0040] (1) The optical module (MAX30101) 400 illuminates the sweaty wrist with red, infrared, and green light, reads the reflected PPG signal for 2 seconds, and calculates the ADC value of the PPG signal. 2 The data is sent via the C bus to the processing module (CC2650) 530. During the same 2 second period, the accelerometer readings are also taken from the CC2650 530. The samples are read and stored in the 6KB sensor controller of the CC2650 530.

[0041] (2) The green PPG signal is preprocessed using DC removal and bandpass filtering. The accelerometer readings are also bandpass filtered. Heart rate is then calculated using the TROIKA framework to remove motion artifacts. A description of TROIKA is included in Zhang et al., IEEE Trans Biomed Eng. 2015 Feb;62(2):522-31, the entire contents of which are incorporated herein by reference.

[0042] (3) The infrared and red PPG signals are read, the DC and AC components are extracted, and the SpO2 value is calculated.

[0043] (4) Finally, the DC components of the infrared and red PPG signals are used to calculate the pH value of the sweat.

[0044] (5) The calculated pH, heart rate and SpO2 values ​​are all displayed on the Watch Devpack LCD display 536 (FIG. 5).

[0045] (6) Repeat step (1).

[0046] Immediately after the calculation, the first 2 seconds of PPG signals and intermediate values ​​are flushed. The CC2650 source code is about 800 lines long and takes about 4 seconds to execute.

[0047] Before starting exercise or engaging in any activity that stimulates sweating, the user can be prompted to press the user button on the CC2650, which will record an average of the DC components of the infrared and red PPG signals reflected from the dry wrist.

[0048] The pH, heart rate and SpO2 determination operations will now be described in more detail.

[0049] [Determination of pH value] When the MAX30101 400 attached with PANI 106 is placed on a sweaty wrist 500, the reflected PPG signal contains five components. The DC component of the PPG signal is derived from four components: reflections from tissue and sweat, non-pulsating arterial blood, reflections from pulsating arterial blood, and reflections from the PANI film 106. The AC component of the PPG signal is derived from pulsating arterial blood. Therefore, the infrared and red components reflected from the PANI film 106, which correspond to the pH value of sweat, are present in the DC component of the signal. Furthermore, a major portion (approximately 80%) of the DC component of the PPG signal is contributed by reflections from tissue. Therefore, it is necessary to separate the DC component reflected from PANI from the total DC component of the infrared and red PPG signals.

[0050] The time series PPG signal recorded by the sensor 510 is defined as follows:

[0051] I=[I(0),I(1),...,I(M-1)],R=[R(0),R(1),...,R(M-1)] where I and R are the reflected infrared and red PPG signals, respectively, recorded by the sensor 510. M is the number of samples. The sensor 510 is sampled at a sampling frequency f s = 100 Hz, the MAX30101 chip 400 can send data by averaging adjacent samples to reduce processing, so samples are averaged by 4 and the pH is calculated every T = 5 seconds over a sliding time window with an overlap of S = 3 seconds. Therefore, M = f s T=125 samples. DCI and DCR are the DC components of the infrared and red PPG signals, respectively. The infrared and red LEDs are switched on alternately in the MAX30101 400 to avoid self-heating and to reduce power consumption. The infrared and red pulse repetition frequency can be as high as 100KHz, but a lower sampling frequency also results in lower power consumption, so a sampling rate of 100Hz has been chosen.

[0052] The purpose is to Iand DC R from

number

number

number

[0053] From the DC component of the PPG signal

number

number

number

number

number

number

number

[0054] from now,

number

[0055] Using the calculated infrared / red ratio, we can use pH-to-infrared / red ratio calibration (see below) to calculate sweat pH every 5-second time window. Because the typical frequency of motion artifacts is 0.5 Hz to 5 Hz, the above pH prediction method does not suffer from motion artifacts caused by hand shaking. This pH prediction method was evaluated in an on-body trial with 10 participants (described later) and demonstrated a real-time pH prediction accuracy of approximately 91%, with an error of ±0.6 pH.

[0056] It will be appreciated that a procedure similar to that described above can be used to determine the concentration of other types of target analytes (e.g., using previously derived calibration curves of target analyte concentration versus ratio of signals at different wavelengths used for calculation).

[0057] [Heart Rate Perception] A pulse oximeter illuminates the skin with red, infrared, and green light and measures heart rate (HR) from the reflected green photoplethysmographic (PPG) signal. For heart rate measurement, the green PPG signal is used instead of the red or infrared PPG signal because the wrist has low blood perfusion and a higher energy PPG signal (green light) is preferred. The MAX30101 has three LEDs: a red LED, an infrared LED, and a green LED. As already explained, the infrared and red PPG signals are used for sweat pH estimation. For heart rate monitoring, the green PPG signal is also used in addition to the infrared and red PPG signals.

[0058] PPG signals are heavily affected by motion artifacts (MA), which interfere with heart rate measurements. Acceleration data has been shown to be an effective solution for removing MA from PPG signals, even during intense exercise. The CC2650 sensor tag is equipped with an accelerometer, so MA is removed from PPG signals using the light-weighting TROIKA framework, utilizing both the accelerometer readings from the CC2650 sensor tag and the green PPG signal obtained from the MAX30101. TROIKA consists of three main parts: (a) signal decomposition, (b) sparse signal reconstruction, and (c) spectral peak tracking.

[0059] In the TROIKA framework, the PPG signal is processed in a sliding time window of T seconds with an increment step of S seconds, preferably S≧T / 2, and HR is predicted in individual time windows of T seconds, where T=5 seconds and S=2 seconds have been chosen.

[0060] Before feeding the PPG signal into the TROIKA framework, the following pre-processing is performed on the signal.

[0061] (1) DC removal is performed on the PPG signal to achieve a baseline detrended signal with zero mean.

[0062] (2) To remove noise and MA outside the heart rate frequency, the PPG signal and triaxial accelerometer data are band-pass filtered between 0.4 Hz and 5 Hz.

[0063] It has been found that pre-processing and TROIKA MA removal can successfully clean up the signal for monitoring HR using sensor 510. Thus, adding PANI membrane 106 to pulse oximeter 400 does not interfere with the ability of pulse oximeter 400 to measure heart rate, even during exercise. Therefore, if PANI membrane 106 were attached to a fitness tracker today, any custom motion artifact removal algorithms (not just TROIKA) developed by the fitness tracker manufacturer would also be able to predict heart rate, even during intense exercise.

[0064] [SpO2 prediction] To predict SpO2, it is necessary to calculate the ratio R from the AC and DC components of the red and infrared signals, respectively, as shown in the following equation:

number

[0065] [Manufacturing method, performance measurement and pH calibration] (sensor manufacturing) To improve skin contact and flexible conformability, PANI was functionalized on a flexible and transparent support substrate prepared using polydimethylsiloxane (PDMS). The PDMS support was prepared by mixing an elastomer with a curing agent in a 10:1 weight ratio. The mixture was poured into a master mold and cured in an oven at 75 °C for 1 hour. Following curing, the PDMS layer was peeled off from the master mold. The PDMS replica (2 mm high) was thoroughly washed with isopropanol. To functionalize the surface of the PDMS 104 with a PANI film 106, the PDMS surface was first exposed to oxygen plasma (Harrick Plasma) for 60 seconds. The activated PDMS was then incubated with a 20 wt% solution of N-[3(trimethoxylsilyl)propyl]aniline in ethanol for 60 minutes. Using this technique, a monolayer of silane-bearing aniline was formed on the substrate by molecular self-assembly. Chemical deposition of PANI onto the surface of PDMS was carried out by immersing the PDMS in a freshly prepared 1 M HCl solution containing an oxidizing agent (0.25 M ammonium persulfate) and 1 M aniline. The pendant aniline on the surface served as an initiation site for polymerization and was also used to covalently anchor the PANI film 106 to the substrate 104. The polymerization time was fixed at 1 day. After polymerization, the PDMS-PANI was washed extensively with water to remove any loose PANI. The resulting film had good adhesion due to the chemical bond between the substrate and the polymer film.

[0066] The manufacturing cost of our PANI-PDMS sensors 104, 106 was less than $1. Furthermore, due to the simplicity of the process and its amenability to castings of various shapes and sizes, this technique allows for the mass production of our pH sensors.

[0067] (sensor performance measurement) First, the optical transmittance spectra of the prepared PANI membranes were evaluated using a plate reader. A plate reader is a device that emits light of a specific wavelength onto the PANI membrane 106 and measures the absorbance / transmittance from the reflected light. The membranes were incubated in different pH solutions for 1 min before collecting UV-Vis transmittance spectra. As shown in Figure 9(a), PANI optical changes are highly sensitive to pH changes. As the pH increased from 2 (acidic) to 12 (alkaline), the PANI membranes exhibited a shift in absorption peak from 420 nm (at pH 2) to 605 nm (at pH 12). This change is consistent with published studies on different degrees of protonation of the imine nitrogen atoms in the polymer chain. Next, the pH dependence of transmittance at 660 nm and 880 nm, respectively, was plotted (Figure 9(b)). A characteristic sigmoidal curve for PANI was obtained with good correlation over a wide range of pH changes, with R2 (660 nm) = 0.989 and R2 (880 nm) = 0.996. To further improve the detection accuracy and material stability, the ratio of the transmittance intensity at 880 nm to that at 660 nm was obtained, as shown in Figure 8(a). This intensity ratio decreased and saturated as the solution pH changed from 2 to 12. It is noteworthy that this ratio change was most pronounced in the pH range of 3 to 8, which corresponds to the physiological pH range of human sweat. To demonstrate the repeatability of the sensor for measuring pH changes, the integrated system was repeatedly tested using two different pH solutions, pH 3.4 and pH 7.0, prepared by mixing different amounts of HCl and NaOH solutions. As shown in Figure 8(b), after alternating four cycles of changing the solution pH, the PANI film maintained a relatively consistent infrared / red ratio at the same pH, demonstrating the repeatability of this material. The sensors not only demonstrated good reproducibility to produce comparable signal outputs when treated with solutions of the same pH, but also showed excellent response and repeatability. + , K. + , Cl - , P 5- and H +Next, the specificity of the integrated sensor to pH changes against a complex ion background was tested. Four different solution mixtures were selected: a sodium-based buffer, a potassium-based buffer, a phosphate-based buffer, and an HCl / NaOH solution (pH 7). These solutions contain various ions, such as those mentioned above (Na + , K. + , Cl - , P 5- and H + These mixtures were prepared to pH = 7, containing varying concentrations of HCl / NaOH (pH 3.5). Separately, a mixture of HCl / NaOH solutions (pH 3.5) was prepared. PANI films were incubated in these solutions for 1 min, and ratiometric measurements were performed. As demonstrated in Figure 8(c), the sensor selectively responded to pH changes. Finally, the reproducibility of the membrane preparation, which yields reliable analysis, was tested. Multiple PANI films were prepared, and the signal ratios of these sensors were measured when incubated in various pH solutions. As shown in Figure 8(d), the sensor provided highly uniform and robust signals throughout the pH monitoring, with a very small relative standard deviation (RSD) of 5.3%, without any significant deviations.

[0068] (pH calibration) To perceive pH values ​​from sweat, it is first necessary to calibrate the ratio of reflected infrared to red light from (for example) a PANI-attached MAX30101 for different pH values. To calibrate the infrared / red light ratio curve using the MAX30101, synthetic sweat solutions with pH values ​​between 3 and 8 were created, since the pH of human sweat is between 3 and 8. 200 μL of each pH solution was placed on the MAX30101, and the average ratio of reflected infrared to red light over a 30-second period was measured. This procedure was repeated three times for each pH solution. For each pH value, the average infrared / red light ratio recorded during the three different trials was used to calibrate the pH vs. infrared / red light ratio curve by performing a fourth-order polynomial fit of the infrared / red light ratio recorded for each pH solution. Because PANI does not produce a linear trend in the infrared / red light ratio, as can be seen in Figure 8(a), and a polynomial fit provides more accurate results than a linear fit, a fourth-order polynomial fit was used instead of a linear fit. Figure 10(a) shows the calibrated pH vs. infrared / red ratio curve, which clearly shows the same increasing trend as the pH vs. infrared / red ratio curve in Figure 8(a) obtained using a plate reader. An ordinary least squares (OLS) fit between the infrared / red ratio obtained from the plate reader and the infrared / red ratio recorded from the MAX30101 was also performed. The resulting fit, shown in Figure 10(b), had a very large R of 0.981. 2 values, indicating that the infrared / red ratios obtained from MAX30101 follow the same trend as the infrared / red ratios obtained using the plate reader.

[0069] [Experimental evaluation] (pH-Watch accuracy for synthetic pH solutions) First, to calculate the accuracy of PANI in predicting pH using the calibration curve in Figure 10(a), the MAX30101 with PANI attached was evaluated without placing it on the skin. Synthetic buffer solutions with different pH values ​​were created by mixing different concentrations of HCl and NaOH. A commercial wireless pH meter, the HI14142-HALO® Wireless pH Meter from Hannah Instruments, was used to measure the pH of each solution. The HI14142 can also measure sweat pH directly from the skin. 200 μL of each solution was placed on the PANI, and the pH value was measured from the reflected infrared / red ratio. Table 2 shows the pH values ​​of the different solutions measured by the commercial pH meter and the MAX30101 with PANI attached. [Table 1] The average error between the pH measured by the commercial pH meter and the predicted pH values ​​was found to be 2.13%. The predicted pH values ​​varied by a maximum of 2.5% from the commercial pH meter readings, which is comparable to the pH change of less than 2.2% reported in the literature for PANI-based electrochemical pH sensors.

[0070] (pH-Watch accuracy in measuring heart rate (HR) and SpO2) The accuracy of the pH Watch in measuring heart rate and SpO2 with and without motion artifact was evaluated. Participants were recruited to wear a pH Watch on their left wrist and a MAX30102 high-sensitivity finger pulse oximeter (with infrared and red LEDs) on their right hand, which served as ground truth for HR and SpO2 measurements. Measurements from the MAX30102 were measured and recorded using the MAX30102ACCEVKIT evaluation board, which reads finger PPG measurements and calculates HR and SpO2. During the experiment, participants restricted the movement of their right hand to ensure that the ground truth HR and SpO2 values ​​were not affected by motion artifacts. Each experiment lasted 120 seconds. The experiment was repeated eight times with the left hand, which held the sensor 510, stationary to test accuracy in the absence of motion. To test accuracy in the presence of motion artifacts, the experiment was repeated eight times with participants required to induce random, continuous movement of the left hand by waving and vigorously shaking their left hand back and forth. For the left hand movement experiment, the user was asked to remain still for the first 3 seconds, as TROIKA requires heart rate initialization. Finally, for each experiment, the resulting HR and SpO2 measurements from sensor 510 were compared against ground truth HR and SpO2 measurements recorded by a finger pulse oximeter.

[0071] (Accuracy of HR and SpO2 measurements without motion artifact) Table 3 shows HR measurements recorded by the sensor 510 with no hand movement that were validated against finger pulse oximeter readings, which served as ground truth. The average percentage errors for HR and SpO2 measurements were 1.22% and 2.75%, respectively, and HR and SpO2 changed by a maximum of less than 3.44% and <3.22%, respectively. The observed errors are negligible and can be attributed to the fact that PPG readings from the finger are inherently more accurate than PPG readings from the wrist due to better blood perfusion in the finger. [Table 2]

[0072] (Accuracy of HR and SpO2 measurements with motion artifact) Table 4 shows HR measurements recorded by the sensor 510 with random hand movements, validated against finger pulse oximeter readings that served as ground truth. The average percentage errors for HR and SpO2 measurements were 4.98% and 4.57%, respectively, with HR and SpO2 varying by less than 6.41% and <6.73% maximum, respectively. This is comparable to the average heart rate error of 1.8% and maximum change of 4.70% originally reported in TROIKA. The slight difference in error rates is due to the sensor 510 being configured to use a very short time window of 5 seconds due to RAM limitations in the CC2650. The study in the original TROIKA document used a 10-second time window with 1250 samples, thus providing more accurate HR measurements than the pH Watch. The average error can be further reduced using a more powerful CPU with larger RAM, which more recent wearables already have. For example, the latest Samsung Galaxy watch has a 1.15GHz Exynos 9110 processor with 768MB RAM. Therefore, the experiment clearly shows that the pH Watch with PANI does not interfere with the normal ability of the pulse oximeter to measure HR and blood oxygen level, thereby confirming that Applicant's pH sensing approach can be easily integrated into any today's smartwatch or fitness tracker with the addition of Applicant's PANI film 106. [Table 3]

[0073] [On-body trial] To evaluate the accuracy and efficiency of the pH Watch for real-time monitoring of pH from human sweat, 10 participants were recruited and an on-body trial was conducted using the sensor 510 to measure the participants' sweat pH values.

[0074] Six participants were recruited for an initial experiment to test the accuracy of the sensor 510 for sensing pH values ​​from sweat. To protect the participants' personal data, randomized IDs were assigned to each participant, and data collected from those participants was mapped to the randomized ID. Each participant was asked to wear the sensor 510 and exercise for 10 to 20 minutes, causing them to sweat. Before starting the exercise, participants were asked to press the user button on the sensor 510 to measure the DC init The tissue DC of the participants was measured over a 3-second period. When the participant began to sweat, the sweat pH measured by the sensor 510 was compared to the sweat pH measured by the HI14142 wireless pH meter to calculate accuracy. The above procedure was repeated twice for each participant.

[0075] In a second experiment to evaluate the sensor 510's ability to continuously sense pH values ​​in real time and detect dehydration during exercise, four participants were recruited to wear the sensor 510 and perform periodic exercises over 80 minutes. HR, SpO2, and pH values ​​measured by the pH Watch were recorded. Because the output speed of our wireless pH meter is limited, ground truth pH values ​​were measured manually every 5 or 10 minutes during exercise using the wireless pH meter. Our wireless pH meter takes approximately 30 seconds to obtain a stable reading. Because sweat pH changes very slowly with exercise and does not change significantly within 10 minutes, this does not affect our evaluation.

[0076] Our evaluation considered participants between the ages of 21 and 30 (Chinese, Malay, and Indian), with a variety of skin tones ranging from Chinese skin types to moderately dark Indian skin. All participants were male. A summary of the participants' skin types and demographics is shown in Table 6. [Table 4]

[0077] (pH Watch accuracy) Table 7 shows pH measurements made by the pH Watch validated against sweat pH values ​​measured by a wireless pH meter during an on-body trial according to the first experimental procedure. The average percentage error was 2.31% from the pH meter reading, with a maximum variation of less than 4.3%. Also shown in Figure 11 is a pH vs. infrared / red ratio curve for the pH values ​​measured during this experiment, which is similar to the calibration curve in Figure 10(a). An R of 0.99 indicates a very high correlation between these two curves. 2 Values ​​were observed that are similar to previously described wearable pH sensing approaches, showing a maximum change of less than 2.2% as reported by (1) PANI-based electrochemical pH sensing approaches and (2) 8.3% (±0.5 pH change) as reported by ORMOSIL textile-based optical pH sensing approaches. This demonstrates that sensor 510 has sensing accuracy comparable to current prior art pH sensors. Furthermore, it is reusable for extended periods of time, making it compatible with today's popular wearables. [Table 5]

[0078] (Real-time continuous monitoring of HR, SpO2 and pH) 12 shows a participant's HR, SpO2, and pH measurements continuously recorded by the sensor 510 during a second experimental procedure to evaluate the real-time perception capabilities of the sensor 510. The real-time heart rate measurements were (1) At the start of exercise, the heart rate is between 70 BPM and 90 BPM for the first 5 minutes, indicating the start of exercise. (2) HR remained fairly constant around 90 BPM for the next 10 minutes, then increased again to 100 BPM as exercise resumed, indicating that the participant was now engaging in faster cyclic exercise. (3) HR remained near 110 BPM for the next 15 minutes, then increased to 120 BPM and stabilized at around 120 BPM. The increase in HR in (3) may be due to a decrease in the body's hydration level as sweating occurs during exercise. This decrease in the body's hydration level reduces strain on the heart by reducing blood volume. Also, as a result of the decrease in hydration level, the blood carries more sodium, which makes it more difficult for the heart to pump blood. Therefore, the heart beats faster, resulting in an increase in HR.

[0079] Meanwhile, real-time SpO2 measurements remain consistent, remaining near 97% for most of the time, with minimum SpO2 near 96% and maximum SpO2 near 98%. This is consistent with the fact that SpO2 levels self-regulate and remain stable during non-vigorous and moderate exercise, provided that a person continues to breathe sufficient oxygen during exercise.

[0080] Figure 12 shows real-time pH values ​​verified against sweat pH values ​​measured by a wireless pH meter. The real-time pH values ​​increased over time during exercise. Because the participant did not sweat during the first 20 minutes, pH readings are not shown on the graph. Because the wrist area does not sweat very quickly, it took the participant 20 minutes to sweat. After 20 minutes, the participant began to sweat, and for the next 20 minutes, the sweat pH value remained stable, near 5.2–5.3. This stability in sweat pH is due to the extremely limited sweat rate during the moderate phase of exercise. As the user increased the intensity of their exercise and began to see results, the sweat pH increased from 5.3 to 5.7, indicating an increase in the participant's sweat rate. This is also aptly shown in the heart rate measurements, which clearly increased to 120 BPM due to increased sweat rate and decreased hydration levels. Then, over the next 20 minutes, the pH slowly increased from 5.7 to 5.8, and increased again to 6.2 within 10 minutes. This also confirms that when a person exercises for a long duration (70-80 minutes), their sweat production increases, and the sodium concentration in their sweat also increases, which in turn increases the pH value of their sweat.

[0081] The same experimental procedure was repeated for three other participants, and the real-time pH, HR, and SpO2 values ​​measured by the sensor 510 are shown in Figure 12. Figure 13 shows the cumulative distribution function of the pH error for each of the four participants.

[0082] The median pH error ranged from 0.2 to 0.28 for all participants. The maximum pH error ranged from 0.4 to 0.6. Thus, the pH values ​​measured by the sensor 510 differed from the wireless pH meter by a maximum of ±0.6 pH variation, which is comparable to previously reported results, and also indicates that the sensor 510 can detect pH with approximately 91% accuracy. This clearly demonstrates the potential of the sensor 510 to detect bodily dehydration levels during exercise, and future studies on the correlation between heart rate and sweat pH values ​​may prove useful for accurately detecting dehydration, at the mere cost of adding a low-cost, off-the-shelf pH-sensitive PANI 106 to pulse oximeters used in current fitness trackers.

[0083] For our real-time continuous perception experiment, participants with Chinese and Indian skin types were included. Figure 14 shows the pH errors reported by Indian and Chinese skin types. It can be observed that Indian and Chinese skin types have similar average pH errors of 0.2 and 0.27, respectively, verifying that our pH measurements show similar errors for different skin types and maintain errors relatively unaffected by skin color. More intuitively, skin color has little effect on our pH measurements because it contributes to the DC portion of the PPG signal, which is measured during the initialization part of our pH perception algorithm and removed while measuring pH.

[0084] In some embodiments, the sensor 100, 200, 300, or 510 can be configured to pair with an external device, such as a smartphone 1500 shown in Figure 15. For example, pairing can occur via a Bluetooth connection 1510.

[0085] Smartphone 1500 can run a mobile application (app) that allows a user wearing a sensor (e.g., sensor 510) to continuously track dehydration risk / skin health by monitoring trends in real-time pH values ​​obtained from sensor 510.

[0086] The mobile app can recommend skin care regimens and cosmetics based on observed trends in sweat pH. The mobile app can also provide a "Drink Water" feature that alerts the user whenever the user's sweat pH rises above the normal range, thereby helping the user stay hydrated. Normal sweat pH ranges between 4.5 and 7.0. In dehydrated states, sweat pH maintains a higher level of 6 to 8 (similar to water). The Drink Water feature implemented in the mobile app can remind the user to drink water whenever a pH value higher than 6 is detected by sensor 510. Once the user's pH returns to normal after drinking water, the mobile app can generate a further notification informing the user that their hydration level is sufficient.

[0087] An exemplary GUI wireframe for the mobile app that may be displayed on the mobile device's display 1502 is shown in expanded view on the right side of Figure 15. This GUI allows for easier tracking of health conditions from sensors 510.

[0088] The sweat pH determined by sensor 510 is an indicator of skin health. Cosmetic products are known to alter the pH of human skin. To suggest the correct cosmetic product for the user's skin, the pH value obtained from sensor 510 can be used to recommend cosmetic products with appropriate corresponding pH values ​​to maintain skin health and ensure that the cosmetic product does not adversely affect skin pH. To that end, the mobile app can maintain a database of widely used cosmetic products and their corresponding pH values.

[0089] The use of pH-sensitive polymers in embodiments of the present invention offers many advantages. These polymers are flexible, easily miniaturized, and biocompatible, allowing for easy integration with wearables. The conductive polymer polyaniline (PANI), in particular, not only offers a wide range of sensitive pH response but also enables real-time optical readout. After proton-mediated post-polymer doping and dedoping, PANI exhibits significant changes in the near-infrared spectrum, a target frequency for pulse oximetry.

[0090] Attaching a sensor layer, such as a pH-sensitive polymer layer, to a pulse oximeter to enable reusable, real-time, continuous monitoring of target analytes (e.g., pH) in sweat offers the following advantages. First, the polymer acts as a dual-substrate support and indicator dye—a substance used to visually indicate the state of a solution (e.g., different concentrations of hydrogen ions) in response to the presence of a specific material—by changing color, allowing for easy interface for safe skin contact. The polymer is biocompatible and does not degrade. This improves long-term stability by eliminating any possible leaching of dyes, a common problem with sensors using other pH-responsive small molecule dyes. Second, when treated with different pH solutions, PANI exhibits significant and rapid responses at wavelengths of 660 nm and 880 nm, which are quite different from the wavelengths illuminated and measured by many pulse oximeters. This makes it highly compatible with direct integration, thus enabling sensitive, reusable monitoring of sweat pH.

[0091] The embodiment also enables effective control of the thickness of the PANI film, adapting to the thickness requirements of different devices. First, from a macroscopic perspective, the substrate-PDMS thickness can be easily altered. Furthermore, for different PANI film thicknesses, the thickness can be controlled by changing the temperature. The film thickens as the temperature increases, reaching thicknesses of several microns. The film growth rate increases with increasing temperature, and the rate accelerates with time. Based on this thickness tunability, the polymer film can be tailored to suit different chip morphologies and appearance designs.

[0092] When fabricating a PANI layer on PDMS, it is advantageous to add a mask to the PDMS so that only specific arrays of silane-loaded aniline molecules are formed on the exposed substrate by molecular self-assembly. By adjusting the exposed area, the transparency of the PANI sensor layer can be varied. Adjusting the transparency of the PANI film results in a more compatible detection signal. For example, under a highly transparent polymer layer, light absorbed by the PANI itself can be filtered out, and the remaining light can be used to detect other targets within other sensors.

[0093] Many modifications will be apparent to those skilled in the art that do not depart from the scope of the invention.

[0094] Throughout this specification, unless the context requires otherwise, the term "comprises" and variations such as "comprising" will be understood to be implicitly meant to encompass a stated integer or step or group of integers or steps, but not to exclude any other integer or step or group of integers or steps.

[0095] Any reference in this specification to any prior publication (or information derived therefrom) or to any known material is not, and should not be taken as, an acknowledgement or admission, or any form of suggestion, that the prior publication (or information derived therefrom) or known material forms part of the common general knowledge in the field of endeavor to which this specification pertains.

[0096] A description of some example embodiments of the present disclosure is contained in one or more of the following numbered descriptions: 1. A wearable sweat sensor for detecting one or more analytes in human sweat, comprising: an optical module comprising at least one light source and at least one light detector mounted on a support; at least one sensor layer optically coupled to the optical module, the at least one sensor layer having an optical absorbance characteristic that depends on a concentration of a target analyte among the one or more analytes; one or more processors in communication with the optical module, transmitting light from at least one light source towards and / or through at least one sensor layer; acquiring one or more optical signals reflected and / or transmitted from the at least one sensor layer from at least one photodetector; Determining a target analyte concentration from at least one wavelength component of the one or more optical signals one or more processors configured to A wearable sweat sensor comprising: 2. The wearable sweat sensor of 1, wherein the optical module comprises a pulse oximeter. 3. A wearable sweat sensor as described in 1 or 2, wherein the optical module includes multiple light sources that emit light of different wavelengths. 4. A wearable sweat sensor as described in any one of claims 1 to 3, wherein the one or more processors are further configured to determine the user's heart rate and / or SpO2 from the one or more optical signals in addition to determining the target analyte concentration. 5. A wearable sweat sensor described in any one of claims 1 to 4, wherein the optical module includes a plurality of photodetectors configured to detect light of different wavelengths. 6. A wearable sweat sensor described in any one of claims 1 to 5, wherein one or more processors are configured to determine a target analyte concentration of a user based on a ratio of two different wavelength components of one or more optical signals. 7. The wearable sweat sensor of any one of claims 1 to 6, wherein one of the analytes is hydrogen ion and at least one sensor layer comprises a pH-sensitive polymer layer. 8. The wearable sweat sensor according to 7, wherein the pH-sensitive polymer is polyaniline. 9. The wearable sweat sensor of any one of claims 1 to 8, wherein one of the analytes is glucose and at least one sensor layer comprises a glucose-responsive hydrogel. 10. The wearable sweat sensor of 9, wherein the hydrogel comprises polyacrylamide, N,N'-methylenebisacrylamide polymerized with 3-(acrylamido)phenylboronic acid. 11. The wearable sweat sensor according to 9 or 10, wherein the glucose-responsive hydrogel contains gold nanoparticles. 12. The wearable sweat sensor according to 11, wherein the gold nanoparticles have an absorbance peak at 600 nm. 13. A wearable sweat sensor described in any one of 1 to 12, wherein at least one sensor layer comprises a polymer layer to which aptamer-conjugated gold nanoparticles are tethered (bound), and the aptamer is capable of specifically binding to a target analyte. 14. The wearable sweat sensor of claim 13, wherein the target analyte is cortisol. 15. A wearable sweat sensor described in any one of claims 1 to 14, comprising multiple sensor layers, each of the multiple sensor layers configured to detect a different target analyte among the one or more analytes. 16. A wearable sweat sensor described in any one of claims 1 to 15, wherein at least one of the plurality of sensor layers comprises a plurality of regions, each of the plurality of regions configured to detect a different target analyte among the one or more analytes. 17. The wearable sweat sensor described in any one of 1 to 16, wherein the wavelength components include a red component and an infrared component. 18. A wearable sweat sensor described in any one of claims 1 to 17, wherein at least one sensor layer comprises multiple regions, each of the multiple regions having a different thickness and / or surface texture and / or dopant level. 19. The wearable sweat sensor of any one of claims 2 to 18, wherein at least one sensor layer is attached to or integral with a pulse oximeter. 20. The wearable sweat sensor according to 19, wherein at least one sensor layer forms a protective layer for the pulse oximeter. 21. A wearable sweat sensor as described in any one of claims 1 to 20, comprising one or more of a band, clip, and adhesive layer for attaching the support to a user. 22. A wearable sweat sensor as described in any one of claims 1 to 21, wherein the one or more processors are configured to issue an alarm based on the determined pH level.

Claims

1. 1. A wearable sweat sensor for detecting one or more analytes in human sweat, comprising: an optical module comprising a pulse oximeter, at least one light source, and at least one photodetector mounted on a support; at least one sensor layer optically coupled to the optical module, the at least one sensor layer having one or more regions responsive to target analytes, each region having an optical absorbance characteristic that depends on a concentration of the target analyte among the one or more analytes; one or more processors in communication with the optical module, transmitting light from the at least one light source towards and / or through the at least one sensor layer; obtaining one or more optical signals reflected and / or transmitted from the one or more regions from the at least one photodetector; determining the optical absorbance characteristic of each region from at least one wavelength component of the one or more optical signals; and determining the target analyte concentration from the optical absorbance characteristic. and one or more processors configured to: the one or more analytes include one or both of glucose and cortisol; Wearable sweat sensor.

2. The wearable sweat sensor of claim 1 , wherein the one or more analytes are hydrogen ions, sodium ions, chloride ions, or potassium ions.

3. The wearable sweat sensor of claim 1 or claim 2, wherein the optical module comprises a plurality of light sources that emit light of different wavelengths.

4. The one or more processors, in addition to determining the target analyte concentration, determine the user's heart rate and / or SpO2 from the one or more optical signals. ２ The wearable sweat sensor of claim 1 , further configured to determine:

5. The wearable sweat sensor of claim 1 , wherein the optical module comprises a plurality of photodetectors configured to detect light of different wavelengths.

6. 6. The wearable sweat sensor of claim 1, wherein the one or more processors are configured to determine the target analyte concentration of the user based on a ratio of two different wavelength components of the one or more optical signals.

7. one of the analytes is glucose, and at least one sensor layer comprises a glucose-responsive hydrogel comprising gold nanoparticles, the aggregation of which is dependent on glucose concentration, and the glucose-responsive hydrogel having optical absorption properties that are dependent on the aggregation; The wearable sweat sensor according to any one of claims 1 to 6.

8. The wearable sweat sensor of claim 7 , wherein the glucose-responsive hydrogel comprises polyacrylamide, N,N′-methylenebisacrylamide polymerized with 3-(acrylamido)phenylboronic acid.

9. The wearable sweat sensor of claim 7 or claim 8, wherein the glucose-responsive hydrogel comprises gold nanoparticles.

10. 10. The wearable sweat sensor of claim 1, wherein the at least one sensor layer comprises a polymer layer having aptamer-conjugated gold nanoparticles attached thereto, the aptamer-conjugated gold nanoparticles comprising gold nanoparticles bound to an aptamer, the aptamer being capable of specifically binding to the target analyte.

11. The wearable sweat sensor of claim 10 , wherein the target analyte is cortisol.

12. 12. The wearable sweat sensor of claim 1, comprising a plurality of sensor layers, each of the plurality of sensor layers configured to detect a different target analyte of the one or more analytes.

13. 13. The wearable sweat sensor of claim 12, wherein at least one of the plurality of sensor layers comprises a plurality of regions, each of the plurality of regions configured to detect a different target analyte of the one or more analytes.

14. The wearable sweat sensor of claim 1 , wherein the wavelength components include a red component and an infrared component.

15. 15. The wearable sweat sensor of claim 1, wherein the at least one sensor layer comprises a plurality of regions, each of the plurality of regions having a different thickness and / or surface texture and / or dopant level.

16. The wearable sweat sensor of any one of claims 1 to 15, wherein the at least one sensor layer is attached to or integral with the pulse oximeter.

17. 17. The wearable sweat sensor of claim 16, wherein at least one sensor layer forms a protective layer of the pulse oximeter.

18. The wearable sweat sensor of any one of claims 1 to 17, comprising one or more of a band, a clip, and an adhesive layer for attaching the support to a user.

Images