Fully integrated wearable sensor arrays for multiplexed in situ perspiration analysis

Abstract

Wearable sensor technologies are essential to the realization of personalized medicine through continuously monitoring an individual's state of health. Sampling human sweat, which is rich in physiological information, could enable non-invasive monitoring. Previously reported sweat-based and other non-invasive biosensors either can only monitor a single analyte at a time or lack on-site signal processing circuitry and sensor calibration mechanisms for accurate analysis of the physiological state. Given the complexity of sweat secretion, simultaneous and multiplexed screening of target biomarkers is critical and requires full system integration to ensure the accuracy of measurements. Here we present a mechanically flexible and fully integrated (that is, no external analysis is needed) sensor array for multiplexed in situ perspiration analysis, which simultaneously and selectively measures sweat metabolites (such as glucose and lactate) and electrolytes (such as sodium and potassium ions), as well as the skin temperature (to calibrate the response of the sensors). Our work bridges the technological gap between signal transduction, conditioning (amplification and filtering), processing and wireless transmission in wearable biosensors by merging plastic-based sensors that interface with the skin with silicon integrated circuits consolidated on a flexible circuit board for complex signal processing. This application could not have been realized using either of these technologies alone owing to their respective inherent limitations. The wearable system is used to measure the detailed sweat profile of human subjects engaged in prolonged indoor and outdoor physical activities, and to make a real-time assessment of the physiological state of the subjects. This platform enables a wide range of personalized diagnostic and physiological monitoring applications.

Extended Data Figure 1. Fabrication process of the flexible sensor array

a, PET cleaning using acetone, isopropanol and O₂ plasma etching. b, Patterning of Cr/Au electrodes using photolithography, electron-beam evaporation and lift-off in acetone. c, Parylene insulating layer deposition. d, Photolithography and O₂ plasma etching of parylene in the electrode areas. e, Electron-beam deposition of the Ag layer followed by lift-off in acetone. f, Ag etching on the Au working electrode area and Ag chloridation on the reference electrode area. g, Optical image of the flexible electrode array. h, Photograph of the multiplexed sensor array after surface modification.

Extended Data Figure 2. The characterizations of the modified electrodes

a, Cyclic voltammetry of the amperometric glucose and lactate sensors using Prussian blue as a mediator in PBS (pH 7.2). Scan range, −0.2 V to 0.5 V; scan rate, 50 mV s⁻¹. b, Potential stability of a PVB-coated Ag/AgCl electrode and a solid-state Ag/AgCl reference electrode (versus commercial aqueous Ag/AgCl electrode) in different NaCl solutions. c, d, The stability of a PVB-coated reference electrode in solutions containing 50 mM NaCl and 10 mM of different anionic (c) and cationic (d) solutions. Data recording was paused for 30 s for each solution change in b–d.

Extended Data Figure 3. The custom-developed mobile application for data display and aggregation

a, The home page of the application after Bluetooth pairing. b, Real-time data display of sweat analyte levels as well as skin temperature during exercise. c, Real-time data progression of individual sensor. d, Available data sharing and uploading options.

Extended Data Figure 4. Schematic diagram of signal-conditioning circuit

a–d, Signal conditioning circuits for (a) glucose, (b) lactate, (c) sodium and (d) potassium channels. VDD and VSS represent the positive and negative power supplies, respectively. LT1462 is the integrated-circuit chip part.

Extended Data Figure 5. The calibration and power delivery of the FISA

a–d, Flexible PCB calibration for glucose (a), lactate (b), sodium (c) and potassium (d) channels. e, Power delivery diagram of the system. f, Photograph of a small rechargeable battery module used in the current work (placed next to a quarter-dollar coin for comparison). g, Representative photograph of the power delivery package inside a transparent wristband on a subject's wrist.

Extended Data Figure 6. Reproducibility and long-term stability of the biosensors

a–d, The reproducibility of the sodium (a), potassium (b), glucose (c) and lactate (d) sensors (eight samples for each kind of sensor). e–h, The long-term stability of the sodium (e), potassium (f), glucose (g) and lactate (h) sensors. Sensitivity is measured in millivolts per decade of concentration. The error bars represent the standard deviations of the measured data for five samples.

Extended Data Figure 7. Selectivity study for electrochemical biosensors

a–d, The interference study for individual glucose (a), lactate (b), sodium (c) and potassium (d) sensors using an electrochemical working station. Data recording was paused for 30 s for the addition of each analyte in c and d. e, f, The real-time system-level interference study (e) and calibration plot (f) of the amperometric glucose and lactate sensor array with a shared solid-state Ag/AgCl reference electrode. g, h, The real-time interference study (g) and calibration plot (h) of the potentiometric Na⁺ and K⁺ sensor array with a shared PVB-coated reference electrode. Data recording was paused for 30 s for each solution change in e and g.

Extended Data Figure 8. Mechanical deformation study of the flexible sensors and the FPCB

a–f, The responses of the sodium (a), potassium (b), glucose (c), lactate (d), temperature (e) sensors and of the FPCB (f) after 0, 30 and 60 cycles of bending. g–l, The responses of the sodium (g), potassium (h), glucose (i), lactate (j),and temperature (k) sensors and of the FPCB (l) during bending. The radii of curvature for the bending study of sensors and the FPCB were 1.5 cm and 3 cm, respectively. Data recording was paused for 30 s to change the conditions and settings.

Extended Data Figure 9. On-body real-time perspiration analysis during stationary cycling using the FISA on a subject's wrist

Conditions are as in Fig. 3c and d.

Extended Data Figure 10. Ex situ measurement of collected sweat samples using the FISA on a subject during stationary cycling at 150 W

a, c, The ex situ results of [Na⁺] (a) and [K⁺] (c) from the sweat samples collected from the subject's forehead without water intake (~2.5% of body weight dehydration). b, d, The ex situ results of [Na⁺] (b) and [K⁺] (d) from the sweat samples collected from the subject's forehead with water intake (150 ml per 5 min).

Figure 1. Images and schematic illustrations of the FISA for multiplexed perspiration analysis

a, Photograph of a wearable FISA on a subject's wrist, integrating the multiplexed sweat sensor array and the wireless FPCB. (All photographs in this paper were taken by the authors.) b, Photograph of a flattened FISA. The red dashed box indicates the location of the sensor array and the white dashed boxes indicate the locations of the integrated circuit components. c, Schematic of the sensor array (including glucose, lactate, sodium, potassium and temperature sensors) for multiplexed perspiration analysis. GOx and LOx, glucose oxidase and lactate oxidase. d, System-level block diagram of the FISA showing the signal transduction (orange) (with potential V, current I and resistance R outputs), conditioning (green), processing (purple) and wireless transmission (blue) paths from sensors to the custom-developed mobile application (numbers in parentheses indicate the corresponding labelled components in b). ADC, analogue-to-digital converter. The inset images show the home page (left) and the real-time data display page (right) of the mobile application.

Figure 2. Experimental characterizations of the wearable sensors

a, b, The chronoamperometric responses of the glucose (a) and lactate (b) sensors to the respective analyte solutions in phosphate-buffered saline (PBS). c, d, The open circuit potential responses of the sodium (c) and potassium (d) sensors in NaCl and KCl solutions. e, The resistance response of the temperature sensor to temperature changes (20–40 °C) in PBS. Insets in a–e show the corresponding calibration plots of the sensors. Data recording was paused for 30 s for each solution change in a–e. f, System-level interference studies of the sensor array. g, The influence of temperature on the responses of the glucose and lactate sensors. h, System-level real-time temperature T compensation for the glucose and lactate sensors in 100-μM glucose and 5-mM lactate solutions, respectively.

Figure 3. On-body real-time perspiration analysis during stationary cycling

a, Photographs of a subject wearing a ‘smart headband’ and a ‘smart wristband’ during stationary cycling. b, Comparison of ex situ calibration data of the sodium and glucose sensors from the collected sweat samples with the on-body readings of the FISA during the stationary cycling exercise detailed in f. c, d, Constant-load exercise at 150 W: power output, heart rate (in beats per minute, b.p.m.), oxygen consumption (V_O2) and pulmonary minute ventilation, as measured by external monitoring systems (c) and the real-time sweat analysis results of the FISA worn on a subject's forehead (d). e, f, Graded-load exercise, involving a dramatic power increase from 75 W to 200 W: power output, heart rate, V_O2 and pulmonary minute ventilation, as measured by external monitoring systems (e) and the real-time analysis results using the FISA worn on a different subject's forehead (f). Data collection for each sensor took place when a sufficient sweat sample was present (see Methods).

Figure 4. Hydration status analysis during group outdoor running using the FISAs

a, Schematic illustration showing the group outdoor running trial based on wearable FISAs (packaged as ‘smart headbands’). The data are transmitted to the user's cell phone and uploaded to cloud servers. b, c, Representative real-time sweat sodium (b) and potassium (c) levels during an endurance run with water intake. d, e, Representative real-time sweat sodium (d) and potassium (e) levels during an endurance run without water intake.