CoChat

Regional and correlative sweat analysis using high-throughput microfluidic sensing patches toward decoding sweat

Open access

The DOI matches its Crossref record; 49 of the 49 references it cites were found in a registry or an index; 27 of the 56 sources its claims cite were checked for support.

Abstract

], [glucose], and sweat rate in exercise and chemically induced sweat. The patch is demonstrated for investigating regional sweat composition, predicting whole-body fluid/electrolyte loss during exercise, uncovering relationships between sweat metrics, and tracking glucose dynamics to explore sweat-to-blood correlations in healthy and diabetic individuals. By enabling a comprehensive sweat analysis, the presented device is a crucial tool for advancing sweat testing beyond the research stage for point-of-care medical and athletic applications.

Overview

CoChat Summary

## TL;DR

This paper presents a high-throughput, roll-to-roll fabricated wearable microfluidic sweat patch for simultaneous measurement of sweat analytes and sweat rate. Using it, the authors show that regional sweat patterns can help estimate whole-body fluid loss, that sweat sodium and potassium are generally positively correlated within individuals, and that iontophoretic sweat glucose does not robustly predict blood glucose across healthy and diabetic cohorts.

## Key findings

  • The patch was fabricated by R2R rotary screen printing and laser cutting, and the authors report production at 60 devices per minute on a 100-m web length.
  • Functional sensors showed calibration performance of 56.2 mV/decade for Na\+ (with a mean of 55.7 mV/decade, SD 1.1 mV/decade across 30 sensors), 51.3 mV/decade for K\+, and 1.0 nA/μM sensitivity for glucose over 50–200 μM.
  • For exercise sweat, sweat [Na\+] and secretion rate were typically positively correlated across regions and subjects; for subject 1, regional sweat rate vs whole-body fluid loss had r = 0.70 (forehead) and r = 0.68 (forearm), with P = 0.012 and 0.014, respectively.
  • In iontophoretic sweat, sweat [Na\+] and [K\+] were generally positively correlated within individuals; one subject’s off-body ICP-MS comparison gave Pearson’s r = 0.70.
  • Across 20 healthy and 28 diabetic subjects, average iontophoretic sweat glucose and blood glucose showed only a mild positive correlation (Pearson’s r = 0.3, P = 0.016); after cohort-wise outlier removal the correlations were 0.28 (healthy, P = 0.167) and 0.42 (diabetic, P = 0.016).
  • The authors also report that multiplying sweat glucose by sweat rate gave an even weaker relationship with blood glucose (Pearson’s coefficient = 0.2).

## Methods

  • The study combines wearable microfluidic sweat patch design/fabrication with on-body and off-body human sweat measurements. The patch integrates electrochemical sensors for Na\+, K\+, and glucose with an impedance-based sweat rate sensor.
  • Human trials included exercise (stationary cycling) and pilocarpine iontophoresis. On-body regional sites included the forehead, underarm, forearm, back, and leg; iontophoresis sweat was collected from the forearm and distal leg.
  • Main analyses included sensor calibration in standard solutions, ICP-MS for Na\+ and K\+, HPAEC-PAD / LC for sweat glucose validation, and correlation analyses between regional sweat variables, whole-body fluid loss, and blood glucose.

## Limitations

  • The paper states that a simple relation does not seem to exist across healthy and diabetic cohorts for predicting blood glucose from iontophoretic sweat glucose and sweat rate, limiting universal diabetes thresholds.
  • The authors explicitly note that their findings do not preclude more complex, individual-specific relations; this means the current analysis is limited by using simple population-level correlations, rather than longitudinal personalized models.
  • Evident from the text: some reported findings rely on small numbers of subjects in specific demonstrations (for example, three subjects in some real-time sweat experiments and one-subject examples for certain off-body validations), which limits generalizability.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1126/sciadv.aaw9906

Cites

49 of 49 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Schematics of R2R fabricated sweat sensing patches for regional and correlative sweat analysis. (A) R2R rotary screen printing of the wearable biosensing patch and (B) optical image of sensing electrode patterns printed via R2R processing are shown. (C) The biosensing patch is assembled by combining the sensing layer, microfluidic adhesive spacer, and PET cover sheet. (D) The assembled biosensing patch includes electrochemical sensors contained within the collection reservoir and electrical sweat rate sensor embedded in the microfluidic channel. (E) The patch, connected to printed circuit board (PCB), can be worn at various parts of the body for regional studies of the composition and dynamics of exercise and iontophoretic sweat. Photo credit: Antti Veijola (VTT) and Hnin Yin Yin Nyein (University of California, Berkeley).
Fig. 1 Schematics of R2R fabricated sweat sensing patches for regional and correlative sweat analysis. (A) R2R rotary screen printing of the wearable biosensing patch and (B) optical image of sensing electrode patterns printed via R2R processing are shown. (C) The biosensing patch is assembled by combining the sensing layer, microfluidic adhesive spacer, and PET cover sheet. (D) The assembled biosensing patch includes electrochemical sensors contained within the collection reservoir and electrical sweat rate sensor embedded in the microfluidic channel. (E) The patch, connected to printed circuit board (PCB), can be worn at various parts of the body for regional studies of the composition and dynamics of exercise and iontophoretic sweat. Photo credit: Antti Veijola (VTT) and Hnin Yin Yin Nyein (University of California, Berkeley).CC BY-NC · PubMed Central
Fig. 2 Characterization of the sweat patch sensors. General performance of (A) Na+, (B) K+, and (C) glucose sensors functionalized on printed Ag/C electrodes is shown. Measurements are paused when testing solution is changed to a new concentration, and only the last 2.5 min of data is displayed [transient portion in (C), as represented in fig. S2, is removed]. (D) Relationship between fluid volume contained in the microfluidic channel and admittance measured across the Ag sweat rate electrodes in 15 to 120 mM NaCl solutions is shown. (E) Calibration curves of 30 Na+ sensors, (F) histogram showing variation in the open-circuit potential of 30 Na+ sensors in 60 mM NaCl solution, (G) sensitivity variation of the 30 sensors, and (H) performance variation across 30 sweat rate sensors in 60 mM NaCl solution are presented.
Fig. 2 Characterization of the sweat patch sensors. General performance of (A) Na+, (B) K+, and (C) glucose sensors functionalized on printed Ag/C electrodes is shown. Measurements are paused when testing solution is changed to a new concentration, and only the last 2.5 min of data is displayed [transient portion in (C), as represented in fig. S2, is removed]. (D) Relationship between fluid volume contained in the microfluidic channel and admittance measured across the Ag sweat rate electrodes in 15 to 120 mM NaCl solutions is shown. (E) Calibration curves of 30 Na+ sensors, (F) histogram showing variation in the open-circuit potential of 30 Na+ sensors in 60 mM NaCl solution, (G) sensitivity variation of the 30 sensors, and (H) performance variation across 30 sweat rate sensors in 60 mM NaCl solution are presented.CC BY-NC · PubMed Central
Fig. 3 Regional analysis of exercise sweat during stationary cycling. (A) Biosensing patches are worn on the subject’s forehead, forearm, underarm, and back, and sweat parameters in these locations are monitored simultaneously. PCB is not attached to the sweat patch in the images. Multiplexed, regional exercise sweat measurements of [Na+] and secretion rate for subjects 1, 2, and 3 are displayed in (B), (C), and (D). Relationship between measured sweat [Na+] and sweat rate for subjects 1, 2, and 3 are plotted in (E), (F), and (G). (H) Relationships between whole-body fluid loss rate and regional sweat rate of subject 1 show strong positive relations on forehead and forearm. (I) Measured and predicted whole-body fluid loss rate of subject 1 at discrete times are close to 1:1. (J) Real-time prediction of whole-body fluid loss from regional forehead sweat rate is displayed. Predictions are made using the relationships obtained in (I). Photo credit: Bizen Maskey, Sunchon National University.
Fig. 3 Regional analysis of exercise sweat during stationary cycling. (A) Biosensing patches are worn on the subject’s forehead, forearm, underarm, and back, and sweat parameters in these locations are monitored simultaneously. PCB is not attached to the sweat patch in the images. Multiplexed, regional exercise sweat measurements of [Na+] and secretion rate for subjects 1, 2, and 3 are displayed in (B), (C), and (D). Relationship between measured sweat [Na+] and sweat rate for subjects 1, 2, and 3 are plotted in (E), (F), and (G). (H) Relationships between whole-body fluid loss rate and regional sweat rate of subject 1 show strong positive relations on forehead and forearm. (I) Measured and predicted whole-body fluid loss rate of subject 1 at discrete times are close to 1:1. (J) Real-time prediction of whole-body fluid loss from regional forehead sweat rate is displayed. Predictions are made using the relationships obtained in (I). Photo credit: Bizen Maskey, Sunchon National University.CC BY-NC · PubMed Central
Fig. 4 Regional analysis of iontophoretic sweat electrolytes and secretion rate. (A) Schematics showing iontophoretic sweat extraction and regional analysis of iontophoretic sweat electrolytes and sweat rate. Sensors are worn on the forearm and leg to measure dynamic [Na+], [K+], and sweat rate profiles of three subjects in (B), (C), and (D). Comparison of electrolyte concentrations and sweat rate show strong positive correlation for two of the three subjects, while [Na+] correlates strongly with [K+] for all subjects in (E to G). (H) Off-body analysis of collected iontophoresis sweat samples corroborates the strong relation between [Na+] and [K+]. (I) Comparison of [Na+] ranges of multiple subjects and (J) [K+] ranges of those subjects show that subjects with higher sodium levels often have lower potassium levels when comparing across the population. Photo credit: Bizen Maskey, Sunchon National University.
Fig. 4 Regional analysis of iontophoretic sweat electrolytes and secretion rate. (A) Schematics showing iontophoretic sweat extraction and regional analysis of iontophoretic sweat electrolytes and sweat rate. Sensors are worn on the forearm and leg to measure dynamic [Na+], [K+], and sweat rate profiles of three subjects in (B), (C), and (D). Comparison of electrolyte concentrations and sweat rate show strong positive correlation for two of the three subjects, while [Na+] correlates strongly with [K+] for all subjects in (E to G). (H) Off-body analysis of collected iontophoresis sweat samples corroborates the strong relation between [Na+] and [K+]. (I) Comparison of [Na+] ranges of multiple subjects and (J) [K+] ranges of those subjects show that subjects with higher sodium levels often have lower potassium levels when comparing across the population. Photo credit: Bizen Maskey, Sunchon National University.CC BY-NC · PubMed Central
Fig. 5 Regional analysis of iontophoretic sweat glucose and secretion rate in healthy and diabetic subjects. Real-time measurements of iontophoretic sweat [glucose] and sweat rate for (A) one healthy and (B) one diabetic subject repeated in two trials each show variation in sweat glucose response to changing blood glucose levels. (C) Off-body measurements of dynamic glucose profiles in the iontophoretic sweat of one subject across four trials show that sweat glucose levels do not consistently evolve with blood glucose. (D) Fasting levels of average sweat glucose versus blood glucose of 20 healthy and 28 diabetic subjects. (E) Average sweat glucose secretion rate versus blood glucose across the healthy and diabetic cohorts.
Fig. 5 Regional analysis of iontophoretic sweat glucose and secretion rate in healthy and diabetic subjects. Real-time measurements of iontophoretic sweat [glucose] and sweat rate for (A) one healthy and (B) one diabetic subject repeated in two trials each show variation in sweat glucose response to changing blood glucose levels. (C) Off-body measurements of dynamic glucose profiles in the iontophoretic sweat of one subject across four trials show that sweat glucose levels do not consistently evolve with blood glucose. (D) Fasting levels of average sweat glucose versus blood glucose of 20 healthy and 28 diabetic subjects. (E) Average sweat glucose secretion rate versus blood glucose across the healthy and diabetic cohorts.CC BY-NC · PubMed Central

Verification

Claims the paper makes

16 Valid · 1 Not valid · 6 Not checked · 90 Own results and interpretations

59 more claims · 1 with their cited sources checked

Create a free account to see all 113 claims and to verify the 1 reference still open.

See all claims — free account
  • Probably not a substantial claim. 18
  • It may state two separate claims as one. 36

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 49 references

Research works printed but not linked: 1

  1. L. B. Baker, Sweat testing methodology in the field: Challenges and best practices. Sports Sci. Exchange 28, 1–6 (2016).

Cited by

Metadata

Authors
Nyein, H. Y. Y., Bariya, M., Kivimäki, L., Uusitalo, S., Liaw, T. S., Jansson, E., Ahn, C. H., Hangasky, J. A., Zhao, J., Lin, Y., Happonen, T., Chao, M., Liedert, C., Zhao, Y., Tai, L.-C., Hiltunen, J., & Javey, A.
Affiliations
  • Lawrence Berkeley National Laboratory (US)
  • VTT Technical Research Centre of Finland (FI)
  • University of California, Berkeley (US)
  • QB3 (US)
Corresponding author
Javey, Ali
Venue
Science Advances
Published
2019-08-02
Volume
5
Issue
8
Pages
eaaw9906-eaaw9906
ISSN
2375-2548 (ISSN-L)
OpenAlex
W2966925761
Cited by
388
References
49
Citations by year
2026: 46 · 2025: 57 · 2024: 56 · 2023: 59 · 2022: 71 · 2021: 47 · 2020: 45 · 2019: 6
FWCI
14.87
Type
journal article
Language
en
Open access
Licence
cc-by-nc
Locations
  • Science Advances (publishedVersion, cc-by-nc)
  • PubMed Central (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
  • eScholarship (California Digital Library) (submittedVersion, cc-by-nc)
  • OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
  • eScholarship (California Digital Library) (submittedVersion)
  • Rare & Special e-Zone (The Hong Kong University of Science and Technology) (submittedVersion)
Funders
  • National Science Foundation Nanomanufacturing Systems for Mobile Computing and Mobile Energy Technologies
  • Berkeley Sensor and Actuator Center
  • Bakar fellowship
Topics
  • Physical Sciences › Engineering › Biomedical Engineering › Advanced Sensor and Energy Harvesting Materials
  • Physical Sciences › Chemistry › Analytical Chemistry › Electrochemical sensors and biosensors
  • Physical Sciences › Engineering › Biomedical Engineering › Microfluidic and Capillary Electrophoresis Applications
Keywords
sweat analysis, high-throughput sensing, sweat glucose, sweat rate, diabetes mellitus, point-of-care testing, athletic performance monitoring
MeSH terms
Diabetes Mellitus, Glucose, Humans, Ions, Potassium, Sodium, Sweat, Biosensing Techniques, Microfluidics, High-Throughput Screening Assays

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.