In recent years, wearable epidermal sweat sensors have received extensive attention owing to their great potential to provide personalized information on the health status of individuals at the molecular level. For on‐demand medical analysis of sweat in sedentary conditions, a cost‐effective wearable integrated platform combining sweat stimulation, sampling, transport, and analysis is highly desirable. In this work, a printed iontophoretic system integrated into a microfluidic sensing platform, which combines sweat induction, collection, and real‐time analysis of sweat‐ions into a single patch for on‐demand sweat monitoring on human subjects in stationary conditions is reported. The incorporation of microfluidics features facilitates sweat sampling, collection, and guiding through capillary effect. The multisensing sensor array exhibits sensitivity close to Nernstian behavior for sodium, potassium, and pH. The correlation between the concentrations of ions measured with the sweat patch and with ion chromatography analysis demonstrates the applicability of the system for real‐time point‐of‐care monitoring of the health status of individuals. Furthermore, the sweat patch electronic interface with wireless transmission enables real‐time data monitoring and storage over a cloud platform. This printed iontophoretic‐integrated fluidic sweat patch provides a cost‐effective solution for the on‐demand analysis of sweat components for healthcare applications.

The dysregulation of the hormone cortisol is related to several pathological states, and its monitoring could help prevent severe stress, fatigue, and mental diseases. While wearable antibody-based biosensors could allow real-time and simple monitoring of antigens, an accurate and low-cost antibody-based cortisol detection through electrochemical methods is considerably challenging due to its low concentration and the high ionic strength of real biofluids. Here, a label-free and fast sensor for cortisol detection is proposed based on antibody-coated organic electrochemical transistors. The developed devices show unprecedented high sensitivities of 50 μA/dec for cortisol sensing in high-ionic-strength solutions with effective cortisol detection demonstrated with real human sweat. The sensing mechanism is analyzed through impedance spectroscopy and confirmed with electrical models. Compared to existing methods requiring bulky and expensive laboratory equipment, these wearable devices enable point-of-care cortisol detection in 5 min with direct sweat collection for personalized well-being monitoring.

Improper hydration routines can reduce athletic performance. Recent studies show that data from noninvasive biomarker recordings can help to evaluate the hydration status of subjects during endurance exercise. These studies are usually carried out on multiple subjects. In this work, we present the first study on predicting hydration status using machine learning models from single-subject experiments, which involve 32 exercise sessions of constant moderate intensity performed with and without fluid intake. During exercise, we measured four noninvasive physiological and sweat biomarkers including heart rate, core temperature, sweat sodium concentration, and whole-body sweat rate. Sweat sodium concentration was measured from six body regions using absorbent patches. We used three machine learning models to determine the percentage of body weight loss as an indicator of dehydration with these biomarkers and compared the prediction accuracy. The results on this single subject show that these models gave similar mean absolute errors, while in general the nonlinear models slightly outperformed the linear model in most of the experiments. The prediction accuracy of using the whole-body sweat rate or heart rate was higher than using core temperature or sweat sodium concentration. In addition, the model trained on the sweat sodium concentration collected from the arms gave slightly better accuracy than from the other five body regions. This exploratory work paves the way for the use of these machine learning models to develop personalized health monitoring together with emerging, noninvasive wearable sensor devices.

Sweat secreted by the human eccrine sweat glands can provide valuable biomarker information during exercise in hot and humid conditions. Real-time noninvasive biomarker recordings are therefore useful for evaluating the physiological conditions of an athlete such as their hydration status during endurance exercise. In this work, we describe a platform that in-cludes different sweat biomonitoring prototypes of cost-effective, smart wearable devices for continuous biomonitoring of sweat during exercise. One prototype is based on conformable and disposable soft sensing patches with an integrated multisensor array requiring the integration of different sensors and printed sensors with their corresponding functionalization protocols on the same substrate. The second is based on silicon based sensors and paper microfluidics. Both platforms integrate a multi-sensor array for measuring sodium, potassium, and pH in sweat. We show preliminary results obtained from the multi-sensor prototypes placed on two athletes during exercise. We also show that the machine learning algorithms can predict the percentage of body weight loss during exercise from biomarkers such as heart rate and sweat sodium concentration collected over multiple subjects.

Much progress has been made in wearable sensors that provide real-time continuous physiological data from non-invasive measurements including heart rate and biofluids such as sweat. This information can potentially be used to identify the health condition of a person by applying machine learning algorithms on the physiological measurements. We present a person identification task that uses machine learning algorithms on a set of biomarkers collected from 30 subjects carrying out a cycling experiment. We compared an SVM and a gated recurrent neural network (RNN) for real-time accuracy using different window sizes of the measured data. Results show that using all biomarkers gave the best results from any of the models. With all biomarkers, the gated RNN model achieved 90% accuracy even in a 30 s time window; and 92.3% accuracy in a 150 s time window. Excluding any of the biomarkers leads to at least 7.4% absolute accuracy drop for the RNN model. The RNN implementation on the Jetson Nano incurs a low latency of 45 ms per inference.