[home] [Personal Program] [Help]

---

Session: MSK & Sweat sensing
Session starts: Friday 27 January, 11:30
Presentation starts: 12:30
Room: Room 530

---

Jelte Haakma (Technische Universiteit Eindhoven)
Elisabetta Peri (Technische Universiteit Eindhoven)
Simona Turco (Technische Universiteit Eindhoven)
Emma Moonen (Technische Universiteit Eindhoven)
Eduard Pelssers (Technische Universiteit Eindhoven)
Jaap Den Toonder (Technische Universiteit Eindhoven)
Massiom Mischi (Technische Universiteit Eindhoven)

Abstract:
Sepsis is a life-threatening condition that affects 30 million people every year [1]. Currently, sepsis monitoring is performed by measuring blood lactate levels via regularly repeated blood draws. Sweat is a relatively unexplored bio-fluid containing information that can provide broad insights into the metabolic activity of the human body [2], and that enables semi-continuous monitoring of biomarkers. In particular, sweat lactate may be a valuable biomarker for the detection of sepsis. The relationship between blood and sweat lactate levels, however, is debated. Previous studies suggested that the main contributor to the lactate in sweat is the lactate produced by the sweat gland metabolism itself. It is hypothesized that an estimation of the sweat rate per gland could be used to determine the amount of lactate produced by the sweat glands, thereby elucidating the relationship between blood and sweat lactate. However, currently, no accurate methods to determine the sweat rate per gland are available. We are developing a microfluidic sweat sensing patch that can determine the sweat rate per gland. To guide the design of the patch, we developed a deterministic model of the sweat generation process as well as the sweat sensing patch to simulate the signals produced by such a sweat patch. An analysis of the obtained signals should enable the estimation the number of active sweat glands as well as the sweat rate per gland. To make this possible, the analysis algorithm decomposes the signal into the contributions of the individual glands, based on the signature of each gland. Interpretation of the results shows that the predominant cause possible errors and uncertainties in determining the number of active glands and sweat rate per sweat gland relates to signals that cannot be decomposed uniquely. When only one gland is simulated, an error-rate of 0% is obtained, while larger numbers of simulated glands yield errors up to 10%. Future studies will include the simulation of different sweat rates, the reduction of the current error-rates by using a probabilistic approach and the addition of stochastic elements to the model to realize a more realistic simulation of the sweat gland activity. [1] Bauer et al. Critical care 2020 [2] Ghaffari et al. Sensors and Actuators B: Chemical 2021