A Smart Wristband Integrated with an IoT-Based Alarming System for Real-Time Sweat Alcohol Monitoring

Abstract

Breathalyzer is a common approach to measuring blood alcohol concentration (BAC) levels of individuals suspected of drunk driving. Nevertheless, this device is relatively high-cost, inconvenient for people with limited breathing capacity, and risky for COVID-19 exposure. Here, we designed and developed a smart wristband integrating a real-time noninvasive sweat alcohol metal oxide (MOX) gas sensor with a Drunk Mate, an Internet of Thing (IoT)-based alarming system. A MOX sensor acquired transdermal alcohol concentration (TAC) which was converted to BAC and sent via the IoT network to the Blynk application platform on a smartphone, triggering alarming messages on LINE Notify. A user would receive an immediate alarming message when his BAC level reached an illegal alcohol concentration limit (BAC 50 mg%; TAC 0.70 mg/mL). The sensor readings showed a high linear correlation with TAC (R2 = 0.9815; limit of detection = 0.045 mg/mL) in the range of 0.10−1.05 mg/mL alcohol concentration in artificial sweat, achieving an accuracy of 94.66%. The sensor readings of ethanol in water were not statistically significantly different (p > 0.05) from the measurements in artificial sweat and other sweat-related solutions, suggesting that the device responded specifically to ethanol and was not affected by other electrolytes in the artificial sweat. Moreover, the device could continuously monitor TAC levels simulated in real-time in an artificial sweat testing system. With the integration of an IoT-based alarming system, the smart wristband developed from a commercial gas sensor presented here offers a promising low-cost MOX gas sensor monitoring technology for noninvasive and real-time sweat alcohol measurement and monitoring.

Keywords: Internet of Things (IoT); metal oxide (MOX) gas sensor; noninvasive sweat alcohol monitoring; wearable sensor.

Figure 1

Schematic illustration of the smart wristband for real-time alcohol monitoring. (a) The smart wristband being worn on a wrist for illustration; (b) Design layout of the sensing system with device enclosure; (c) Components used in the device electronic system; (d) A schematic diagram of the sensing system, data acquisition, and the IoT-based alarming system; (e) The isometric and cross-sectional views with a dimension of the device enclosure; (f) The user interfaces of Drunk Mate, consisting of Blynk IoT platform and the LINE Notify messaging platform for real-time alarming notification.

Figure 2

The illustration of an artificial sweat generating system (right) mimicking the human sweating system (left).

Figure 3

The sensitivity analysis (a) Sensor reading outputs in various ethanol concentrations in the range of 0.10–1.05 mg/mL recorded from the beginning until the system reached an equilibrium state (b) Correlation between the measured alcohol concentration versus the actual alcohol concentration in artificial sweat. Each data point represents mean ± standard deviation (n = 3).

Figure 4

The alcohol specificity analysis (a) Raw sensor reading output of DI, AS, 87 ppb acetone in AS, 0.42 mg/mL ethanol in artificial sweat, 87 ppb acetone, KCl, lactic acid, NaCl, and urea solutions. (b) Processed sensor output with t-test analysis, suggesting that the device specifically responded to ethanol only. “ns” means no significant difference (p > 0.05). Data are presented as mean and standard deviation (n = 3).

Figure 5

The comparison of measured (diagonal strip bars) and actual alcohol concentration (solid fill bars) from unknown samples with error percentages for sensor accuracy analysis. Data are presented as mean and standard deviation (n = 3).

Figure 6

Result of real-time sweat alcohol monitoring in the artificial sweat generating system in two drinking behaviors (a) A no time-gap behavior from one drink and two drinks (b) A 15-min interval drinking behavior in multiple drinks.