A microcontroller-based signal conditioning circuitry for acetone concentration detection using a metal oxide-based gas sensor

The presence of acetone gas in human exhaled breath is an important test for the detection of diabetes. Much research has been focused on developing a noninvasive technique for diabetes detection. Among the various gas sensors, semiconductor metal oxide-based gas sensors have garnered research interest because of advantages such as fabrication simplicity, low cost, and low power consumption. In this work, nickel oxide (NiO) is used as the sensing material for detecting acetone. The sensitivity of NiO towards acetone gas was analyzed using COMSOL Multiphysics and the results were compared with experimental data. For an acetone concentration in the range of 5–40 ppm, the results from the simulated sensor showed good agreement with the real sensor, with some anomalies due to the practical conditions in the real chamber and the simulation tool. Comprehensive finite element analysis of the sensor was performed for different operating temperatures and gas concentrations ranging from 1 to 40 ppm in order to analyze diverse diabetes mellitus conditions. Analytical modeling was also developed to analyze the effect of the change in acetone concentration on the resistance of the sensing layer. For a user-friendly interface, a signal conditioning unit and an alarm system for diabetes detection were also built. The COMSOL Multiphysics simulation was extended using a MATLAB script to integrate the results with the signal conditioning unit in the Proteus software program. The findings also demonstrated that this approach might be used to describe and determine the performance of acetone gas sensors prior to fabrication.