<p>Major public health concerns include obesity, diabetes, and related metabolic disorders. These conditions impact physical health, cognitive function, and psychological well-being. Acetone in human breath serves as a key biomarker. It offers a non-invasive, painless, and cost-effective means of monitoring fat metabolism and early signs of diabetes. This study introduces a microelectromechanical systems (MEMS)-based gas sensor. It is developed using the standard Polysilicon Multi-User MEMS Process (PolyMUMPs) process, for sensitive and selective breath acetone detection. The sensor includes a suspended microstructure with a gold micro-heater and temperature sensor, actuated electrothermally to produce displacement. A nanostructured TiO₂ coating enhances acetone adsorption, increasing the sensor’s mass and altering its resonance. These changes are converted into electrical signals using a capacitive readout circuit (MS3110). Modeling and simulations were used to optimize performance, followed by experimental validation. The sensor detected acetone concentrations from 0.01 to 4 ppm, with a minimum detection limit of 44 ppb. Output voltage ranged from 0.176 to 0.295 V, showing a sensitivity of 0.029 V/ppm. It responded in about 90 seconds and recovered in 15 seconds. The sensor also showed strong selectivity against other VOCs, making it suitable for non-invasive breath analysis. Overall, the device offers a promising solution for portable, low-power acetone detection in applications like obesity screening, metabolic monitoring, and personalized healthcare.</p>