Identification of Urine Odour Using CMOS-Based Metal Oxide Resistive Gas Sensors

In this work, three custom-made metal oxide (MOX) sensors were used to identify gaseous headspace of synthetic urine from acetone, ammonia and ethyl acetate. The sensors have chemical coatings of WO
₃
, Pd/Pt doped SnO
₂
and CuO. Gases were trialled at 1 to 10 ppm for ammonia and ethyl acetate, 1 to 50 ppm for acetone and 5 to 100% for gaseous synthetic urine. A gas sensing unit was used for experiments and the collected data were processed in a two-layered pattern recognition model for gas classification. Five datasets were collected, four used for training and one for testing. The model has an overall accuracy of 91.1% and 100% for identifying synthetic urine.