Bayesian Inference using Spike Latency Codes for Quantification of Health Endangering Formaldehyde

Recently, the exposure to formaldehyde has appeared as a major concern since it is listed as a human carcinogen. Co nventional methods for its long-term monitoring are not feasible due to their high operational cost, long analysis time and the requirement of specialized equipment and staff. In this paper, we use an electronic nose, containing an array of commerciallyavailable Figaro gas sensors, to estimate formaldehyde concentr a ion. A hardware friendly bio-inspired spike latency coding scheme has recently been employed for gas classification by using re lative time between spikes. We use this scheme to estimate formalde hyde concentration by utilizing absolute spike timings. However, there is no straightforward relationship between the spike latency and the formaldehyde concentration. Instead, stochastic vari ability in the sensor array response, corresponding to repeated expos ure to the same formaldehyde concentration, implies that laten cy patterns of the sensor array encode probability distribution over the formaldehyde strength. We use a Bayesian inference appr oach to estimate the formaldehyde concentration, and its perfor mance is successfully validated by acquiring data for formaldehyde with our sensor array at twenty different concentrations in the laboratory environment. Keywords–Formaldehyde exposure; Sensor array; Spike latency coding; Bayesian inference.