Relating sensor responses of odorants to their organoleptic properties by means of a biologically-inspired model of receptor neuron convergence onto olfactory bulb

We present a neuromorphic approach to study the relationship between the response of a sensor/instrument to odorant molecules and the perceptual characteristics of the odors. Clearly, such correlations are only possible if the sensing instrument captures information about molecular properties (e.g., functional group, carbon chain-length) to which biological receptors have affinity. Given that information about some of these molecular features can be extracted from their infrared absorption spectra, an attractive candidate for this study is infrared (IR) spectroscopy. In our proposed model, high-dimensional IR absorption spectra of analytes are converted into compact, spatial odor maps using a feature clustering scheme that mimics the chemotopic convergence of receptor neurons onto the olfactory bulb. Cluster analysis of the generated IR odor maps reveals chemical groups with members that have similar perceptual characteristics e.g. fruits, nuts, etc. Further, the generated clusters match those obtained from a similar analysis of olfactory bulb odor maps obtained in rats for the same set of chemicals. Our results suggest that convergence mapping combined with IR absorption spectra may be an appropriate method to capture perceptual characteristics of certain classes of odorants.