enhancement-method-surface-acoustic-wave-atomizer-efficiency-for

Enhancement Method of Surface Acoustic Wave-Atomizer Efficiency for Olfactory Display

Since olfaction is an important sense in human interfaces, we have developed an olfactory display using a surface acoustic wave (SAW) atomizer and ...

Tokyo Institute of Technology

Study of odor recorder for dynamical change of odor.

Chemical sensing in spatial/temporal domains.

Cooking up an interactive olfactory game display.

In this review, an odor sensing system and an olfactory display are introduced into people in pharmacy. An odor sensing system consists of an array of sensors with partially overlapping specificities and pattern recognition technique. One of examples of odor sensing systems is a halitosis sensor which quantifies the mixture composition of three volatile sulfide compounds. A halitosis sensor was realized using a preconcentrator to raise sensitivity and an electrochemical sensor array to suppress the influence of humidity. Partial least squares (PLS) method was used to quantify the mixture composition. The experiment reveals that the sufficient accuracy was obtained. Moreover, the olfactory display, which present scents to human noses, is explained. A multi-component olfactory display enables the presentation of a variety of smells. The two types of multi-component olfactory display are described. The first one uses many solenoid valves with high speed switching. The valve ON frequency determines the concentration of the corresponding odor component. The latter one consists of miniaturized liquid pumps and a surface acoustic wave (SAW) atomizer. It enables the wearable olfactory display without smell persistence. Finally, the application of the olfactory display is demonstrated. Virtual ice cream shop with scents was made as a content of interactive art. People can enjoy harmony among vision, audition and olfaction. In conclusion, both odor sensing system and olfactory display can contribute to the field of human health care.

[Odor sensing system and olfactory display].

The sense of smell arises from the perception of odors from chemicals. However, the relationship between the impression of odor and the numerous physicochemical parameters has yet to be understood owing to its complexity. As such, there is no established general method for predicting the impression of odor of a chemical only from its physicochemical properties. In this study, we designed a novel predictive model based on an artificial neural network with a deep structure for predicting odor impression utilizing the mass spectra of chemicals, and we conducted a series of computational analyses to evaluate its performance. Feature vectors extracted from the original high-dimensional space using two autoencoders equipped with both input and output layers in the model are used to build a mapping function from the feature space of mass spectra to the feature space of sensory data. The results of predictions obtained by the proposed new method have notable accuracy (R≅0.76) in comparison with a conventional method (R≅0.61). 

Odor Impression Prediction from Mass Spectra.

Recent studies on machine learning technology have reported successful performances in some visual and auditory recognition tasks, while little has been reported in the field of olfaction. In this paper we report computational methods to predict the odor impression of a chemical from its physicochemical properties. Our predictive model utilizes nonlinear dimensionality reduction on mass spectra data and performs the clustering of descriptors by natural language processing. Sensory evaluation is widely used to measure human impressions to smell or taste by using verbal descriptors, such as "spicy" and "sweet". However, as it requires significant amounts of time and human resources, a large-scale sensory evaluation test is difficult to perform. Our model successfully predicts a group of descriptors for a target chemical through a series of computer simulations. Although the training text data used in the language modeling is not specialized for olfaction, the experimental results show that our method is useful for analyzing sensory datasets. This is the first report to combine machine olfaction with natural language processing for odor character prediction.

Article	Year
Study of odor recorder for dynamical change of odor. Chemical senses\| PubMed ID: 15738143	2005
Chemical sensing in spatial/temporal domains. Chemical reviews\| PubMed ID: 18220426	2008
Cooking up an interactive olfactory game display. IEEE computer graphics and applications\| PubMed ID: 18240789	2008
[Odor sensing system and olfactory display]. Yakugaku zasshi : Journal of the Pharmaceutical Society of Japan\| PubMed ID: 24584010	2014
Odor Impression Prediction from Mass Spectra. PloS one\| PubMed ID: 27326765	2016
Predictive modeling for odor character of a chemical using machine learning combined with natural language processing. PloS one\| PubMed ID: 29902194	2018

Takamichi Nakamoto

Biography

JoVE Journal Publications

Other Publications