Recent Progress in Smart Electronic Nose Technologies Enabled with Machine Learning Methods

Abstract

Machine learning methods enable the electronic nose (E-Nose) for precise odor identification with both qualitative and quantitative analysis. Advanced machine learning methods are crucial for the E-Nose to gain high performance and strengthen its capability in many applications, including robotics, food engineering, environment monitoring, and medical diagnosis. Recently, many machine learning techniques have been studied, developed, and integrated into feature extraction, modeling, and gas sensor drift compensation. The purpose of feature extraction is to keep robust pattern information in raw signals while removing redundancy and noise. With the extracted feature, a proper modeling method can effectively use the information for prediction. In addition, drift compensation is adopted to relieve the model accuracy degradation due to the gas sensor drifting. These recent advances have significantly promoted the prediction accuracy and stability of the E-Nose. This review is engaged to provide a summary of recent progress in advanced machine learning methods in E-Nose technologies and give an insight into new research directions in feature extraction, modeling, and sensor drift compensation.

Machine Learning-Assisted Volatile Organic Compound Gas Classification Based on Polarized Mixed-Potential Gas Sensors

“…To get important features as the criteria for the gas type or concentration, the artificial olfaction procedure needed feature extraction of gas-sensitive signals. Common feature extraction methods include the differential method, FastFourier transform, fitting function method, and so forth. , The polynomial ( f ( x ) = ∑ i = 0 4 a i · x i ) was utilized to fit the response recovery curve in this study, and the parameters of this function ( a 0 , a 1 , a 2 , a 3 , a 4 ) were used as output features. At the same time, the maximum values of the first-order and second-order derivatives of the response curve, the stable response value of the sensor, the response value at the middle time of the response, the maximum value of the response, and the amplitude of the response recovery curve after Fourier transform were extracted as a total of 12 features as the output features (Figure S5).…”

A Virtual Electronic Nose for the Efficient Classification and Quantification of Volatile Organic Compounds

“…This process is related to each gas or set of gases and the concentrations being studied and, therefore, needs to be customised on a case-by-case basis to find the parameters that best highlight the particularities in the sensor signals for the different exposures under evaluation. In general, many features are extracted, and, among these, the most relevant ones are considered for gas discrimination, classification, and/or quantification [ 34 , 35 , 43 , 44 , 45 ]. This process is called a top-down dimensionality reduction and is the second step we implemented in our data evaluation process.…”

“…Electronic noses appear to be promising candidates for selectivity enhancement, thanks to elaborated “fingerprint” patterns produced by a group of robust features that are unique for each gas and concentration studied [ 24 , 25 , 26 , 27 , 28 , 29 , 30 ]. By combining manifold features and applying machine learning (ML) techniques, it is possible to identify and differentiate between gases of interest, i.e., to enhance selectivity [ 31 , 32 , 33 , 34 , 35 ]. Unfortunately, despite being a potential pathway to overcoming selectivity problems, multi-sensor electronic noses suffer from high manufacturing costs, complexity, and high levels of power consumption.…”

Hybrid learning method based on feature clustering and scoring for enhanced COVID-19 breath analysis by an electronic nose

“…Despite these success stories, in the materials science community, MOS sensors also possess a cross-sensitivity drawback in detecting different types of VOCs and other gases. Basically, two optimization strategies can be opted, focusing on either the sensing active materials (e.g., employing hybrid organic–inorganic functional nanomaterials or molecular imprinting technique to increase the sensor selectivity [46] , [47] ) or post-processing of MOS sensor output signals by machine learning [48] , [49] , [50] . The latter approach has been favorable for commercial e-nose developers because no modification is needed in the hardware assembly and setup, which consequently lowers the product development cost.…”