This paper provides a review of the most recent works in machine olfaction as applied to the identification of Chinese Herbal Medicines (CHMs). Due to the wide variety of CHMs, the complexity of growing sources and the diverse specifications of herb components, the quality control of CHMs is a challenging issue. Much research has demonstrated that an electronic nose (E-nose) as an advanced machine olfaction system, can overcome this challenge through identification of the complex odors of CHMs. E-nose technology, with better usability, high sensitivity, real-time detection and non-destructive features has shown better performance in comparison with other analytical techniques such as gas chromatography-mass spectrometry (GC-MS). Although there has been immense development of E-nose techniques in other applications, there are limited reports on the application of E-noses for the quality control of CHMs. The aim of current study is to review practical implementation and advantages of E-noses for robust and effective odor identification of CHMs. It covers the use of E-nose technology to study the effects of growing regions, identification methods, production procedures and storage time on CHMs. Moreover, the challenges and applications of E-nose for CHM identification are investigated. Based on the advancement in E-nose technology, odor may become a new quantitative index for quality control of CHMs and drug discovery. It was also found that more research could be done in the area of odor standardization and odor reproduction for remote sensing.

Indoor harmful gases are a considerable threat to the health of residents. In order to improve the accuracy of indoor harmful gas component identification, we propose an indoor toxic gas component analysis method that is based on the combination of bionic olfactory and convolutional neural network. This method uses the convolutional neural network’s ability to extract nonlinear features and identify each component of bionic oflactory respense signal. A comparison with the results of other methods verifies the improvement of recognition rate while with the same level of time cost, which proved the effectiveness of the proposed model. The experimental results showed that the recognition rate of different types and concentrations of harmful gas components reached 90.96% and it solved the problem of mutual interference between gases.

Machine olfaction is a novel technology and has been developed for many years. The electronic nose with an array of gas sensors, a crucial application form of the machine olfaction, is capable of sensing not only odorous compounds, but also odorless chemicals. Because of its fast response, mobility and easy of use, the electronic nose has been applied to scientific and commercial uses such as environment monitoring and food processing inspection. Additionally, odor characterization and reproduction are the two novel parts of machine olfaction, which extend the field of machine olfaction. Odor characterization is the technique that characterizes odorants as some form of general odor information. At present, there have already been odor characterizations by means of the electronic nose. Odor reproduction is the technique that re-produces an odor by some form of general odor information and displays the odor by the olfactory display. It enhances the human ability of controlling odors just as the control of light and voice. In analogy to visual and auditory display technologies, is it possible that the olfactory display will be used in our daily life? There have already been some efforts toward odor reproduction and olfactory displays.

Recently, various deep learning methods have been applied to Modulation Format Identification (MFI). The interpretability of deep learning models is important. However, this interpretability is challenged due to the black-box nature of deep learning. To deal with this difficulty, we propose an Adaptive Mask-Based Interpretable Convolutional Neural Network (AMI-CNN) that utilizes a mask structure for feature selection during neural network training and feeds the selected features into the classifier for decision making. During training, the masks are updated dynamically with parameters to optimize feature selection. The extracted mask serves as interpretable weights, with each weight corresponding to a feature, reflecting the contribution of each feature to the model’s decision. We validate the model on two datasets—Power Spectral Density (PSD) and constellation phase histogram—and compare it with three classical interpretable methods: Gradient-Weighted Class Activation Mapping (Grad-CAM), Local Interpretable Model-Agnostic Explanations (LIME), and Shapley Additive exPlanations (SHAP). The MSE values are as follows: AMI-CNN achieves the lowest MSE of 0.0246, followed by SHAP with 0.0547, LIME with 0.0775, and Grad-CAM with 0.1995. Additionally, AMI-CNN achieves the highest PG-Acc of 1, whether on PSD or on constellation phase histogram. Experimental results demonstrate that the AMI-CNN model outperforms compared methods in both qualitative and quantitative analyses.