Reliable e-nose for air toxicity monitoring by filter diagonalization method

This paper introduces a compact, affordable electronic nose (e-nose) device devoted to detect the presence of toxic compounds that could affect human health, such as carbon monoxide, combustible gas, hydrogen, methane, and smoke, among others. Such artificial olfaction device consists of an array of six metal oxide semiconductor (MOS) sensors and a computer-based information system for signal acquisition, processing, and visualization. This study further proposes the use of the filter diagonalization method (FDM) to extract the spectral contents of the signals obtained from the sensors. Preliminary results show that the prototype is functional and that the FDM approach is suitable for a later classification stage. Example deployment scenarios of the proposed e-nose include indoor facilities (buildings and warehouses), compromised air quality places (mines and sanitary landfills), public transportation, mobile robots, and wireless sensor networks.

Machine–learning-enabled metasurface for direction of arrival estimation

Metasurfaces, interacted with artificial intelligence, have now been motivating many contemporary research studies to revisit established fields, e.g., direction of arrival (DOA) estimation. Conventional DOA estimation techniques typically necessitate bulky-sized beam-scanning equipment for signal acquisition or complicated reconstruction algorithms for data postprocessing, making them ineffective for in-situ detection. In this article, we propose a machine-learning-enabled metasurface for DOA estimation. For certain incident signals, a tunable metasurface is controlled in sequence, generating a series of field intensities at the single receiving probe. The perceived data are subsequently processed by a pretrained random forest model to access the incident angle. As an illustrative example, we experimentally demonstrate a high-accuracy intelligent DOA estimation approach for a wide range of incident angles and achieve more than 95% accuracy with an error of less than 0.5 ° $0.5{\degree}$ . The reported strategy opens a feasible route for intelligent DOA detection in full space and wide band. Moreover, it will provide breakthrough inspiration for traditional applications incorporating time-saving and equipment-simplified majorization.

Identification and Modeling of College Students’ Psychological Stress Indicators for Deep Learning

Aiming at the problems of low accuracy of recognition results, long recognition time, and easy interference in traditional methods, a deep learning-oriented recognition modeling method of college students' psychological stress indicators is proposed. First, the ECG signal is collected by the ECG signal acquisition system, and the wavelet transform method is used to denoise the collected ECG signal. Then, the sequential backward selection algorithm is used to select the features of psychological stress indicators to reduce the feature dimension. Finally, based on the convolutional neural network in deep learning technology, a mental pressure indicator recognition model is established and the model parameters are optimized to realize the recognition of college students’ mental pressure indicators. Experimental results show that the method in this paper has high recognition accuracy, has high recognition efficiency, is not susceptible to interference, and has certain feasibility and effectiveness.

Design of sEMG Acquisition Circuit and Its Adaptive EEMD Denosing Research

The Surface electromyography (sEMG) signal is a kind of electrical signal which generated by human muscles during contraction. It is prone to being affected by noise because of its small amplitude, so it is necessary to remove the noise in its original signal with an appropriate algorithm. Based on the traditional signal denoising indicators, a new complex indicator r has been proposed in this paper which combines three different indicator parameters, that is, Signal to Noise Ratio (SNR), correlation coefficient (R), and standard error (SE). At the same time, an adaptive ensemble empirical mode decomposition (EEMD) method named AIO-EEMD which based on the proposed indicator is represented later. To verify the effective of the proposed algorithm, an electromyography signal acquisition circuit is designed firstly for collecting the original sEMG signal. Then, the denosing performance from the designed method is been compared with empirical mode decomposition (EMD) method and wavelet transform noise reduction method, respectively. The experiment results shown that the designed algorithm can not only automatically get the numbers of the reconstructed signal numbers, but also obtain the best reduction performance.

Development of fabric electrode for bio-potential signal acquisition in wearable health monitoring and effect of perspiration on signal acquisition

Development of a gel-free bio-potential electrode for the wearable health monitoring applications is a challenging goal. A conductive fabric electrode can replace the traditional conductive gel electrode. This paper describes the development of a conductive fabric electrode with regard to its potential use for electrocardiogram (ECG) acquisition. Since direct contact between the conductive fabric and human skin will be involved, an investigation on the effect of perspiration on the electrical conductivity of fabric is critical. Hence, the developed electrode was treated with alkaline (pH=8.0) and acidic (pH=4.3) perspiration for 3, 8 and 40 h to study the effect of perspiration on the conductivity and surface morphology. The acquired ECG signals were analysed with respect to morphology and frequency distribution. Conductivity tests were carried out on the perspiration-treated test electrodes by two probe method and surface resistivity meter. The ECG signals of volunteers were also recorded. The results showed a slight decrease in conductivity but without affecting the morphology and the quality of ECG signal. Leached silver content in the acid perspiration-treated solution was found to be 0.117 ppm as determined by Atomic absorption spectroscopy. The result shows that soft conducting textile materials can indeed be used as an electrode for ECG acquisition. This is a novel type of gel-free fabric electrode for long term wearable health monitoring applications including space application.

1H detection and dynamic nuclear polarization–enhanced NMR of Aβ1-42 fibrils

Several publications describing high-resolution structures of amyloid-β (Aβ) and other fibrils have demonstrated that magic-angle spinning (MAS) NMR spectroscopy is an ideal tool for studying amyloids at atomic resolution. Nonetheless, MAS NMR suffers from low sensitivity, requiring relatively large amounts of samples and extensive signal acquisition periods, which in turn limits the questions that can be addressed by atomic-level spectroscopic studies. Here, we show that these drawbacks are removed by utilizing two relatively recent additions to the repertoire of MAS NMR experiments—namely, 1H detection and dynamic nuclear polarization (DNP). We show resolved and sensitive two-dimensional (2D) and three-dimensional (3D) correlations obtained on 13C,15N-enriched, and fully protonated samples of M0Aβ1-42 fibrils by high-field 1H-detected NMR at 23.4 T and 18.8 T, and 13C-detected DNP MAS NMR at 18.8 T. These spectra enable nearly complete resonance assignment of the core of M0Aβ1-42 (K16-A42) using submilligram sample quantities, as well as the detection of numerous unambiguous internuclear proximities defining both the structure of the core and the arrangement of the different monomers. An estimate of the sensitivity of the two approaches indicates that the DNP experiments are currently ∼6.5 times more sensitive than 1H detection. These results suggest that 1H detection and DNP may be the spectroscopic approaches of choice for future studies of Aβ and other amyloid systems.

Efficient and Explainable Deep Neural Networks for Airway Symptom Detection in Support of Wearable Health Technology

Mobile health wearables are often embedded with small processors for signal acquisition and analysis. These embedded wearable systems are, however, limited with low available memory and computational power. Advances in machine learning, especially deep neural networks (DNNs), have been adopted for efficient and intelligent applications to overcome constrained computational environments. In this study, evolutionary optimized DNNs were analyzed to classify three common airway-related symptoms, namely coughs, throat clears and dry swallows. As opposed to typical microphone-acoustic signals, mechano-acoustic data signals, which did not contain identifiable speech information for better privacy protection, were acquired from laboratory-generated and publicly available datasets. The optimized DNNs had a low footprint of less than 150 kB and predicted airway symptoms of interests with 83.7% accuracy on unseen data. By performing explainable AI techniques, namely occlusion experiments and class activation maps, mel-frequency bands up to 8,000 Hz were found as the most important feature for the classification. We further found that DNN decisions were consistently relying on these specific features, fostering trust and transparency of proposed DNNs. Our proposed efficient and explainable DNN is expected to support edge computing on mechano-acoustic sensing wearables for remote, long-term monitoring of airway symptoms.

Modal Identification of Ultralow-Frequency Flexible Structures Based on Digital Image Correlation Method

Traditional modal testing has difficult accurately identifying the ultralow-frequency modes of flexible structures. Ultralow-frequency excitation and vibration signal acquisition are two main obstacles. Aiming at ultralow-frequency modal identification of flexible structures, a modal testing method based on Digital Image Correlation method and Eigensystem Realization Algorithm is proposed. Considering impulse and shaker excitation are difficult to make generate ultralow-frequency vibration of structures, the initial displacement is applied to the structure for excitation. The ultralow-frequency accelerometer always has a large mass, which will change the dynamics performance of the flexible structure, so a structural vibration response was obtained through the Digital Image Correlation method. After collecting the free-decay vibration signal, the ultralow-frequency mode of the structure was identified by using the Eigensystem Realization Algorithm. Ground modal tests were conducted to verify the proposed method. Firstly, a solar wing structure was adopted, from which it was concluded that the signal acquisition using Digital Image Correlation method had high feasibility and accuracy. Secondly, an ultralow-frequency flexible cantilever beam structure which had the theoretical solution was employed to verify the proposed method and the theoretical fundamental frequency of the structure was 0.185 Hz. Results show that the Digital Image Correlation method can effectively measure the response signal of the ultralow-frequency flexible structure, and obtain the dynamics characteristics.