AMGCN-L: an adaptive multi-time-window graph convolutional network with long-short-term memory for depression detection.

Objective.Depression is a common chronic mental disorder characterized by high rates of prevalence, recurrence, suicide, and disability as well as heavy disease burden. An accurate diagnosis of depression is a prerequisite for treatment. However, existing questionnaire-based diagnostic methods are limited by the innate subjectivity of medical practitioners and subjects. In the search for a more objective diagnostic methods for depression, researchers have recently started to use deep learning approaches.Approach.In this work, a deep-learning network, named adaptively multi-time-window graph convolutional network (GCN) with long-short-term memory (LSTM) (i.e. AMGCN-L), is proposed. This network can automatically categorize depressed and non-depressed people by testing for the existence of inherent brain functional connectivity and spatiotemporal features contained in electroencephalogram (EEG) signals. AMGCN-L is mainly composed of two sub-networks: the first sub-network is an adaptive multi-time-window graph generation block with which adjacency matrices that contain brain functional connectivity on different time periods are adaptively designed. The second sub-network consists of GCN and LSTM, which are used to fully extract the innate spatial and temporal features of EEG signals, respectively.Main results.Two public datasets, namely the patient repository for EEG data and computational tools, and the multi-modal open dataset for mental-disorder analysis, were used to test the performance of the proposed network; the depression recognition accuracies achieved in both datasets (using tenfold cross-validation) were 90.38% and 90.57%, respectively.Significance.This work demonstrates that GCN and LSTM have eminent effects on spatial and temporal feature extraction, respectively, suggesting that the exploration of brain connectivity and the exploitation of spatiotemporal features benefit the detection of depression. Moreover, the proposed method provides effective support and supplement for the detection of clinical depression and later treatment procedures.

Pleasantness Recognition Induced by Different Odor Concentrations Using Olfactory Electroencephalogram Signals.

Olfactory-induced emotion plays an important role in communication, decision-making, multimedia, and disorder treatment. Using electroencephalogram (EEG) technology, this paper focuses on (1) exploring the possibility of recognizing pleasantness induced by different concentrations of odors, (2) finding the EEG rhythm wave that is most suitable for the recognition of different odor concentrations, (3) analyzing recognition accuracies with concentration changes, and (4) selecting a suitable classifier for this classification task. To explore these issues, first, emotions induced by five different concentrations of rose or rotten odors are divided into five kinds of pleasantness by averaging subjective evaluation scores. Then, the power spectral density features of EEG signals and support vector machine (SVM) are used for classification tasks. Classification results on the EEG signals collected from 13 participants show that for pleasantness recognition induced by pleasant or disgusting odor concentrations, considerable average classification accuracies of 93.5% or 92.2% are obtained, respectively. The results indicate that (1) using EEG technology, pleasantness recognition induced by different odor concentrations is possible; (2) gamma frequency band outperformed other EEG rhythm-based frequency bands in terms of classification accuracy, and as the maximum frequency of the EEG spectrum increases, the pleasantness classification accuracy gradually increases; (3) for both rose and rotten odors, the highest concentration obtains the best classification accuracy, followed by the lowest concentration.

A novel channel selection scheme for olfactory EEG signal classification on Riemannian manifolds.

Objective.The classification of olfactory-induced electroencephalogram (olfactory EEG) signals has potential applications in disease diagnosis, emotion regulation, multimedia, and so on. To achieve high-precision classification, numerous EEG channels are usually used, but this also brings problems such as information redundancy, overfitting and high computational load. Consequently, channel selection is necessary to find and use the most effective channels.Approach.In this study, we proposed a multi-strategy fusion binary harmony search (MFBHS) algorithm and combined it with the Riemannian geometry classification framework to select the optimal channel sets for olfactory EEG signal classification. MFBHS was designed by simultaneously integrating three strategies into the binary harmony search algorithm, including an opposition-based learning strategy for generating high-quality initial population, an adaptive parameter strategy for improving search capability, and a bitwise operation strategy for maintaining population diversity. It performed channel selection directly on the covariance matrix of EEG signals, and used the number of selected channels and the classification accuracy computed by a Riemannian classifier to evaluate the newly generated subset of channels.Main results.With five different classification protocols designed based on two public olfactory EEG datasets, the performance of MFBHS was evaluated and compared with some state-of-the-art algorithms. Experimental results reveal that our method can minimize the number of channels while achieving high classification accuracy compatible with using all the channels. In addition, cross-subject generalization tests of MFBHS channel selection show that subject-independent channels obtained through training can be directly used on untrained subjects without greatly compromising classification accuracy.Significance.The proposed MFBHS algorithm is a practical technique for effective use of EEG channels in olfactory recognition.

[Application of PDCA circulation in reduction of the turnaround time of semen samples in the andrology laboratory].

OBJECTIVE: To reduce the out-of-threshold (OOT) value of the turnaround time (TAT) of semen samples in the andrology laboratory and improve the clinical diagnosis and patients' satisfaction. METHODS: We retrospectively analyzed the defect rate of TAT of semen samples in the andrology laboratory in the first two quarters of 2018. In the second two quarters, we made a table of countermeasures targeting the causes of the defects using plan-do-check-act (PDCA) circulation and the fishbone diagram drawn with the brainstorm method, followed by supervision of the implementation of the measures and observation of the changes in the OOT value of TAT of semen samples. RESULTS: The OOT rate of TAT of semen samples before seminal examination was significantly lower in the third and fourth than in the first and second quarters of 2018 (0.83% and 0.78% vs 3.43% and 2.07%, P < 0.01), and so was the total OOT rate of TAT (6.36% and 0.87% vs 7.00% and 7.15%, P < 0.01). The median of TAT of semen samples before computer assisted semen analysis was decreased from 22 min in the first to 17 min in the fourth quarter, and the 90th percentile from 54 min to 40 min. The median of total TAT in biochemical analysis was reduced from 387 min in the first to 315 min in the fourth quarter, and the 90th percentile from 1415 min to 1179 min. CONCLUSIONS: The application of PDCA circulation can significantly shorten the turnaround time of semen samples and improve the efficiency of diagnosis and treatment and quality control in the andrology laboratory.

Decoding olfactory EEG signals for different odor stimuli identification using wavelet-spatial domain feature.

BACKGROUND: Decoding olfactory-induced electroencephalography (olfactory EEG) signals has gained significant attention in recent years, owing to its potential applications in several fields, such as disease diagnosis, multimedia applications, and brain-computer interaction (BCI). Extracting discriminative features from olfactory EEG signals with low spatial resolution and poor signal-to-noise ratio is vital but challenging for improving decoding accuracy. NEW METHODS: By combining discrete wavelet transform (DWT) with one-versus-rest common spatial pattern (OVR-CSP), we develop a novel feature, named wavelet-spatial domain feature (WSDF), to decode the olfactory EEG signals. First, DWT is employed on EEG signals for multilevel wavelet decomposition. Next, the DWT coefficients obtained at a specific level are subjected to OVR-CSP for spatial filtering. Correspondingly, the variance is extracted to generate a discriminative feature set, labeled as WSDF. RESULTS: To verify the effectiveness of WSDF, a classification of olfactory EEG signals was conducted on two data sets, i.e., a public EEG dataset 'Odor Pleasantness Perception Dataset (OPPD)', and a self-collected dataset, by using support vector machine (SVM) trained based on different cross-validation methods. Experimental results showed that on OPPD dataset, the proposed method achieved a best average accuracy of 100% and 94.47% for the eyes-open and eyes-closed conditions, respectively. Moreover, on our own dataset, the proposed method gave a highest average accuracy of 99.50%. COMPARISON WITH EXISTING METHODS: Compared with a wide range of EEG features and existing works on the same dataset, our WSDF yielded superior classification performance. CONCLUSIONS: The proposed WSDF is a promising candidate for decoding olfactory EEG signals.

Odor-induced emotion recognition based on average frequency band division of EEG signals.

BACKGROUND: Emotion recognition plays a key role in multimedia. To enhance the sensation of reality, smell has been incorporated into multimedia systems because it can directly stimulate memories and trigger strong emotions. NEW METHOD: For the recognition of olfactory-induced emotions, this study explored a combination method using a support vector machine (SVM) with an average frequency band division (AFBD) method, where the AFBD method was proposed to extract the power-spectral-density (PSD) features from electroencephalogram (EEG) signals induced by smelling different odors. The so-called AFBD method means that each PSD feature was calculated based on equal frequency bandwidths, rather than the traditional EEG rhythm-based bandwidth. Thirteen odors were used to induce olfactory EEGs and their corresponding emotions. These emotions were then divided into two types of emotions, pleasure and disgust, or five types of emotions that were very unpleasant, slightly unpleasant, neutral, slightly pleasant, and very pleasant. RESULTS: Comparison between the proposed SVM plus AFBD method and other methods found average accuracies of 98.9 % and 88.5 % for two- and five-emotion recognition, respectively. These values were considerably higher than those of other combination methods, such as the combinations of AFBD or EEG rhythm-based features with naive Bayesian, k-nearest neighbor classification, voting-extreme learning machine, and backpropagation neural network methods. CONCLUSIONS: The SVM plus AFBD method represents a useful contribution to olfactory-induced emotion recognition. Classification of the five-emotion categories was generally inferior to the classification of the two-emotion categories, suggesting that the recognition performance decreased as the number of emotions in the category increased.

Latticed mode: A new control strategy for wind field simulation in a multiple-fan wind tunnel.

On the basis of the actively controlled multiple-fan wind tunnel we designed, this paper proposes a latticed mode-based control strategy for the fan array. The so-called latticed mode is the overall topology of the working fans. In order to investigate the simulation ability of the latticed mode, several latticed modes are designed to analyze the temporal and spatial properties of generated wind fields. Airflow data under different latticed modes are measured using nine two-dimensional anemometers, and then, statistical indicators of wind speed and wind direction, as well as the spatial uniformity and complexity of generated wind fields, are calculated to analyze the characteristics of wind records. The results indicate that the distribution of active fans in the latticed mode has significant influence on the statistical properties and spatial evolutions of wind speed and direction. Besides, the latticed modes can regulate the spatial uniformity and complexity of the wind fields. The latticed modes with a high clustering degree of active fans can generate wind fields with low spatial uniformity and high complexity. In addition to the fan voltage and the rotation angle of swivel plates, the proposed latticed mode provides more possibilities for wind field simulation.

Electronic nose using a bio-inspired neural network modeled on mammalian olfactory system for Chinese liquor classification.

The simplification of data processing is the frontier domain for electronic nose (e-nose) applications, whereas there are a lot of manual operations in a traditional processing procedure. To solve this problem, we propose a novel data processing method using the bio-inspired neural network modeled on the mammalian olfactory system. Through a neural coding scheme with multiple squared cosine receptive fields, continuous sensor data are simplified as the spike pattern in virtual receptor units. The biologically plausible olfactory bulb, which mimics the structure and function of main olfactory pathways, is designed to refine the olfactory information embedded in the encoded spikes. As a simplified presentation of cortical function, the bionic olfactory cortex is established to further analyze olfactory bulb's outputs and perform classification. The proposed method can automatically learn features without tedious steps such as denoising, feature extraction and reduction, which significantly simplifies the processing procedure for e-noses. To validate algorithm performance, comparison studies were performed for seven kinds of Chinese liquors using the proposed method and traditional data processing methods. The experimental results show that squared cosine receptive fields and the olfactory bulb model are crucial for improving classification performance, and the proposed method has higher classification rates than traditional methods when the sensor quantity and type are changed.

A gas source declaration scheme based on a tetrahedral sensor structure in three-dimensional airflow environments.

A gas source declaration scheme based on a tetrahedral sensor structure in three-dimensional airflow environments is proposed. First, a tetrahedral sensor structure was established. Based on the tetrahedral structure, the gas source declaration problem was converted into a two-class classification issue. Then a classification algorithm combining an extreme learning machine (ELM, a fast neural network classifier) with a gas mass flux criterion is proposed. A novel calculation method for the mass flux through a closed tetrahedral surface is presented, and a mass flux criterion was developed which acts as a training sample filter for the ELM. The source declaration scheme was validated by using both regular and irregular tetrahedron experiments.

Chemical Source Searching by Controlling a Wheeled Mobile Robot to Follow an Online Planned Route in Outdoor Field Environments.

In this paper, we present an estimation-based route planning (ERP) method for chemical source searching using a wheeled mobile robot and validate its effectiveness with outdoor field experiments. The ERP method plans a dynamic route for the robot to follow to search for a chemical source according to time-varying wind and an estimated chemical-patch path (C-PP), where C-PP is the historical trajectory of a chemical patch detected by the robot, and normally different from the chemical plume formed by the spatial distribution of all chemical patches previously released from the source. Owing to the limitations of normal gas sensors and actuation capability of ground mobile robots, it is quite hard for a single robot to directly trace the intermittent and rapidly swinging chemical plume resulting from the frequent and random changes of wind speed and direction in outdoor field environments. In these circumstances, tracking the C-PP originating from the chemical source back could help the robot approach the source. The proposed ERP method was tested in two different outdoor fields using a wheeled mobile robot. Experimental results indicate that the robot adapts to the time-varying airflow condition, arriving at the chemical source with an average success rate and approaching effectiveness of about 90% and 0.4~0.6, respectively.

A Wind Estimation Method with an Unmanned Rotorcraft for Environmental Monitoring Tasks.

Wind velocity (strength and direction) is an important parameter for unmanned aerial vehicle (UAV)-based environmental monitoring tasks. A novel wind velocity estimation method is proposed for rotorcrafts. Based on an extended state observer, this method derives the wind disturbance from rotors' speeds and rotorcraft's acceleration and position. Then the wind disturbance is scaled to calculate the airspeed vector, which is substituted into a wind triangle to obtain the wind velocity. Easy-to-implement methods for calculating the rotorcraft's thrust and drag coefficient are also proposed, which are important parameters to obtain the wind drag and the airspeed, respectively. Simulations and experiments using a quadrotor in both hovering and flight conditions have validated the proposed method.

A multiple-fan active control wind tunnel for outdoor wind speed and direction simulation.

This article presents a new type of active controlled multiple-fan wind tunnel. The wind tunnel consists of swivel plates and arrays of direct current fans, and the rotation speed of each fan and the shaft angle of each swivel plate can be controlled independently for simulating different kinds of outdoor wind fields. To measure the similarity between the simulated wind field and the outdoor wind field, wind speed and direction time series of two kinds of wind fields are recorded by nine two-dimensional ultrasonic anemometers, and then statistical properties of the wind signals in different time scales are analyzed based on the empirical mode decomposition. In addition, the complexity of wind speed and direction time series is also investigated using multiscale entropy and multivariate multiscale entropy. Results suggest that the simulated wind field in the multiple-fan wind tunnel has a high degree of similarity with the outdoor wind field.

Design of a portable electronic nose for real-fake detection of liquors.

Portability is a major issue that influences the practical application of electronic noses (e-noses). For liquors detection, an e-nose must preprocess the liquid samples (e.g., using evaporation and thermal desorption), which makes the portable design even more difficult. To realize convenient and rapid detection of liquors, we designed a portable e-nose platform that consists of hardware and software systems. The hardware system contains an evaporation/sampling module, a reaction module, a control/data acquisition and analysis module, and a power module. The software system provides a user-friendly interface and can achieve automatic sampling and data processing. This e-nose platform has been applied to the real-fake recognition of Chinese liquors. Through parameter optimization of a one-class support vector machine classifier, the error rate of the negative samples is greatly reduced, and the overall recognition accuracy is improved. The results validated the feasibility of the designed portable e-nose platform.

Stacked Sparse Auto-Encoders (SSAE) Based Electronic Nose for Chinese Liquors Classification.

This paper presents a stacked sparse auto-encoder (SSAE) based deep learning method for an electronic nose (e-nose) system to classify different brands of Chinese liquors. It is well known that preprocessing; feature extraction (generation and reduction) are necessary steps in traditional data-processing methods for e-noses. However, these steps are complicated and empirical because there is no uniform rule for choosing appropriate methods from many different options. The main advantage of SSAE is that it can automatically learn features from the original sensor data without the steps of preprocessing and feature extraction; which can greatly simplify data processing procedures for e-noses. To identify different brands of Chinese liquors; an SSAE based multi-layer back propagation neural network (BPNN) is constructed. Seven kinds of strong-flavor Chinese liquors were selected for a self-designed e-nose to test the performance of the proposed method. Experimental results show that the proposed method outperforms the traditional methods.

Learning to rapidly re-contact the lost plume in chemical plume tracing.

Maintaining contact between the robot and plume is significant in chemical plume tracing (CPT). In the time immediately following the loss of chemical detection during the process of CPT, Track-Out activities bias the robot heading relative to the upwind direction, expecting to rapidly re-contact the plume. To determine the bias angle used in the Track-Out activity, we propose an online instance-based reinforcement learning method, namely virtual trail following (VTF). In VTF, action-value is generalized from recently stored instances of successful Track-Out activities. We also propose a collaborative VTF (cVTF) method, in which multiple robots store their own instances, and learn from the stored instances, in the same database. The proposed VTF and cVTF methods are compared with biased upwind surge (BUS) method, in which all Track-Out activities utilize an offline optimized universal bias angle, in an indoor environment with three different airflow fields. With respect to our experimental conditions, VTF and cVTF show stronger adaptability to different airflow environments than BUS, and furthermore, cVTF yields higher success rates and time-efficiencies than VTF.

Distributed least-squares estimation of a remote chemical source via convex combination in wireless sensor networks.

This paper investigates the problem of locating a continuous chemical source using the concentration measurements provided by a wireless sensor network (WSN). Such a problem exists in various applications: eliminating explosives or drugs, detecting the leakage of noxious chemicals, etc. The limited power and bandwidth of WSNs have motivated collaborative in-network processing which is the focus of this paper. We propose a novel distributed least-squares estimation (DLSE) method to solve the chemical source localization (CSL) problem using a WSN. The DLSE method is realized by iteratively conducting convex combination of the locally estimated chemical source locations in a distributed manner. Performance assessments of our method are conducted using both simulations and real experiments. In the experiments, we propose a fitting method to identify both the release rate and the eddy diffusivity. The results show that the proposed DLSE method can overcome the negative interference of local minima and saddle points of the objective function, which would hinder the convergence of local search methods, especially in the case of locating a remote chemical source.

Adapting an ant colony metaphor for multi-robot chemical plume tracing.

We consider chemical plume tracing (CPT) in time-varying airflow environments using multiple mobile robots. The purpose of CPT is to approach a gas source with a previously unknown location in a given area. Therefore, the CPT could be considered as a dynamic optimization problem in continuous domains. The traditional ant colony optimization (ACO) algorithm has been successfully used for combinatorial optimization problems in discrete domains. To adapt the ant colony metaphor to the multi-robot CPT problem, the two-dimension continuous search area is discretized into grids and the virtual pheromone is updated according to both the gas concentration and wind information. To prevent the adapted ACO algorithm from being prematurely trapped in a local optimum, the upwind surge behavior is adopted by the robots with relatively higher gas concentration in order to explore more areas. The spiral surge (SS) algorithm is also examined for comparison. Experimental results using multiple real robots in two indoor natural ventilated airflow environments show that the proposed CPT method performs better than the SS algorithm. The simulation results for large-scale advection-diffusion plume environments show that the proposed method could also work in outdoor meandering plume environments.

Collective odor source estimation and search in time-variant airflow environments using mobile robots.

This paper addresses the collective odor source localization (OSL) problem in a time-varying airflow environment using mobile robots. A novel OSL methodology which combines odor-source probability estimation and multiple robots' search is proposed. The estimation phase consists of two steps: firstly, the separate probability-distribution map of odor source is estimated via Bayesian rules and fuzzy inference based on a single robot's detection events; secondly, the separate maps estimated by different robots at different times are fused into a combined map by way of distance based superposition. The multi-robot search behaviors are coordinated via a particle swarm optimization algorithm, where the estimated odor-source probability distribution is used to express the fitness functions. In the process of OSL, the estimation phase provides the prior knowledge for the searching while the searching verifies the estimation results, and both phases are implemented iteratively. The results of simulations for large-scale advection-diffusion plume environments and experiments using real robots in an indoor airflow environment validate the feasibility and robustness of the proposed OSL method.

Improvement in the accuracy of estimating the time-of-flight in an ultrasonic ranging system using multiple square-root unscented Kalman filters.

The square-root unscented Kalman filter (SRUKF) is applied to identify the shape parameters of an ultrasonic echo envelope. The SRUKF has better stability than the normal unscented Kalman filter (UKF) because the square-root of the error covariance matrix used in the SRUKF guarantees positive semidefiniteness. Considering the effect of the initial state on the convergence speed of filters, the multi-SRUKF is used to estimate the time-of-flight (TOF). Each SRUKF has a different initial state. The result estimated in a limited time with minimum mean square error is finally adopted. Simulation experiments for various couples of shape parameters and signal-to-noise ratios validate the improvement in the TOF accuracy. Real experiments using the echo signals of a SensComp 600 ultrasonic transducer show that the relative means and standard deviations of the TOF error obtained using the multi-SRUKF method are less than 0.2% and 0.15%, respectively.

Improvement of energy efficiency via spectrum optimization of excitation sequence for multichannel simultaneously triggered airborne sonar system.

Both the energy efficiency and correlation characteristics are important in airborne sonar systems to realize multichannel ultrasonic transducers working together. High energy efficiency can increase echo energy and measurement range, and sharp autocorrelation and flat cross correlation can help eliminate cross-talk among multichannel transducers. This paper addresses energy efficiency optimization under the premise that cross-talk between different sonar transducers can be avoided. The nondominated sorting genetic algorithm-II is applied to optimize both the spectrum and correlation characteristics of the excitation sequence. The central idea of the spectrum optimization is to distribute most of the energy of the excitation sequence within the frequency band of the sonar transducer; thus, less energy is filtered out by the transducers. Real experiments show that a sonar system consisting of eight-channel Polaroid 600 series electrostatic transducers excited with 2 ms optimized pulse-position-modulation sequences can work together without cross-talk and can measure distances up to 650 cm with maximal 1% relative error.