Quantitative Analysis of Mother Wavelet Function Selection for Wearable Sensors-Based Human Activity Recognition.

Recent advancements in the Internet of Things (IoT) wearable devices such as wearable inertial sensors have increased the demand for precise human activity recognition (HAR) with minimal computational resources. The wavelet transform, which offers excellent time-frequency localization characteristics, is well suited for HAR recognition systems. Selecting a mother wavelet function in wavelet analysis is critical, as optimal selection improves the recognition performance. The activity time signals data have different periodic patterns that can discriminate activities from each other. Therefore, selecting a mother wavelet function that closely resembles the shape of the recognized activity's sensor (inertial) signals significantly impacts recognition performance. This study uses an optimal mother wavelet selection method that combines wavelet packet transform with the energy-to-Shannon-entropy ratio and two classification algorithms: decision tree (DT) and support vector machines (SVM). We examined six different mother wavelet families with different numbers of vanishing points. Our experiments were performed on eight publicly available ADL datasets: MHEALTH, WISDM Activity Prediction, HARTH, HARsense, DaLiAc, PAMAP2, REALDISP, and HAR70+. The analysis demonstrated in this paper can be used as a guideline for optimal mother wavelet selection for human activity recognition.

Hybrid WT-CNN-GRU-based model for the estimation of reservoir water quality variables considering spatio-temporal features.

Water quality indicators (WQIs), such as chlorophyll-a (Chl-a) and dissolved oxygen (DO), are crucial for understanding and assessing the health of aquatic ecosystems. Precise prediction of these indicators is fundamental for the efficient administration of rivers, lakes, and reservoirs. This research utilized two unique DL algorithms-namely, convolutional neural network (CNNs) and gated recurrent units (GRUs)-alongside their amalgamation, CNN-GRU, to precisely gauge the concentration of these indicators within a reservoir. Moreover, to optimize the outcomes of the developed hybrid model, we considered the impact of a decomposition technique, specifically the wavelet transform (WT). In addition to these efforts, we created two distinct machine learning (ML) algorithms-namely, random forest (RF) and support vector regression (SVR)-to demonstrate the superior performance of deep learning algorithms over individual ML ones. We initially gathered WQIs from diverse locations and varying depths within the reservoir using an AAQ-RINKO device in the study area to achieve this. It is important to highlight that, despite utilizing diverse data-driven models in water quality estimation, a significant gap persists in the existing literature regarding implementing a comprehensive hybrid algorithm. This algorithm integrates the wavelet transform, convolutional neural network (CNN), and gated recurrent unit (GRU) methodologies to estimate WQIs accurately within a spatiotemporal framework. Subsequently, the effectiveness of the models that were developed was assessed utilizing various statistical metrics, encompassing the correlation coefficient (r), root mean square error (RMSE), mean absolute error (MAE), and Nash-Sutcliffe efficiency (NSE) throughout both the training and testing phases. The findings demonstrated that the WT-CNN-GRU model exhibited better performance in comparison with the other algorithms by 13% (SVR), 13% (RF), 9% (CNN), and 8% (GRU) when R-squared and DO were considered as evaluation indices and WQIs, respectively.

Mattress-Based Non-Influencing Sleep Apnea Monitoring System.

A mattress-type non-influencing sleep apnea monitoring system was designed to detect sleep apnea-hypopnea syndrome (SAHS). The pressure signals generated during sleep on the mattress were collected, and ballistocardiogram (BCG) and respiratory signals were extracted from the original signals. In the experiment, wavelet transform (WT) was used to reduce noise and decompose and reconstruct the signal to eliminate the influence of interference noise, which can directly and accurately separate the BCG signal and respiratory signal. In feature extraction, based on the five features commonly used in SAHS, an innovative respiratory waveform similarity feature was proposed in this work for the first time. In the SAHS detection, the binomial logistic regression was used to determine the sleep apnea symptoms in the signal segment. Simulation and experimental results showed that the device, algorithm, and system designed in this work were effective methods to detect, diagnose, and assist the diagnosis of SAHS.

Deep Learning with LPC and Wavelet Algorithms for Driving Fault Diagnosis.

Vehicle fault detection and diagnosis (VFDD) along with predictive maintenance (PdM) are indispensable for early diagnosis in order to prevent severe accidents due to mechanical malfunction in urban environments. This paper proposes an early voiceprint driving fault identification system using machine learning algorithms for classification. Previous studies have examined driving fault identification, but less attention has focused on using voiceprint features to locate corresponding faults. This research uses 43 different common vehicle mechanical malfunction condition voiceprint signals to construct the dataset. These datasets were filtered by linear predictive coefficient (LPC) and wavelet transform(WT). After the original voiceprint fault sounds were filtered and obtained the main fault characteristics, the deep neural network (DNN), convolutional neural network (CNN), and long short-term memory (LSTM) architectures are used for identification. The experimental results show that the accuracy of the CNN algorithm is the best for the LPC dataset. In addition, for the wavelet dataset, DNN has the best performance in terms of identification performance and training time. After cross-comparison of experimental results, the wavelet algorithm combined with DNN can improve the identification accuracy by up to 16.57% compared with other deep learning algorithms and reduce the model training time by up to 21.5% compared with other algorithms. Realizing the cross-comparison of recognition results through various machine learning methods, it is possible for the vehicle to proactively remind the driver of the real-time potential hazard of vehicle machinery failure.

Spatiotemporal multiscale ICA could invariantly extract task (motor) modes from wavelet subbands of fMRI data.

BACKGROUND AND OBJECTIVE: . Given a timeseries of task-evoked functional MRI (fMRI) images (4D spatiotemporal data), we can extract the task mode by statistical independent component analysis (ICA). If the 4D data are spatiotemporally decomposed into subbands (multiresolutions in both time and space), is ICA still capable of extracting the task modes at multiscales? We answer this question using the well-established fingertapping motor-task experiments at 3T and 7T. The positive answer informs that a brain task is spatiotemporal separable at ICA decomposition and shift invariant at multiscales during activation over a finite region. METHODS: . We collected a set of task fMRI datasets from sixteen subjects performing fingertapping at 3T and one single dataset from a different subject at 7T. For each 4D fMRI dataset, we first performed temporal wavelet transform (1D WT) at 3 levels using different wavelets (e.g. 'db1','db2', and 'sym4'), then extracted the task modes from the WT subbands via ICA (as called multi-timescale ICA). Meanwhile, we also performed task mode extraction by applying ICA to 3D spatial WT subbands (as called multi-spacescale ICA). Upon the multiscale ICA results, we identified the primary motor task modes in the motor cortex, in comparison to the raw fMRI data analysis (at level 0). RESULTS: . In the 7T experiment, the multiscale ICA across 3 timescale levels and 2 spacescale levels could extract the primary task modes at a tasktcorr of 0.90 and 0.86, respectively, compared to 0.87 for the ICA task extraction from raw data. In the 3T experiment, the multiscale could extract the primary task mode with 0.92 and 0.91, while the ICA task extraction from raw data was 0.91. CONCLUSION: . ICA could extract the primary motor task modes from wavelet-decomposed multi-timescale and multi-spacescale subbands, construing the broad spatial activation (extent >>voxel size) of the brain motor task performed in a long duration (>>TR). Our experimental results show the brain functional activity signal is spatiotemporal separable as well as shift invariant at multiscales in both time and space.

Groundwater level modeling framework by combining the wavelet transform with a long short-term memory data-driven model.

Developing models that can accurately simulate groundwater level is important for water resource management and aquifer protection. In particular, machine learning tools provide a new and promising approach to efficiently forecast long-term groundwater table fluctuations without the computational burden of building a detailed flow model. This study proposes a multistep modeling framework for simulating groundwater levels by combining the wavelet transform (WT) with the long short-term memory (LSTM) network; the framework is named the combined WT-multivariate LSTM (WT-MLSTM) method. First, the WT decomposes the groundwater level time series (i.e., the training stage) into a self-control term and a set of external-control terms. Second, Pearson correlation analysis reveals the correlations between the influencing factors (i.e., river stage) and the groundwater table, and the multivariate LSTM model incorporating external factors is built to simulate the external-control terms. Third, the spatiotemporal evolution of the groundwater level is modeled by reconstructing the sequence of each term of the groundwater level time series. Methodological applications in the Liangshui River Basin, Beijing, China and the Cibola National Wildlife Refuge along the lower Colorado River, United States, show that the combined WT-MLSTM model has a higher simulation accuracy than the standard LSTM, MLSTM, and WT-LSTM models. A comparison between the combined WT-MLSTM model and support vector machine (SVM) also demonstrates the advantage of the proposed model. Additional comparison between model forecasts and observed groundwater levels shows the model predictability for short-term time series. Further analysis reveals that the applicability of the combined WT-MLSTM model decreases with increasing distance between the groundwater well and adjacent river channel, or with the increasing complexity of the changing groundwater level patterns, which may be driven by additional controlling factors. This study therefore provides a new methodology/approach for the rapid and accurate simulation and prediction of groundwater level.

Motor Subtypes of Parkinson's Disease Can Be Identified by Frequency Component of Postural Stability.

Parkinson's disease (PD) can be divided into two subtypes based on clinical features-namely tremor dominant (TD) and postural instability and gait difficulty (PIGD). This categorization is important at the early stage of PD, since identifying the subtypes can help to predict the clinical progression of the disease. Accordingly, correctly diagnosing subtypes is critical in initiating appropriate early interventions and tracking the progression of the disease. However, as the disease progresses, it becomes increasingly difficult to further distinguish those attributes that are relevant to the subtypes. In this study, we investigated whether a method using the standing center of pressure (COP) time series data can separate two subtypes of PD by looking at the frequency component of COP (i.e., COP position and speed). Thirty-six participants diagnosed with PD were evaluated, with their bare feet on the force platform, and were instructed to stand upright with their arms by their sides for 20 s (with their eyes open and closed), which is consistent with the traditional COP measures. Fast Fourier transform (FFT) and wavelet transform (WT) were performed to distinguish between the motor subtypes using the COP measures. The TD group exhibited larger amplitudes at the frequency range of 3-7 Hz when compared to the PIGD group. Both the FFT and WT methods were able to differentiate the subtypes. COP time series information can be used to differentiate between the two motor subtypes of PD, using the frequency component of postural stability.

Odorant recognition using biological responses recorded in olfactory bulb of rats.

In this study we applied pattern recognition (PR) techniques to extract odorant information from local field potential (LFP) signals recorded in the olfactory bulb (OB) of rats subjected to different odorant stimuli. We claim that LFP signals registered on the OB, the first stage of olfactory processing, are stimulus specific in animals with normal sensory experience, and that these patterns correspond to the neural substrate likely required for perceptual discrimination. Thus, these signals can be used as input to an artificial odorant classification system with great success. In this paper we have designed and compared the performance of several configurations of artificial olfaction systems (AOS) based on the combination of four feature extraction (FE) methods (Principal Component Analysis (PCA), Fisher Transformation (FT), Sammon NonLinear Map (NLM) and Wavelet Transform (WT)), and three PR techniques (Linear Discriminant Analysis (LDA), Multilayer Perceptron (MLP) and Support Vector Machine (SVM)), when four different stimuli are presented to rats. The best results were reached when PCA extraction followed by SVM as classifier were used, obtaining a classification accuracy of over 95% for all four stimuli.

Independent component analysis and multiresolution asymmetry ratio for brain-computer interface.

This study proposes a brain-computer interface (BCI) system for the recognition of single-trial electroencephalogram (EEG) data. With the combination of independent component analysis (ICA) and multiresolution asymmetry ratio, a support vector machine (SVM) is used to classify left and right finger lifting or motor imagery. First, ICA and similarity measures are proposed to eliminate the electrooculography (EOG) artifacts automatically. The features are then extracted from the wavelet data by means of an asymmetry ratio. Finally, the SVM classifier is used to discriminate between the features. Compared to the EEG data without EOG artifact removal, band power, and adoptive autoregressive (AAR) parameter features, the proposed system achieves promising results in BCI applications.

Noise smoothing for structural vibration test signals using an improved wavelet thresholding technique.

In structural vibration tests, one of the main factors which disturb the reliability and accuracy of the results are the noise signals encountered. To overcome this deficiency, this paper presents a discrete wavelet transform (DWT) approach to denoise the measured signals. The denoising performance of DWT is discussed by several processing parameters, including the type of wavelet, decomposition level, thresholding method, and threshold selection rules. To overcome the disadvantages of the traditional hard- and soft-thresholding methods, an improved thresholding technique called the sigmoid function-based thresholding scheme is presented. The procedure is validated by using four benchmarks signals with three degrees of degradation as well as a real measured signal obtained from a three-story reinforced concrete scale model shaking table experiment. The performance of the proposed method is evaluated by computing the signal-to-noise ratio (SNR) and the root-mean-square error (RMSE) after denoising. Results reveal that the proposed method offers superior performance than the traditional methods no matter whether the signals have heavy or light noises embedded.

Análise comparativa de desempenho das transformadas Wavelet e Hilbert na detecção do QRS em ECG / Performance comparison analysis of Wavelet and Hilbert transforms for QRS detection in ECG

O processo de detecção do complexo QRS é o primeiro passo de um processo de extração de parâmetros do sinal eletrocardiograma (ECG) em sistemas de auxílio ao diagnóstico médico. O presente trabalho apresenta resultados detalhados de comparação da aplicação de duas transformadas matemáticas, Wavelet e Hilbert, em um algoritmo de detecção de QRS em termos de taxas de detecções corretas (sensibilidade e preditividade positiva) e de uma medida de frequência de recorrência a processos de filtragem (pré-processamento). Uma abordagem inovadora é implementada, na qual as rotinas de filtragem são inseridas dentro do estágio de decisão, ou seja, é realizada a supressão da etapa de pré-processamento. As transformadas são aplicadas no algoritmo, que é baseado em um limiar adaptativo, com o objetivo de realçar, apenas quando necessário, os picos (pontos fiduciais)do QRS. Em uma primeira abordagem, apenas a transformada Wavelet é utilizada neste realce e, numa segunda abordagem, a transformada de Hilbert é inserida em série à aplicação da Wavelet em dois possíveis arranjos. São realizados experimentos dos algoritmos sobre os exames da base de dados Arrhythmia Database, pertencente ao conjunto de bases de dados do MIT-BIH. É composta por 48 gravações de ECG com duração de trinta minutos, amostrados a uma frequência de 360 Hz com resolução de 4,88 μV sobre uma faixa de variação de 10 mV. Ao todo, contabilizam-se 109.662 complexos QRS. Taxas de 98,85% de sensibilidade e 95,10% de preditividade positiva são obtidas com a aplicação exclusiva da transformada Wavelet, enquanto que 98,89% de sensibilidade e 98,52% de preditividade positiva são obtidas com aaplicação em série das transformadas Wavelet e de Hilbert.

The process of QRS detection is the first stage of a greater process: the feature extraction in the electrocardiogram (ECG). This work presents detailed results on the performance of two mathematical transforms, Hilbert and Wavelet, which are applied in QRS detection. The evaluation parameters are the detection rates and a measure of frequency of recurrence to filtering processes. An innovative approach is implemented: the filtering routines are inserted in the decision stage, i.e. the preprocessing stage is removed. The algorithm is based on adaptive threshold technique and the two transforms are applied in order to emphasize, only when necessary, the QRS fiducial points. In a first approach, only the Wavelet transform is applied, and in a second approach, the Hilbert transform is inserted before the Wavelet transform or after it. We evaluate these approaches on the well-known MIT-BIH Arrhythmia Database. It contains 48 half-hour recordings of annotated ECG with a sampling rate of 360 Hz and 4.88 μV resolution over a 10 mV range, totalizing 109,662 QRS complexes. Sensitivity rates of 98.85% and 98.89% are respectively attained when the Wavelet transform is applied in the filtering processes and both Hilbert and Wavelet transforms are applied. Predictability rates of 95.10% and 98.52% are also attained respectively using Wavelet transform and the simultaneous application of Hilbert and Wavelet transforms in the filtering processes.