RESUMO
Objective.Due to the difficulty in acquiring motor imagery electroencephalography (MI-EEG) data and ensuring its quality, insufficient training data often leads to overfitting and inadequate generalization capabilities of deep learning-based classification networks. Therefore, we propose a novel data augmentation method and deep learning classification model to enhance the decoding performance of MI-EEG further.Approach.The raw EEG signals were transformed into the time-frequency maps as the input to the model by continuous wavelet transform. An improved Wasserstein generative adversarial network with gradient penalty data augmentation method was proposed, effectively expanding the dataset used for model training. Additionally, a concise and efficient deep learning model was designed to improve decoding performance further.Main results.It has been demonstrated through validation by multiple data evaluation methods that the proposed generative network can generate more realistic data. Experimental results on the BCI Competition IV 2a and 2b datasets and the actual collected dataset show that classification accuracies are 83.4%, 89.1% and 73.3%, and Kappa values are 0.779, 0.782 and 0.644, respectively. The results indicate that the proposed model outperforms state-of-the-art methods.Significance.Experimental results demonstrate that this method effectively enhances MI-EEG data, mitigates overfitting in classification networks, improves MI classification accuracy, and holds positive implications for MI tasks.
Assuntos
Interfaces Cérebro-Computador , Eletroencefalografia , Imaginação , Redes Neurais de Computação , Eletroencefalografia/métodos , Eletroencefalografia/classificação , Humanos , Imaginação/fisiologia , Aprendizado Profundo , Análise de OndaletasRESUMO
Early fault detection and diagnosis of grid-connected photovoltaic systems (GCPS) is imperative to improve their performance and reliability. Low-cost edge devices have emerged as innovative solutions for real-time monitoring, reducing latency, and improving response times. In this work, a lightweight Convolutional Neural Network (CNN) is designed and fine-tuned using Energy Valley Optimizer (EVO) for fault diagnosis. The CNN input consists of two-dimensional scalograms generated using Continuous Wavelet Transform (CWT). The proposed diagnosis technique demonstrated superior performance compared to benchmark architectures, namely MobileNet, NASNetMobile, and InceptionV3, achieving higher test accuracies and lower losses on binary and multi-fault classification tasks on balanced, unbalanced, and noisy datasets. Further, a quantitative comparison is conducted with similar recent studies. The obtained results indicate good performance and high reliability of the proposed fault diagnosis method.
RESUMO
In response to the issues of low model recognition accuracy and weak generalization in mechanical equipment fault diagnosis due to scarce data, this paper proposes an innovative solution, a cross-device secondary transfer-learning method based on EGRUN (efficient gated recurrent unit network). This method utilizes continuous wavelet transform (CWT) to transform source domain data into images. The EGRUN model is initially trained, and shallow layer weights are frozen. Subsequently, random overlapping sampling is applied to the target domain data to enhance data and perform secondary transfer learning. The experimental results demonstrate that this method not only significantly improves the model's ability to learn fault features but also enhances its classification accuracy and generalization performance. Compared to current state-of-the-art algorithms, the model proposed in this study shows faster convergence speed, higher diagnostic accuracy, and superior robustness and generalization, providing an effective approach to address the challenges arising from scarce data and varying operating conditions in practical engineering scenarios.
RESUMO
Human Activity Recognition (HAR), alongside Ambient Assisted Living (AAL), are integral components of smart homes, sports, surveillance, and investigation activities. To recognize daily activities, researchers are focusing on lightweight, cost-effective, wearable sensor-based technologies as traditional vision-based technologies lack elderly privacy, a fundamental right of every human. However, it is challenging to extract potential features from 1D multi-sensor data. Thus, this research focuses on extracting distinguishable patterns and deep features from spectral images by time-frequency-domain analysis of 1D multi-sensor data. Wearable sensor data, particularly accelerator and gyroscope data, act as input signals of different daily activities, and provide potential information using time-frequency analysis. This potential time series information is mapped into spectral images through a process called use of 'scalograms', derived from the continuous wavelet transform. The deep activity features are extracted from the activity image using deep learning models such as CNN, MobileNetV3, ResNet, and GoogleNet and subsequently classified using a conventional classifier. To validate the proposed model, SisFall and PAMAP2 benchmark datasets are used. Based on the experimental results, this proposed model shows the optimal performance for activity recognition obtaining an accuracy of 98.4% for SisFall and 98.1% for PAMAP2, using Morlet as the mother wavelet with ResNet-101 and a softmax classifier, and outperforms state-of-the-art algorithms.
Assuntos
Atividades Humanas , Análise de Ondaletas , Humanos , Atividades Humanas/classificação , Algoritmos , Aprendizado Profundo , Dispositivos Eletrônicos Vestíveis , Atividades Cotidianas , Redes Neurais de Computação , Processamento de Imagem Assistida por Computador/métodosRESUMO
This study proposes an innovative expert system that uses exclusively EEG signals to diagnose schizophrenia in its early stages. For diagnosing psychiatric/neurological disorders, electroencephalogram (EEG) testing is considered a financially viable, safe, and reliable alternative. Using the reconstructed phase space (RPS) and the continuous wavelet transform, the researchers maximized the differences between the EEG nonstationary signals of normal and schizophrenia individuals, which cannot be observed in the time, frequency, or time-frequency domains. This reveals significant information, highlighting more distinguishable features. Then, a deep learning network was trained to enhance the accuracy of the resulting image classification. The algorithm's efficacy was confirmed through three distinct methods: employing 70% of the dataset for training, 15% for validation, and the remaining 15% for testing. This was followed by a 5-fold cross-validation technique and a leave-one-out classification approach. Each method was iterated 100 times to ascertain the algorithm's robustness. The performance metrics derived from these tests - accuracy, precision, sensitivity, F1 score, Matthews correlation coefficient, and Kappa - indicated remarkable outcomes. The algorithm demonstrated steady performance across all evaluation strategies, underscoring its relevance and reliability. The outcomes validate the system's accuracy, precision, sensitivity, and robustness by showcasing its capability to autonomously differentiate individuals diagnosed with schizophrenia from those in a state of normal health.
Assuntos
Aprendizado Profundo , Eletroencefalografia , Esquizofrenia , Análise de Ondaletas , Esquizofrenia/diagnóstico , Esquizofrenia/fisiopatologia , Humanos , Eletroencefalografia/métodos , Algoritmos , Adulto , Reprodutibilidade dos Testes , Processamento de Sinais Assistido por Computador , Redes Neurais de ComputaçãoRESUMO
BACKGROUND: Autism Spectrum Disorder (ASD) is a condition with social interaction, communication, and behavioral difficulties. Diagnostic methods mostly rely on subjective evaluations and can lack objectivity. In this research Machine learning (ML) and deep learning (DL) techniques are used to enhance ASD classification. OBJECTIVE: This study focuses on improving ASD and TD classification accuracy with a minimal number of EEG channels. ML and DL models are used with EEG data, including Mu Rhythm from the Sensory Motor Cortex (SMC) for classification. METHODS: Non-linear features in time and frequency domains are extracted and ML models are applied for classification. The EEG 1D data is transformed into images using Independent Component Analysis-Second Order Blind Identification (ICA-SOBI), Spectrogram, and Continuous Wavelet Transform (CWT). RESULTS: Stacking Classifier employed with non-linear features yields precision, recall, F1-score, and accuracy rates of 78%, 79%, 78%, and 78% respectively. Including entropy and fuzzy entropy features further improves accuracy to 81.4%. In addition, DL models, employing SOBI, CWT, and spectrogram plots, achieve precision, recall, F1-score, and accuracy of 75%, 75%, 74%, and 75% respectively. The hybrid model, which combined deep learning features from spectrogram and CWT with machine learning, exhibits prominent improvement, attained precision, recall, F1-score, and accuracy of 94%, 94%, 94%, and 94% respectively. Incorporating entropy and fuzzy entropy features further improved the accuracy to 96.9%. CONCLUSIONS: This study underscores the potential of ML and DL techniques in improving the classification of ASD and TD individuals, particularly when utilizing a minimal set of EEG channels.
RESUMO
The present work describes a developed analytical method based on a colorimetric assay using gold nanoparticles (AuNPs) along with chemometric techniques for the simultaneous estimation of sofosbuvir (SOF) and ledipasvir (LED) in their synthetic mixtures and tablet dosage form. The applied chemometric approaches were continuous wavelet transform (CWT) and least squares support vector machine (LS-SVM). Characterization of AuNPs and AuNPs in combination with the drug was performed by UV-vis spectrophotometer, transmission electron microscopy (TEM), dynamic light scattering (DLS), and Fourier transform infrared (FTIR) spectroscopy. In the CWT method, the zero amplitudes were determined at 427â¯nm with Daubechies wavelet family for SOF (zero crossing point of LED) and 440â¯nm with Symlet wavelet family for LED (zero crossing point of SOF) over the concentration range of 7.5-90.0⯵g/L and 40.0-100.0⯵g/L with coefficients of determination (R2) of 0.9974 and 0.9907 for SOF and LED, respectively. The limit of detection (LOD) and limit of quantification (LOQ) of this method were found to be 7.92, 9.96⯵g/L and 12.02, 30.2⯵g/L for SOF and LED, respectively. In the LS-SVM model, the mean percentage recovery of SOF and LED in synthetic mixtures was 98.29â¯% and 99.25â¯% with root mean square error of 2.392 and 1.034, which were obtained by the optimization of regularization parameter (γ) and width of the function (σ) based on the cross-validation method. The proposed methods were also applied for the determination concentration of SOF and LED in the combined dosage form, recoveries were higher than 95â¯%, and relative standard deviation (RSD) values were lower than 0.4â¯%. The achieved results were statistically compared with those obtained from the high-performance liquid chromatography (HPLC) technique for the concurrent estimation of components through one-way analysis of variance (ANOVA), and no significant difference was found between the suggested approaches and the reference one. According to these results, simplicity, high speed, lack of time-consuming process, and cost savings are considerable benefits of colorimetry along with chemometrics methods compared to other ways.
Assuntos
Antivirais , Benzimidazóis , Colorimetria , Fluorenos , Ouro , Nanopartículas Metálicas , Sofosbuvir , Ressonância de Plasmônio de Superfície , Nanopartículas Metálicas/química , Ouro/química , Colorimetria/métodos , Antivirais/análise , Antivirais/química , Cromatografia Líquida de Alta Pressão/métodos , Sofosbuvir/análise , Sofosbuvir/química , Benzimidazóis/análise , Benzimidazóis/química , Fluorenos/análise , Fluorenos/química , Ressonância de Plasmônio de Superfície/métodos , Limite de Detecção , Comprimidos , Máquina de Vetores de Suporte , Quimiometria/métodos , Combinação de Medicamentos , Análise dos Mínimos Quadrados , Reprodutibilidade dos Testes , Hepacivirus/efeitos dos fármacos , Espectroscopia de Infravermelho com Transformada de Fourier/métodosRESUMO
In this study, the spectrophotometric method integrated with continuous wavelet transform (CWT) and coupled discrete wavelet transform (DWT) with fuzzy inference system (FIS) was developed for the simultaneous determination of ethinyl estradiol (EE) and drospirenone (DP) in combined oral contraceptives (COCs). The CWT approach was performed in the linearity range of 0.6-6 µg/mL for EE and 0.9 to 18 µg/mL for DP. Biorthogonal with an order of 1.3 (bior1.3) at a wavelength of 216 nm and Daubechies with an order of 2 (db2) at a wavelength of 278 nm were selected as the best wavelet families for obtaining the best zero crossing point for EE and DP, respectively. The limit of detection (LOD) of 0.7677 and 0.3222 µg/mL and the limit of quantification (LOQ) of 2.326 and 0.9765 µg/mL were obtained for EE and DP, respectively. The mean recovery of 103.24% and 99.77%, as well as root mean square error (RMSE) of 0.1896 and 0.1969, were found for EE and DP, respectively. In the DWT, the absorption of the mixtures was decomposed using different wavelets named db4, db2, Symlet2 (sym2), and bior1.3. Each of the wavelet outputs was dimension reduced by the principal component analysis (PCA) method and considered as FIS input. The wavelet of db4 with the coefficient of determination (R2) of 0.9979, RMSE of 0.0968, and mean recovery of 100.63% was chosen as the best one for the EE, while bior1.3 with R2 of 0.9955, RMSE of 0.4055, and mean recovery of 101.93% was selected for DP. These methods were successfully used to analyze the EE and DP simultaneously in tablet pharmaceutical formulation without any separation step. The suggested methods were compared with a reference method (HPLC) using analysis of variance (ANOVA) at a 95% confidence level, and no significant difference was observed in terms of accuracy. The suggested chemometric methods are reliable, rapid, and inexpensive, and can be used as an environmentally friendly alternative to HPLC for the simultaneous estimation of the mentioned drugs in commercial pharmaceutical products.
Assuntos
Androstenos , Anticoncepcionais Orais Combinados , Etinilestradiol , Lógica Fuzzy , Limite de Detecção , Análise de Componente Principal , Análise de Ondaletas , Etinilestradiol/análise , Androstenos/análise , Anticoncepcionais Orais Combinados/análise , HumanosRESUMO
Respiratory diseases are among the leading causes of death, with many individuals in a population frequently affected by various types of pulmonary disorders. Early diagnosis and patient monitoring (traditionally involving lung auscultation) are essential for the effective management of respiratory diseases. However, the interpretation of lung sounds is a subjective and labor-intensive process that demands considerable medical expertise, and there is a good chance of misclassification. To address this problem, we propose a hybrid deep learning technique that incorporates signal processing techniques. Parallel transformation is applied to adventitious respiratory sounds, transforming lung sound signals into two distinct time-frequency scalograms: the continuous wavelet transform and the mel spectrogram. Furthermore, parallel convolutional autoencoders are employed to extract features from scalograms, and the resulting latent space features are fused into a hybrid feature pool. Finally, leveraging a long short-term memory model, a feature from the latent space is used as input for classifying various types of respiratory diseases. Our work is evaluated using the ICBHI-2017 lung sound dataset. The experimental findings indicate that our proposed method achieves promising predictive performance, with average values for accuracy, sensitivity, specificity, and F1-score of 94.16%, 89.56%, 99.10%, and 89.56%, respectively, for eight-class respiratory diseases; 79.61%, 78.55%, 92.49%, and 78.67%, respectively, for four-class diseases; and 85.61%, 83.44%, 83.44%, and 84.21%, respectively, for binary-class (normal vs. abnormal) lung sounds.
RESUMO
We utilize Laser Speckle Contrast Imaging (LSCI) for visualizing cerebral blood flow in mice during and post-cardiac arrest. Analyzing LSCI images, we noted temporal blood flow variations across the brain surface for hours postmortem. Fast Fourier Transform (FFT) analysis depicted blood flow and microcirculation decay post-death. Continuous Wavelet Transform (CWT) identified potential cerebral hemodynamic synchronization patterns. Additionally, non-negative matrix factorization (NMF) with four components segmented LSCI images, revealing structural subcomponent alterations over time. This integrated approach of LSCI, FFT, CWT, and NMF offers a comprehensive tool for studying cerebral blood flow dynamics, metaphorically capturing the 'end of the tunnel' experience. Results showed primary postmortem hemodynamic activity in the olfactory bulbs, followed by blood microflow relocations between somatosensory and visual cortical regions via the superior sagittal sinus. This method opens new avenues for exploring these phenomena, potentially linking neuroscientific insights with mysteries surrounding consciousness and perception at life's end.
Assuntos
Encéfalo , Hemodinâmica , Animais , Camundongos , Encéfalo/irrigação sanguínea , Encéfalo/diagnóstico por imagem , Circulação Cerebrovascular , Imagem de Contraste de Manchas a Laser , Masculino , AutopsiaRESUMO
The identification of mixed solutions is a challenging and important subject in chemical analysis. In this paper, we propose a novel workflow that enables rapid qualitative and quantitative detection of mixed solutions. We use a methanol-ethanol mixed solution as an example to demonstrate the superiority of this workflow. The workflow includes the following steps: (1) converting Raman spectra into Raman images through CWT; (2) using MobileNetV3 as the backbone network, improved multi-label and multi-channel synchronization enables simultaneous prediction of multiple mixture concentrations; and (3) using transfer learning and multi-stage training strategies for training to achieve accurate quantitative analysis. We compare six traditional machine learning algorithms and two deep learning models to evaluate the performance of our new method. The experimental results show that our model has achieved good prediction results when predicting the concentration of methanol and ethanol, and the coefficient of determination R2 is greater than 0.999. At different concentrations, both MAPE and RSD outperform other models, which demonstrates that our workflow has outstanding analytical capabilities. Importantly, we have solved the problem that current quantitative analysis algorithms for Raman spectroscopy are almost unable to accurately predict the concentration of multiple substances simultaneously. In conclusion, it is foreseeable that this non-destructive, automated, and highly accurate workflow can further advance Raman spectroscopy.
RESUMO
Thin copper plate is widely used in architecture, transportation, heavy equipment, and integrated circuit substrates due to its unique properties. However, it is challenging to identify surface defects in copper strips arising from various manufacturing stages without direct contact. A laser ultrasonic inspection system was developed based on the Lamb wave (LW) produced by a laser pulse. An all-fiber laser heterodyne interferometer is applied for measuring the ultrasonic signal in combination with an automatic scanning system, which makes the system flexible and compact. A 3-D model simulation of an H62 brass specimen was carried out to determine the LW spatial-temporal wavefield by using the COMSOL Multiphysics software. The characteristics of the ultrasonic wavefield were extracted through continuous wavelet transform analysis. This demonstrates that the A0 mode could be used in defect detection due to its slow speed and vibrational direction. Furthermore, an ultrasonic wave at the center frequency of 370 kHz with maximum energy is suitable for defect detection. In the experiment, the size and location of the defect are determined by the time difference of the transmitted wave and reflected wave, respectively. The relative error of the defect position is 0.14% by averaging six different receiving spots. The width of the defect is linear to the time difference of the transmitted wave. The goodness of fit can reach 0.989, and it is in good agreement with the simulated one. The experimental error is less than 0.395 mm for a 5 mm width of defect. Therefore, this validates that the technique can be potentially utilized in the remote defect detection of thin copper plates.
RESUMO
Microneurographic recordings of muscle sympathetic nerve activity (MSNA) reflect postganglionic sympathetic axonal activity directed toward the skeletal muscle vasculature. Recordings are typically evaluated for spontaneous bursts of MSNA; however, the filtering and integration of raw neurograms to obtain multiunit bursts conceals the underlying c-fiber discharge behavior. The continuous wavelet transform with matched mother wavelet has permitted the assessment of action potential discharge patterns, but this approach uses a mother wavelet optimized for an amplifier that is no longer commercially available (University of Iowa Bioengineering Nerve Traffic Analysis System; Iowa NTA). The aim of this project was to determine the morphology and action potential detection performance of mother wavelets created from the commercially available NeuroAmp (ADinstruments), from distinct laboratories, compared with a mother wavelet generated from the Iowa NTA. Four optimized mother wavelets were generated in a two-phase iterative process from independent datasets, collected by separate laboratories (one Iowa NTA, three NeuroAmp). Action potential extraction performance of each mother wavelet was compared for each of the NeuroAmp-based datasets. The total number of detected action potentials was not significantly different across wavelets. However, the predictive value of action potential detection was reduced when the Iowa NTA wavelet was used to detect action potentials in NeuroAmp data, but not different across NeuroAmp wavelets. To standardize approaches, we recommend a NeuroAmp-optimized mother wavelet be used for the evaluation of sympathetic action potential discharge behavior when microneurographic data are collected with this system.NEW & NOTEWORTHY The morphology of custom mother wavelets produced across laboratories using the NeuroAmp was highly similar, but distinct from the University of Iowa Bioengineering Nerve Traffic Analysis System. Although the number of action potentials detected was similar between collection systems and mother wavelets, the predictive value differed. Our data suggest action potential analysis using the continuous wavelet transform requires a mother wavelet optimized for the collection system.
Assuntos
Potenciais de Ação , Análise de Ondaletas , Potenciais de Ação/fisiologia , Animais , Sistema Nervoso Simpático/fisiologia , Músculo Esquelético/fisiologia , MasculinoRESUMO
Background: Accurate classification of electrocardiogram (ECG) signals is crucial for automatic diagnosis of heart diseases. However, existing ECG classification methods often require complex preprocessing and denoising operations, and traditional convolutional neural network (CNN)-based methods struggle to capture complex relationships and high-level time-series features. Method: In this study, we propose an ECG classification method based on continuous wavelet transform and multi-branch transformer. The method utilizes continuous wavelet transform (CWT) to convert the ECG signal into time-series feature map, eliminating the need for complicated preprocessing. Additionally, the multi-branch transformer is introduced to enhance feature extraction during model training and improve classification performance by removing redundant information while preserving important features. Results: The proposed method was evaluated on the CPSC 2018 (6877 cases) and MIT-BIH (47 cases) ECG public datasets, achieving an accuracy of 98.53% and 99.38%, respectively, with F1 scores of 97.57% and 98.65%. These results outperformed most existing methods, demonstrating the excellent performance of the proposed method. Conclusion: The proposed method accurately classifies the ECG time-series feature map, which holds promise for the diagnosis of cardiac arrhythmias. The findings of this study are valuable for advancing the field of automatic ECG diagnosis.
RESUMO
Among unmanned surface vehicle (USV) components, underwater thrusters are pivotal in their mission execution integrity. Yet, these thrusters directly interact with marine environments, making them perpetually susceptible to malfunctions. To diagnose thruster faults, a non-invasive and cost-effective vibration-based methodology that does not require altering existing systems is employed. However, the vibration data collected within the hull is influenced by propeller-fluid interactions, hull damping, and structural resonant frequencies, resulting in noise and unpredictability. Furthermore, to differentiate faults not only at fixed rotational speeds but also over the entire range of a thruster's rotational speeds, traditional frequency analysis based on the Fourier transform cannot be utilized. Hence, Continuous Wavelet Transform (CWT), known for attributions encapsulating physical characteristics in both time-frequency domain nuances, was applied to address these complications and transform vibration data into a scalogram. CWT results are diagnosed using a Vision Transformer (ViT) classifier known for its global context awareness in image processing. The effectiveness of this diagnosis approach was verified through experiments using a USV designed for field experiments. Seven cases with different fault types and severity were diagnosed and yielded average accuracy of 0.9855 and 0.9908 at different vibration points, respectively.
RESUMO
Objective.Although emotion recognition has been studied for decades, a more accurate classification method that requires less computing is still needed. At present, in many studies, EEG features are extracted from all channels to recognize emotional states, however, there is a lack of an efficient feature domain that improves classification performance and reduces the number of EEG channels.Approach.In this study, a continuous wavelet transform (CWT)-based feature representation of multi-channel EEG data is proposed for automatic emotion recognition. In the proposed feature, the time-frequency domain information is preserved by using CWT coefficients. For a particular EEG channel, each CWT coefficient is mapped into a strength-to-entropy component ratio to obtain a 2D representation. Finally, a 2D feature matrix, namely CEF2D, is created by concatenating these representations from different channels and fed into a deep convolutional neural network architecture. Based on the CWT domain energy-to-entropy ratio, effective channel and CWT scale selection schemes are also proposed to reduce computational complexity.Main results.Compared with previous studies, the results of this study show that valence and arousal classification accuracy has improved in both 3-class and 2-class cases. For the 2-class problem, the average accuracies obtained for valence and arousal dimensions are 98.83% and 98.95%, respectively, and for the 3-class, the accuracies are 98.25% and 98.68%, respectively.Significance.Our findings show that the entropy-based feature of EEG data in the CWT domain is effective for emotion recognition. Utilizing the proposed feature domain, an effective channel selection method can reduce computational complexity.
Assuntos
Algoritmos , Eletroencefalografia , Emoções , Redes Neurais de Computação , Análise de Ondaletas , Humanos , Eletroencefalografia/métodos , Processamento de Sinais Assistido por Computador , Entropia , Nível de Alerta/fisiologiaRESUMO
OBJECTIVES: Electrocardiogram (ECG) signals are extensively utilized in the identification and assessment of diverse cardiac conditions, including congestive heart failure (CHF) and cardiac arrhythmias (ARR), which present potential hazards to human health. With the aim of facilitating disease diagnosis and assessment, advanced computer-aided systems are being developed to analyze ECG signals. METHODS: This study proposes a state-of-the-art ECG data pattern recognition algorithm based on Continuous Wavelet Transform (CWT) as a novel signal preprocessing model. The Motif Transformation (MT) method was devised to diminish the drawbacks and limitations inherent in the CWT, such as the issue of boundary effects, limited localization in time and frequency, and overfitting conditions. This transformation technique facilitates the formation of diverse patterns (motifs) within the signals. The patterns (motifs) are constructed by comparing the amplitudes of each individual sample value in the ECG signals in terms of their largeness and smallness. In the subsequent stage, the obtained one-dimensional signals from the MT transformation were subjected to CWT to obtain scalogram images. In the last stage, the obtained scalogram images were subjected to classification using DenseNET deep transfer learning techniques. RESULTS AND CONCLUSIONS: The combined approach of MT + CWT + DenseNET yielded an impressive success rate of 99.31â¯%.
Assuntos
Algoritmos , Eletrocardiografia , Análise de Ondaletas , Humanos , Eletrocardiografia/métodos , Aprendizado Profundo , Processamento de Sinais Assistido por Computador , Insuficiência Cardíaca/diagnóstico por imagem , Insuficiência Cardíaca/fisiopatologia , Arritmias Cardíacas/diagnóstico por imagem , Arritmias Cardíacas/fisiopatologia , Cardiopatias/fisiopatologiaRESUMO
Obstructive sleep apnea (OSA), a prevalent sleep disorder, is intimately associated with various other diseases, particularly cardiovascular conditions. The conventional diagnostic method, nocturnal polysomnography (PSG), despite its widespread use, faces challenges due to its high cost and prolonged duration. Recent developments in electrocardiogram-based diagnostic techniques have opened new avenues for addressing these challenges, although they often require a deep understanding of feature engineering. In this study, we introduce an innovative method for OSA classification that combines a composite deep convolutional neural network model with a multimodal strategy for automatic feature extraction. This approach involves transforming the original dataset into scalogram images that reflect heart rate variability attributes and Gramian angular field matrix images that reveal temporal characteristics, aiming to enhance the diversity and richness of data features. The model comprises automatic feature extraction and feature enhancement components and has been trained and validated on the PhysioNet Apnea-ECG database. The experimental results demonstrate the model's exceptional performance in diagnosing OSA, achieving an accuracy of 96.37%, a sensitivity of 94.67%, a specificity of 97.44%, and an AUC of 0.96. These outcomes underscore the potential of our proposed model as an efficient, accurate, and convenient tool for OSA diagnosis.
Assuntos
Apneia Obstrutiva do Sono , Humanos , Apneia Obstrutiva do Sono/diagnóstico , Eletrocardiografia/métodos , Redes Neurais de Computação , Polissonografia , Frequência CardíacaRESUMO
Electric power utilities must ensure a consistent and undisturbed supply of power, with the voltage levels adhering to specified ranges. Any deviation from these supply specifications can lead to malfunctions in equipment. Monitoring the quality of supplied power is crucial to minimize the impact of fluctuations in voltage. Variations in voltage or current from their ideal values are referred to as "power quality (PQ) disturbances," highlighting the need for vigilant monitoring and management. Signal processing methods are widely used for power system applications which include understanding of voltage disturbance signals and used for retrieval of signal information from the signals Different signal processing methods are used for extracting information about a signal. The method of Fourier analysis involves application of Fourier transform giving frequency information. The method of Short-Time Fourier analysis involves application of Short-Time Fourier transform (STFT) giving time-frequency information. The method of continuous wavelet analysis involves application of Continuous Wavelet transform (CWT) giving signal information in terms of scale and time where frequency is inversely related to scale. The method of discrete wavelet analysis involves application of Discrete Wavelet transform (DWT) giving signal information in terms of approximations and details where approximations and details are low and high frequency representation of original signal. In this paper, an attempt is made to perceive power quality disturbances in MATLAB using Fourier, Short-Time Fourier, Continuous Wavelet and Discrete Wavelet Transforms. Proper understanding of the signals can be possible by transforming the signals into different domains. An emphasis on application of signal processing techniques can be laid for power quality studies. The paper compares the results of each transform using MATLAB-based visualizations. The discussion covers the advantages and disadvantages of each technique, providing valuable insights into the interpretation of power quality disturbances. As the paper delves into the complexities of each method, it takes the reader on a journey of signal processing complexities, culminating in a nuanced understanding of power quality disturbances and their representations across various domains. The outcomes of this research, elucidated through energy values, 3D plots, and comparative analyses, contribute to a comprehensive understanding of power quality disturbances. The findings not only traverse theoretical domains but also find practical utility in real-world scenarios.
RESUMO
Herein, two different sustainable and green signal processing spectrophotometric approaches, namely, derivative spectroscopy and wavelet transform, have been utilized for effective measurement of the antiretroviral therapy abacavir and lamivudine in their pharmaceutical formulations. These methods were used to enhance the spectral data and differentiate between the absorption bands of abacavir and lamivudine in order to accurately measure their concentrations. For determining abacavir and lamivudine, the first derivative spectrophotometric method has been applied to the zero-order and ratio spectra of both drugs. The same approach has been tested using the continuous wavelet transform method where a second order 2.4 of rbio and bior wavelet families were found to be optimum for measuring both drugs. Validation of the proposed methods affirmed their reliability in terms of linearity over the concentration range 1.5-30 µg/mL and 1.5-36 µg/mL for abacavir and lamivudine, respectively, precision (RSD < 2 %), and accuracy with mean recoveries ranging between 98 % and 102 %. Additionally, these spectrophotometric methodologies were applied to real pharmaceutical preparations and yielded results congruent with a prior chromatographic method. Most prominently, the proposed methods stood out for their greenness and sustainability with 97 points as evaluated by the analytical eco-scale method and a score value of 0.79 as analyzed by AGREE method, thereby making them suitable for resource-limited settings and highlighting the potential for broader application of green analytical methods in pharmaceutical analysis.