One-dimensional Convolutional Neural Network Research Articles

Predictive modeling and insight into protein fouling in microfiltration and ultrafiltration through one-dimensional convolutional models

Membrane fouling, a critical issue in membrane performance associated with proteins, is often difficult to analyze due to limited knowledge of contamination scenarios. This study uses a one-dimensional convolutional neural network (ODCNN) to examine the complex relationships among membrane properties, protein characteristics, and operational parameters, aiming to elucidate the dynamics of protein-related contamination. To improve the predictive performance of the deep learning model, we propose a novel strategy that learns numerical and categorical features in sequence. Compared with the other seven conventional models, our results reveal that the ODCNN model shows the best performance, which achieves impressive coefficients of determination, reaching 0.833 for flux and 0.723 for protein rejection, indicating the model’s robustness even when dealing with incomplete information on membrane fouling. To mitigate the issue of insufficient data, we applied the Synthetic Minority Over-Sampling Technique to increase the dataset’s diversity, validating this approach with experimental evidence. To improve the interpretability of the model, we used the Shapley Additive exPlanations method to assess the contribution of various features, which revealed that membrane pore size, protein concentration, transmembrane pressure, and crossflow velocity are key factors influencing both flux and protein retention. We also analyzed the causes and assessed the degree of contamination according to the degree of their influence. By leveraging machine learning techniques to analyze and predict membrane filtration performance, particularly concerning protein fouling, this study demonstrates significant potential for practical applications in environments with limited information. This innovative approach is poised to contribute to ongoing advancements in membrane filtration technology and offers a promising pathway for enhancing the understanding and management of membrane fouling.

Bending induced polarization charges in non-polar porous polymer for stroke rehabilitation

Owing to the flexibility, lightweight, long-life and low cost, polymers are promising candidates to realize electromechanical conversion in wearable electronics. The flexoelectric effect enables non-piezoelectric materials to achieve electromechanical coupling, thereby broadening application ranges of nonpolar polymers. In this work, the porous samples of polydimethylsiloxane (PDMS) with various pore sizes and porosities are fabricated by the sacrificial salt template method and the chemical gas foaming method. The flexoelectric polarization response can be tuned directly by the structural characters of the porous PDMS. The transverse flexoelectric coefficient of porous PDMS can reach approximately 2.9 times the counterpart of the bulk one. The competitive mechanism between strain gradient and permittivity is proposed on flexoelectricity of random porous media by dielectric impedance spectrums and finite element analyses. The present regulation of flexoelectricity provides an alternative to electromechanical conversion. With the deep learning technique based on one-dimensional convolution neural networks, actions of both sick and normal legs of various stroke patients can be well recognized by the designed device of rehabilitation. The present rehabilitation monitoring system, which is an innovative application of the flexoelectricity in the porous polymers, offers a new approach to support physical therapy for post-stroke patients.

Analyzing the opening and closing of windows in residential for predicting the energy consumption using optimized multi-scale convolution networks

PurposeThis proposal aims to forecast energy consumption in residential buildings based on the effect of opening and closing windows by the deep architecture approach. In this task, the developed model has three stages: (1) collection of data, (2) feature extraction and (3) prediction. Initially, the data for the closing and opening frequency of the window are taken from the manually collected datasets. After that, the weighted feature extraction is performed in the collected data. The attained weighted feature is fed to predict energy consumption. The prediction uses the efficient hybrid multi-scale convolution networks (EHMSCN), where two deep structured architectures like a deep temporal context network and one-dimensional deep convolutional neural network. Here, the parameter optimization takes place with the hybrid algorithm named jumping rate-based grasshopper lemur optimization (JR-GLO). The core aim of this energy consumption model is to predict the consumption of energy accurately based on the effect of opening and closing windows. Therefore, the offered energy consumption prediction approach is analyzed over various measures and attains an accurate performance rate than the conventional techniques.Design/methodology/approachAn EHMSCN-aided energy consumption prediction model is developed to forecast the amount of energy usage during the opening and closing of windows accurately. The emission of CO2 in indoor spaces is highly reduced.FindingsThe MASE measure of the proposed model was 52.55, 43.83, 42.01 and 36.81% higher than ANN, CNN, DTCN and 1DCNN.Originality/valueThe findings of the suggested model in residences were attained high-quality measures with high accuracy, precision and variance.

Tool Wear Classification in Chipboard Milling Processes Using 1-D CNN and LSTM Based on Sequential Features

The paper presents the comparative analysis of Long short-term memory (LSTM) and one-dimensional convolutional neural networks (1-D CNNs) for tool wear classification in chipboard milling processes. The complexity of sequence data in various fields makes selecting the right model for sequence classification very important. This research aims to show the distinct capabilities and performance nuances of LSTM and 1-D CNN models, leveraging their inherent strengths in understanding temporal dependencies and feature extraction, respectively. Through a series of experiments, the study unveils that while both models demonstrate competencies in handling sequence data, the 1-D CNN model, with its superior feature extraction capabilities, achieved the best performance, boasting an accuracy of 94.5% on the test dataset. The insights gained from this comparison not only help to understand LSTM and 1-D CNN models better, but also open the door for future improvements in using neural networks for complex sequence classification challenges.

An explainable dual-mode convolutional neural network for multivariate time series classification

Multivariate time series classification (MTSC) is a crucial machine learning problem prevalent across various real-life domains. Traditional deep learning approaches with high accuracy in MTSC are often criticized for their “black box” nature, offering no insight into their operational mechanisms or decision-making processes. Explainable artificial intelligence (XAI) becomes a key idea for dealing with such limitations in decision-sensitive areas. While some researchers have explored the interpretability of MTSC, the majority have focused on elucidating the relationship between variables over time, neglecting the intricate connections among different variables. To address the interpretability issue in MTSC, we introduce an explainable dual-mode convolutional neural network (XDM-CNN) designed specifically for MTSC. The proposed XDM-CNN framework comprises two modules: a classification module and an explanation module. The classification module can ensure exceptional classification performance by combining both one-dimensional and two-dimensional convolutional neural networks, while the explanation module can mine the underlying logical relationships among variables during the classification process by utilizing a synergy of visualization and quantification techniques. We validate our approach using 26 datasets sourced from the University of East Anglia (UEA) archive. The experimental results demonstrate that XDM-CNN not only exhibits excellent performance in classification accuracy, but also has strong explainability. By combining visual and numerical explanation, the hidden logical relationships among multivariate time series are explored and interpreted, providing human users with a decidable basis for classification decisions.

Optical routing: link failure detection-aided optimal routing in elastic optical networks using hybrid deep learning with target-based feature pool formation

ABSTRACT The Elastic Optical Network (EON) is a kind of optical network where a large amount of data signals are given into the channel of optical fibre by modulating with the optical signal. One of the most crucial tasks for optical networks is resource allocation and routing, which has a significant impact on resource effectiveness and network capacity. Deep learning-based routing and resource allocation have piqued the interest of researchers to a considerable extent. Thus, this paper aims to develop efficient link failure detection in EONs with hybrid deep learning strategies. Initially, the data aggregation is performed to combine the multiple link information from EONs. The aggregated link data is given into the hybrid detection model named as Enhanced Deep Learning Network (EDLNet), where the One-Dimensional Convolutional Neural Network (1D-CNN) is used for extracting the features and used for generating the target-based feature pool. This feature pool is utilised in the Long Short-Term Memory (LSTM) network for detecting link failures. The parameter optimization is done by the hybrid 1DCNN-LSTM network with the support of the Hybrid Red Deer with Sewing Training-based Optimisation (HRD-STO) to achieve accurate detection results. The detected failed link is mitigated at the time of routing, and only reliable links are used to attain certain objective constraints. These objective constraints include shortest path, link reliability and stability, throughput and delay for enhancing the efficiency of the EONs. The replication analysis reveals the effectiveness of the proposed link failure detection model by comparing it with the conventional models.

Identification of hypermucoviscous Klebsiella pneumoniae K1, K2, K54 and K57 capsular serotypes by Raman spectroscopy

Antimicrobial resistance poses a significant challenge in modern medicine, affecting public health. Klebsiella pneumoniae infections compound this issue due to their broad range of infections and the emergence of multiple antibiotic resistance mechanisms. Efficient detection of its capsular serotypes is crucial for immediate patient treatment, epidemiological tracking and outbreak containment. Current methods have limitations that can delay interventions and increase the risk of morbidity and mortality. Raman spectroscopy is a promising alternative to identify capsular serotypes in hypermucoviscous K. pneumoniae isolates. It provides rapid and in situ measurements with minimal sample preparation. Moreover, its combination with machine learning tools demonstrates high accuracy and reproducibility. This study analyzed the viability of combining Raman spectroscopy with one-dimensional convolutional neural networks (1-D CNN) to classify four capsular serotypes of hypermucoviscous K. pneumoniae: K1, K2, K54 and K57. Our approach involved identifying the most relevant Raman features for classification to prevent overfitting in the training models. Simplifying the dataset to essential information maintains accuracy and reduces computational costs and training time. Capsular serotypes were classified with 96 % accuracy using less than 30 Raman features out of 2400 contained in each spectrum. To validate our methodology, we expanded the dataset to include both hypermucoviscous and non-mucoid isolates and distinguished between them. This resulted in an accuracy rate of 94 %. The results obtained have significant potential for practical healthcare applications, especially for enabling the prompt prescription of the appropriate antibiotic treatment against infections.

Blind source separation algorithm for noisy hydroacoustic signals based on decoupled convolutional neural networks

Wireless communication technology has been widely used in marine engineering, marine ranching and Marine environmental monitoring. However, structural redundancy and functional confusion exist in applying neural networks in signal separation technology in underwater communication environments, which can result in a slower rate of signal separation and lead to confusion of parameters during transfer learning. Based on this, an end-to-end, internal functionally structured decoupled neural network (D-CNN) blind source separation (BSS) model is proposed in this paper, which can realize a neural network BSS algorithm with a well-defined structure and function. The one-dimensional convolutional neural network layer is used in algorithm to automatically extract observed signal's features, based on the features, and there are two generation modules of separation matrix and scaling coefficients. Then the two modules can be used to separate the observed signal and adjust the signal coefficients to obtain the separated signal. Finally the transfer learning technique is used to generalize the model, which reduces the transfer cost of the model in different application scenarios. Experimental results show that when the communication distance is set to 0.02 km–2 km, the MSE of independent signal and related signal can be reduced by 14.24% and 14.95% respectively compared with the nearest Neural FCA algorithm. The results prove that the proposed algorithm can accurately estimate the source signal and improve the signal reception quality.

Abnormal Detection and Fault Diagnosis of Adjustment Hydraulic Servomotor Based on Genetic Algorithm to Optimize Support Vector Data Description with Negative Samples and One-Dimensional Convolutional Neural Network

Because of the difficulty in fault detection for and diagnosing the adjustment hydraulic servomotor, this paper uses feature extraction technology to extract the time domain and frequency domain features of the pressure signal of the adjustment hydraulic servomotor and splice the features of multiple pressure signals through the Multi-source Information Fusion (MSIF) method. The comprehensive expression of device status information is obtained. After that, this paper proposes a fault detection Algorithm GA-SVDD-neg, which uses Genetic Algorithm (GA) to optimize Support Vector Data Description with negative examples (SVDD-neg). Through joint optimization with the Mutual Information (MI) feature selection algorithm, the features that are most sensitive to the state deterioration of the adjustment hydraulic servomotor are selected. Experiments show that the MI algorithm has a better performance than other feature dimensionality reduction algorithms in the field of the abnormal detection of adjustment hydraulic servomotors, and the GA-SVDD-neg algorithm has a stronger robustness and generality than other anomaly detection algorithms. In addition, to make full use of the advantages of deep learning in automatic feature extraction and classification, this paper realizes the fault diagnosis of the adjustment hydraulic servomotor based on 1D Convolutional Neural Network (1DCNN). The experimental results show that this algorithm has the same superior performance as the traditional algorithm in feature extraction and can accurately diagnose the known faults of the adjustment hydraulic servomotor. This research is of great significance for the intelligent transformation of adjustment hydraulic servomotors and can also provide a reference for the fault warning and diagnosis of the Electro-Hydraulic (EH) system of the same type of steam turbine.

Multi-Perception Graph Convolution Transfer Network Bearing Fault Diagnosis Method

Targeting the challenge of variable working conditions in bearing fault diagnosis, most of the fault diagnosis methods based on transfer learning focus on the transfer of knowledge, resulting in a poor diagnosis effect in the target domain. To solve the problem of transfer performance degradation, a multi-perception graph convolution transfer network (MPGCTN) is proposed. The MPGCTN is composed of a graph generation module, graph perception module, and domain discrimination module. In the graph generation module, a one-dimensional convolution neural network (1-D CNN) is used to extract features from the input, and then the structural features of samples are mined in the graph generation layer to construct the sample graph. In the following graph perception module, a multi-perception graph convolution network is designed to model the sample graph and learn the data structure information of the sample. Finally, in the domain discrimination module, the method is used to align the structural differences of the case graphs in different domains. Experimental results from experiments on Case Western Reserve University (CWRU) and Paderborn University (PU) bearing datasets show that the proposed method is effective and superior.

Machine learning can predict implantable cardioverter defibrillator therapy: a development study

Abstract Introduction Remote monitoring (RM) data has been used to develop predictive algorithms that warn of decompensated heart failure before a heart failure episode. The same RM data could also be used to predict other adverse events in heart failure patients, such as ventricular arrhythmias. Machine learning (ML) models can be used to process complex multivariate time series RM data. Purpose The aim of this analysis was to generate a clinically useful model to predict defibrillator therapy (shock or anti-tachycardia pacing (ATP)). The output of the model is the likelihood of a shock in the next 24 hours. Methods RM data of all 292 implantable cardioverter defibrillators (ICDs) at our centre were collected (male 81%, 66±12 years). Electrograms of ICD therapies were manually verified. 63 time series variables were collated. Atrial lead-dependent RM parameters and ones with >30% missing data were excluded leaving 23 variables for analysis. After the exclusion of inappropriate therapies and patients with missing RM data, 218 patients were analysed. Of these, 41 (18.8%) had at least one ICD therapy. Three types of ML models were generated (1) random forest (RF), (2) extreme gradient boosted model (XGBoost), (3) one-dimensional convolutional neural network (1D-CNN) with three different intervals preceding ICD therapy: 30, 10, and 4 days. The model accuracy was measured utilising stratified grouped 5-fold cross-validation, with 20% of data used for testing in each fold. Results 4-day decision tree-based models (RF, XGBoost) demonstrated higher area under the curve (AUC) scores, compared with 10- and 30-day models. Percentage accuracy was similar across all three models and interval lengths. RF: The 4-day model had the highest AUC of 0.86 and an accuracy of 83.8%, compared to the 10-day (AUC 0.85, 83.0%) and 30-day (AUC 0.80, 82.2%) models. XGBoost: The 4-day model had an AUC of 0.85 and an accuracy of 83.4%, compared to the 10-day (AUC 0.84, 83.1%) and 30-day (AUC 0.81, 82.4%) models. 1D-CNN: The 4-day model had an AUC of 0.78 and an accuracy of 83.5%, compared to the 10-day (AUC 0.74, 83.1%) and 30-day (AUC 0.71, 82.2%) models. Episodes of prior ventricular tachycardia (VT) in the VT2 zone and prior ICD therapies contributed the most to model predictions, with their importance highest in the days immediately preceding ICD therapy (Figure 1). To better identify ‘first ICD therapy’, prior ICD therapies were excluded, and the models were re-run. AUC was lower at 0.70 with an accuracy of 72.9% in the highest-performing model (10-day XGBoost). The most important features were patient activity, RV pacing impedance, and mean ventricular heart rate (Figure 2). Conclusion We were able to predict ICD therapy with an AUC of 0.86. 4-day models were better than the 10-day and 30-day models. Model performance remained high (AUC 0.70), even after exclusion of prior ventricular arrhythmias from the predictors.1.Importance of top variables over time2.Top features without prior ICD therapy

Detection Method of Epileptic Seizures Using a Neural Network Model Based on Multimodal Dual-Stream Networks.

Epilepsy is a common neurological disorder, and its diagnosis mainly relies on the analysis of electroencephalogram (EEG) signals. However, the raw EEG signals contain limited recognizable features, and in order to increase the recognizable features in the input of the network, the differential features of the signals, the amplitude spectrum and the phase spectrum in the frequency domain are extracted to form a two-dimensional feature vector. In order to solve the problem of recognizing multimodal features, a neural network model based on a multimodal dual-stream network is proposed, which uses a mixture of one-dimensional convolution, two-dimensional convolution and LSTM neural networks to extract the spatial features of the EEG two-dimensional vectors and the temporal features of the signals, respectively, and combines the advantages of the two networks, using the hybrid neural network to extract both the temporal and spatial features of the signals at the same time. In addition, a channel attention module was used to focus the model on features related to seizures. Finally, multiple sets of experiments were conducted on the Bonn and New Delhi data sets, and the highest accuracy rates of 99.69% and 97.5% were obtained on the test set, respectively, verifying the superiority of the proposed model in the task of epileptic seizure detection.

Comparative analysis of classical artificial intelligence and deep learning models for atrial fibrillation detection in electrocardiograms

Abstract Background/Introduction Early diagnosis of atrial fibrillation (AF) presents a challenging yet critical task for appropriate interventions aimed at reducing disease-related burden. In this context, strategies employing classical artificial intelligence (CAI) and deep learning (DL) have emerged as promising approaches to optimize cardiac disorder screening and detection. Purpose This study aimed to compare a CAI model and a DL model for the detection of AF in patients undergoing electrocardiographic (ECG) examinations in tertiary healthcare centers. Methods Between December 2022 and November 2023, a total of 135,476 ECGs were performed, comprising 5,067 with AF and 130,409 without AF. The ECGs were analyzed using both artificial intelligence models. The obtained results were then compared to the gold standard (cardiologist's report). In the CAI model, signals were extracted from ECG images, analyzing five key parameters: cardiac rhythm, atrial depolarization, atrioventricular conduction, ventricular depolarization, and ventricular repolarization (figure 1A). These parameters were benchmarked against the standard values from the Brazilian Society of Cardiology guidelines for detecting cardiac anomalies. Conversely, the DL model utilized a one-dimensional ResNet-based Convolutional Neural Network (CNN). This model was trained using ADAM optimization and binary cross-entropy loss, enabling the learning of complex patterns in the data (figure 1B). Results The mean age was 54.6 years (71.9 years with AF and 53.9 without AF). In the AF population, 52.2% were male (46% were male in the overall sample). In the analysis conducted, the CAI model showed a sensitivity and specificity of 90% and 62%, respectively, while the DL model had 90% and 69%, respectively. ROC curves were generated for both models, demonstrating the superior performance of the DL model (figure 2A). Conclusions Although the sensitivity remained similar between the models, the DL model distinguished itself with higher specificity. These results suggest that artificial intelligence, particularly the deep learning approach, holds promise as a supportive resource in AF diagnosis. However, further studies are needed to evaluate the models more thoroughly and determine their clinical applicability in a broader context.Artificial intelligence modelsROC curves

Recognition Algorithm of AE Signal of Rock Fracture Based on Multiscale 1DCNN-BLSTM

Acoustic emission (AE) signals produced by different types of rocks have different characteristics of information. Determining the brittle mineral content of rock according to the acoustic emission characteristics of rock is helpful to understand the mechanical behavior of rock in field monitoring. This article constructs a deep learning algorithm model to identify acoustic emission signals released from rock fractures with different brittle mineral contents. In response to the interference characteristics of acoustic emission signal data, a multiscale one-dimensional convolutional neural network embedded with efficient channel attention (ECA) module was incorporated into the model, and multiscale convolutional kernels were used to extract features of different levels of precision. In the latter half of the model, the BLSTM network was incorporated to extract time series-related features, local spatial uncorrelated features, and weak periodic pattern features from the acoustic emission signal data. To solve the problem that the recognition accuracy of minority samples decreases, this study replaces ReLU activation function with SELU. The results show that the multiscale 1DCNN-BLSTM model embedded in ECA module has a good antinoise performance, and the recognition accuracy can reach over 90%. The discovery of this work provides a new idea for exploring the mechanism of rock mass instability.

1D-CNN: Classification of normal delivery and cesarean section types using cardiotocography time-series signals

Abstract Cardiotocography (CTG) is considered the gold standard for monitoring fetal heart rate (FHR) during pregnancy and labor to estimate the danger of oxygen deprivation. Visual interpretation of CTG traces is complex and frequently results in high rates of false positives and false negatives, leading to unfavorable and unwanted outcomes such as fetal mortality or needless cesarean surgery. If the data are well-balanced, which is uncommon in medical datasets, machine learning techniques can be helpful in interpretation. This study is designed to determine classification performance under various data balance approaches. We propose a robust methodology for the automated extraction of features that use a deep learning model based on the one-dimensional convolutional neural network (1D-CNN). We used a public database containing 552 intrapartum CTG recordings. Due to the imbalance in the dataset, the experiments were conducted under a variety of conditions such as (i) an unbalanced dataset, (ii) undersampling, (iii) a weighted binary cross-entropy approach, and (iv) oversampling utilizing the synthetic minority oversampling technique (SMOTE). We found an excellent sensitivity (99.80% for the unbalanced dataset, 96.25% for the weighted binary cross-entropy approach, and 99.81% with SMOTE) except for the under sampling situation, in which the sensitivity was 85.71%. Moreover, the 1D-CNN model incorporating SMOTE yielded promising results in 88% specificity, 93.72% quality index (QI), and 95.10% area under the curve. The model exhibited excellent performance in terms of sensitivity in every scenario except for undersampling. The oversampling of training data with SMOTE yielded a decent level of specificity, demonstrating the model’s strong predictive capacity. In addition, the SMOTE scenario resulted in fewer training epochs, which is another accomplishment.

A Multi-Step-Ahead Photovoltaic Power Forecasting Approach Using One-Dimensional Convolutional Neural Networks and Transformer

Due to environmental concerns about the use of fossil fuels, renewable energy, especially solar energy, is increasingly sought after for its ease of installation, cost-effectiveness, and versatile capacity. However, the variability in environmental factors poses a significant challenge to photovoltaic (PV) power generation forecasting, which is crucial for maintaining power system stability and economic efficiency. In this paper, a novel muti-step-ahead PV power generation forecasting model by integrating single-step and multi-step forecasts from various time resolutions was developed. One-dimensional convolutional neural network (CNN) layers were used for single-step forecasting to capture specific temporal patterns, with the transformer model improving multi-step forecasting by leveraging the combined outputs of the CNN. This combination can provide accurate and immediate forecasts as well as the ability to identify longer-term generation trends. Using the DKASC-ASA-1A and 1B datasets for empirical validation, several preprocessing methods were applied and a series of experiments were conducted to compare the performance of the model with other widely used deep learning models. The framework proved to be capable of accurately predicting multi-step-ahead PV power generation at multiple time resolutions.

Deep learning-based prediction of structural responses of RC slabs subjected to blast loading

Considering the risk of explosion, studying the dynamic structural behavior of reinforced concrete (RC) slabs subjected to blast loads is crucial for enhancing their resistance. Conventional approaches, including costly experimental tests, highly hypothetical analytical methods, and time-consuming numerical simulations, have their limitations. This study presents two deep learning-based models for rapid and accurate prediction of explosion-induced responses in RC slabs. Available literature data and supplemented numerical simulation data are used for training and testing. First, a multi-layer perceptron (MLP) model is established to predict the maximum displacement of RC slabs subjected to explosions. The input features include 10 parameters. The training results show that the MLP model exhibits good prediction performance. Second, a one-dimensional convolutional neural network (1D-CNN) model is constructed to predict the failure modes of RC slabs under explosive loads. The input features of the model are consistent with those of the MLP model. The 1D-CNN model outperforms the other five conventional machine learning models in classification. Furthermore, a permutation feature importance analysis is conducted to determine the effects of the 10 input features on the predicted outcomes. This analysis makes the predictions interpretable, thereby bolstering the credibility of the models. Results show that the proposed deep learning models offer an efficient and reliable method to predict the structural response of RC slabs under explosion loads.

Online Detection of Dry Matter in Potatoes Based on Visible Near-Infrared Transmission Spectroscopy Combined with 1D-CNN

More efficient resource utilization and increased crop utilization rate are needed to address the growing demand for food. The efficient quality testing of key agricultural products such as potatoes, especially the rapid testing of key nutritional indicators, has become an important strategy for ensuring their quality and safety. In this study, visible and near infrared (Vis/NIR) transmittance spectroscopy (600–900 nm) was used for the online analysis of multiple quality parameters in potatoes. The study concentrated on comparing three one-dimensional convolutional neural network (1D-CNN) models, specifically, the fine-tuned DeepSpectra, the fine-tuned 1D-AlexNet, and classic CNN, with UVE-PLS (uninformative variable elimination–partial least squares) models. These models utilized spectral data for the real-time detection of dry matter (DM) content in potatoes. To address the challenges posed by limited data from Vis/NIR, this study strategically implemented data augmentation techniques. This approach significantly enhanced the robustness and generalization capabilities of the models. The 1D-AlexNet and DeepSpectra models achieved 0.934 and 0.913 R2P and 0.0603 and 0.0695 g/100 g RMSEP for DM, respectively. Compared to UVE-PLS, the R2P value improved by 21.31% (0.770 to 0.934) for the 1D-AlexNet model and 18.64% (0.770 to 0.913) for the DeepSpectra model. The RMSEP value was reduced by 47.31% (0.114 to 0.0603) for 1D-AlexNet, and 39.30% (0.114 to 0.0695) for the DeepSpectra model. As a result, this study would be helpful for researching the online Vis/NIR transmission determination of potato DM using deep learning. These results highlighted the immense potential of employing specific spectral features in deep-learning models for a more precise and efficient online assessment of agricultural quality. This advancement provided some insight and reference for further contributing to the evolution of more targeted and efficient quality assessment methods in agricultural products.

Deep Learning-Based Acoustic Emission Signal Filtration Model in Reinforced Concrete

AbstractAcoustic emission is a nondestructive testing (NDT) technique, widely used to monitor the condition of structures for safety reasons especially in real time. The method utilizes the electrical signals generated by the elastic waves in a material under load to detect and locate damage in structures. However, identifying the sources of AE signals in concrete or composite materials can be challenging due to the anisotropic properties of materials and interpreting a large amount of AE data, leading to data misinterpretation and inaccurate detection of damage. Hence, the need for filtering out noise-induced signals from recorded data and emphasizing the actual AE source is crucial for monitoring and source localization of damage in real time. This study proposed a one-dimensional convolutional neural network (1D-CNN) deep learning approach to filter around 22,000 AE data in a reinforced concrete (RC) beam. The model utilizes significant AE parameters identified through neighborhood component analysis (NCA) to classify true AE signals from noise-induced signals. By using the optimized network parameters, a high classification accuracy of 97% and 96.29% was achieved during the training and testing phases, respectively. To check the reliability of the proposed AE filtering model in the real world, it was evaluated and verified using source location AE activities collected during a four-point bending test on a shear-deficient beam. The outcomes suggest that the proposed AE filtration model has the potential to accurately classify AE signals with an accuracy of 92.8% and proved that the filtration model provides accurate and valuable insight into source location determination.

Self-supervised motor imagery EEG recognition model based on 1-D MTCNN-LSTM network

Objective. Aiming for the research on the brain–computer interface (BCI), it is crucial to design a MI-EEG recognition model, possessing a high classification accuracy and strong generalization ability, and not relying on a large number of labeled training samples. Approach. In this paper, we propose a self-supervised MI-EEG recognition method based on self-supervised learning with one-dimensional multi-task convolutional neural networks and long short-term memory (1-D MTCNN-LSTM). The model is divided into two stages: signal transform identification stage and pattern recognition stage. In the signal transform recognition phase, the signal transform dataset is recognized by the upstream 1-D MTCNN-LSTM network model. Subsequently, the backbone network from the signal transform identification phase is transferred to the pattern recognition phase. Then, it is fine-tuned using a trace amount of labeled data to finally obtain the motion recognition model. Main results. The upstream stage of this study achieves more than 95% recognition accuracy for EEG signal transforms, up to 100%. For MI-EEG pattern recognition, the model obtained recognition accuracies of 82.04% and 87.14% with F1 scores of 0.7856 and 0.839 on the datasets of BCIC-IV-2b and BCIC-IV-2a. Significance. The improved accuracy proves the superiority of the proposed method. It is prospected to be a method for accurate classification of MI-EEG in the BCI system.