Search

• A novel optimization algorithm integrating random and grid search mechanisms as function calls. • A cost-effective low inference time deep learning model for detecting 2D ionospheric plasma structures from all-sky airglow images. • The issue of uncertainty quantification has been addressed using the approach of bootstrapping. The ionosphere, part of the Earth's atmosphere, where different plasma irregularities/structures are generated through various electrodynamical processes. Airglow imaging is one of the tools to observe various morphological features of these irregularities. We aim to help all-sky airglow imaging-based ionospheric research by proposing novel deep learning at the edge framework for the detection of different types of mid-latitude ionospheric plasma structures (medium scale traveling ionospheric disturbances with single and multiple bands, and plasma bubbles). The primary challenge was to find an optimal deep neural network by considering the trade-off between accuracy and inference time. In order to address this, a novel hyperparameter optimization technique has been used that integrates the approaches of random and grid search mechanisms as a compact function. The random function generates the subspace for hyperparameters to check the convergence of the model while the grid search creates possible combinations to tune these hyperparameters. Our novel optimization method for deep learning inference at the edge led us to a low-cost, high-accuracy convolutional neural network (CNN) model, which outperformed the complex state-of-the-art deep learning models such as Inception-v3, DenseNet169, VGG16, and VGG19. For airglow image-based plasma structure detection, the proposed model recorded an accuracy of 99.9% with an inference time of 5.8 s. This was an improvement of about 4% in accuracy while 85% reduction in inference time over the best-performing baseline. There was about 93% reduction in the number of model parameters with respect to the best-performing baseline. Uncertainty quantification has also been performed in the present work through bootstrapping to validate the robustness of the proposed model. The findings of our study show the promise of deep learning at the edge for all-sky airglow imaging systems by demonstrating an alternate low-cost, high-accuracy deep neural network. [ABSTRACT FROM AUTHOR]

One of the significant challenges in Brain–Computer Interface (BCI) is to develop a classifier that can decode users' mental states based on electroencephalogram (EEG) data collected from independent subjects. The focus of such subject-independent (SI) classification is justified because it can lead to BCIs that do not require a user-specific calibration process. In recent years, the emergence of deep neural networks (DNNs) has significantly improved the performance of EEG classification. Among various deep learning techniques, the training efficiency and the performance of Convolutional Neural Networks (CNNs), in particular, have led to several state-of-the-art architectures for accurate classification of EEG. Not surprisingly, the efforts to improve the performance of these architectures for EEG classification have been ramped up in recent years. In this regard, a trivial approach is to train and tune a large number of architectures and hyperparameters and hope to improve upon the existing results. In contrast with this ad hoc approach, here we put forward a systematic method inspired by the jackknife estimation to improve the performance of existing CNN architectures. Using EEGNet and ShallowConvNet as archetypical, our empirical results show that the proposed "Delete-a-Subject Jackknife" (DASJ) technique can potentially improve the performance of existing CNN architectures for SI classification of EEG. • Proposes DASJ-CNN which is a jackknife-inspired approach with Convolutional Neural Network (CNN) base classifiers. • Presents an analytical motivation of the DASJ ensemble classification. • Evaluates the effectiveness of the method using two state-of-the-art CNN architectures for EEG classification. • Demonstrates the capability of the method to improve the performance of existing CNN architectures. [ABSTRACT FROM AUTHOR]

High angular resolution diffusion imaging (HARDI) is a promising method for advanced analysis of brain microstructure. However, comprehensive HARDI analysis requires multiple acquisitions of diffusion images (multi-shell HARDI), which is time consuming and often impractical in clinical settings. This study aimed to establish neural network models that can predict new diffusion datasets from clinically feasible brain diffusion MRI for multi-shell HARDI. The development included 2 algorithms: multi-layer perceptron (MLP) and convolutional neural network (CNN). Both followed a voxel-based approach for model training (70%), validation (15%), and testing (15%). The investigations involved 2 multi-shell HARDI datasets: 1) 11 healthy subjects from the Human Connectome Project (HCP); and 2) 10 local subjects with multiple sclerosis (MS). To assess outcomes, we conducted neurite orientation dispersion and density imaging using both predicted and original data and compared their orientation dispersion index (ODI) and neurite density index (NDI) in different brain tissues with 2 measures: peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). Results showed that both models achieved robust predictions, which provided competitive ODI and NDI, especially in brain white matter. The CNN outperformed MLP with the HCP data on both PSNR (p < 0.001) and SSIM (p < 0.01). With the MS data, the models performed similarly. Overall, the optimized neural networks can help generate non-acquired brain diffusion MRI, which will make advanced HARDI analysis possible in clinical practice following further validation. Enabling detailed characterization of brain microstructure will allow enhanced understanding of brain function in both health and disease. [ABSTRACT FROM AUTHOR]

In recent years, deep learning super-resolution models for progressive reconstruction have achieved great success. However, these models which refer to multi-resolution analysis basically ignore the information contained in the lower subspaces and do not explore the correlation between features in the wavelet and spatial domain, resulting in not fully utilizing the auxiliary information brought by multi-resolution analysis with multiple domains. Therefore, we propose a super-resolution network based on the wavelet multi-resolution framework (WMRSR) to capture the auxiliary information contained in multiple subspaces and to be aware of the interdependencies between spatial domain and wavelet domain features. Initially, the wavelet multi-resolution input (WMRI) is generated by combining wavelet sub-bands obtained from each subspace through wavelet multi-resolution analysis and the corresponding spatial domain image content, which serves as input to the network. Then, the WMRSR captures the corresponding features from the WMRI in the wavelet domain and spatial domain, respectively, and fuses them adaptively, thus learning fully explored features in multi-resolution and multi-domain. Finally, the high-resolution images are gradually reconstructed in the wavelet multi-resolution framework by our convolution-based wavelet transform module which is suitable for deep neural networks. Extensive experiments conducted on two public datasets demonstrate that our method outperforms other state-of-the-art methods in terms of objective and visual qualities. • A super-resolution network considering the transform of wavelet multi-resolution. • A convolution-based wavelet module supporting decomposition and reconstruction. • Features extracted from the wavelet domain and spatial domain are fused adaptively. • Extensive experiments on two datasets demonstrate the advantages of our method. [ABSTRACT FROM AUTHOR]

In this paper, a three-step model based on the integration of Deep Neural Networks (DNN) and Decision Models is introduced for image classification which is inspired by the human visual system. To make a decision about an object, many actions should be done in a hierarchical process in the brain. First, the retina receives visual stimuli and transfers them to the visual cortex in the brain. The information extracted in the visual cortex, is accumulated over time to select an appropriate response. Many of the current decision-making models do not show how each image is converted into useful information for the decision model. Some models have used neural networks to convert each image into the information needed in the decision-making model; however, the role of the retina is ignored among these models. In this paper, a combination of retina inspired filters, CNN-based description and accumulator-based decision model is used to classify images. This model's structure resembles the human brain due to the usage of the DoG filter bank as retina inspired filter in the first stage of it. This model shows a significant improvement in accuracy in comparison to other models; furthermore, its performance is acceptable even with the small sample training set. [ABSTRACT FROM AUTHOR]

This work explores the use of constant-Q transform based modulation spectral features (CQT-MSF) for speech emotion recognition (SER). The human perception and analysis of sound comprise of two important cognitive parts: early auditory analysis and cortex-based processing. The early auditory analysis considers spectrogram-based representation whereas cortex-based analysis includes extraction of temporal modulations from the spectrogram. This temporal modulation representation of spectrogram is called modulation spectral feature (MSF). As the constant-Q transform (CQT) provides higher resolution at emotion salient low-frequency regions of speech, we find that CQT-based spectrogram, together with its temporal modulations, provides a representation enriched with emotion-specific information. We argue that CQT-MSF when used with a 2-dimensional convolutional network can provide a time-shift invariant and deformation insensitive representation for SER. Our results show that CQT-MSF outperforms standard mel-scale based spectrogram and its modulation features on two popular SER databases, Berlin EmoDB and RAVDESS. We also show that our proposed feature outperforms the shift and deformation invariant scattering transform coefficients, hence, showing the importance of joint hand-crafted and self-learned feature extraction instead of reliance on complete hand-crafted features. Finally, we perform Grad-CAM analysis to visually inspect the contribution of constant-Q modulation features over SER. • Use of modulation spectral feature that represents cortical processing of speech in humans. • Introduction of constant-Q transform-based modulation spectrum feature. • Performance comparison showing large improvement with the proposed feature over standard features. • Transform domain analysis confirming the importance of low-frequency speech information for emotion recognition. [ABSTRACT FROM AUTHOR]

Pruning is a useful technique for decreasing the memory consumption and floating point operations (FLOPs) of deep convolutional neural network (CNN) models. Nevertheless, at modest pruning levels, current structured pruning approaches often lead to considerable declines in accuracy. Furthermore, existing approaches often treat pruning rates as super parameters, neglecting the sensitivity of different convolution layers. In this study, we propose a novel sensitivity-based method for channel pruning that utilizes second-order sensitivity as a criterion. The essential concept is to prune insensitive filters while retaining sensitive ones. We quantify the sensitivity of the filter using the sum of the sensitivities of all weights in the filter, rather than the magnitude-based metric frequently applied in the literature. Furthermore, a layer sensitivity approach based on the Hessian eigenvalues of each layer is introduced into the process of automatically choosing the most appropriate pruning rate for each layer. Experiments on a variety of modern CNN architectures demonstrate that we can considerably enhance the pruning rate while sacrificing a small amount of accuracy, resulting in a reduction of more than 60% in FLOPs on CIFAR-10. Notably, on ImageNet, pruning based on ResNet50 decreased the FLOPs by 56.3% while losing only 0.92% of accuracy. [ABSTRACT FROM AUTHOR]

Lithium-ion batteries have been used in all aspects of life for environmental and resource reasons. The accurate estimation of the state of charge (SOC) ensures the proper operation of the battery. However, few methods have focused on the problem of SOC estimation at low temperatures. A hybrid method based on the CNN-BWGRU network is proposed in this paper. The method optimizes the influence of battery information on the results through a "multi-moment input" structure and the bidirectional network. The convolutional neural network (CNN) is used to learn the feature parameters in the input. The bidirectional weighted gated recurrent unit (BWGRU) can improve the fitting performance of the network at low temperatures by changing the weights. The proposed network has a strong generalization capability, estimation accuracy, and robustness. SOC estimation is performed under different conditions to verify the plausibility of the network. The experimental results show that the method has higher accuracy and stability than other networks. In addition, the proposed method can overcome the effect of different initial SOC on the estimation results. Therefore, the CNN-BWGRU network provides a new method in battery SOC estimation while providing helpful guidance for safe and stable battery operation in natural environments. [Display omitted] [ABSTRACT FROM AUTHOR]

The growing threat of global climate change stemming from the huge carbon footprint left behind by fossil fuels has prompted interest in exploring and utilizing renewable energy resources. Several statistical, Machine and Deep Learning techniques exist and have been used for many years for a range of forecasting problems. This study is based on the data recorded for 4 years and 9 months using precise instruments, in Islamabad, Pakistan. For this purpose we use statistical and Deep Learning architectures for forecasting solar Global Horizontal Irradiance which not only helps in grid management and power distribution, but also brings attention towards the potential of solar power production in Pakistan and its part to play in tackling global climate change. We have used statistical methods such as Seasonal Auto-Regressive Integrated Moving Average Exogenous (SARIMAX) and Prophet, and Machine Learning methods such as Long Short-Term Memory (LSTM) which is an extension of Recurrent Neural Networks (RNN), Convolutional Neural Network (CNN) and Artificial Neural Network (ANN). The selected forecast methods in our study are based on their ability to work with time series data and we have used different models configurations to see which performs best for our dataset. The performance of every model is studied using different error metrics such as Coefficient of Determination (R2), Mean Absolute Error (MAE), Mean Square Error (MSE), and Root Mean Square Error (RMSE). The major contribution of this study is the data collected to carry out research towards the goal of renewable energy future, and from the test methods used on the data in this study, it can be intuitively determined that ANN, CNN, and LSTM architectures perform best for short-term forecasts, while SARIMAX and Prophet are efficient for long-term forecasts. [ABSTRACT FROM AUTHOR]

• Hybrid multi-scale block is constructed. • Weighted soft-voting rule of decision fusion strategy is proposed. • Fault diagnosis method driven by multi-scale information fusion is developed. The ability of engineering systems to process multi-scale information is a crucial requirement in the development of an intelligent fault diagnosis model. This study develops a hybrid multi-scale convolutional neural network model coupled with multi-attention capability (HMS-MACNN) to solve both the inefficient and insufficient extrapolation problems of multi-scale models in fault diagnosis of a system operating in complex environments. The model's capabilities are demonstrated by its ability to capture the rich multi-scale characteristics of a gearbox including time and frequency multi-scale information. The capabilities of the Multi-Attention Module, which consists of an adaptive weighted rule and a novel weighted soft-voting rule, are respectively integrated to efficiently consider the contribution of each characteristic with different scales-to-faults at both feature- and decision-levels. The model is validated against experimental gearbox fault results and offers robustness and generalization capability with F1 value that is 27% higher than other existing multi-scale CNN-based models operating in a similar environment. Furthermore, the proposed model offers higher accuracy than other generic models and can accurately assign attention to features with different scales. This offers an excellent generalization performance due to its superior capability in capturing multi-scale information and in fusing advanced features following different fusion strategies by using Multi-Attention Module and the hybrid MS block compared to conventional CNN-based models. [ABSTRACT FROM AUTHOR]

Harmful algal bloom (HAB) caused by Magalefidinium polykrikoides becomes frequent in Korean coastal waters during the mid-1990s and is now annual events on the southern coast of Korea. HAB often leads to high rates of fish mortality and subsequent economic losses in aquaculture. In addition, non-harmful algal blooms (non-HABs) caused by the dinoflagellate Alexandrium sp., Mesodinium rubrum , and the diatom Skeletonema sp. occur simultaneously in time and space. Because HAB and non-HABs are difficult to discriminate using multi-band satellite data, most previous studies have attempted only detection or qualitative classification with limited data. In contrast, in this current study, we aimed to quantitatively discriminate M. polykrikoides bloom associated HAB from non-HABs around the southern coast of Korea using a convolutional neural network (CNN) model with Sentinel-3 Ocean and Land Colour Instrument (OLCI) imagery with a spatial resolution of 300 m and 16 spectral bands for the first time. To identify the effect of non-HAB patches on the performance of the CNN model, five CNN models were trained with OLCI images as input and ground-truth HAB maps as output data. The appropriate figure-of-merits values (FOMs) with sensitivity of 0.53, precision of 0.92, and F-measure of 0.67 were reasonable when the CNN model trained using the dataset with the highest ratio of non-HABs patches was applied to HAB images. Even when non-HAB images were applied to the models, the CNN model exhibited the lowest error pixel count. Therefore, we confirmed that the CNN model, which can discriminate red tide blooms with subtle differences between the spectrum bands and spatial characteristics, helps solve the complexity and ambiguity in discriminating HAB from non-HABs. [ABSTRACT FROM AUTHOR]

• A novel application of hybrid method based on EDA for CNN training in the text categorization task. • The CNN training is divided into two parts, allowing the combination of several algorithms for weights optimization. • Accuracy of optimization via EDA improves in comparison with two gradient-based methods. Convolutional Neural Networks for text categorization allows the extraction of features from the text represented through word embedding. The high dimensionality of the texts themselves implies a larger number of network parameters and a more complex optimization surface. Artificial neural network training is an NP-Hard optimization problem, which has been addressed by methods based on partial derivatives of the objective function and presents several theoretical and practical limitations, such as the probability of convergence to local minimums. In this work, we propose a hybrid method based on the Estimation of Distribution Algorithms for training a Convolutional Neural Network. For this, we train together gradient-based methods with the Estimation of Multivariate Normal Algorithm and Univariate Marginal Distribution Algorithm by dividing the training process into two stages. The different variants obtained with the proposed method are compared with gradient-based methods on public benchmark datasets and statistical differences are analyzed by nonparametric tests. The proposed method increases the accuracy of the convolutional network applied to the text categorization task and overcome in about 0.22%–24% the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Currently, the most widely used site response analysis methods include linear, equivalent linear and nonlinear approaches. However, the results of these analyses gradually deviate from the true values as the site softens or the earthquake intensity increases. To overcome this issue, this study employs a 1D Convolutional Neural Network model to simulate the relationships between the surface and downhole motions. To ensure comprehensive coverage of various sites and earthquake conditions, a careful selection of stations and seismic intensities resulted in 56 stations and 34,538 seismic records from KiK-net. Additionally, 30 seismic intensity measures and three site condition parameters were chosen. The convolutional operation was utilized to consider the correlations between the parameters, enabling the prediction of surface motion intensity measures. The performance of the neural network model was assessed based on residual analysis, and the model's generalization ability was tested using samples that were not included in the training dataset. The results demonstrate that the neural network can effectively predict surface motion intensity measures. Furthermore, in comparison with the equivalent linear and nonlinear methods that do not rely on site-specific detailed soil data, the 1D Convolutional Neural Network exhibits superior performance and maintains high prediction accuracy even for soft soil and strong earthquake scenarios, confirming the enhanced predictive capability of the proposed model. [ABSTRACT FROM AUTHOR]

Date fruit (Phoenix dactylifera L.) is known for its diversity in terms of genotypes that differ in shape, color, texture, taste, and other physical and sensory attributes, in addition to biochemical compounds. This richness generates complex, long, and tedious work at the scale of farm as well as in handling, processing, and storage facilities to prepare standardized, and homogenous final products that deal with consumer preferences and needs. To alleviate this, an increased importance and particular trend were observed recently for the use of artificial intelligence (AI) techniques and computer vision technology since they are considered as accurate, objective, non-destructive, fast, and efficient emerging technologies. In this context, this manuscript provides comprehensive insights into key applications of AI techniques and computer vision technology, using different classification models and algorithms developed on the basis (for example) of neural network classifiers, such as convolutional and artificial neural network, on harvest and postharvest assessment of date fruit. Thus, the importance of date fruit dataset used in the AI domain was discussed with link to all accessible datasets in the field of date palm. In addition, available and relevant AI tools and approaches applied to the postharvest of date fruit from maturity distinguishing until storage through other key handling operations, sorting, grading, etc., were detailed in the five sections of this paper. In terms of conclusion, this review showed that these smart technologies can play a substantial role and offer new opportunities to improve the whole value chain of date palm sector. • Harvesting and post-harvest handling impact quality and storability of dates. • These operations are tedious and subjective leading to quantitative/qualitative losses. • Artificial intelligence (AI) techniques are emergent technologies that can reduce these losses. • AI techniques are more accurate, fast, non-destructive and objective. [ABSTRACT FROM AUTHOR]

A new model of wind pressure prediction for low-rise buildings is built based on deep convolutional neural network (CNN), and the model is trained and tested by the aerodynamic data from international open database. The prediction accuracy is compared with that of the literature, and then comprehensive error analysis on all 335 predicted taps under 3 untrained wind directions are carried out for mean and RMS pressure coefficients by various error metrics. The study shows that the distributions of mean and RMS coefficients predicted by the CNN model are closer to the experimental results in comparison with those predicted by the traditional artificial neural network (ANN) model and the common deep neural network (DNN) model without the convolution and pooling layers. The relative error on the corner tap is 13.5%, 4.3% and 0.1% by the ANN, DNN, and CNN models respectively. The mean-square-error (MSE) of the corner bay and the whole roof are also eminently reduced by the CNN model. By introducing the convolutional and pooling layers, more prediction errors are confined within 5% and basically no errors exceed 20%. Consequently, the CNN algorithm can be applied in solving various wind pressure prediction problems on buildings in the future work. • Deep convolutional neural network is introduced in wind pressure prediction. • Prediction accuracy is higher than that of common deep neural network in literature. • The errors on the corner tap, corner bay or the whole roof are all reduced. • Relative errors exceeding 20% disappear for both mean and RMS pressure coefficients. [ABSTRACT FROM AUTHOR]

Existing methods for simulating human visual attention primarily focus on 2D displays and limited research has been conducted on predicting visual attention in three-dimensional (3D) light field content. 3D light field displays provide a heightened sense of stereoscopic realism to viewers. To ensure that the content of the 3D light field display appears more consistent with human visual characteristics, we proposed a novel method for predicting human eye fixation in 3D light field display images. Firstly, we collected real eye movement data and utilized it to create an eye movement dataset based on 3D light field display images. This solves the problem of missing datasets in the field of human gaze based on three-dimensional light field images. Then, we proposed a convolutional neural network model with multiple inputs and outputs, integrating attention modules. This model was trained and used to predict eye fixation within the constructed eye movement dataset. A correlation exists between predicted human gaze of multiple distinct views of same light field image. Finally, we predicted the human gaze area of light field multi-view images based on our model. Experimental results demonstrate that our model accurately predicts human gaze regions across different views of a 3D light field image. The human gaze predicted by the model on each view is basically consistent and relatively accurate. By leveraging proposed method, we can effectively anticipate where viewers will focus their attention on the 3D light field display, which is beneficial for targeted improvement of 3D light field display content. • A convolutional network model suitable for predicting 3D light field images is proposed. • The model has a multi-input multi-output structure and a fused attention module, which improves prediction accuracy. • An eye movement dataset based on 3D light field device is constructed. • Design a data annotation method based on light field distribution to ensure the accuracy of the dataset. [ABSTRACT FROM AUTHOR]

Alternating current optimal power flow (OPF) analysis is critical for efficient and reliable operation of power systems. For large systems or repetitive computations, the traditional methods such as the direct and gradient methods, or non-traditional methods, such as the genetic algorithm and simulating annealing, are time-consuming and unsuitable for real-time computing. The work in this paper proposes a novel framework to obtain the optimal solution of power flow in real-time using a combination of convolutional neural networks and a self-attention mechanism. All parameters of the power networks are rearranged in an image-like shape of a multi-channel image where each channel is a two-dimensional matrix. The proposed approach is adaptive with every input size of power systems as well as frequent variations of network topologies without intervention to the framework core. The encompassment of all power system contexts in which all parameters of internal elements, generation costs, and topology information are included, contributes to the higher accuracy of inference compared to other current machine-learning-based OPF-solving methods. Besides, the proposed framework established on ubiquitous platforms is effortlessly integrated into current infrastructures of power systems, and the great efficiency along with the computation speed may serve as a critical point for practical implications, such as enabling faster decision-making during real-time operations, predicting system contingencies, and remedial actions based on an offline pre-trained model. This supervised learning process is applied to the dataset of four case studies of meshed power systems: the IEEE 5-bus system (IEEE-5), the IEEE 30-bus system (IEEE-30), the IEEE 39-bus system (IEEE-39), and the IEEE 57-bus system (IEEE-57) to prove the efficacy of the proposed method. • Developing a novel approach to solve the OPF problems in real-time. • Wrap power system characteristics under a mold of an image-like shape. • Overcome the outburst of trainable parameters when training with large-scale systems. • Do not require looking up a new set of hyperparameters or re-coding a framework. • All time-consuming computations are shifted to the offline training process. [ABSTRACT FROM AUTHOR]

Flow regime identification is crucial for design and optimization of gas-solid fluidized bed devices. However, how to quickly and accurately identify flow regimes remains a formidable challenge. This study presents a machine learning-aided method using convolutional neural network (CNN) techniques aiming to identify gas-solid flow regimes efficiently and precisely. A database of images depicting various flow regimes is constructed through highly resolved simulations. Subsequently, we compare and evaluate the performance of six commonly used CNNs, namely LeNet-5, AlexNet, VGG-16, InceptionV1, ResNet-50, and MobileNetV2, in identifying typical gas-solid flow regimes. The results demonstrate that, compared to conventional methods, CNNs can identify flow regimes quickly and accurately. Among them, AlexNet achieves an identification accuracy of 100% and an identification speed of 55.86 ms per image. This study enhances the comprehension of gas-solid flow mechanisms, thereby fostering the advancement of reactor engineering. [Display omitted] • The convolutional neural network (CNN) techniques are adopted. • Six convolutional neural network models are used to identify flow regimes. • The highly resolved fine-grid simulations are used to generated images. • Flow regimes can be identified quickly and accurately by using CNN models. [ABSTRACT FROM AUTHOR]

Buried pipelines is easily subjected to external corrosion affected by complex underground environment, such as soil resistivity, pH, dissolved ion concentration, water content and coatings condition. Therefore, external corrosion rate prediction is essential to ensure pipelines intrinsic safety. Generally numerous researches focused on mathematical models and degradation analysis are limited to considering comprehensive corrosion factors. In this work, a novel proposed deep learning method involves a deep neural network (DNN) and attention mechanism which has good robust generalization ability. DNN was used to predict external corrosion rate of buried pipelines with extracting corrosion characteristics data, and attention mechanism is carried on weighting operations of corrosion factors furtherly. The results show that main controlling factors are pipeline age and soil bulk density, sulfate ion concentration and redox potential have less effect by analyzing external corrosion factors dataset, and data reduction and standardization of DNN-attention mechanism model are carried out. The external corrosion rate value predicted by the DNN-attention mechanism model is the closest prediction with the real truth, which the evaluation index root mean square error (RMSE) is 0.0064, and prediction accuracy is 91.91 % compared with convolutional neural network (CNN), CNN-residual, DNN. It proves that it is reasonable to predict pipeline external corrosion rate using DNN-attention mechanism based on publicly pipeline corrosion dataset. • A deep learning method for predicting the external corrosion rate of buried pipelines is proposed. • The importance and correlation of corrosion parameters are assessed to achieve data dimensionality reduction. • The attention mechanism improves the model performance by evaluating the importance of each feature channel. • The proposed composite prediction model has strong generalization ability and high prediction accuracy. [ABSTRACT FROM AUTHOR]

The accurate state of health (SOH) estimation of lithium-ion batteries (LIBs) is crucial for the operation and maintenance of new energy electric vehicles. To address this current problem, an improved hybrid neural network model for SOH prediction based on a sparrow search algorithm (SSA) optimized convolutional bi-directional long short-term memory neural network (CNN-Bi-LSTM) is proposed. Firstly, by analyzing the battery aging data, several feature indicators with highly correlated battery life degradation are constructed. Secondly, the CNN-Bi-LSTM model is used to extract the battery aging data features and the latent timing laws. Finally, the SSA optimizes the parameters to improve the model accuracy. Experimental results based on the NASA-Pcoe battery dataset show that the SSA-CNN-Bi-LSTM model outperforms other models, and the root-mean-square errors of the SOH prediction results are all less than 0.6 %. It indicates that the proposed SSA-CNN-Bi-LSTM model is capable of predicting SOH accurately and with high precision. [ABSTRACT FROM AUTHOR]

The effective detection of synthetic antioxidants by surface-enhanced Raman scattering (SERS) is still challenging due to the presence of phenolic hydroxyl groups in their molecular structures which is difficult to be adsorbed on the conventional SERS substrates, resulting in weak Raman scattering signals. A diazo derivatization reaction was introduced to convert antioxidant molecules into azo derivatives through a diazo derivatization reaction, enabling the amplification of weak Raman signals via the N=N double bond with strong vibrational modes. And excellent linearity was achieved for the antioxidants by a deep learning algorithm called the convolutional neural networks regression model to automatically extract features from the entire spectrum and reduce the signal fluctuation mitigation. [Display omitted] • Diazo derivatization boosts Raman signals of antioxidants by inducing strong N=N vibrations. • A CNN regression model was applied to facilitate the automatic extraction of features from the spectrum and obtain the enlarged linear range. • The SERS method with diazo derivatization detects antioxidants quickly and sensitively in 20 mins. Synthetic antioxidants serve as essential protectors against oxidation and deterioration of edible oils, however, prudent evaluation is necessary regarding potential health risks associated with excessive intake. The direct adsorption of antioxidants onto conventional surface-enhanced Raman scattering (SERS) substrates is challenging due to the presence of phenolic hydroxyl groups in their molecular structures, resulting in weak Raman scattering signals and rendering direct SERS detection difficult. In this study, a diazo derivatization reaction was employed to enhance SERS signals by converting antioxidant molecules into azo derivatives, enabling the amplification of the weak Raman scattering signals through the strong vibrational modes induced by the N = N double bond. The resulting diazo derivatives were characterized using UV–visible absorption and infrared spectroscopy, confirming the occurrence of diazo derivatization of the antioxidants. The proposed method successfully achieved the rapid detection of three commonly used synthetic antioxidants, namely butylated hydroxyanisole (BHA), tert -butylhydroquinone (TBHQ), and propyl gallate (PG) on interfacial self-assembled gold nanoparticles. Furthermore, rapid predictions of BHA, PG, and TBHQ within the concentration range of 1 × 10−6 to 2 × 10−3 mol/L were achieved by integrating a convolutional neural network model. The predictive range of this model surpassed the traditional quantitative method of manually selecting characteristic peaks, with linear coefficients (R2) of 0.9992, 0.9997, and 0.9997, respectively. The recovery of antioxidants in real soybean oil samples ranged from 73.0 % to 126.4 %. Based on diazo derivatization, the proposed SERS method eliminates the need for complex substrates and enables the analysis and determination of synthetic antioxidants in edible oils within 20 min, providing a convenient analytical approach for quality control in the food industry. [ABSTRACT FROM AUTHOR]

In personal identity recognition systems, detecting a person's age, gender, and language using voice signal characteristics is a crucial issue, especially because of the importance of security considerations. Age, gender, and language classification problems are important in signal processing because they are used to analyze and understand human behavior, interactions, and preferences. This can be especially useful in the fields of human-computer interaction, psychology, and social science research. In this paper, a new system for detecting a speaker's age, gender, and language based on deep learning models is presented. Deep learning models have shown great efficacy in various fields of signal processing. For this paper, a range of deep models were tested, including convolutional neural networks (CNNs), recurrent neural network (RNN), and a fine-tuning ResNet34 architecture. Additionally, techniques such as transfer learning were applied to improve the efficiency of the proposed system. The input voice signals are preprocessed by applying the spectro-temporal transform to obtain additional features that can be fed to the ResNet34 model, which is designed specifically for the task of voice signal processing. The dataset used in this paper was sourced from the Mozilla common voice initiative, which is dedicated to advancing speech recognition and language identification technologies. The performance of the proposed algorithm was evaluated in the presence of Gaussian noise to determine its robustness. The experimental results demonstrated that the proposed algorithm significantly outperformed basic algorithms and other deep neural networks in terms of age and gender recognition from voice signals. [ABSTRACT FROM AUTHOR]

Cropping an image is a common image editing technique that aims to find viewpoints with suitable image composition. It is also a frequently used post-processing technique to reduce the evidence of tampering in an image. Detecting cropped images poses a significant challenge in the field of digital image forensics, as the distortions introduced by image cropping are often imperceptible to the human eye. Although deep neural networks achieve state-of-the-art performance, due to their ability to encode large-scale data and handle billions of model parameters. However, due to their high computational complexity and substantial storage requirements, it is difficult to deploy these large deep learning models on resource-constrained devices such as mobile phones and embedded systems. To address this issue, we propose a lightweight deep learning framework for cropping detection in spatial domain, based on knowledge distillation. Initially, we constructed four datasets containing a total of 60,000 images cropped using various tools. We then used Efficient-Net-B0, pre-trained on ImageNet with significant surgical adjustments, as the teacher model, which makes it more robust and faster to converge in this downstream task. The model was trained on 20,000 cropped and uncropped images from our own dataset, and we then applied its knowledge to a more compact model called the student model. Finally, we selected the best-performing lightweight model as the final prediction model, with a testing accuracy of 98.44% on the test dataset, which outperforms other methods. Extensive experiments demonstrate that our proposed model, distilled from Efficient-Net-B0, achieves state-of-the-art performance in terms of detection accuracy, training parameters, and FLOPs, outperforming existing methods in detecting cropped images. • First lightweight knowledge distillation-based deep learning framework for cropped images in spatial domain, suitable for deployment on edge devices. • Created four distinct datasets with 60,000 cropped images, the first publicly available dataset for cropped image detection. • Proposed framework outperformed state-of-the-art methods in detection accuracy, training parameters, and FLOP on various cropped image datasets. • Experiments on diverse datasets demonstrate greater robustness of our proposed framework in real-world scenarios. [ABSTRACT FROM AUTHOR]

The most studied sub-category of the retrovirus is the human immunodeficiency virus (HIV), which is a type of virus of the Retroviridae family. The HIV integration site (HIV IS)/ integration sites (HIV ISs) denote a crucial entity in the entire process of infection and its rebound if there is an interruption in therapy. It determines the steps involved in the formation of latent viral reserve. This work proposes a very deep neural network framework, where each of the layers of the multi-layered network is itself a neural network (NN). The attention mechanism is used for the extraction of the importance of positions in terms of an attention map. This framework will use the Capsule Network along with the attention mechanism to increase the explainability about the presence of local features. Convolutional neural networks (CNNs), which are specialized for image-based recognition and classification, have many drawbacks with one of these being its invariance to translation. This drawback has been overcome by the Capsule Networks. The proposed model also identifies the HIV ISs achieving a performance better than the State-of-the-art methods. Support Vector Machine (SVM), Random Forest (RF) and Logistic Regression (LR) classifiers have been used. Only two state-of-the-art methods are present in the literature that achieve these two goals and a comparison of this work with those two works has been provided here. This work performs way better than the state-of-the-art work in detecting the HIV IS. The comparison is presented here in terms of AUC-ROC, AUC-PR, accuracy, confusion matrix, and F β score. • A very Deep Learning architecture in order to advance towards HIV cure is proposed. • A fusion of capsule network into a multi-layered DL architecture is proposed. • Integration sites (IS) are pivotal in the whole infection process. • IS are unique for each retrovirus. • Proposed architecture identifies positions of HIV IS along with the importance. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have demonstrated expert-level performance in cutaneous tumour classification using clinical images, but most previous studies have focused on dermatologist-versus-CNN comparisons rather than their combination. The objective of our study was to evaluate the potential impact of CNN assistance on dermatologists for clinical image interpretation. A multi-class CNN was trained and validated using a dataset of 25,773 clinical images comprising 10 categories of cutaneous tumours. The CNN's performance was tested on an independent dataset of 2107 images. A total of 400 images (40 per category) were randomly selected from the test dataset. A fully crossed, self-control, multi-reader multi-case (MRMC) study was conducted to compare the performance of 18 board-certified dermatologists (experience: 13/18 ≤ 10 years; 5/18＞10 years) in interpreting the 400 clinical images with or without CNN assistance. The CNN achieved an overall accuracy of 78.45% and kappa of 0.73 in the classification of 10 types of cutaneous tumours on 2107 images. CNN-assisted dermatologists achieved a higher accuracy (76.60% vs. 62.78%, P < 0.001) and kappa (0.74 vs. 0.59, P < 0.001) than unassisted dermatologists in interpreting the 400 clinical images. Dermatologists with less experience benefited more from CNN assistance. At the binary classification level (malignant or benign), the sensitivity (89.56% vs. 83.21%, P < 0.001) and specificity (87.90% vs. 80.92%, P < 0.001) of dermatologists with CNN assistance were also significantly improved than those without. CNN assistance improved dermatologist accuracy in interpreting cutaneous tumours and could further boost the acceptance of this new technique. • A multi-class CNN was developed using 25,773 clinical images. • An MRMC study was conducted to evaluate performance of CNN-assisted dermatologists. • CNN-assisted dermatologists achieved a higher accuracy and kappa than unassisted. • Dermatologists with less experience benefited more from CNN assistance. [ABSTRACT FROM AUTHOR]

Natural gas leakage can impose significant danger on a facility and its surrounding communities. Methods for early detection and diagnosis of such leakages have been developed and widely used for gas pipelines and storage tanks. Most techniques include inspection of sensor-aided mathematical models. Application of machine learning techniques to gas leakage detection has been rarely explored. In the present work, convolutional network (to model spatial likelihood of leak) is combined with bi-directional long short-term memory layer network, or BiLSTM (to model temporal dependence of leak likelihood) to perform leak detection and diagnosis. The developed model was trained and tested using sequence of concentration profiles generated using open-source simulated data. The model learned successfully to predict gas leakage and classify its size. The study also explores the flexibility of this network to perform quick detection and diagnose with the limited data. While the networks did not require parameter adjustments to achieve high prediction accuracy, further optimization is possible through data selection and pre-processing. The model needs to be further tested for wide range of leak scenarios. At its present condition, the combined apph cation of convolutional network and BiLSTM shows promising results for early and accurate leak detection in natural gas facilities. Experimental results are needed to confirm the effectiveness of the model and data uncertainty. [ABSTRACT FROM AUTHOR]

Advances in biomedical artificial intelligence may introduce or perpetuate sex and gender discriminations. Convolutional neural networks (CNN) have proven a dermatologist-level performance in image classification tasks but have not been assessed for sex and gender biases that may affect training data and diagnostic performance. In this study, we investigated sex-related imbalances in training data and diagnostic performance of a market-approved CNN for skin cancer classification (Moleanalyzer Pro®, Fotofinder Systems GmbH, Bad Birnbach, Germany). We screened open-access dermoscopic image repositories widely used for CNN training for distribution of sex. Moreover, the sex-related diagnostic performance of the market-approved CNN was tested in 1549 dermoscopic images stratified by sex (female n = 773; male n = 776). Most open-access repositories showed a marked under-representation of images originating from female (40%) versus male (60%) patients. Despite these imbalances and well-known sex-related differences in skin anatomy or skin-directed behaviour, the tested CNN achieved a comparable sensitivity of 87.0% [80.9%–91.3%] versus 87.1% [81.1%–91.4%], specificity of 98.7% [97.4%–99.3%] versus 96.9% [95.2%–98.0%] and ROC-AUC of 0.984 [0.975–0.993] versus 0.979 [0.969–0.988] in dermoscopic images of female versus male origin, respectively. In the sample at hand, sex-related differences in ROC-AUCs were not statistically significant in the per-image analysis nor in an additional per-individual analysis (p ≥ 0.59). Design and training of artificial intelligence algorithms for medical applications should generally acknowledge sex and gender dimensions. Despite sex-related imbalances in open-access training data, the diagnostic performance of the tested CNN showed no sex-related bias in the classification of skin lesions. • Artificial intelligence systems may introduce sex and gender discriminations. • Convolutional neural networks (CNN) are increasingly used for skin cancer diagnosis. • We investigated CNN training data and diagnostic performance for sex-related biases. • Training data: Images of females were under-represented in open-access archives. • Diagnostic performance: The tested marked-approved CNN showed no sex-related bias. [ABSTRACT FROM AUTHOR]

Energy consumption forecasting is essential for energy system integration and management. However, existing studies mainly focus on temporal features of energy consumption, which neglects the spatial correlation of variables with time information. Capturing the spatio-temporal relationships helps to improve forecasting accuracy and further promote energy dispatch. To tackle this problem, an explainable Convolutional Neural Network-Long Short Term Memory forecasting model is employed to effectively predict the total energy consumption by capturing the spatial and temporal features of multivariate time series. In the model, the autoencoder is used to achieve the nonlinear dimensionality reduction and transfer the data to a low-dimensional space. Furthermore, a Convolutional Neural Network is used to extract more effective features from the decoded data, and long short-term memory is employed to identify the temporal dependencies between extracted features and total energy consumption. Shapley additive explanation is introduced to interpret the outputs of the black-box model. The superior performance of the proposed method with high accuracy and good adaptability is verified by the comparisons with conventional forecasting models. This method provides an insight into the regional energy consumption analyzing contributions of weather variables to energy consumption, which helps administers in understanding regional energy performance for enhancing energy efficiency. • Deep learning techniques are utilized for multivariate time series forecasting. • Deep neural network for spatio-temporal forecasting is proposed. • Shapley additive explanations is introduced to interpret the black-box model. • The superior performance and high-stability of the proposed method were proved. [ABSTRACT FROM AUTHOR]

[Display omitted] • Three types of PD signals (corona, internal, and surface) are collected experimentally and denoised using Symlet4 before extracting input features for classification study. • ML approaches incorporate statistical characteristics, while the CNN model emphasizes CWT-based scalogram images. • Enhancing the PD classification with the help of CWT-based scalogram images using customize CNN model with an appropriate hyperparameter setting. • Classification performance is also analyzed using ML models (ANN, SVM, KNN, and RF) with the proposed CNN model. The purpose of this work is to classify different types of partial discharge (PD) signals including corona, surface, and internal discharges. The collected PD signals are de-noised using symlet4 mother wavelet and then subjected to classification using machine learning (ML) and convolutional neural network (CNN) architecture. Statistical features (skewness, kurtosis, standard deviation, variance, etc.) are extracted which are used to train the ML models, whereas the CNN model is emphasized automated feature extraction from a novel approach based on scalogram images. From the denoised signals, continuous wavelet transforms (CWT) based scalogram images are obtained which highlight the resolution of the signal's energy and are used for feature extraction in the classification study. In this work, PD signals are used for both ML and CNN model training, validation, and testing. Classification results demonstrate that the proposed CNN model achieves a recognition accuracy of up to 100% compared with support vector machines (SVM), artificial neural network (ANN), k- nearest neighbors (KNN), and random forest (RF) based ML classifiers which are 94.26%, 94.26%, 91%, and 95.1% respectively. [ABSTRACT FROM AUTHOR]

[Display omitted] • The first time using CBAM-CNN to predict particle size distribution during fluidized bed granulation. • The CBAM-CNN model was optimized by introducing Bayesian optimization and C-Mixup algorithms. • The performance of CBAM-CNN was compared to other classic models. • The modeling convenience of CBAM-CNN was evaluated. Monitoring the particle size distribution (PSD) is crucial for controlling product quality during fluidized bed granulation. This paper proposed a rapid analytical method that quantifies the D10, D50, and D90 values using a Convolutional Block Attention Module-Convolutional Neural Network (CBAM-CNN) framework tailored for deep learning with near-infrared (NIR) spectroscopy. This innovative framework, which fuses CBAM with CNN, excels at extracting intricate features while prioritizing crucial ones, thereby facilitating the creation of a robust multi-output regression model. To expand the training dataset, we incorporated the C-Mixup algorithm, ensuring that the deep learning model was trained comprehensively. Additionally, the Bayesian optimization algorithm was introduced to optimize the hyperparameters, improving the prediction performance of the deep learning model. Compared with the commonly used Partial Least Squares (PLS), Support Vector Machine (SVM), and Artificial Neural Network (ANN) models, the CBAM-CNN model yielded higher prediction accuracy. Furthermore, the CBAM-CNN model avoided spectral preprocessing, preserved the spectral information to the maximum extent, and returned multiple predicted values at one time without degrading the prediction accuracy. Therefore, the CBAM-CNN model showed better prediction performance and modeling convenience for analyzing PSD values in fluidized bed granulation. [ABSTRACT FROM AUTHOR]

Nutrition plays a vital role in maintaining human health. Traditional methods used for assessing food composition & nutritional content often require destructive sample preparation, which can be time-consuming and costly. Therefore, computer vision-based approaches have emerged as promising alternatives that enable rapid and non-destructive analysis of various nutritional parameters in foods. In this review, we summarized computer vision applications in meat processing, grains, fruits and vegetables, and seafood. We reviewed recent advancements in computer vision and deep learning-based algorithms employed for food recognition and nutrient estimation. Various existing food recognition and nutrient estimation datasets are also reviewed. Conventional methods offer some limitations, while vision-based technologies provide quick and non-destructive analysis of food composition & nutritional content. Computer vision and deep neural network architectures provide remarkable accuracy for food nutrient measurement. In conclusion, deep learning-based models are paving the way for a promising future in nutritional and health optimization research. In the future, vision-based technologies are expected to transform food classification and detection by enabling more rapid, affordable, and accurate nutritional analyses. Therefore, computer vision is developing into a useful tool for fast and precise evaluation of food nutrients without enabling samples to be damaged. [Display omitted] • Integrating nutrition analysis with computer vision for non-destructive food quality assessment are reviewed. • Transformer-based and CNN-based hybrid approaches are the accurate nutrient recognition methods in the food industry. • Vision-based technologies have revolutionized food analysis by facilitating faster, cost-effective, and precise methods. [ABSTRACT FROM AUTHOR]

Heat pipes are highly efficient heat transfer devices that have found widespread use in various fields. However, accurately predicting their heat transfer performance under wide range conditions remains a challenge. In this study, we propose a deep learning-based approach for predicting the heat transfer performance of heat pipes, considering various input parameters such as wick type, nanoparticle type, and operating conditions. A comprehensive dataset was obtained from the literature to increase the range of prediction. The deep learning architectures including Artificial Neural Networks (ANNs), Deep Neural Networks (DNNs), and Convolutional Neural Networks (CNNs) were utilized. The CNN architecture with skip connections demonstrated the best prediction performance, outperforming other architectures considered in this study. The prediction performance of a skip connection-based CNN architecture shows a maximum percentage error of 7 % and a mean absolute percentage error of 0.39 % in the training dataset range. Experimental validation was conducted to assess the model's performance beyond the range of the training dataset. Experimental validation confirmed the accuracy and versatility of the proposed approach with approximately 20 % of the maximum percentage error. The proposed deep learning-based approach shows promise in efficiently predicting the heat transfer performance of heat pipes, offering potential cost and time savings in various industrial applications. • The proposed deep learning-based approach predicts heat pipe heat transfer performance with wick type. • CNN architecture with a skip connection method yields outstanding prediction performance within 7 % of percentage error. • Reasonable prediction performance was observed for the outer range of the dataset within 20 % of percentage error. • This method could reduce heat pipe manufacturing time and costs without compromising validation. [ABSTRACT FROM AUTHOR]

• Neurobiological-inspired CNN models accurately predict sound location. • Models of binaural integration in human brainstem simulate human sound localization. • Objective functions minimizing angular or Euclidean distance lead to distinct errors. • Feature representations of shallow layers resemble brainstem neuronal spatial tuning. • Neurobiological-inspired CNN models generate testable predictions for neuroscience. The human brain effortlessly solves the complex computational task of sound localization using a mixture of spatial cues. How the brain performs this task in naturalistic listening environments (e.g. with reverberation) is not well understood. In the present paper, we build on the success of deep neural networks at solving complex and high-dimensional problems [1] to develop goal-driven, neurobiological-inspired convolutional neural network (CNN) models of human spatial hearing. After training, we visualize and quantify feature representations in intermediate layers to gain insights into the representational mechanisms underlying sound location encoding in CNNs. Our results show that neurobiological-inspired CNN models trained on real-life sounds spatialized with human binaural hearing characteristics can accurately predict sound location in the horizontal plane. CNN localization acuity across the azimuth resembles human sound localization acuity, but CNN models outperform human sound localization in the back. Training models with different objective functions - that is, minimizing either Euclidean or angular distance - modulates localization acuity in particular ways. Moreover, different implementations of binaural integration result in unique patterns of localization errors that resemble behavioral observations in humans. Finally, feature representations reveal a gradient of spatial selectivity across network layers, starting with broad spatial representations in early layers and progressing to sparse, highly selective spatial representations in deeper layers. In sum, our results show that neurobiological-inspired CNNs are a valid approach to modeling human spatial hearing. This work paves the way for future studies combining neural network models with empirical measurements of neural activity to unravel the complex computational mechanisms underlying neural sound location encoding in the human auditory pathway. [ABSTRACT FROM AUTHOR]

Sequential digital dermoscopy (SDD) is applied for early melanoma detection by uncovering dynamic changes of monitored lesions. Convolutional neural networks (CNN) are capable of high diagnostic accuracies similar to trained dermatologists. To investigate the capability of CNN to correctly classify melanomas originally diagnosed by mere dynamic changes during SDD. A retrospective cross-sectional study using image quartets of 59 high-risk patients each containing one melanoma diagnosed by dynamic changes during SDD and three nevi (236 lesions). Two validated CNN classified quartets at baseline or after SDD follow-up at the time of melanoma diagnosis. Moreover, baseline quartets were rated by 26 dermatologists. The main outcome was the number of quartets with correct classifications. CNN-1 correctly classified 9 (15.3%) and CNN-2 8 (13.6%) of 59 baseline quartets. In baseline images, CNN-1 attained a sensitivity of 25.4% (16.1%–37.8%) and specificity of 92.7% (87.8%–95.7%), whereas CNN-2 of 28.8% (18.8%–41.4%) and 75.7% (68.9%–81.4%). Expectedly, after SDD follow-up CNN more readily detected melanomas resulting in improved sensitivities (CNN-1: 44.1% [32.2%–56.7%]; CNN-2: 49.2% [36.8%–61.6%]). Dermatologists were told that each baseline quartet contained one melanoma, and on average, correctly classified 24 (22–27) of 59 quartets. Correspondingly, accepting a baseline quartet to be appropriately classified whenever the highest malignancy score was assigned to the melanoma within, CNN-1 and CNN-2 correctly classified 28 (47.5%) and 22 (37.3%) of 59 quartets, respectively. The tested CNN could not replace the strategy of SDD. There is a need for CNN capable of integrating information on dynamic changes into analyses. • Differentiating early melanomas from atypical nevi is challenging. • Sequential digital dermoscopy helps to uncover the dynamic changes in early melanomas. • CNN are equally accurate at diagnosing skin cancer than trained dermatologists. • The CNN of our study may not replace the strategy of sequential digital dermoscopy. • Development of CNN, including information on lesion evolution, seems indispensable. [ABSTRACT FROM AUTHOR]

Modern complex chemical process operations produce a large amount of process monitoring and control time series data, which are potentially useful to extract valuable operating experiences and manipulating rules to improve the intelligence of process operations. Previous researches show that time series clustering is an effective method to mine historical control sequences. However, the actual working conditions often deviate from the historical data, making it difficult to reconstruct accurate process regulation manipulating strategies. In response to this problem, this paper proposes a manipulation strategies reconstruction method using a convolutional neural network model (MSR-CNN). Therein, the time series are hierarchically clustered according to the Levenshtein distance to obtain different classes of process disturbance states, and the corresponding manipulated sequences fragments obtained are treated with a symbolic aggregation approximation. Then the improved convolutional neural network is used for deep learning and manipulation strategies reconstruction of the process disturbance states and the corresponding manipulating sequences strings. Finally, applying to a numerical example and an ethanol–water distillation tower unit, the proposed method is proved to be effective. Compared with traditional supervisory control methods, our contributions can help well construct complex chemical process control strategies less dependent on human operators' operating experiences. • Agglomerative hierarchical time series clustering based on levenshtein distance. • Fragmentation of effective manipulation sequences. • Deep learning and reconstruction of manipulation strategies with convolutional neural networks. • Favorable comparison with the supervisory control method. [ABSTRACT FROM AUTHOR]

• Developing a new recursive convolutional layer (RCL) able to stop the iteration when the hidden state stabilize. • First network to have different depth for different batch of samples. • A comprehensive study on the best locations and numbers of RCL in a multilayer network called C-FRPN. To overcome problems with the design of large networks, particularly with respect to the depth of the network, this paper presents a new model of convolutional neural networks (CNN) which features fully recursive convolutional layers (RCLs). An RCL is a generalization of the classic one-stage feedforward convolutional layer (CL) to fully direct feedback connections between the outputs of the CL and its inputs. A traditional deep CNN consisting of many CLs, can then be generalized to include some CLs, and some RCLs in the intermediate stages. We call the corresponding network a Convolutional Neural Network with Fully Recursive Perceptron Network (C-FRPN). Through an analysis of results obtained from applications of the C-FRPN to three benchmark image classification datasets: CIFAR-10, SVHN, ISIC, it is found that (i) in general, the performance of a C-FRPN, even with only one RCL, is better than the performance of the corresponding deep CNN with all CLs, under the constraint of having the same number of unknown parameters; (ii) the performance of the C-FRPN varies with respect to (a) where the RCLs are located, and (b) the number of RCLs in the C-FRPN; and, (iii) the effectiveness of the RCLs depends on the size of the training dataset. The results suggest that: (a) it is advisable to use RCLs particularly when training very large sets of data, (b) it is best to prioritize placement of RCLs close to the input layer of the C-FRPN, and (c) it is advisable to increase the number of RCLs as long as the training dataset can sustain without overfitting being observed. [ABSTRACT FROM AUTHOR]

Despite the superiority of convolutional neural networks demonstrated in time series modeling and forecasting, it has not been fully explored on the design of the neural network architecture and the tuning of the hyper-parameters. Inspired by the incremental construction strategy for building a random multilayer perceptron, we propose a novel Error-feedback Stochastic Modeling (ESM) strategy to construct a random Convolutional Neural Network (ESM-CNN) for time series forecasting task, which builds the network architecture adaptively. The ESM strategy suggests that random filters and neurons of the error-feedback fully connected layer are incrementally added to steadily compensate the prediction error during the construction process, and then a filter selection strategy is introduced to enable ESM-CNN to extract the different size of temporal features, providing helpful information at each iterative process for the prediction. The performance of ESM-CNN is justified on its prediction accuracy of one-step-ahead and multi-step-ahead forecasting tasks respectively. Comprehensive experiments on both the synthetic and real-world datasets show that the proposed ESM-CNN not only outperforms the state-of-art random neural networks, but also exhibits stronger predictive power and less computing overhead in comparison to trained state-of-art deep neural network models. [ABSTRACT FROM AUTHOR]

• Developing a new recursive convolutional layer (RCL) able to stop the iteration when the hidden state stabilize. • First network to have different depth for different batch of samples. • A comprehensive study on the best locations and numbers of RCL in a multilayer network called C-FRPN. To overcome problems with the design of large networks, particularly with respect to the depth of the network, this paper presents a new model of convolutional neural networks (CNN) which features fully recursive convolutional layers (RCLs). An RCL is a generalization of the classic one-stage feedforward convolutional layer (CL) to fully direct feedback connections between the outputs of the CL and its inputs. A traditional deep CNN consisting of many CLs, can then be generalized to include some CLs, and some RCLs in the intermediate stages. We call the corresponding network a Convolutional Neural Network with Fully Recursive Perceptron Network (C-FRPN). Through an analysis of results obtained from applications of the C-FRPN to three benchmark image classification datasets: CIFAR-10, SVHN, ISIC, it is found that (i) in general, the performance of a C-FRPN, even with only one RCL, is better than the performance of the corresponding deep CNN with all CLs, under the constraint of having the same number of unknown parameters; (ii) the performance of the C-FRPN varies with respect to (a) where the RCLs are located, and (b) the number of RCLs in the C-FRPN; and, (iii) the effectiveness of the RCLs depends on the size of the training dataset. The results suggest that: (a) it is advisable to use RCLs particularly when training very large sets of data, (b) it is best to prioritize placement of RCLs close to the input layer of the C-FRPN, and (c) it is advisable to increase the number of RCLs as long as the training dataset can sustain without overfitting being observed. [ABSTRACT FROM AUTHOR]

Gastrointestinal cancers account for approximately 20% of all cancer diagnoses and are responsible for 22.5% of cancer deaths worldwide. Artificial intelligence–based diagnostic support systems, in particular convolutional neural network (CNN)–based image analysis tools, have shown great potential in medical computer vision. In this systematic review, we summarise recent studies reporting CNN-based approaches for digital biomarkers for characterization and prognostication of gastrointestinal cancer pathology. Pubmed and Medline were screened for peer-reviewed papers dealing with CNN-based gastrointestinal cancer analyses from histological slides, published between 2015 and 2020.Seven hundred and ninety titles and abstracts were screened, and 58 full-text articles were assessed for eligibility. Sixteen publications fulfilled our inclusion criteria dealing with tumor or precursor lesion characterization or prognostic and predictive biomarkers: 14 studies on colorectal or rectal cancer, three studies on gastric cancer and none on esophageal cancer. These studies were categorised according to their end-points: polyp characterization, tumor characterization and patient outcome. Regarding the translation into clinical practice, we identified several studies demonstrating generalization of the classifier with external tests and comparisons with pathologists, but none presenting clinical implementation. Results of recent studies on CNN-based image analysis in gastrointestinal cancer pathology are promising, but studies were conducted in observational and retrospective settings. Large-scale trials are needed to assess performance and predict clinical usefulness. Furthermore, large-scale trials are required for approval of CNN-based prediction models as medical devices. • Computational medical image analysis is rising in gastrointestinal cancer pathology. • Several studies deal with CNN–based identification of biomarkers on HE slides. • Results are promising but studies are at early stages. • Further studies are needed to assess performance and potential clinical usefulness. [ABSTRACT FROM AUTHOR]

• We analyzed brain imaging and genetic data from more than 15,000 UK Biobank subjects. • We trained a convolutional neural network model to accurately predict brain age. • We identified novel genetic factors significantly associated with brain aging. To study genetic factors associated with brain aging, we first need to quantify brain aging. Statistical models have been created for estimating the apparent age of the brain, or predicted brain age (PBA), using imaging data. Recent studies have refined these models to obtain a more accurate PBA, but research has yet to demonstrate the scientific value of doing so. Here, we show that a more accurate PBA leads to better characterization of genetic factors associated with brain aging. We trained a convolutional neural network (CNN) model on 16,998 UK Biobank subjects to derive PBA, then conducted a genome-wide association study on the PBA, in which we identified single nucleotide polymorphisms from four independent loci significantly associated with brain aging, three of which were novel. By comparing association results based on the CNN-derived PBA to those based on a linear regression-derived PBA, we concluded that a more accurate PBA enables the discovery of novel genetic associations. Our results may be valuable for identifying other lifestyle factors associated with brain aging. [ABSTRACT FROM AUTHOR]

[Display omitted] • A data-driven approach using synchrotron radiation imaging experiments to explore the materials interface of PV ribbon. • Morphological gradient operation identified as suitable data augmentation technique. • Data of augmented images visualized with PCA technique. • Bubble(s) and IMC identifying machine learning model developed using CNN. Designing means and methods to detect the presence of interfacial bubbles and intermetallic compounds (IMCs) during hot dipping solder coating of Cu ribbon, can help in the production of defect-free PV ribbons. A mechanistic study of Cu 6 Sn 5 IMC grain growth and bubble morphology evolution at the solder-substrate interface is performed with phase field simulation. A machine learning model is utilized to identify the occurrence of bubble(s) and IMC at the material interface of liquid solder and solid Cu. The datasets for the microstructural images consisting of bubble(s), IMC and planar solder/Cu interface are generated using in situ synchrotron radiation (SR) imaging experiment techniques. The integration of in situ SR radiography based non-destructive testing experiments with convolutional neural network model to intelligently detect the interfacial microstructures paves the path for potential industrial application of this technique in the smart manufacturing of defect free and reliable PV ribbon material. [ABSTRACT FROM AUTHOR]

Foams are relevant in many food products and food production processes. During the bottling of beverages, foam can severely impair the process as overflowing foam causes underfilled bottles and poses a high contamination risk. Consequently, the filling speed is limited by the foaming properties of the beverage. Several filling experiments with different juices were executed to improve the understanding of the influence of flow and material properties on the foaming behavior. The foam evolution in time was evaluated using a convolutional neural network (CNN) to detect visible foam regions in video recordings of the experiments. The use of image data has the advantage that its acquisition is noninvasive and no supplementary sensor hardware with direct product contact has to be installed into the process. The CNN was able to detect image regions that contain foam with errors of only a few millimeters. The detection worked for different juice colors, regardless of the lighting and noise in the image. Based on the filling speed, viscosity, surface tension, and density, the foaming behavior during the filling process was modeled with a recurrent neural network (RNN). The RNN was able to predict the foam height evolution with average errors below five millimeters. Both models are easily transferable to new use cases by retraining the networks. [ABSTRACT FROM AUTHOR]

• HELP improves the efficiency of hyperparameter tunning with random exploration. • HELP avoids bad hyperparameters in future exploration by learning with an RNN. • HELP finds the optimal hyperparameters at iteration by information during training. Hyperparameter selection is very important for the success of deep neural network training. Random search of hyperparameters for deep neural networks may take a long time to converge and yield good results because the training of deep neural networks with a huge number of parameters for every selected hyperparameter is very time-consuming. In this work, we propose the Hyperparameter Exploration LSTM-Predictor (HELP) which is an improved random exploring method using a probability-based exploration with an LSTM-based prediction. The HELP has a higher probability to find a better hyperparameter with less time. The HELP uses a series of hyperparameters in a time period as input and predicts the fitness values of these hyperparameters. Then, exploration directions in the hyperparameter space yielding higher fitness values will have higher probabilities to be explored in the next turn. Experimental results for training both the Generative Adversarial Net and the Convolution Neural Network show that the HELP finds hyperparameters yielding better results and converges faster. [ABSTRACT FROM AUTHOR]