Exploring AI in metasurface structures with forward and inverse design

As an artificially manufactured planar device, a metasurface structure can produce unusual electromagnetic responses by harnessing four basic characteristics of the light wave. Traditional design processes rely on numerical algorithms combined with parameter optimization. However, such methods are often time-consuming and struggle to match actual responses. This paper aims to give a unique perspective to classify the artificial intelligence(AI)-enabled design, dividing it into forward and inverse designs according to the mapping relationship between variables and performance. Forward designs are driven by intelligent algorithms; neural networks are one of the principal ways to realize reverse design. This paper reviews recent progress in AI-enabled metasurface design, examining the principles, advantages, and potential applications. A rich content and detailed comparison can help build a holistic understanding of metasurface design. Moreover, the authors believe that this systematic and detailed review will pave the way for future research and the selection of practical applications.

Adaptive genetic algorithm based deep feature selector for cancer detection in lung histopathological images

Cancer is a global health concern because of a significant mortality rate and a wide range of affected organs. Early detection and accurate classification of cancer types are crucial for effective treatment. Imaging tests on different image modalities such as Histopathology images, provide valuable insights into the cellular and architectural features of tissues, allowing pathologists to make diagnosis, determine disease stages, and guide treatment decisions. They are an essential tool in the study and understanding of diseases, aiding in research, education, and patient care. Convolutional neural network based pretrained deep learning models can be used successfully to detect lung cancer. In this study, we have used a channel attention-enabled deep learning model as a feature extractor followed by an adaptive Genetic Algorithm (GA) based feature selector. Here, we calculate the fitness score of each chromosome (i.e., a candidate solution) using a filter method, instead of a classifier. Further, the GA optimized feature vector is fed to the K-nearest neighbors classifier for final classification. The proposed method shows a promising result with an overall accuracy of 99.75% on the LC25000 dataset, which is a publicly available dataset of lung histopathological images. The source code for this work can be found https://github.com/priyam-03/GA-Feature-Selector-Lung-Cancer .

PC-CS-YOLO: High-Precision Obstacle Detection for Visually Impaired Safety

The issue of obstacle avoidance and safety for visually impaired individuals has been a major topic of research. However, complex street environments still pose significant challenges for blind obstacle detection systems. Existing solutions often fail to provide real-time, accurate obstacle avoidance decisions. In this study, we propose a blind obstacle detection system based on the PC-CS-YOLO model. The system improves the backbone network by adopting the partial convolutional feed-forward network (PCFN) to reduce computational redundancy. Additionally, to enhance the network's robustness in multi-scale feature fusion, we introduce the Cross-Scale Attention Fusion (CSAF) mechanism, which integrates features from different sensory domains to achieve superior performance. Compared to state-of-the-art networks, our system shows improvements of 2.0%, 3.9%, and 1.5% in precision, recall, and mAP50, respectively. When evaluated on a GPU, the inference speed is 20.6 ms, which is 15.3 ms faster than YOLO11, meeting the real-time requirements for blind obstacle avoidance systems.

AttentionTTE: a deep learning model for estimated time of arrival

Estimating travel time (ETA) for arbitrary paths is crucial in urban intelligent transportation systems. Previous studies primarily focus on constructing complex feature systems for individual road segments or sub-segments, which fail to effectively model the influence of each road segment on others. To address this issue, we propose an end-to-end model, AttentionTTE. It utilizes a self-attention mechanism to capture global spatial correlations and a recurrent neural network to capture temporal dependencies from local spatial correlations. Additionally, a multi-task learning module integrates global spatial correlations and temporal dependencies to estimate the travel time for both the entire path and each local path. We evaluate our model on a large trajectory dataset, and extensive experimental results demonstrate that AttentionTTE achieves state-of-the-art performance compared to other methods.

Prediction of PM2.5 concentration based on a CNN-LSTM neural network algorithm

Fine particulate matter (PM2.5) is a major air pollutant affecting human survival, development and health. By predicting the spatial distribution concentration of PM2.5, pollutant sources can be better traced, allowing measures to protect human health to be implemented. Thus, the purpose of this study is to predict and analyze the PM2.5 concentration of stations based on the integrated deep learning of a convolutional neural network long short-term memory (CNN-LSTM) model. To solve the complexity and nonlinear characteristics of PM2.5 time series data problems, we adopted the CNN-LSTM deep learning model. We collected the PM2.5data of Qingdao in 2020 as well as meteorological factors such as temperature, wind speed and air pressure for pre-processing and characteristic analysis. Then, the CNN-LSTM deep learning model was integrated to capture the temporal and spatial features and trends in the data. The CNN layer was used to extract spatial features, while the LSTM layer was used to learn time dependencies. Through comparative experiments and model evaluation, we found that the CNN-LSTM model can achieve excellent PM2.5 prediction performance. The results show that the coefficient of determination (R2) is 0.91, and the root mean square error (RMSE) is 8.216 µg/m3. The CNN-LSTM model achieves better prediction accuracy and generalizability compared with those of the CNN and LSTM models (R2 values of 0.85 and 0.83, respectively, and RMSE values of 11.356 and 14.367, respectively). Finally, we analyzed and explained the predicted results. We also found that some meteorological factors (such as air temperature, pressure, and wind speed) have significant effects on the PM2.5 concentration at ground stations in Qingdao. In summary, by using deep learning methods, we obtained better prediction performance and revealed the association between PM2.5 concentration and meteorological factors. These findings are of great significance for improving the quality of the atmospheric environment and protecting public health.

AISOA-SSformer: An Effective Image Segmentation Method for Rice Leaf Disease Based on the Transformer Architecture

Rice leaf diseases have an important impact on modern farming, threatening crop health and yield. Accurate semantic segmentation techniques are crucial for segmenting diseased leaf parts and assisting farmers in disease identification. However, the diversity of rice growing environments and the complexity of leaf diseases pose challenges. To address these issues, this study introduces an innovative semantic segmentation algorithm for rice leaf pests and diseases based on the Transformer architecture AISOA-SSformer. First, it features the sparse global-update perceptron for real-time parameter updating, enhancing model stability and accuracy in learning irregular leaf features. Second, the salient feature attention mechanism is introduced to separate and reorganize features using the spatial reconstruction module (SRM) and channel reconstruction module (CRM), focusing on salient feature extraction and reducing background interference. Additionally, the annealing-integrated sparrow optimization algorithm fine-tunes the sparrow algorithm, gradually reducing the stochastic search amplitude to minimize loss. This enhances the model's adaptability and robustness, particularly against fuzzy edge features. The experimental results show that AISOA-SSformer achieves an 83.1% MIoU, an 80.3% Dice coefficient, and a 76.5% recall on a homemade dataset, with a model size of only 14.71 million parameters. Compared with other popular algorithms, it demonstrates greater accuracy in rice leaf disease segmentation. This method effectively improves segmentation, providing valuable insights for modern plantation management. The data and code used in this study will be open sourced at https://github.com/ZhouGuoXiong/Rice-Leaf-Disease-Segmentation-Dataset-Code.

Floating on groundwater: Insight of multi-source remote sensing for Qaidam basin

Situated in the north of the Qinghai-Tibet Plateau, the Qaidam Basin experiences limited precipitation and significant evaporation. Despite these conditions, it stands out as one of the most densely distributed lakes in China. The formation of these lakes is controversial: whether the lake water primarily originates from local precipitation or external water sources. To address this issue, this paper explores the recharge sources of lakes in the Qaidam Basin and the circulation patterns of groundwater from a remote sensing perspective. Based on deep learning networks, we optimized the soft object regions of the Object-Contextual Representations Network (OCRNet) and proposed the Remote·Sensing Adaptive-Improved OCRNet (RSA-IOCRNet). Compared with seven other networks, RSA-IOCRNet obtained better experimental results and was used to construct an area sequence of 16 major lakes in the Qaidam Basin. Combined with multi-source data, the comprehensive analysis indicates no significant correlation between climatic factors and lake changes, while an obvious correlation between lakes and groundwater changes in the eastern Qaidam, consisting with the results of the field survey. Deep-circulating groundwater recharges numerous Qaidam lakes through upwelling from fault zones, such as Gasikule Lake and Xiaochaidan Lake. Groundwater in the Qaidam Basin is more depleted in hydrogen-oxygen isotope characteristics than surface water in the basin, but similar to some river water in the endorheic Tibetan Plateau. This indicates that Tibetan seepage water, estimated at approximately 540 billion m3/a, is transported through the Qaidam Basin via deep circulation. Moreover, it rises to recharge the groundwater and lakes within this basin through fracture zones, extending to various arid and semi-arid regions such as Taitema Lake. This work provides a new perspective on the impact of deep groundwater on lakes and water circulation in these areas.

Developing deep learning methods for classification of teeth in dental panoramic radiography

OBJECTIVES

We aimed to develop an artificial intelligence-based clinical dental decision-support system using deep-learning methods to reduce diagnostic interpretation error and time and increase the effectiveness of dental treatment and classification.

STUDY DESIGN

We compared the performance of 2 deep-learning methods, You Only Look Once V4 (YOLO-V4) and Faster Regions with the Convolutional Neural Networks (R-CNN), for tooth classification in dental panoramic radiography for tooth classification in dental panoramic radiography to determine which is more successful in terms of accuracy, time, and detection ability. Using a method based on deep-learning models trained on a semantic segmentation task, we analyzed 1200 panoramic radiographs selected retrospectively. In the classification process, our model identified 36 classes, including 32 teeth and 4 impacted teeth.

RESULTS

The YOLO-V4 method achieved a mean 99.90% precision, 99.18% recall, and 99.54% F1 score. The Faster R-CNN method achieved a mean 93.67% precision, 90.79% recall, and 92.21% F1 score. Experimental evaluations showed that the YOLO-V4 method outperformed the Faster R-CNN method in terms of accuracy of predicted teeth in the tooth classification process, speed of tooth classification, and ability to detect impacted and erupted third molars.

CONCLUSIONS

The YOLO-V4 method outperforms the Faster R-CNN method in terms of accuracy of tooth prediction, speed of detection, and ability to detect impacted third molars and erupted third molars. The proposed deep learning based methods can assist dentists in clinical decision making, save time, and reduce the negative effects of stress and fatigue in daily practice.

Infrared and Visible Image Fusion Algorithm Based on Double-Domain Transform Filter and Contrast Transform Feature Extraction

Current challenges in visible and infrared image fusion include color information distortion, texture detail loss, and target edge blur. To address these issues, a fusion algorithm based on double-domain transform filter and nonlinear contrast transform feature extraction (DDCTFuse) is proposed. First, for the problem of incomplete detail extraction that exists in the traditional transform domain image decomposition, an adaptive high-pass filter is proposed to decompose images into high-frequency and low-frequency portions. Second, in order to address the issue of fuzzy fusion target caused by contrast loss during the fusion process, a novel feature extraction algorithm is devised based on a novel nonlinear transform function. Finally, the fusion results are optimized and color-corrected by our proposed spatial-domain logical filter, in order to solve the color loss and edge blur generated in the fusion process. To validate the benefits of the proposed algorithm, nine classical algorithms are compared on the LLVIP, MSRS, INO, and Roadscene datasets. The results of these experiments indicate that the proposed fusion algorithm exhibits distinct targets, provides comprehensive scene information, and offers significant image contrast.

A Comparative Review on Enhancing Visual Simultaneous Localization and Mapping with Deep Semantic Segmentation

Visual simultaneous localization and mapping (VSLAM) enhances the navigation of autonomous agents in unfamiliar environments by progressively constructing maps and estimating poses. However, conventional VSLAM pipelines often exhibited degraded performance in dynamic environments featuring mobile objects. Recent research in deep learning led to notable progress in semantic segmentation, which involves assigning semantic labels to image pixels. The integration of semantic segmentation into VSLAM can effectively differentiate between static and dynamic elements in intricate scenes. This paper provided a comprehensive comparative review on leveraging semantic segmentation to improve major components of VSLAM, including visual odometry, loop closure detection, and environmental mapping. Key principles and methods for both traditional VSLAM and deep semantic segmentation were introduced. This paper presented an overview and comparative analysis of the technical implementations of semantic integration across various modules of the VSLAM pipeline. Furthermore, it examined the features and potential use cases associated with the fusion of VSLAM and semantics. It was found that the existing VSLAM model continued to face challenges related to computational complexity. Promising future research directions were identified, including efficient model design, multimodal fusion, online adaptation, dynamic scene reconstruction, and end-to-end joint optimization. This review shed light on the emerging paradigm of semantic VSLAM and how deep learning-enabled semantic reasoning could unlock new capabilities for autonomous intelligent systems to operate reliably in the real world.

Image-based molecular representation learning for drug development: a survey

Artificial intelligence (AI) powered drug development has received remarkable attention in recent years. It addresses the limitations of traditional experimental methods that are costly and time-consuming. While there have been many surveys attempting to summarize related research, they only focus on general AI or specific aspects such as natural language processing and graph neural network. Considering the rapid advance on computer vision, using the molecular image to enable AI appears to be a more intuitive and effective approach since each chemical substance has a unique visual representation. In this paper, we provide the first survey on image-based molecular representation for drug development. The survey proposes a taxonomy based on the learning paradigms in computer vision and reviews a large number of corresponding papers, highlighting the contributions of molecular visual representation in drug development. Besides, we discuss the applications, limitations and future directions in the field. We hope this survey could offer valuable insight into the use of image-based molecular representation learning in the context of drug development.

Underwater object detection method based on learnable query recall mechanism and lightweight adapter

With the rapid development of ocean observation technology, underwater object detection has begun to occupy an essential position in the fields of aquaculture, environmental monitoring, marine science, etc. However, due to the problems unique to underwater images such as severe noise, blurred objects, and multi-scale, deep learning-based target detection algorithms lack sufficient capabilities to cope with these challenges. To address these issues, we improve DETR to make it well suited for underwater scenarios. First, a simple and effective learnable query recall mechanism is proposed to mitigate the effect of noise and can significantly improve the detection performance of the object. Second, for underwater small and irregular object detection, a lightweight adapter is designed to provide multi-scale features for the encoding and decoding stages. Third, the regression mechanism of the bounding box is optimized using the combination loss of smooth L1 and CIoU. Finally, we validate the designed network against other state-of-the-art methods on the RUOD dataset. The experimental results show that the proposed method is effective.

Named entity recognition of rice genes and phenotypes based on BiGRU neural networks

Named Entity Recognition (NER) is a fundamental but crucial task in natural language processing (NLP) and big data analysis, with wide application range. NER for rice genes and phenotypes is a technique to identify genes and phenotypes from a large amount of text. NER for rice genes and phenotypes can facilitate the acquisition of information in the field of crops and provide references for our research on higher quality crops. At the same time, named entity recognition still faces many challenges. In this paper, we propose an improved bidirectional gated recurrent unit neural network (BI-GRU) method, which is used to automatically identify the required entities (i.e. gene names, rice phenotypes) from relevant rice literature and patents. The neural network model is combined with the Softmax function to directly output the probabilities of labels, forming the BI-GRU-SF model. With the ability of deep learning methods, the semantic information in the context can be learned without the need for feature engineering. Finally, we conducted experiments, and the results showed that our proposed model provided better performance compared to other models. All datasets and resource codes of BI-GRU-SF are available at https://github.com/qqeeqq/NER for academic use.

Preassigned-time synchronization for complex-valued memristive neural networks with reaction-diffusion terms and Markov parameters

This study addresses the preassigned-time synchronization for complex-valued memristive neural networks with reaction-diffusion terms and Markov parameters. Employing a preassigned-time stable control strategy, two distinct controllers with varying power exponent parameters are designed to ensure that synchronization can be achieved within a predefined time frame. Unlike existing finite/fixed-time results, a priori specification of the settling time is addressed. Furthermore, Green's formula and boundary conditions are efficiently applied to overcome potential symmetry loss. Additionally, the activation function's constraint range is more lenient compared to existing constraints. Finally, the effectiveness of the presented methods are demonstrated through two examples.

Forecasting water quality variable using deep learning and weighted averaging ensemble models

Water quality variables, including chlorophyll-a (Chl-a), play a pivotal role in comprehending and evaluating the condition of aquatic ecosystems. Chl-a, a pigment present in diverse aquatic organisms, notably algae and cyanobacteria, serves as a valuable indicator of water quality. Thus, the objectives of this study encompass: (1) the assessment of the predictive capabilities of four deep learning (DL) models - namely, recurrent neural network (RNN), long short-term memory (LSTM), gated recurrence unit (GRU), and temporal convolutional network (TCN) - in forecasting Chl-a concentrations; (2) the incorporation of these DL models into ensemble models (EMs) employing genetic algorithm (GA) and non-dominated sorting genetic algorithm (NSGA-II) to harness the strengths of each standalone model; and (3) the evaluation of the efficacy of the developed EMs. Utilizing data collected at 15-min intervals from Small Prespa Lake (SPL) in Greece, the models employed hourly Chl-a concentration lag times, extending up to 6 h, as models' inputs to forecast Chla (t+1). The proposed models underwent training on 70% of the dataset and were subsequently validated on the remaining 30%. Among the standalone DL models, the GRU model exhibited superior performance in Chl-a forecasting, surpassing the RNN, LSTM, and TCN models by 8%, 2%, and 2%, respectively. Furthermore, the integration of DL models through single-objective GA and multi-objective NSGA-II optimization algorithms yielded hybrid models adept at effectively forecasting both low and high Chl-a concentrations. The ensemble model based on NSGA-II outperformed standalone DL models as well as the GA-based model across a range of evaluation indices. For instance, considering the R-squared metric, the study's findings demonstrated that the EM-NSGA-II stands out with exceptional effectiveness compared to DL and EM-GA models, showcasing improvements of 14% (RNN), 8% (LSTM), 6% (GRU), 8% (TCN), and 3% (EM-GA) during the testing phase.

Stacking-ac4C: an ensemble model using mixed features for identifying n4-acetylcytidine in mRNA

N4-acetylcytidine (ac4C) is a modification of cytidine at the nitrogen-4 position, playing a significant role in the translation process of mRNA. However, the precise mechanism and details of how ac4C modifies translated mRNA remain unclear. Since identifying ac4C sites using conventional experimental methods is both labor-intensive and time-consuming, there is an urgent need for a method that can promptly recognize ac4C sites. In this paper, we propose a comprehensive ensemble learning model, the Stacking-based heterogeneous integrated ac4C model, engineered explicitly to identify ac4C sites. This innovative model integrates three distinct feature extraction methodologies: Kmer, electron-ion interaction pseudo-potential values (PseEIIP), and pseudo-K-tuple nucleotide composition (PseKNC). The model also incorporates the robust Cluster Centroids algorithm to enhance its performance in dealing with imbalanced data and alleviate underfitting issues. Our independent testing experiments indicate that our proposed model improves the Mcc by 15.61% and the ROC by 5.97% compared to existing models. To test our model's adaptability, we also utilized a balanced dataset assembled by the authors of iRNA-ac4C. Our model showed an increase in Sn of 4.1%, an increase in Acc of nearly 1%, and ROC improvement of 0.35% on this balanced dataset. The code for our model is freely accessible at https://github.com/louliliang/ST-ac4C.git, allowing users to quickly build their model without dealing with complicated mathematical equations.

Detection of Pedestrians in Reverse Camera Using Multimodal Convolutional Neural Networks

In recent years, the application of artificial intelligence (AI) in the automotive industry has led to the development of intelligent systems focused on road safety, aiming to improve protection for drivers and pedestrians worldwide to reduce the number of accidents yearly. One of the most critical functions of these systems is pedestrian detection, as it is crucial for the safety of everyone involved in road traffic. However, pedestrian detection goes beyond the front of the vehicle; it is also essential to consider the vehicle's rear since pedestrian collisions occur when the car is in reverse drive. To contribute to the solution of this problem, this research proposes a model based on convolutional neural networks (CNN) using a proposed one-dimensional architecture and the Inception V3 architecture to fuse the information from the backup camera and the distance measured by the ultrasonic sensors, to detect pedestrians when the vehicle is reversing. In addition, specific data collection was performed to build a database for the research. The proposed model showed outstanding results with 99.85% accuracy and 99.86% correct classification performance, demonstrating that it is possible to achieve the goal of pedestrian detection using CNN by fusing two types of data.

Prediction model with multi-point relationship fusion via graph convolutional network: A case study on mining-induced surface subsidence

Accurate prediction of surface subsidence is of significance for analyzing the pattern of mining-induced surface subsidence, and for mining under buildings, railways, and water bodies. To address the problem that the existing prediction models ignore the correlation between subsidence points, resulting in large prediction errors, a Multi-point Relationship Fusion prediction model based on Graph Convolutional Networks (MRF-GCN) for mining-induced subsidence was proposed. Taking the surface subsidence in 82/83 mining area of Yuandian No. 2 Mine in Anhui Province in eastern China as an example, the surface deformation data obtained from 250 InSAR images captured by Sentinel-1A satellite from 2018 to 2022, combined with GNSS observation data, were used for modeling. The deformation pattern of each single observation point was obtained by feeding their deformation observation data into the LSTM encoder, after that, the relationship graph was created based on the correlation between points in the observation network and MRF-GCN was established. Then the prediction results came out through a nonlinear activation function of neural network. The research shows that the R2R2 value of MRF-GCN model was 0.865 0, much larger than that of Long-Short Term Memory (LSTM) and other conventional models, while mean square error (MSE) of MRF-GCN model was 1.59 899, much smaller than that of LSTM and other conventional models. Therefore, the MRF-GCN model has better prediction accuracy than other models and can be applied to predicting surface subsidence in large areas.

Recent Advances in Explainable Artificial Intelligence for Magnetic Resonance Imaging

Advances in artificial intelligence (AI), especially deep learning (DL), have facilitated magnetic resonance imaging (MRI) data analysis, enabling AI-assisted medical image diagnoses and prognoses. However, most of the DL models are considered as "black boxes". There is an unmet need to demystify DL models so domain experts can trust these high-performance DL models. This has resulted in a sub-domain of AI research called explainable artificial intelligence (XAI). In the last decade, many experts have dedicated their efforts to developing novel XAI methods that are competent at visualizing and explaining the logic behind data-driven DL models. However, XAI techniques are still in their infancy for medical MRI image analysis. This study aims to outline the XAI applications that are able to interpret DL models for MRI data analysis. We first introduce several common MRI data modalities. Then, a brief history of DL models is discussed. Next, we highlight XAI frameworks and elaborate on the principles of multiple popular XAI methods. Moreover, studies on XAI applications in MRI image analysis are reviewed across the tissues/organs of the human body. A quantitative analysis is conducted to reveal the insights of MRI researchers on these XAI techniques. Finally, evaluations of XAI methods are discussed. This survey presents recent advances in the XAI domain for explaining the DL models that have been utilized in MRI applications.

An Active Multi-Object Ultrafast Tracking System with CNN-Based Hybrid Object Detection

This study proposes a visual tracking system that can detect and track multiple fast-moving appearance-varying targets simultaneously with 500 fps image processing. The system comprises a high-speed camera and a pan-tilt galvanometer system, which can rapidly generate large-scale high-definition images of the wide monitored area. We developed a CNN-based hybrid tracking algorithm that can robustly track multiple high-speed moving objects simultaneously. Experimental results demonstrate that our system can track up to three moving objects with velocities lower than 30 m per second simultaneously within an 8-m range. The effectiveness of our system was demonstrated through several experiments conducted on simultaneous zoom shooting of multiple moving objects (persons and bottles) in a natural outdoor scene. Moreover, our system demonstrates high robustness to target loss and crossing situations.

Stability analysis of chaotic generalized Lotka-Volterra system via active compound difference anti-synchronization method

This work deals with a systematic approach for the investigation of compound difference anti-synchronization (CDAS) scheme among chaotic generalized Lotka-Volterra biological systems (GLVBSs). First, an active control strategy (ACS) of nonlinear type is described which is specifically based on Lyapunov's stability analysis (LSA) and master-slave framework. In addition, the biological control law having nonlinear expression is constructed for attaining asymptotic stability pattern for the error dynamics of the discussed GLVBSs. Also, simulation results through MATLAB environment are executed for illustrating the efficacy and correctness of considered CDAS approach. Remarkably, our attained analytical outcomes have been in outstanding conformity with the numerical outcomes. The investigated CDAS strategy has numerous significant applications to the fields of encryption and secure communication.

A method for finding the background potential of quantum devices from scanning gate microscopy data using machine learning

The inverse problem of estimating the background potential from measurements of the local density of states is a challenging issue in quantum mechanics. Even more difficult is to do this estimation using approximate methods such as scanning gate microscopy (SGM). Here, we propose a machine-learning-based solution by exploiting adaptive cellular neural networks (CNNs). In the paradigmatic setting of a quantum point contact, the training data consist of potential-SGM functional relations represented by image pairs. These are generated by the recursive Green’s function method. We demonstrate that the CNN-based machine learning framework can predict the background potential corresponding to the experimental image data. This is confirmed by analyzing the estimated potential with image processing techniques based on the comparison between the charge densities and those obtained using different techniques. Correlation analysis of the images suggests the possibility of estimating different contributions to the background potential. In particular, our results indicate that both charge puddles and fixed impurities contribute to the spatial patterns found in the SGM data. Our work represents a timely contribution to the rapidly evolving field of exploiting machine learning to solve difficult problems in physics.

Domain generated algorithms detection applying a combination of a deep feature selection and traditional machine learning models

The use of command and control (C2) servers in cyberattacks has risen considerably, attackers frequently employ the domain generated algorithm (DGA) technique to conceal their C2 servers. Various machine learning models have been suggested for binary identification of domain names as either benign or DGA domain. The Existing techniques are inefficient and have real-time detection issues and are also very data hypersensitive, therefore, they can be circumvented by the attackers. The main problem this article addresses is how to automatically detect DGA in a way that does not rely solely on reverse engineering, not strongly affected by data size, and allows detection of this DGA in real time. This paper presents DTFS-DGA model that combine neural networks models with traditional machine learning models and maintains its performance even if the data size changes to detect DGA in real time. The model uses 15 linguistics and networks features with the features extracted by long short-term memory and convolutional neural network to classify domain names using random forest and support vector machines. The comprehensive experimental findings confirm the suggested model’s accuracy. To be precise, the model achieve an average accuracy of 99.8 % for the classification.

Deep learning in macroscopic diffuse optical imaging

SIGNIFICANCE

Biomedical optics system design, image formation, and image analysis have primarily been guided by classical physical modeling and signal processing methodologies. Recently, however, deep learning (DL) has become a major paradigm in computational modeling and has demonstrated utility in numerous scientific domains and various forms of data analysis.

AIM

We aim to comprehensively review the use of DL applied to macroscopic diffuse optical imaging (DOI).

APPROACH

First, we provide a layman introduction to DL. Then, the review summarizes current DL work in some of the most active areas of this field, including optical properties retrieval, fluorescence lifetime imaging, and diffuse optical tomography.

RESULTS

The advantages of using DL for DOI versus conventional inverse solvers cited in the literature reviewed herein are numerous. These include, among others, a decrease in analysis time (often by many orders of magnitude), increased quantitative reconstruction quality, robustness to noise, and the unique capability to learn complex end-to-end relationships.

CONCLUSIONS

The heavily validated capability of DL's use across a wide range of complex inverse solving methodologies has enormous potential to bring novel DOI modalities, otherwise deemed impractical for clinical translation, to the patient's bedside.

India perspective: CNN-LSTM hybrid deep learning model-based COVID-19 prediction and current status of medical resource availability

The epidemic situation may cause severe social and economic impacts on a country. So, there is a need for a trustworthy prediction model that can offer better prediction results. The forecasting result will help in making the prevention policies and remedial action in time, and thus, we can reduce the overall social and economic impacts on the country. This article introduces a CNN-LSTM hybrid deep learning prediction model, which can correctly forecast the COVID-19 epidemic across India. The proposed model uses convolutional layers, to extract meaningful information and learn from a given time series dataset. It is also enriched with the LSTM layer's capability, which means it can identify long-term and short-term dependencies. The experimental evaluation has been performed to gauge the performance and suitability of our proposed model among the other well-established time series forecasting models. From the empirical analysis, it is also clear that the use of extra convolutional layers with the LSTM layer may increase the forecasting model's performance. Apart from this, the deep insides of the current situation of medical resource availability across India have been discussed.

A Review of Remote Sensing Image Dehazing

Remote sensing (RS) is one of the data collection technologies that help explore more earth surface information. However, RS data captured by satellite are susceptible to particles suspended during the imaging process, especially for data with visible light band. To make up for such deficiency, numerous dehazing work and efforts have been made recently, whose strategy is to directly restore single hazy data without the need for using any extra information. In this paper, we first classify the current available algorithm into three categories, i.e., image enhancement, physical dehazing, and data-driven. The advantages and disadvantages of each type of algorithm are then summarized in detail. Finally, the evaluation indicators used to rank the recovery performance and the application scenario of the RS data haze removal technique are discussed, respectively. In addition, some common deficiencies of current available methods and future research focus are elaborated.

A Fault Diagnosis and Visualization Method for High-Speed Train Based on Edge and Cloud Collaboration

Safety is the most important aspect of railway transportation. To ensure the safety of high-speed trains, various train components are equipped with sensor devices for real-time monitoring. Sensor monitoring data can be used for fast intelligent diagnosis and accurate positioning of train faults. However, existing train fault diagnosis technology based on cloud computing has disadvantages of long processing times and high consumption of computing resources, which conflict with the real-time response requirements of fault diagnosis. Aiming at the problems of train fault diagnosis in the cloud environment, this paper proposes a train fault diagnosis model based on edge and cloud collaboration. The model first utilizes a SAES-DNN (stacked auto-encoders deep neural network) fault recognition method, which can integrate automatic feature extraction and type recognition and complete fault classification over deep hidden features in high-dimensional data, so as to quickly locate faults. Next, to adapt to the characteristics of edge computing, the model applies a SAES-DNN model trained in the cloud and deployed in the edge via the transfer learning strategy and carries out real-time fault diagnosis on the vehicle sensor monitoring data. Using a motor fault as an example, when compared with a similar intelligent learning model, the proposed intelligent fault diagnosis model can greatly improve diagnosis accuracy and significantly reduce training time. Through the transfer learning approach, adaptability of the fault diagnosis algorithm for personalized applications and real-time performance of the fault diagnosis is enhanced. This paper also proposes a visual analysis method of train fault data based on knowledge graphs, which can effectively analyze fault causes and fault correlation.

Deep Flexible Sequential (DFS) Model for Air Pollution Forecasting

Growing metropolitan areas bring rapid urbanization and air pollution problems. As diseases and mortality rates increase because of the air pollution problem, it becomes a necessity to estimate the air pollution density and inform the public to protect the health. Air pollution problem displays contextual characteristics such as meteorological conditions, industrial and technological developments, traffic problem etc. that change from country to country and also from city to city. In this study, we determined PM[Formula: see text] as the target pollutant and designed a new deep learning based air quality forecasting model, namely DFS (Deep Flexible Sequential). Our study uses real world hourly data from Istanbul, Turkey between 2014 and 2018 to forecast the air pollution 4, 12, and 24 hours before. DFS model is a hybrid & flexible deep model including Long Short Term Memory (LSTM) and Convolutional Neural Network (CNN). The proposed model also is capable of generalization with standard and flexible Dropout layers. Through flexible Dropout layer, the model also obtains flexibility to adapt changing window sizes in sequential modelling. Moreover, this model can be applied to other air pollution time series data problems with small modifications on parameters by taking into account the nature of the data set.

Machine Learning in Nanoscience: Big Data at Small Scales

Recent advances in machine learning (ML) offer new tools to extract new insights from large data sets and to acquire small data sets more effectively. Researchers in nanoscience are experimenting with these tools to tackle challenges in many fields. In addition to ML's advancement of nanoscience, nanoscience provides the foundation for neuromorphic computing hardware to expand the implementation of ML algorithms. In this Mini Review, we highlight some recent efforts to connect the ML and nanoscience communities by focusing on three types of interaction: (1) using ML to analyze and extract new insights from large nanoscience data sets, (2) applying ML to accelerate material discovery, including the use of active learning to guide experimental design, and (3) the nanoscience of memristive devices to realize hardware tailored for ML. We conclude with a discussion of challenges and opportunities for future interactions between nanoscience and ML researchers.

Soft tissue deformation modelling through neural dynamics-based reaction-diffusion mechanics

Soft tissue deformation modelling forms the basis of development of surgical simulation, surgical planning and robotic-assisted minimally invasive surgery. This paper presents a new methodology for modelling of soft tissue deformation based on reaction-diffusion mechanics via neural dynamics. The potential energy stored in soft tissues due to a mechanical load to deform tissues away from their rest state is treated as the equivalent transmembrane potential energy, and it is distributed in the tissue masses in the manner of reaction-diffusion propagation of nonlinear electrical waves. The reaction-diffusion propagation of mechanical potential energy and nonrigid mechanics of motion are combined to model soft tissue deformation and its dynamics, both of which are further formulated as the dynamics of cellular neural networks to achieve real-time computational performance. The proposed methodology is implemented with a haptic device for interactive soft tissue deformation with force feedback. Experimental results demonstrate that the proposed methodology exhibits nonlinear force-displacement relationship for nonlinear soft tissue deformation. Homogeneous, anisotropic and heterogeneous soft tissue material properties can be modelled through the inherent physical properties of mass points. Graphical abstract Soft tissue deformation modelling with haptic feedback via neural dynamics-based reaction-diffusion mechanics.

Convolutional Neural Network-Based Embarrassing Situation Detection under Camera for Social Robot in Smart Homes

Recent research has shown that the ubiquitous use of cameras and voice monitoring equipment in a home environment can raise privacy concerns and affect human mental health. This can be a major obstacle to the deployment of smart home systems for elderly or disabled care. This study uses a social robot to detect embarrassing situations. Firstly, we designed an improved neural network structure based on the You Only Look Once (YOLO) model to obtain feature information. By focusing on reducing area redundancy and computation time, we proposed a bounding-box merging algorithm based on region proposal networks (B-RPN), to merge the areas that have similar features and determine the borders of the bounding box. Thereafter, we designed a feature extraction algorithm based on our improved YOLO and B-RPN, called F-YOLO, for our training datasets, and then proposed a real-time object detection algorithm based on F-YOLO (RODA-FY). We implemented RODA-FY and compared models on our MAT social robot. Secondly, we considered six types of situations in smart homes, and developed training and validation datasets, containing 2580 and 360 images, respectively. Meanwhile, we designed three types of experiments with four types of test datasets composed of 960 sample images. Thirdly, we analyzed how a different number of training iterations affects our prediction estimation, and then we explored the relationship between recognition accuracy and learning rates. Our results show that our proposed privacy detection system can recognize designed situations in the smart home with an acceptable recognition accuracy of 94.48%. Finally, we compared the results among RODA-FY, Inception V3, and YOLO, which indicate that our proposed RODA-FY outperforms the other comparison models in recognition accuracy.

Designing a model to detect the brain connections abnormalities in children with autism using 3D-cellular neural networks

In neuropsychological disorders, the significant abnormalities in the brain connections in some regions are observed. This paper presents a novel model to demonstrate the connections between different regions in children with autism. The proposed model first conducts the wavelet decomposition of electroencephalography signals by wavelet transform then the features are extracted, such as relative energy and entropy. These features are fed to the 3D-cellular neural network model as inputs to indicate the brain connections. The results showed that there are significant differences and abnormalities in the left hemisphere, (p<0.05) at the electrodes AF3, F3, P7, T7 and O1 in alpha band, AF3, F7, T7 and O1 in beta band, T7 and P7 in gamma band for children with autism compared with the control children. Also, the evaluation of the obtained connections values between brain regions indicated that there are more abnormalities in the connectivity of frontal and parietal lobes and the relations of the neighboring regions in all three bands especially in gamma band for autistic children. Evaluation of the analysis demonstrated that alpha frequency band had the best distinction level of 96.6% based on the obtained values of the cellular neural network using support vector machine method.

A Memristive Multilayer Cellular Neural Network With Applications to Image Processing

The memristor has been extensively studied in electrical engineering and biological sciences as a means to compactly implement the synaptic function in neural networks. The cellular neural network (CNN) is one of the most implementable artificial neural network models and capable of massively parallel analog processing. In this paper, a novel memristive multilayer CNN (Mm-CNN) model is presented along with its performance analysis and applications. In this new CNN design, the memristor crossbar circuit acts as the synapse, which realizes one signed synaptic weight with a pair of memristors and performs the synaptic weighting compactly and linearly. Moreover, the complex weighted summation is executed in an efficient way with a proper design of Mm-CNN cell circuits. The proposed Mm-CNN has several merits, such as compactness, nonvolatility, versatility, and programmability of synaptic weights. Its performance in several image processing applications is illustrated through simulations.

Retinal Circuit Emulator With Spatiotemporal Spike Outputs at Millisecond Resolution in Response to Visual Events

To gain insights on how visual information of the real world is filtered, compressed, and encoded by the vertebrate retinas, emulating in silico the spatiotemporal patterns of the graded and action potentials of neuronal responses to natural visual scenes on biological time scale is a feasible approach. As a basic platform for such an emulation, we here developed a compact hardware system comprising an analog silicon retina and a field-programmable gate array module. With utilizing the Izhikevich formalism, a retinal circuit model that emulates spiking of ganglion cells was implemented in this system. The emulated spike timing had the resolution of about 2 ms relative to the stimulus onset and was little affected by timings of the synchronous frame sampling in the silicon retina. Thus, the emulator can mimic the event-driven spike outputs of biological retinas. The system was useful for simultaneously visualizing neural images of both the graded potentials and the spikes in response to real live visual scenes. Since our emulator system is reconfigurable, it provides a flexible platform for investigating visual functions of retinal circuits under natural visual environment.

Speeding Up Cellular Neural Network Processing Ability by Embodying Memristors

Cellular neural networks (CNNs) are an efficient tool for image analysis and pattern recognition. Based on elementary cells connected to neighboring units, they are easy to install in hardware, carrying out massively parallel processes. This brief presents a new model of CNN with memory devices, which enhances further CNN performance. By introducing a memristive element in basic cells, we carry out different experiments, allowing the analysis of the functions traditionally carried out by the standard CNN. Without modifying the templates considered by the scientific literature, this simple variation originates a significant improvement in  ∼ 30 % of performances in pattern recognition and image processing. These progresses were experimentally calculated on the time the system requires to reach a fixed point. Moreover, the different role that each parameter has in the developed method was also analyzed to better understand the complex processing ability of these systems.

Intrinsic optimization using stochastic nanomagnets

This paper draws attention to a hardware system which can be engineered so that its intrinsic physics is described by the generalized Ising model and can encode the solution to many important NP-hard problems as its ground state. The basic constituents are stochastic nanomagnets which switch randomly between the ±1 Ising states and can be monitored continuously with standard electronics. Their mutual interactions can be short or long range, and their strengths can be reconfigured as needed to solve specific problems and to anneal the system at room temperature. The natural laws of statistical mechanics guide the network of stochastic nanomagnets at GHz speeds through the collective states with an emphasis on the low energy states that represent optimal solutions. As proof-of-concept, we present simulation results for standard NP-complete examples including a 16-city traveling salesman problem using experimentally benchmarked models for spin-transfer torque driven stochastic nanomagnets.

An Evaluation of a Sensor Fusion System to Improve Drivers' Nighttime Detection of Road Hazards

Objective: To determine whether combination of two spectral bands within a single sensor-fused image can improve drivers' detection of road hazards. Methods: Images were collected with visible and short wave infrared sensors, and were combined by an image fusion algorithm derived from a computational model of human retinal processing (Werblin et al., 1997). Test stimuli were single-band and fused images of a nighttime scene, collected with sensors mounted atop a vehicle and facing down a stretch of road into an opposing vehicle's headlights. The intensity of the opposing headlights was varied to produce images of low and of high glare. The subjects' task was to detect the presence of a pedestrian within each image. Results: Sensor-fused imagery reliably produced performance better than or equivalent to that produced by either format of single-band imagery. Conclusions: Sensor fusion may provide an effective method of facilitating the detection of road hazards under low visibility conditions.