Search

Copyright of Electrotechnical Review / Elektrotehniski Vestnik is the property of Electrotechnical Society of Slovenia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

SDN has the ability to transform network design by providing increased versatility and effective regulation. Its programmable centralized controller gives network administration employees more authority, allowing for more seamless supervision. However, centralization makes it vulnerable to a variety of attack vectors, with distributed denial of service (DDoS) attacks posing a serious concern. Feature selection-based Machine Learning (ML) techniques are more effective than traditional signature-based Intrusion Detection Systems (IDS) at identifying new threats in the context of defending against distributed denial of service (DDoS) attacks. In this study, NGBoost is compared with four additional machine learning (ML) algorithms: convolutional neural network (CNN), Stochastic Gradient Descent (SGD), Decision Tree, and Random Forest, in order to assess the effectiveness of DDoS detection on the CICDDoS2019 dataset. It focuses on important measures such as F1 score, recall, accuracy, and precision. We have examined NeTBIOS, a layer-7 attack, and SYN, a layer-4 attack, in our paper. Our investigation shows that Natural Gradient Boosting and Convolutional Neural Networks, in particular, show promise with tabular data categorization. In conclusion, we go through specific study results on protecting against attacks using DDoS. These experimental findings offer a framework for making decisions. [ABSTRACT FROM AUTHOR]

Based on the rapid development of big data, cloud computing, Internet of things and other technologies in recent years, intelligent hardware devices has been applied to all aspects of life. Under this background, some scholars have put forward relevant concepts such as "Smart Life". In the field of mass sports life, through the development and application of science and technology, there has been new changes related to the application of neural network algorithm technology and intelligent hardware devices. Therefore, artificial intelligence wearable devices based on wearable technology came into being. This paper analyzes the application of this device in mass sports activities. Then, this paper describes the key research technologies of motion data processing based on neural network algorithm, including: depth frame differential convolution neural network structure, motion data extraction method, human motion signal processing algorithm, etc.; then it analyzes the action recognition and interaction system design based on Intelligent wearable devices. Finally, it analyzes the recognition results of human action system, the accuracy of human action recognition system and the factors that affect the performance of the recognition system. It is concluded that the artificial intelligent wearable devices designed in this paper can be well used in popular sports activities. Finally, it introduces the research on the evaluation strategy of popular sports activities based on artificial intelligence, and hopes that this equipment can help public sports activities. This paper studies the neural network algorithm and applies it to the design process of artificial intelligence wearable devices, which promotes the development of mass sports activity evaluation. [ABSTRACT FROM AUTHOR]

With the rapid development of deep learning, computer vision has also become a rapidly developing field in the field of artificial intelligence. Combining the physical training of deep learning will bring good practical value. Physical training has different effects on people's body shape, physical function, and physical quality. It is mainly reflected in the changes of relevant physical indicators after physical training. Therefore, the purpose of this article is to study the method of evaluating the impact of sports training on physical indicators based on deep learning. This paper mainly uses the convolutional neural network in deep learning to design sports training, then constructs the evaluation system of physical index impact, and finally uses the deep learning algorithm to evaluate the impact of physical index. The experimental results show that the accuracy of the algorithm proposed in this paper is significantly higher than that of the other three algorithms. Firstly, in the angular motion, the accuracy of the mean algorithm is 0.4, the accuracy of the variance algorithm is 0.2, the accuracy of the RFE algorithm is 0.4, and the accuracy of the DLA algorithm is 0.6. Similarly, in foot racing and skill sports, the accuracy of the algorithm proposed in this paper is significantly higher than that of other algorithms. Therefore, the method proposed in this paper is more effective in the evaluation of the impact of physical training on physical indicators. [ABSTRACT FROM AUTHOR]

With the continuous mining and gradual reduction of shallow deposits, deep prospecting has become a new global prospecting trend. In addition, with the development of artificial intelligence, deep learning provides a favorable means for geological big data analysis. This paper, researches the No. II Orebody of the Xiongcun deposit. First, based on previous research results and metallogenic regularity, prospecting information, namely, lithology, Au-Ag-Cu chemical elements and wall rock alteration is extracted, and the block model is established by combining the Kriging interpolation structure. Second, the datasets are divided into dataset I and dataset II according to "randomness" and "depth". Third, deep prospecting prediction models based on deep neural networks (DNN) and the convolutional neural networks (CNN) is constructed, and the model parameters are optimized. Finally, the models are applied to the deep prediction of the Xiongcun No. II Orebody. The results show that the accuracy rate and recall rate of the prediction model based on the DNN algorithm are 96.15% and 89.23%, respectively, and the AUC is 96.39%, which are higher values than those of the CNN algorithm, indicating that the performance of the prediction model based on the DNN algorithm is better. The accuracy of prediction model based on dataset I is higher than that of dataset II. The accuracy of deep metallogenic prediction based on the DNN algorithm is approximately 89%, that based on the CNN is approximately 87%, and that based on prospecting information method is approximately 61.27%. The prediction results of the DNN algorithm are relatively consistent in the spatial location and scale of the orebody. Therefore, based on the work done in this paper, it is feasible to use a deep learning method to carry out deep mineral prediction. [ABSTRACT FROM AUTHOR]

With the continuous mining and gradual reduction of shallow deposits, deep prospecting has become a new global prospecting trend. In addition, with the development of artificial intelligence, deep learning provides a favorable means for geological big data analysis. This paper, researches the No. II Orebody of the Xiongcun deposit. First, based on previous research results and metallogenic regularity, prospecting information, namely, lithology, Au-Ag-Cu chemical elements and wall rock alteration is extracted, and the block model is established by combining the Kriging interpolation structure. Second, the datasets are divided into dataset I and dataset II according to "randomness" and "depth". Third, deep prospecting prediction models based on deep neural networks (DNN) and the convolutional neural networks (CNN) is constructed, and the model parameters are optimized. Finally, the models are applied to the deep prediction of the Xiongcun No. II Orebody. The results show that the accuracy rate and recall rate of the prediction model based on the DNN algorithm are 96.15% and 89.23%, respectively, and the AUC is 96.39%, which are higher values than those of the CNN algorithm, indicating that the performance of the prediction model based on the DNN algorithm is better. The accuracy of prediction model based on dataset I is higher than that of dataset II. The accuracy of deep metallogenic prediction based on the DNN algorithm is approximately 89%, that based on the CNN is approximately 87%, and that based on prospecting information method is approximately 61.27%. The prediction results of the DNN algorithm are relatively consistent in the spatial location and scale of the orebody. Therefore, based on the work done in this paper, it is feasible to use a deep learning method to carry out deep mineral prediction. [ABSTRACT FROM AUTHOR]

The merging of neural radiance fields with generative adversarial networks (GANs) can synthesize novel views of objects from latent code (noise). However, the challenge for generative neural radiance fields (NERFs) is that a single multiple layer perceptron (MLP) network represents a scene or object, and the shape and appearance of the generated object are unpredictable, owing to the randomness of latent code. In this paper, we propose a stylized generative radiance field (SGRF) to produce 3D-aware images with explicit control. To achieve this goal, we manipulated the input and output of the MLP in the model to entangle and disentangle label codes into/from the latent code, and incorporated an extra discriminator to differentiate between the class and color mode of the generated object. Based on the labels provided, the model could generate images of prohibited items varying in class, pose, scale, and color mode, thereby significantly increasing the quantity and diversity of images in the dataset. Through a systematic analysis of the results, the method was demonstrated to be effective in improving the detection performance of deep learning algorithms during security screening. [ABSTRACT FROM AUTHOR]

Accurate and effective building energy consumption prediction is an important basis for carrying out energy-saving evaluation and the main basis for building energy-saving optimization design. However, due to the influence of environmental and human factors, energy consumption prediction is often inaccurate. Therefore, this paper presents a building energy consumption prediction model based on an attention mechanism, time convolutional neural (TCN) network fusion, and a bidirectional gated cycle unit (BIGRU). First, t-distributed stochastic neighbor embedding (T-SNE) was used to preprocess the data and extract the key features, and then a BIGRU was employed to acquire past and future data while capturing immediate connections. Then, to catch the long-term dependence, the dataset was partitioned into the TCN network, and the extended sequence was transformed into several short sequences. Consequently, the gradient explosion or vanishing problem is mitigated when the BIGRU handles lengthy sequences while reducing the spatial complexity. Second, the self-attention mechanism was introduced to enhance the model's capability to address data periodicity. The proposed model is superior to the other four models in accuracy, with an mean absolute error of 0.023, an mean-square error of 0.029, and an coefficient of determination of 0.979. Experimental results indicate that T-SNE can significantly improve the model performance, and the accuracy of predictions can be improved by the attention mechanism and the TCN network. [ABSTRACT FROM AUTHOR]

Imaging is one of the main tools of modern astronomy—many images are collected each day, and they must be processed. Processing such a large amount of images can be complex, time-consuming, and may require advanced tools. One of the techniques that may be employed is artificial intelligence (AI)-based image detection and classification. In this paper, the research is focused on developing such a system for the problem of the Magellan dataset, which contains 134 satellite images of Venus's surface with individual volcanoes marked with circular labels. Volcanoes are classified into four classes depending on their features. In this paper, the authors apply the You-Only-Look-Once (YOLO) algorithm, which is based on a convolutional neural network (CNN). To apply this technique, the original labels are first converted into a suitable YOLO format. Then, due to the relatively small number of images in the dataset, deterministic augmentation techniques are applied. Hyperparameters of the YOLO network are tuned to achieve the best results, which are evaluated as mean average precision (mAP@0.5) for localization accuracy and F1 score for classification accuracy. The experimental results using cross-vallidation indicate that the proposed method achieved 0.835 mAP@0.5 and 0.826 F1 scores, respectively. [ABSTRACT FROM AUTHOR]

Among all usual formats of representing 3D objects, including depth image, mesh and volumetric grid, point cloud is the most commonly used and preferred format, because it preserves the original geometric information in 3D space without any discretization and can provide a comprehensive understanding of the target objects. However, due to their unordered and unstructured nature, conventional deep learning methods such as convolutional neural networks cannot be directly applied to point clouds, which poses a challenge for extracting semantic features from them. This paper proposes a feature fusion algorithm based on attention graph convolution and error feedback, which considers global features, local features and the problem of the features loss during the learning process. Comparison experiments are conducted on the ModelNet40 and ShapeNet datasets to verify the performance of the proposed algorithm, and experimental results show that the proposed method achieves a classification accuracy of 93.1% and a part segmentation mIoU (mean Intersection over Union) of 85.4%. Our algorithm outperforms state-of-the-art algorithms, and effectively improves the accuracy of point cloud classification and segmentation with faster convergence speed. [ABSTRACT FROM AUTHOR]

Facility Management is a multi-disciplinary task in which coordination is key to attaining success during the building life cycle and for which technology assumes an increasing role. This sector is demanding more available and accurate tools to optimize the management process, decrease the probability of failure, and reduce the time spent on anomaly analysis. So, the present paper presents work developed to improve access to building anomaly recognition and to predict the building degradation state in an automatized way. The methodology applied to achieve this goal started with a survey and digital data acquisition from a case study, followed by the automatized detection of building anomalies using supervised classification in Deep Learning; then, the early diagnosis of threatening conditions for building degradation took place using degradation curves based on data records and regression algorithms. The results drive this study a step forward toward obtaining advanced tools for Facility Management based in Artificial Intelligence, able to provide the most appropriate moment at which to intervene according to the cost-benefit. The present work provided better results on the harmonic mean of precision and recall when compared with previous studies of image classification for the construction sector. Moreover, the mathematical functions for the prediction of future degradation based on the data field for each construction system were presented and can be applied to the typologies of other buildings. In the end, future developments and limitations are highlighted. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI), machine learning (ML) and deep learning (DL) along with big data (BD) management are currently viable approaches that can significantly help gas turbine components' design and development. Optimizing microstructures of hot section components such as thermal barrier coatings (TBCs) to improve their durability has long been a challenging task in the gas turbine industry. In this paper, a literature review on ML principles and its various associated algorithms was presented first and then followed by its application to investigate thermal conductivity of TBCs. This combined approach can help better understand the physics behind thermal conductivity, and on the other hand, can also boost the design of low thermal conductivity of the TBCs system in terms of microstructure–property relationships. Several ML models and algorithms such as support vector regression (SVR), Gaussian process regression (GPR) and convolution neural network and regression algorithms were used via Python. A large volume of thermal conductivity data was compiled and extracted from the literature for TBCs using PlotDigitizer software and then used to test and validate ML models. It was found that the test data were strongly associated with five key factors as identifiers. The prediction of thermal conductivity was performed using three approaches: polynomial regression, neural network (NN) and gradient boosting regression (GBR). The results suggest that NN using the BR model and GBR have better prediction capability. [ABSTRACT FROM AUTHOR]

Wind patterns can change due to climate change, causing more storms, hurricanes, and quiet spells. These changes can dramatically affect wind power system performance and predictability. Researchers and practitioners are creating more advanced wind power forecasting algorithms that combine more parameters and data sources. Advanced numerical weather prediction models, machine learning techniques, and real-time meteorological sensor and satellite data are used. This paper proposes a Recurrent Neural Network (RNN) forecasting model incorporating a Dynamic Fitness Al-Biruni Earth Radius (DFBER) algorithm to predict wind power data patterns. The performance of this model is compared with several other popular models, including BER, Jaya Algorithm (JAYA), Fire Hawk Optimizer (FHO), Whale Optimization Algorithm (WOA), Grey Wolf Optimizer (GWO), and Particle Swarm Optimization (PSO)-based models. The evaluation is done using various metrics such as relative root mean squared error (RRMSE), Nash Sutcliffe Efficiency (NSE), mean absolute error (MAE), mean bias error (MBE), Pearson's correlation coefficient (r), coefficient of determination (R2), and determination agreement (WI). According to the evaluation metrics and analysis presented in the study, the proposed RNN-DFBER-based model outperforms the other models considered. This suggests that the RNN model, combined with the DFBER algorithm, predicts wind power data patterns more effectively than the alternative models. To support the findings, visualizations are provided to demonstrate the effectiveness of the RNN-DFBER model. Additionally, statistical analyses, such as the ANOVA test and the Wilcoxon Signed-Rank test, are conducted to assess the significance and reliability of the results. [ABSTRACT FROM AUTHOR]

In this paper, a structural health monitoring (SHM) system is proposed to provide automatic early warning for detecting damage and its location in composite pipelines at an early stage. The study considers a basalt fiber reinforced polymer (BFRP) pipeline with an embedded Fiber Bragg grating (FBG) sensory system and first discusses the shortcomings and challenges with incorporating FBG sensors for accurate detection of damage information in pipelines. The novelty and the main focus of this study is, however, a proposed approach that relies on designing an integrated sensing-diagnostic SHM system that has the capability to detect damage in composite pipelines at an early stage via implementation of an artificial intelligence (AI)-based algorithm combining deep learning and other efficient machine learning methods using an Enhanced Convolutional Neural Network (ECNN) without retraining the model. The proposed architecture replaces the softmax layer by a k-Nearest Neighbor (k-NN) algorithm for inference. Finite element models are developed and calibrated by the results of pipe measurements under damage tests. The models are then used to assess the patterns of the strain distributions of the pipeline under internal pressure loading and under pressure changes due to bursts, and to find the relationship of strains at different locations axially and circumferentially. A prediction algorithm for pipe damage mechanisms using distributed strain patterns is also developed. The ECNN is designed and trained to identify the condition of pipe deterioration so the initiation of damage can be detected. The strain results from the current method and the available experimental results in the literature show excellent agreement. The average error between the ECNN data and FBG sensor data is 0.093%, thus confirming the reliability and accuracy of the proposed method. The proposed ECNN achieves high performance with 93.33 % accuracy (P%), 91.18 % regression rate (R%) and a 90.54 % F1-score (F%). [ABSTRACT FROM AUTHOR]

In order to solve the declining influence of traditional cultural symbols, the research on traditional cultural symbols has become more meaningful. This article aims to study the application of traditional cultural symbols in art design under the background of artificial intelligence. In this paper, a fractal model with self-combined nonlinear function changes is constructed. By combining nonlinear transformations and multiparameter adjustments, various types of fractal models can be automatically rendered. The convolutional neural network algorithm is used to extract the characteristics of the style picture, and it is compared with the trained picture many times to avoid the problem of excessive tendency of the image with improper weight. The improved L-BFGS algorithm is also used to optimize the loss of the traditional L-BFGS, which improves the quality of the generated pictures and reduces the noise of the chessboard. The experimental results in this paper show that the improved L-BFGS algorithm has the least loss and the shortest time in the time used for more than 500 s. Compared with the traditional AdaGrad method, its loss is reduced by about 62%; compared with the traditional AdaDelta method, its loss is reduced by 46%. Its loss is reduced by about 8% compared with the newly optimized Adam method, which is a great improvement. [ABSTRACT FROM AUTHOR]

Since the onset of civilization, sketches have been used to portray our visual world, and they continue to do so in many different disciplines today. As in specific government agencies, establishing similarities between sketches is a crucial aspect of gathering forensic evidence in crimes, in addition to satisfying the user's subjective requirements in searching and browsing for specific sorts of images (i.e., clip art images), especially with the proliferation of smartphones with touchscreens. With such a kind of search, quickly and effectively drawing and retrieving sketches from databases can occasionally be challenging, when using keywords or categories. Drawing some simple forms and searching for the image in that way could be simpler in some situations than attempting to put the vision into words, which is not always possible. Modern techniques, such as Content-Based Image Retrieval (CBIR), may offer a more useful solution. The key engine of such techniques that poses various challenges might be dealt with using effective visual feature representation. Object edge feature detectors are commonly used to extract features from different image sorts. However, they are inconvenient as they consume time due to their complexity in computation. In addition, they are complicated to implement with real-time responses. Therefore, assessing and identifying alternative solutions from the vast array of methods is essential. Scale Invariant Feature Transform (SIFT) is a typical solution that has been used by most prevalent research studies. Even for learning-based methods, SIFT is frequently used for comparison and assessment. However, SIFT has several downsides. Hence, this research is directed to the utilization of handcrafted-feature-based Oriented FAST and Rotated BRIEF (ORB) to capture visual features of sketched images to overcome SIFT limitations on small datasets. However, handcrafted-feature-based algorithms are generally unsuitable for large-scale sets of images. Efficient sketched image retrieval is achieved based on content and separation of the features of the black line drawings from the background into precisely-defined variables. Each variable is encoded as a distinct dimension in this disentangled representation. For representation of sketched images, this paper presents a Sketch-Based Image Retrieval (SBIR) system, which uses the information-maximizing GAN (InfoGAN) model. The establishment of such a retrieval system is based on features acquired by the unsupervised learning InfoGAN model to satisfy users' expectations for large-scale datasets. The challenges with the matching and retrieval systems of such kinds of images develop when drawing clarity declines. Finally, the ORB-based matching system is introduced and compared to the SIFT-based system. Additionally, the InfoGAN-based system is compared with state-of-the-art solutions, including SIFT, ORB, and Convolutional Neural Network (CNN). [ABSTRACT FROM AUTHOR]

In this research paper, we focused on the developing a secure and efficient time-series forecasting of nuclear reactions using swarm intelligence (SI) algorithm. Nuclear radioactive management and efficient time series for casting of nuclear reactions is a problem to be addressed if nuclear power is to deliver a major part of our energy consumption. This problem explains how SI processing techniques can be used to automate accurate nuclear reaction forecasting. The goal of the study was to use swarm analysis to understand patterns and reactions in the dataset while forecasting nuclear reactions using swarm intelligence. The results obtained by training the SI algorithm for longer periods of time for predicting the efficient time series events of nuclear reactions with 94.58 percent accuracy, which is higher than the deep convolution neural networks (DCNNs) 93% accuracy for all predictions, such as the number of active reactions, to see how the results can improve. Our earliest research focused on determining the best settings and preprocessing for working with a certain nuclear reaction, such as fusion and fusion task: forecasting the time series as the reactions took 0-500 ticks being trained on 300 epochs. [ABSTRACT FROM AUTHOR]

Convolutional neural networks also encountered some problems in the development of image recognition. The most prominent problem is that it is costly and time-consuming to collect data sets and train models. Limited data sets will cause the trained models to overfit. This paper proposes two methods to reduce overfitting based on the residual neural network architecture. The first type of method proposes a method of cross-combining waivers, reducing the size of the convolution kernel, and reducing the number of convolution kernels. The fitting method uses cross-combination to make the accuracy of Kaggle cat and dog data on the validation data set reach 95.37% and 90.31% on 30 types of engineering practice verification data set. The second method is based on the finetune residual neural network. A method of recurrent finetune residual neural network is proposed to improve the accuracy of the model. The accuracy of the finetune residual neural network on the Kaggle cat and dog validation dataset is 99.37%, and the accuracy of the dataset is verified in 30 types of engineering practice. The accuracy is 99.30%. The residual neural network method achieves 99.68% accuracy in the Kaggle cat and dog validation dataset and 99.61% in the validation dataset for 30 types of engineering practice. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) has been widely used all over the world. AI can be applied not only in mechanical learning and expert system but also in knowledge engineering and intelligent information retrieval and has achieved amazing results. This article aims to study the relevant knowledge of deep learning algorithms and multisensor information fusion and how to use deep learning algorithms and multisensor information fusion to study AI algorithms. This paper raises the question of whether the improved multisensor information fusion will affect the AI algorithm. From the data in the experiment of this article, the accuracy of the neural network before the improvement was 4.1%. With the development of society, the traditional algorithm finally dropped to 1.3%. The accuracy of the multisensor information fusion algorithm before the improvement was 3.1% at the beginning; with the development of society, it finally dropped to 1%; it can be known that the accuracy of the improved neural network is 4.6%, and with continuous improvement, it finally increased to 9.8%. The improved multisensor information fusion algorithm is the same, the accuracy at the beginning was 3.9%, and gradually increased to 9.5%. From this set of data, it can be known that the improved convolutional neural network (CNN) algorithm, and the improved multisensor information fusion algorithm should be used to study AI algorithms. [ABSTRACT FROM AUTHOR]

In e-commerce and on social media, identifying fake opinions has become a tremendous challenge. Such opinions are widely generated on the internet by fake viewers, also called fraudsters. They write deceptive reviews that purport to reflect actual user experience either to promote some products or to defame others. They also target the reputations of e-businesses. Their aim is to mislead customers to make a wrong purchase decision by selecting undesired products. Such reviewers are often paid by rival e-business companies to compose positive reviews of their products and/or negative reviews of other companies’ products. The main objective of this paper is to detect, analyze and calculate the difference between fake and truthful product reviews. To do this, the methodology has planned to have seven phases: reviewing online products, analyzing features through linguistic enquiry and word count (LIWC), preprocessing the data to clean and normalize them, embedding words (Word2Vec) and analyzing performance using artificial deep-learning algorithms for classifying fake and truthful reviews. Two deep-learning neural network models have been evaluated based on standard Yelp product reviews. These models are bidirectional long-short term memory (BiLSTM) and convolutional neural network (CNN). The results from comparing the performance of the two models showed that the BiLSTM model provided higher accuracy for detecting fake reviews than the CNN model. [ABSTRACT FROM AUTHOR]

The occurrence or change in the diseases in a specific area can be predicted in advance with the help of plant disease forecasting model. This helps to undertake suitable management measures to avoid the losses well in advance. Disease forecasting predicts probable outbreaks or increased disease intensity over a period in a particular area. This technique helps in timely application of chemicals to plants, which also involve all activities of crop protection and intimate the farmers in the community via text messages or e-mail etc. means of communication. Environment controls the evolution and survival period of various pathogens. Environmental conditions like minimum leaf wetness duration, soil moisture, micro-level relative humidity etc. contribute in evolution of disease causing pathogens. Disease forecasting system thus helps in predicting and avoiding evolution and spread of diseases. This paper uses Machine Learning (ML) and Deep Learning (DL) algorithms to detect, classify and predict the possible pathogens/diseases in the particular type of crop/plant considering based on weather conditions. Temperature, moisture and humidity are the parameters taken into consideration. Convolution Neural Networks (CNN), Recurrent Neural Network (RNN), Artificial Neural Network (ANN), Support Vector Machines (SVM) and K-Nearest Neighbours (KNN) are the five algorithms implemented and compared based on the obtained output accuracy. ANN outperforms all the other algorithms compared in this paper with accuracy of 90.79%. [ABSTRACT FROM AUTHOR]

Neural network algorithms and intelligent algorithms are hot topics in the field of deep learning. In this study, the neural network algorithm and intelligence are optimized, and it is used in simulation experiments to improve the target image recognition ability of the algorithm in the machine vision environment. First, this paper introduces the application of neural networks in the field of machine vision. Second, in the experiment, the improved VGG-16 convolutional neural network (CNN) model is applied to metal block defect detection. Experimental results show that the optimized network can classify metal block defects with the maximum accuracy of 99.28%. Then, the intelligent algorithm based on neural network is studied, and the CIFAR-10 data set is taken as the experimental target for training test and verification test. Using BP algorithm, particle swarm optimization algorithm (PSO-BP), and improved neural network algorithm, respectively, the convergence speed of ICS algorithm based on BP neural network is compared. In contrast, ICS-BP algorithm has the fastest convergence speed and converges when the number of iterations is 32, followed by PSO-BP algorithm. [ABSTRACT FROM AUTHOR]

Recently, unmanned aerial vehicles (UAVs) enhance connectivity and accessibility for civilian and military applications. A group of UAVs with on-board cameras usually monitors or collects information about designated areas. The UAVs can build a distributed network to share/exchange and to process collected sensing data before sending to a data processing center. A huge data transmission among them may cause latency and high-energy consumption. This paper deploys artificial intelligent (AI) techniques to process the video data streaming among the UAVs. Thus, each distributed UAV only needs to send a certain required information to each other. Each UAV processes data utilizing AI and only sends the data that matters to the others. The UAVs, formed as a connected network, communicate within a short communication range and share their own data to each other. Convolution neural network (CNN) technique extracts feature from images automatically that the UAVs only send the moving objects instead of the whole frames. This significantly reduces redundant information for either each UAV or the whole network and saves a huge energy consumption for the network. The UAVs can also save energy for their motion in the sensing field. In addition, a flocking control algorithm is deployed to lead the group of UAVs in the working fields and to avoid obstacles if needed. Simulation and experimental results are provided to verify the proposed algorithms in either AI-based data processing or controlling the UAVs. The results show promising points to save energy for the networks. [ABSTRACT FROM AUTHOR]

In this study, we demonstrate the viability of applying artificial intelligence (AI) techniques to conduct inspections at the OLED cell level using simulated data. The implementation of AI technologies necessitates training data, which we addressed by generating an OLED dataset via our proprietary A2G algorithm, integrating the finite element method among concerns over data security. Our A2G algorithm is designed to produce time-dependent datasets and establish threshold conditions for the expansion of dark spots based on OLED parameters and predicted lifespan. We explored the potential integration of AI in the inspection phase, performing cell-based evaluations using three distinct convolutional neural network models. The test results yielded a promising 95% recognition rate when classifying OLED data into pass and fail categories, demonstrating the practical effectiveness of this approach. Through this research, we not only confirmed the feasibility of using simulated OLED data in place of actual data but also highlighted the potential for the automation of manual inspection processes. Furthermore, by introducing OLED defect detection models at the cell level, as opposed to the traditional panel level during inspections, we anticipate higher classification rates and improved yield. This forward-thinking approach underscores significant advancement in OLED inspection processes, indicating a potential shift in industry standards. [ABSTRACT FROM AUTHOR]

The accuracy of regional convolutional neural network (R‐CNN) algorithms on image recognition detection remains to be improved. The authors optimised the Mask R‐CNN algorithm and tested it through experiments on the automatic recognition of building targets in urban remote sensing images. It was found that the improved Mask R‐CNN algorithm recognised more complete building targets and clearer edges than the original algorithm with a precision of 95.75%, a recall rate of 96.28% and a mean average precision (mAP) of 0.9403, and it also reduced the detection time per image to 0.264 s, all of which were better than other R‐CNN algorithms. The ablation experiments showed that compared with the original Mask R‐CNN algorithm, the improvement in the mAP of the Mask R‐CNN algorithms with an improved feature pyramid network and an improved non‐maximum suppression (NMS) algorithm was 0.0206 and 0.0119, respectively, while the improvement in the mAP of the improved Mask R‐CNN algorithm was 0.0376. The two improvement methods adopted for the Mask R‐CNN algorithm were proved to be feasible and can effectively improve the automatic recognition and detection accuracy and efficiency of building targets in urban remote sensing images. [ABSTRACT FROM AUTHOR]

There has been growing scientific interest in the research field of deep learning techniques applied to skin cancer diagnosis in the last decade. Though encouraging data have been globally reported, several discrepancies have been observed in terms of study methodology, result presentations and validation in clinical settings. The present review aimed to screen the scientific literature on the application of DL techniques to dermoscopic melanoma/nevi differential diagnosis and extrapolate those original studies adequately by reporting on a DL model, comparing them among clinicians and/or another DL architecture. The second aim was to examine those studies together according to a standard set of statistical measures, and the third was to provide dermatologists with a comprehensive explanation and definition of the most used artificial intelligence (AI) terms to better/further understand the scientific literature on this topic and, in parallel, to be updated on the newest applications in the medical dermatologic field, along with a historical perspective. After screening nearly 2000 records, a subset of 54 was selected. Comparing the 20 studies reporting on convolutional neural network (CNN)/deep convolutional neural network (DCNN) models, we have a scenario of highly performant DL algorithms, especially in terms of low false positive results, with average values of accuracy (83.99%), sensitivity (77.74%), and specificity (80.61%). Looking at the comparison with diagnoses by clinicians (13 studies), the main difference relies on the specificity values, with a +15.63% increase for the CNN/DCNN models (average specificity of 84.87%) compared to humans (average specificity of 64.24%) with a 14,85% gap in average accuracy; the sensitivity values were comparable (79.77% for DL and 79.78% for humans). To obtain higher diagnostic accuracy and feasibility in clinical practice, rather than in experimental retrospective settings, future DL models should be based on a large dataset integrating dermoscopic images with relevant clinical and anamnestic data that is prospectively tested and adequately compared with physicians. [ABSTRACT FROM AUTHOR]

Recently, advances in sensor technologies, data communication paradigms, and data processing algorithms all affect the feasibilities of the bridges structural health monitoring and deterioration detection, and other implementations of monitoring operations. The paper proposes a method to design an irregularity detection and monitoring system for road bridges that combines internet of things (IoT) and artificial intelligence (AI) technologies. Raspberry Pi 4 embedded computer integrating IoT and AI technology with convolutional neural network (CNN) is employed to simultaneously monitor remote bridges on websites and apps via Google Firebase cloud database. The first step of successful testing in the laboratory showed that the system can work stably and coincide with the proposed goals. [ABSTRACT FROM AUTHOR]

In this paper, we have proposed an algorithm to design Convolutional Neural Networks (CNN) structures using Genetic Alghorithms (GAs) that are able to learn the best CNN architecture in a completely automatic manner based on limited computing resources. A coding strategy based on sophisticated, hand- designed, modern network blocks has been proposed. So that, the proposed algorithm does not require users with prior knowledge of CNNs, the problem being addressed or even GAs. The performance of the proposed algorithm was evaluated by conducting a series of experiments with widely used reference datasets for image classification tasks and comparing results with modern algorithms that have shown promising performance in this field. The experimental results showed that the proposed algorithm can be used to automatically find a competitive CNN structure compared to modern models, as this algorithm achieved the best classification accuracy among manually and automatically designed CNNs as well as competitive classification accuracy for semi-automatic algorithms. [ABSTRACT FROM AUTHOR]

Apical periodontitis (AP) is one of the most prevalent disorders in dentistry. However, it can be underdiagnosed in asymptomatic patients. In addition, the perioperative evaluation of 3-dimensional (3D) lesion volume is of great clinical relevance, but the required slice-by-slice manual delineation method is time- and labor-intensive. Here, for quickly and accurately detecting and segmenting periapical lesions (PALs) associated with AP on cone beam computed tomography (CBCT) images, we proposed and geographically validated a novel 3D deep convolutional neural network algorithm, named PAL-Net. On the internal 5-fold cross-validation set, our PAL-Net achieved an area under the receiver operating characteristic curve (AUC) of 0.98. The algorithm also improved the diagnostic performance of dentists with varying levels of experience, as evidenced by their enhanced average AUC values (junior dentists: 0.89–0.94; senior dentists: 0.91–0.93), and significantly reduced the diagnostic time (junior dentists: 69.3 min faster; senior dentists: 32.4 min faster). Moreover, our PAL-Net achieved an average Dice similarity coefficient over 0.87 (0.85–0.88), which is superior or comparable to that of other existing state-of-the-art PAL segmentation algorithms. Furthermore, we validated the generalizability of the PAL-Net system using multiple external data sets from Central, East, and North China, showing that our PAL-Net has strong robustness. Our PAL-Net can help improve the diagnostic performance and speed of dentists working from CBCT images, provide clinically relevant volume information to dentists, and can potentially be applied in dental clinics, especially without expert-level dentists or radiologists. [ABSTRACT FROM AUTHOR]

Environmental comfort takes a central role in the well-being and health of people. In modern industrial, commercial, and residential buildings, passive energy sources (such as solar irradiance and heat exchangers) and heating, ventilation, and air conditioning (HVAC) systems are usually employed to achieve the required comfort. While passive strategies can effectively enhance the livability of indoor spaces with limited or no energy cost, active strategies based on HVAC machines are often preferred to have direct control over the environment. Commonly, the working parameters of such machines are manually tuned to a fixed set point during working hours or throughout the whole day, leading to inefficiencies in terms of comfort and energy consumption. Albeit effective, previous works that tackle the comfort–energy tradeoff are tailored to the specific environment under study (in terms of geometry, characteristics of the building, etc.) and thus cannot be applied on a large industrial scale. We address the problem from a different angle and propose an adaptive and practical solution for comfort optimisation. It does not require the intervention of expert personnel or any customisations around the environment while it implicitly analyses the influence of different agents (e.g., passive phenomena) on the monitored parameters. A convolutional neural network (CNN) predicts the long-term impact on thermal comfort and energy consumption of a range of possible actuation strategies for the HVAC system. The decision on the best HVAC settings is taken by choosing the combination of ON/OFF and set point (SP), which optimises thermal comfort and, at the same time, minimises energy consumption. We validate our solution in a real-world scenario and through software simulations, providing a performance comparison against the fixed set point strategy and a greedy approach. The evaluation results show that our solution achieves the desired thermal comfort while reducing the energy footprint by up to approximately 16% in a real environment. [ABSTRACT FROM AUTHOR]

To equip computers with human communication skills and to enable natural interaction between the computer and a human, intelligent solutions are required based on artificial intelligence (AI) methods, algorithms, and sensor technology. This study aimed at identifying and analyzing the state-of-the-art AI methods and algorithms and sensors technology in existing human–computer intelligent interaction (HCII) research to explore trends in HCII research, categorize existing evidence, and identify potential directions for future research. We conduct a systematic mapping study of the HCII body of research. Four hundred fifty-four studies published in various journals and conferences between 2010 and 2021 were identified and analyzed. Studies in the HCII and IUI fields have primarily been focused on intelligent recognition of emotion, gestures, and facial expressions using sensors technology, such as the camera, EEG, Kinect, wearable sensors, eye tracker, gyroscope, and others. Researchers most often apply deep-learning and instance-based AI methods and algorithms. The support sector machine (SVM) is the most widely used algorithm for various kinds of recognition, primarily an emotion, facial expression, and gesture. The convolutional neural network (CNN) is the often-used deep-learning algorithm for emotion recognition, facial recognition, and gesture recognition solutions. [ABSTRACT FROM AUTHOR]

The field of explainable artificial intelligence (XAI) aims to build explainable and interpretable machine learning (or deep learning) methods without sacrificing prediction performance. Convolutional neural networks (CNNs) have been successful in making predictions, especially in image classification. These popular and well-documented successes use extremely deep CNNs such as VGG16, DenseNet121, and Xception. However, these well-known deep learning models use tens of millions of parameters based on a large number of pretrained filters that have been repurposed from previous data sets. Among these identified filters, a large portion contain no information yet remain as input features. Thus far, there is no effective method to omit these noisy features from a data set, and their existence negatively impacts prediction performance. In this paper, a novel interaction-based convolutional neural network (ICNN) is introduced that does not make assumptions about the relevance of local information. Instead, a model-free influence score (I-score) is proposed to directly extract the influential information from images to form important variable modules. This innovative technique replaces all pretrained filters found by trial-and-error with explainable, influential, and predictive variable sets (modules) determined by the I-score. In other words, future researchers need not rely on pretrained filters; the suggested algorithm identifies only the variables or pixels with high I-score values that are extremely predictive and important. The proposed method and algorithm were tested on real-world data set and a state-of-the-art prediction performance of 99.8% was achieved without sacrificing the explanatory power of the model. This proposed design can efficiently screen patients infected by COVID-19 before human diagnosis and can be a benchmark for addressing future XAI problems in large-scale data sets. [ABSTRACT FROM AUTHOR]

Digital pathology is vital for the correct diagnosis of kidney before transplantation or kidney disease identification. One of the key challenges in kidney diagnosis is glomerulus detection in kidney tissue segments. In this study, we propose a deep learning–based method for glomerulus detection from digitized kidney slide segments. The proposed method applies models based on convolutional neural networks to detect image segments containing the glomerulus region. We employ various networks such as ResNets, UNet, LinkNet, and EfficientNet to train the models. In our experiments on a network trained on the NIH HuBMAP kidney whole slide image dataset, the proposed method achieves the highest scores with Dice coefficient of 0.942. [ABSTRACT FROM AUTHOR]

In large datasets, irrelevant, redundant, and noisy attributes are often present. These attributes can have a negative impact on the classification model accuracy. Therefore, feature selection is an effective pre-processing step intended to enhance the classification performance by choosing a small number of relevant or significant features. It is important to note that due to the NP-hard characteristics of feature selection, the search agent can become trapped in the local optima, which is extremely costly in terms of time and complexity. To solve these problems, an efficient and effective global search method is needed. Sand cat swarm optimization (SCSO) is a newly introduced metaheuristic algorithm that solves global optimization algorithms. Nevertheless, the SCSO algorithm is recommended for continuous problems. bSCSO is a binary version of the SCSO algorithm proposed here for the analysis and solution of discrete problems such as wrapper feature selection in biological data. It was evaluated on ten well-known biological datasets to determine the effectiveness of the bSCSO algorithm. Moreover, the proposed algorithm was compared to four recent binary optimization algorithms to determine which algorithm had better efficiency. A number of findings demonstrated the superiority of the proposed approach both in terms of high prediction accuracy and small feature sizes. [ABSTRACT FROM AUTHOR]

The modeling of channel and temporal information is of crucial importance for action recognition tasks. To build a high-performance action recognition network by effectively capturing channel and temporal information, we propose CLS-Net: an action recognition algorithm based on channel-temporal information modeling. The proposed CLS-Net characterizes channel and temporal information by inserting multiple modules to an end-to-end backbone network, including a channel attention module (CA module) for modeling channel information, a long-term temporal module (LT module) and a short-term temporal module (ST module) for modeling temporal information. Specifically, the CA module extracts the correlation between feature channels so the network can learn to selectively strengthen the features containing useful information and suppress the useless features through global information. The LT module moves some channels in the temporal dimension to realize information interaction across time domains and model global temporal information. The ST module enhances the motion-sensitive features by calculating the feature-level frame difference information and realizes the representation of local motion information. Since the multi-module insertion mode directly affects the whole model's final performance, we propose a novel multi-module insertion mode instead of a simple series or parallel connection to ensure that the multiple modules can complement one another and cooperate with each other more efficiently. CLS-Net achieves SOTA performance on the EgoGesture and Jester dataset in the same type of network and achieves competitive results on the Something-Something V2 dataset. [ABSTRACT FROM AUTHOR]

The purpose of this study is to explore how machine learning technologies can improve healthcare operations management. A machine learning-based model to solve a specific medical problem is developed to achieve this research purpose. Specifically, this study presents an AI solution for malaria infection diagnosis by applying the CNN (convolutional neural network) algorithm. Based on malaria microscopy image data from the NIH National Library of Medicine, a total of 24,958 images were used for deep learning training, and 2600 images were selected for final testing of the proposed diagnostic architecture. The empirical results indicate that the CNN diagnostic model correctly classified most malaria-infected and non-infected cases with minimal misclassification, with performance metrics of precision (0.97), recall (0.99), and f1-score (0.98) for uninfected cells, and precision (0.99), recall (0.97), and f1-score (0.98) for parasite cells. The CNN diagnostic solution rapidly processed a large number of cases with a high reliable accuracy of 97.81%. The performance of this CNN model was further validated through the k-fold cross-validation test. These results suggest the advantage of machine learning-based diagnostic methods over conventional manual diagnostic methods in improving healthcare operational capabilities in terms of diagnostic quality, processing costs, lead time, and productivity. In addition, a machine learning diagnosis system is more likely to enhance the financial profitability of healthcare operations by reducing the risk of unnecessary medical disputes related to diagnostic errors. As an extension for future research, propositions with a research framework are presented to examine the impacts of machine learning on healthcare operations management for safety and quality of life in global communities. [ABSTRACT FROM AUTHOR]

The tire sidewall is the weakest part of the entire tire. Although the tire sidewall is not directly in contact with the ground, it often undergoes great deformation. Weather, road conditions, and driving habits can also affect the tire life. Cracking is one of the earliest signs of tire aging and deterioration. If a driver does not regularly inspect their vehicle, damage to a tire may remain undetected and an uncontrolled tire explosion may occur. In this study, we use deeplearning-based artificial intelligence computer vision to train a deep neural network model using a large number of digital images to detect tire sidewall cracks instead of traditional sensors, inspection devices, or visual inspection methods. In this study, tire sidewall crack images were preprocessed and annotated using the annotation program VGG Image Annotator (VIA). Residual network 50 (ResNet50) is used as the backbone of mask-region-based convolutional neural networks (Mask R-CNNs). The preprocessing training and test results of our dataset show that the improved Mask R-CNN has better mean accuracy (mAP) and detection accuracy than the original Mask R-CNN and Faster-R-CNN and can not only reduce inspection costs and time, but also improve the efficiency of tire crack analysis. [ABSTRACT FROM AUTHOR]

Simple Summary: Histology sample images are usually diagnosed definitively based on the radiologist's extensive knowledge, yet, owing to the highly gritty visual appearance of such images, specialists sometimes differ on their evaluations. Automating the image diagnostic process and decreasing the analysis time may be achieved via the use of advanced deep learning algorithms. Diagnostic objectivity may be improved with the use of more effective and accurate automated technologies by lessening the differences between the humans. In this research, we propose a CNN model architecture for cancer image classification by accumulating layers closer together to further merge the semantic and spatial features. Regarding precision, our suggested cutting-edge model improves upon the current state-of-the-art approaches. The definitive diagnosis of histology specimen images is largely based on the radiologist's comprehensive experience; however, due to the fine to the coarse visual appearance of such images, experts often disagree with their assessments. Sophisticated deep learning approaches can help to automate the diagnosis process of the images and reduce the analysis duration. More efficient and accurate automated systems can also increase the diagnostic impartiality by reducing the difference between the operators. We propose a FabNet model that can learn the fine-to-coarse structural and textural features of multi-scale histopathological images by using accretive network architecture that agglomerate hierarchical feature maps to acquire significant classification accuracy. We expand on a contemporary design by incorporating deep and close integration to finely combine features across layers. Our deep layer accretive model structure combines the feature hierarchy in an iterative and hierarchically manner that infers higher accuracy and fewer parameters. The FabNet can identify malignant tumors from images and patches from histopathology images. We assessed the efficiency of our suggested model standard cancer datasets, which included breast cancer as well as colon cancer histopathology images. Our proposed avant garde model significantly outperforms existing state-of-the-art models in respect of the accuracy, F1 score, precision, and sensitivity, with fewer parameters. [ABSTRACT FROM AUTHOR]

The present work focuses on the investigation of an artificial intelligence (AI) algorithm for brightest objects segmentation in night sky images' field of view (FOV). This task is mandatory for many applications that want to focus on the brightest objects in an optical sensor image with a particular shape: point-like or streak. The algorithm is developed as a dedicated application for star sensors both for attitude determination (AD) and onboard space surveillance and tracking (SST) tasks. Indeed, in the former, the brightest objects of most concern are stars, while in the latter they are resident space objects (RSOs). Focusing attention on these shapes, an AI-based segmentation approach can be investigated. This will be carried out by designing, developing and testing a convolutional neural network (CNN)-based algorithm. In particular, a U-Net will be used to tackle this problem. A dataset for the design process of the algorithm, network training and tests is created using both real and simulated images. In the end, comparison with traditional segmentation algorithms will be performed, and results will be presented and discussed together with the proposal of an electro-optical payload for a small satellite for an in-orbit validation (IOV) mission. [ABSTRACT FROM AUTHOR]

As is well known, image segmentation is widely used in the fields of echocardiography and diagnostic and interventional radiology. The delineation of structural components of various organs from 2D images is a technique used in the medical field in order to identify intervention targets with increasing precision and accuracy. In recent decades, the automation of this task has been the subject of intensive research. In particular, to improve the segmentation of such images, investigations have focused on the use of neural networks, and in particular convolutional neural networks (CNNs). However, most existing CNN-based methods can produce unsatisfactory segmentation masks without precise object boundaries (Wang, Chen, Ji, Fan & Ye Li, 2022); this is mainly due to the shadows and high noise in these images. To address the problem of automated image segmentation, this work proposes a pipeline technique with two stages (applied primarily to the echocardiographic domain): the first consisting of a Self-normalising Neural Networks (SNNs) performs image denoising, while the second applies a Watershed segmentation algorithm on the cleaned image. The latter is a technique successfully applied in geomatics and land surveying. The proposed methodology may be of interest both in the medical field and in the field of Geomatics where segmentation and classification operations are required in different application areas. [ABSTRACT FROM AUTHOR]

Machine-learning (ML) requires human-labeled "truth" data to train and test. Acquiring and labeling this data can often be the most time-consuming and expensive part of developing trained models of convolutional neural networks (CNN). In this work, we show that an automated workflow using automatically labeled synthetic data can be used to drastically reduce the time and effort required to train a machine learning algorithm for detecting buildings in aerial imagery acquired with low-flying unmanned aerial vehicles. The MSU Autonomous Vehicle Simulator (MAVS) was used in this work, and the process for integrating MAVS into an automated workflow is presented in this work, along with results for building detection with real and simulated images. [ABSTRACT FROM AUTHOR]

Analyzing vibration data using deep neural networks is an effective way to detect damages in rotating machinery at an early stage. However, the black-box approach of these methods often does not provide a satisfactory solution because the cause of classifications is not comprehensible to humans. Therefore, this work investigates the application of the explainable AI (XAI) algorithms to convolutional neural networks for vibration-based condition monitoring. Thus, the three XAI algorithms GradCAM, LRP and LIME with a modified perturbation strategy are applied to classifications based on the Fourier transform as well as the order analysis of the vibration signal. The following visualization as frequency-RPM maps and order-RPM maps allows for an effective assessment of saliency values for variable periodicity of the data, which translates to a varying rotation speed of a real-world machine. To compare the explanatory power of the XAI methods, investigations are first carried out with a synthetic data set with known class-specific characteristics. Both a visual and a quantitative analysis of the resulting saliency maps are presented. Then, a real-world data set for vibration-based imbalance classification on an electric motor, which runs at a broad range of rotation speeds, is used. The results indicate that the investigated algorithms are each partially successful in providing sample-specific saliency maps which highlight class-specific features and omit features which are not relevant for classification. [ABSTRACT FROM AUTHOR]

The modern urban environment is becoming more and more complex. In helping us identify surrounding objects, vehicle vision sensors rely more on the semantic segmentation ability of deep learning networks. The performance of a semantic segmentation network is essential. This factor will directly affect the comprehensive level of driving assistance technology in road environment perception. However, the existing semantic segmentation network has a redundant structure, many parameters, and low operational efficiency. Therefore, to reduce the complexity of the network and reduce the number of parameters to improve the network efficiency, based on the deep learning (DL) theory, a method for efficient image semantic segmentation using Deep Convolutional Neural Network (DCNN) is deeply studied. First, the theoretical basis of the convolutional neural network (CNN) is briefly introduced, and the real-time semantic segmentation technology of urban scenes based on DCNN is recommended in detail. Second, the atrous convolution algorithm and the multi-scale parallel atrous spatial pyramid model are introduced. On the basis of this, an Efficient Symmetric Network (ESNet) of real-time semantic segmentation model for autonomous driving scenarios is proposed. The experimental results show that: (1) On the Cityscapes dataset, the ESNet structure achieves 70.7% segmentation accuracy for the 19 semantic categories set, and 87.4% for the seven large grouping categories. Compared with other algorithms, the accuracy has increased to varying degrees. (2) On the CamVid dataset, compared with segmentation networks of multiple lightweight real-time images, the parameters of the ESNet model are around 1.2 m, the highest FPS value is around 90 Hz, and the highest mIOU value is around 70%. In seven semantic categories, the segmentation accuracy of the ESNet model is the highest at around 98%. From this, we found that the ESNet significantly improves segmentation accuracy while maintaining faster forward inference speed. Overall, the research not only provides technical support for the development of real-time semantic understanding and segmentation of DCNN algorithms but also contributes to the development of artificial intelligence technology. [ABSTRACT FROM AUTHOR]

This work presents a high-performance approach for 3D lung segmentation tasks in computer tomography images using a new intelligent machine learning algorithm called Floor of Log(FoL). The Support Vector Machine was used to learn the better parameter of the FoL algorithm using the parenchyma and its border as labels. Sensitivity, Matthews Correlation Coefficient (MCC), Hausdorff Distance (HD), Dice, Accuracy (ACC), and Jaccard were used to evaluate the proposed algorithm. The FoL was compared with recent 3D region growing, 3D Adaptive Crisp Active Contour, 3D OsiriX toolbox, and Level-set algorithm based on the coherent propagation method algorithms. The FoL algorithm achieves good results with approximately 19 s in the most significant result in an exam with 430 slices and presents similarity indexes achieving HD 3.5, DICE 83.63, and Jaccard 99.73 and qualitative indexes achieving Sensitivity 83.87, MCC 83.08, and ACC 99.62. The proposed approach of this work showed a simple and powerful algorithm to segment lung in computer tomography images of the chest region by combining similar textures, highlighting the lung structure. The FoL was presented as a new supervised clustering algorithm which can be trained to achieve better results and attached to other approaches as Convolutional Deep Neural Networks applications. [ABSTRACT FROM AUTHOR]

Edge Analytics and Artificial Intelligence are important features of the current smart connected living community. In a society where people, homes, cities, and workplaces are simultaneously connected through various devices, primarily through mobile devices, a considerable amount of data is exchanged, and the processing and storage of these data are laborious and difficult tasks. Edge Analytics allows the collection and analysis of such data on mobile devices, such as smartphones and tablets, without involving any cloud-centred architecture that cannot guarantee real-time responsiveness. Meanwhile, Artificial Intelligence techniques can constitute a valid instrument to process data, limiting the computation time, and optimising decisional processes and predictions in several sectors, such as healthcare. Within this field, in this article, an approach able to evaluate the voice quality condition is proposed. A fully automatic algorithm, based on Deep Learning, classifies a voice as healthy or pathological by analysing spectrogram images extracted by means of the recording of vowel /a/, in compliance with the traditional medical protocol. A light Convolutional Neural Network is embedded in a mobile health application in order to provide an instrument capable of assessing voice disorders in a fast, easy, and portable way. Thus, a straightforward mobile device becomes a screening tool useful for the early diagnosis, monitoring, and treatment of voice disorders. The proposed approach has been tested on a broad set of voice samples, not limited to the most common voice diseases but including all the pathologies present in three different databases achieving F1-scores, over the testing set, equal to 80%, 90%, and 73%. Although the proposed network consists of a reduced number of layers, the results are very competitive compared to those of other “cutting edge” approaches constructed using more complex neural networks, and compared to the classic deep neural networks, for example, VGG-16 and ResNet-50. [ABSTRACT FROM AUTHOR]

Background: Transcriptome sequencing has been broadly available in clinical studies. However, it remains a challenge to utilize these data effectively for clinical applications due to the high dimension of the data and the highly correlated expression between individual genes.Methods: We proposed a method to transform RNA sequencing data into artificial image objects (AIOs) and applied convolutional neural network (CNN) algorithms to classify these AIOs. With the AIO technique, we considered each gene as a pixel in an image and its expression level as pixel intensity. Using the GSE96058 (n = 2976), GSE81538 (n = 405), and GSE163882 (n = 222) datasets, we created AIOs for the subjects and designed CNN models to classify biomarker Ki67 and Nottingham histologic grade (NHG).Results: With fivefold cross-validation, we accomplished a classification accuracy and AUC of 0.821 ± 0.023 and 0.891 ± 0.021 for Ki67 status. For NHG, the weighted average of categorical accuracy was 0.820 ± 0.012, and the weighted average of AUC was 0.931 ± 0.006. With GSE96058 as training data and GSE81538 as testing data, the accuracy and AUC for Ki67 were 0.826 ± 0.037 and 0.883 ± 0.016, and that for NHG were 0.764 ± 0.052 and 0.882 ± 0.012, respectively. These results were 10% better than the results reported in the original studies. For Ki67, the calls generated from our models had a better power for prediction of survival as compared to the calls from trained pathologists in survival analyses.Conclusions: We demonstrated that RNA sequencing data could be transformed into AIOs and be used to classify Ki67 status and NHG with CNN algorithms. The AIO method could handle high-dimensional data with highly correlated variables, and there was no need for variable selection. With the AIO technique, a data-driven, consistent, and automation-ready model could be developed to classify biomarkers with RNA sequencing data and provide more efficient care for cancer patients. [ABSTRACT FROM AUTHOR]

This study aims to develop a novel automated computer vision algorithm for quality inspection of surfaces with complex patterns. The proposed algorithm is based on both an autoencoder (AE) and a fully convolutional neural network (FCN). The AE is adopted for the self-generation of templates from test targets for defect detection. Because the templates are produced from the test targets, the position alignment issues for the matching operations between templates and test targets can be alleviated. The FCN is employed for the segmentation of a template into a number of coherent regions. Because the AE has the limitation that its capacities for the regeneration of each coherent region in the template may be different, the segmentation of the template by FCN is beneficial for allowing the inspection of each region to be independently carried out. In this way, more accurate detection results can be achieved. Experimental results reveal that the proposed algorithm has the advantages of simplicity for training data collection, high accuracy for defect detection, and high flexibility for online inspection. The proposed algorithm is therefore an effective alternative for the automated inspection in smart factories with a growing demand for the reliability for high quality production. [ABSTRACT FROM AUTHOR]

Objectives: To develop and evaluate a deep learning–based model capable of detecting primary hepatic malignancies in multiphase CT images of patients at high risk for hepatocellular carcinoma (HCC). Methods: A total of 1350 multiphase CT scans of 1280 hepatic malignancies (1202 HCCs and 78 non-HCCs) in 1320 patients at high risk for HCC were retrospectively analyzed. Following the delineation of the focal hepatic lesions according to reference standards, the CT scans were categorized randomly into the training (568 scans), tuning (193 scans), and test (589 scans) sets. Multiphase CT information was subjected to multichannel integration, and livers were automatically segmented before model development. A deep learning–based model capable of detecting malignancies was developed using a mask region–based convolutional neural network. The thresholds of the prediction score and the intersection over union were determined on the tuning set corresponding to the highest sensitivity with < 5 false-positive cases per CT scan. The sensitivity and the number of false-positives of the proposed model on the test set were calculated. Potential causes of false-negatives and false-positives on the test set were analyzed. Results: This model exhibited a sensitivity of 84.8% with 4.80 false-positives per CT scan on the test set. The most frequent potential causes of false-negatives and false-positives were determined to be atypical enhancement patterns for HCC (71.7%) and registration/segmentation errors (42.7%), respectively. Conclusions: The proposed deep learning–based model developed to automatically detect primary hepatic malignancies exhibited an 84.8% of sensitivity with 4.80 false-positives per CT scan in the test set. Key Points: • Image processing, including multichannel integration of multiphase CT and automatic liver segmentation, enabled the application of a deep learning–based model to detect primary hepatic malignancy. • Our model exhibited a sensitivity of 84.8% with a false-positive rate of 4.80 per CT scan. [ABSTRACT FROM AUTHOR]