Deep learning is an emerging area of machine learning (ML) research. It comprises multiple hidden layers of artificial neural networks. The deep learn- ing methodology applies nonlinear transformations and model abstractions of high level in large databases. The recent advancements in deep learning architec- tures within numerous fields have already provided significant contributions in artificial intelligence. This article presents a state of the art survey on the contri- butions and the novel applications of deep learning. The following review chron- ologically presents how and in what major applications deep learning algorithms have been utilized. Furthermore, the superior and beneficial of the deep learning methodology and its hierarchy in layers and nonlinear operations are presented and compared with the more conventional algorithms in the common applica- tions. The state of the art survey further provides a general overview on the novel concept and the ever-increasing advantages and popularity of deep learning.

"Deep fake" technologies are often associated with face-swapping algorithms based on images or videos. Although much research is focused on developing these techniques, there is still room for improvement, especially in human and visual assessment. In this study, we present an innovative approach to the problem of face replacement based on one model while preserving background features. Our model uses the Adaptive Attentional Denormalization method from FaceShifter while integrating identity features from the ArcFace and BiSeNet models to support the attribute extraction process. We used Fast GAN as a generative model to optimize and accelerate the training process on smaller data sets. Our research proves the effectiveness of the proposed approach by combining innovative methods and architectures in one coherent solution.

The popularity of deep reinforcement learning (DRL) methods in economics have been exponentially increased. DRL through a wide range of capabilities from reinforcement learning (RL) and deep learning (DL) for handling sophisticated dynamic business environments offers vast opportunities. DRL is characterized by scalability with the potential to be applied to high-dimensional problems in conjunction with noisy and nonlinear patterns of economic data. In this work, we first consider a brief review of DL, RL, and deep RL methods in diverse applications in economics providing an in-depth insight into the state of the art. Furthermore, the architecture of DRL applied to economic applications is investigated in order to highlight the complexity, robustness, accuracy, performance, computational tasks, risk constraints, and profitability. The survey results indicate that DRL can provide better performance and higher accuracy as compared to the traditional algorithms while facing real economic problems at the presence of risk parameters and the ever-increasing uncertainties.

To date, deep-tech entrepreneurship and sustainable entrepreneurship are both attracting the interest of researchers. Indeed, deep-tech is being employed to address future sustainable challenges in the present. Nevertheless, the process of deep-tech startups is filled with distinct obstacles. These types of startups may necessitate more targeted sustainable entrepreneurial process and specialised knowledge compared to conventional technological startups or general sustainable entrepreneurial process. Currently, there are only publications discussing either a sustainable en-trepreneurial process or conventional technical startups. Therefore, the aim of this article is to in-vestigate the process of sustainable entrepreneurs as they transition from startup ventures to scaleup enterprises within the context of deep-tech industry. Based on previous sustainable entre-preneurial processes, a multiple case study was conducted within six deep-tech startups in order to find out the sustainable process pattern. A conceptual model of sustainable entrepreneurial process in deep-tech industry is established including five phases with six activities: (i) Sustainable idea definition; (ii) Sustainable opportunity recognition and evaluation; (iii) Venture launch; (iv.i) Sustainable products/ services development; (iv.ii) New sustainable enterprise financing; and (v) Sustainable enterprise scaleup. This process enables sustainable deep-tech entrepreneurs to necessarily have an overview of key entrepreneurial stages to navigate from the startup to scaleup process.

Diabetes mellitus is a popular life-threatening disease and patients may gradually have started suffering from other diabetes-causing diseases such as heart attacks, stroke, hypertension, blurry vision, blindness, foot ulcer, amputation, kidney damage and other organ failures before diagnosis. Early detection can help reduce the fatality of this disease. Deep learning models have proven very useful in disease detection and computer-aided diagnosis. In this work, we proposed a deep unsupervised machine learning model for early detection of diabetes using voting ensemble feature selection and deep belief neural networks (DBN). Dataset was obtained from an online repository containing responses of prediagnosed patients to direct questionnaires administered in Sylhet Diabetes Hospital in Sylhet, Bangladesh. The dataset was preprocessed and preprocessed. Features were reduced using the ensemble feature selector. The DBN model was pretrained and tuned to obtain optimal performance. The model was also compared with other models with no multiple hidden layers. The DBN performed at its relative best with F1-measure, precision and recall of 1.00, 0.92 and 1.00 respectively. We conclude that DBN is a useful tool for an unsupervised early prediction of Type II diabetes mellitus.

We propose a deep-learning approach for the joint MIMO detection and channel decoding problem. Conventional MIMO receivers adopt a model-based approach for MIMO detection and channel decoding in linear or iterative manners. However, due to the complex MIMO signal model, the optimal solution to the joint MIMO detection and channel decoding problem (i.e., the maximum likelihood decoding of the transmitted codewords from the received MIMO signals) is computationally infeasible. As a practical measure, the current model-based MIMO receivers all use suboptimal MIMO decoding methods with affordable computational complexities. This work applies the latest advances in deep learning for the design of MIMO receivers. In particular, we leverage deep neural networks (DNN) with supervised training to solve the joint MIMO detection and channel decoding problem. We show that DNN can be trained to give much better decoding performance than conventional MIMO receivers do. Our simulations show that a DNN implementation consisting of seven hidden layers can outperform conventional model-based linear or iterative receivers. This performance improvement points to a new direction for future MIMO receiver design.

In this paper, target detection of car images in roads is performed based on the YOLOv8 model of YOLO family of models, which improves the accuracy and generalisation of the target detection task by combining multi-scale prediction, CSPNet structure and optimisation techniques such as BoF and BoS. The input images contain five types of vehicles such as Ambulance, Bus, Car, Motorcycle and Truck, which are analysed and learnt to have a classification accuracy of 75.4% on Ambulance, 53.5% on Bus, 55.1% on Car, 51.1% on Motorcycle and 42.5% on Truck. Despite the gap in specific classification accuracy, the YOLOv8 model can detect 100% of vehicles on the road, demonstrating good target detection capability. This research is of great significance for improving road traffic safety, intelligent traffic management, and the development of future autonomous driving technology. By optimising the deep learning model to achieve more accurate and efficient vehicle target detection, it can help to improve road safety and traffic efficiency, and promote the progress of intelligent transportation systems.

Deep networks-based hashing has gained significant popularity in recent years, particularly in the field of image retrieval. However, most existing methods only focus on extracting semantic information from the final layer, disregarding valuable structural information that contains important semantic details crucial for effective hash learning. To address this limitation and improve image retrieval accuracy, we propose a novel deep hashing method called Deep Supervised Hashing by Fusing Multiscale Deep Features (DSHFMDF). Our approach involves extracting multiscale features from multiple convolutional layers and fusing them to generate more robust representations for efficient image retrieval. Experimental results on CIFAR10 and NUS-WIDE datasets demonstrate that our method surpasses the performance of state-of-the-art hashing techniques.

Groundwater is the most extracted raw material in the world with global annual withdrawal rates of 800–1500 km3/year. In East and Southern Africa, 70 % of the population are reliant on shallow groundwater as their primary drinking water source. With increased population growth, intensification of agriculture and industrialization, conflicting demands on groundwater present a challenge to achieve the Sustainable Development Goals (6,3,11,12,15). Between 2015 and 2018, the Horn of Africa was affected by a series of climatic induced events, namely El Nino, La Nina and the Indian Ocean Diopole. These events modified the variability of rainfall patterns and resulted in long periods of low rainfall, low recharge and high evapotranspiration. As a result, shallow aquifers in alluvial deposits of Somali region have low yields and produce brackish and saline water. That situation prompted humanitarian water professionals to finance the transportation of water from selected locations with high groundwater potential through water trucks to areas facing groundwater depletion and drought. To address this challenge, UNICEF explored alternative, sustainable deeper groundwater sources that could be extracted using solar water pumping technology for multi water use. This paper describes a three-phase methodology of deep groundwater development of wells in the Ogaden Jesoma sandstone aquifers of the Somali region of the Horn of Africa to a depth of 600 meters below ground level. The results concluded that the deep sandstone aquifer of Jesoma can provide fresh water with yields of 15 l/s to the local population of Somali region. to the study provided insights into deep groundwater identification and development as well as adaptive deep boreholes drilling as a source for climate resilient water supplies.

Speech Emotion Recognition (SER) is an interesting and difficult problem to handle. In this paper, we deal with it through the implementation of deep learning networks. We have designed and implemented six different deep learning networks, a Deep Belief Network (DBN), a simple deep neural network (SDNN), a LSTM network (LSTM), a LSTM network with the addition of an attention mechanism (LSTM-ATN), a Convolutional neural network (CNN), and a Convolutional neural network with the addition of an attention mechanism (CNN-ATN), having in mind, apart from solving the SER problem, to test the impact of attention mechanism to the results. Dropout and Batch Normalization techniques are also used to improve the generalization ability (prevention of overfitting) of the models as well as to speed up the training process. The Surrey Audio-Visual Expressed Emotion database (SAVEE), and the Ryerson Audio-Visual Database (RAVDESS) database were used for training and evaluation of our models. The results showed that networks with the addition of the attention mechanism did better than the others. Furthermore, they showed that CNN-ATN was the best among tested networks, achieving an accuracy of 74% for the SAVEE and 77% for the RAVDESS dataset, and exceeded existing state-of-the-art systems for the same datasets.

Artificial intelligence (AI) is a current trend in computer science, which extends itself its amazing capacities to other technologies such as mechatronics and robotics. Going beyond technological applications, the philosophy behind AI is that there is a vague and potential convergence of artificial manufacture and natural world although the limiting approach may be still very far away, but why? The implicit problem is that Darwin theory of evolution focuses on natural world where breeding conservation is the cornerstone of the existence of creature world but there is no similar concept of breeding conservation in artificial world whose things are created by human. However, after developing for a long time until now, AI issues an interesting concept of generation in which artifacts created by computer science can derive their new generations inheriting their aspects / characteristics. Such generated artifacts make us look back on offsprings by the process of breeding conservation in natural world. Therefore, it is possible to think that AI generation, which is a recent subject of AI, is a significant development in computer science as well as high-tech domain. AI generation does not help us to reach near biological evolution even in the case that AI can combine with biological technology but, AI generation can help us to extend our viewpoint about Darwin theory of evolution as well as there may exist some uncertain relationship between man-made world and natural world. Anyhow AI generation is a current important subject in AI and there are two main generative models in computer science: 1) generative model that applies large language model into generating natural language texts understandable by human and 2) generative model that applies deep neural network into generating digital content such as sound, image, and video. This technical report focuses on deep generative model (DGM) for digital content generation, which is a short summary of approaches to implement DGMs. Researchers can read this work as an introduction to DGM with easily understandable explanations.

The rise of the Internet has enabled people to express their opinions on various subjects through social media, blogs, and website comments. As a result, there has been a significant increase in research on sentiment analysis. However, most of the research efforts have focused on analyzing sentiment in English-language data, neglecting the wealth of information available in other languages. In this paper, we provide a comprehensive review of the current state-of-the-art in multilingual sentiment analysis. The survey investigates techniques for data preprocessing, representation learning, and feature extraction in multilingual sentiment analysis. It explores cross-lingual transfer learning, domain adaptation, and data augmentation methods that enhance the performance of sentiment analysis models, particularly in low-resource languages and domains. It provides insights into the state-of-the-art approaches, challenges, and opportunities in this evolving field and encourages further advancements in multilingual sentiment analysis research.

When training deep neural networks tasks, the most popular choices are cross-entropy loss. On the other hand, in general speaking, a decent loss function can take on shapes that are considerably more flexible and ought to be adapted for different activities and datasets. In most of the classification tasks, generally if the true class is not correctly recognized by the network (top1), that class is placed among the five classes with the highest probability (top5). This shows that the network does not necessarily recognize the correct class with a low probability, but a class similar to it (such as 3 vs. 8 in mnist) assigns a higher probability and this causes a mistake in that task. Accordingly, we proposed a loss function deals with the error of class that the neural network incorrectly recognized as correct, in addition to the correct class error. We call our proposed loss as False Positive Loss (FPL), with the intention of viewing and designing loss functions not only through the utilization of true class but also through the utilization of the value of false positive classes. One of the core properties of our proposed loss is full adaptability, which makes False Positive Loss be fully capable of getting reformulated by using other widely used loss functions formulas based on the task or the need of the users. Extensive experimental results demonstrate that our suggested loss function outperforms other well-known losses on a variety of tasks and datasets. As can be observed, the performance of our False Positive Loss is superior to that of the cross-entropy loss when it comes to tasks involving 2D picture classification. We have compared our loss with cross entropy as the most common classification loss function on some models (such as ResNet-18, ResNet-50 and Efficientnet-V2) through classification known as a basic computer vision task. with both random or pre-trained initial weights. As a result, in some cases the models with our loss outperform the same tasks with cross entropy from the viewpoint of some metric (i.e. accuracy and FP). For example, the resnet-50 on cifar-10 dataset with random initialization indicated a top1 accuracy of 94.93 with cross entropy and 95.25 with our loss, while for top5 accuracy the results are 99.86 and 99.87, respectively.

We study the brain segmentation by dividing the brain into multiple tissues. Given possible brain segmentation by deep, machine learning can be efficiently exploited to expedite the segmentation process in the clinical practice. To accomplish segmentation process, a MRI and tissues transfer using generative adversarial networks is proposed. Given the better result, we propose the transfer model using GAN. For the case of the brain tissues, white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF) are segmented. Empirical results show that this proposed model significantly improved segmentation results compared to the stat-of-the-art results. Furthermore, a dice coefficient (DC) metric is used to evaluate the model performance.

This paper compares a series of traditional and deep learning methodologies for the segmentation of textures. Six well-known texture composites first published by Randen and Hus{\o}y were used to compare traditional segmentation techniques (co-occurrence, filtering, local binary patterns, watershed, multiresolution sub-band filtering) against a deep-learning approach based on the U-Net architecture. For the latter, the effects of depth of the network, number of epochs and different optimisation algorithms were investigated. Overall, the best results were provided by the deep-learning approach. However, the best results were distributed within the parameters, and many configurations provided results well below the traditional techniques.

In this paper we propose a computational framework that merges mechanistic modeling with deep neural networks obeying the Systems Biology Markup Language (SBML) standard. Over the last 20 years, the systems biology community has developed a large number of mechanistic models in SBML that are currently stored in public databases. With the proposed framework, existing SBML mechanistic models may be upgraded to hybrid systems through the incorporation of deep neural networks into the model core, using a freely available python tool. The so-formed hybrid mechanistic/neural network models are trained with a deep learning algorithm based on the adaptive moment estimation method (ADAM), stochastic regularization and semidirect sensitivity equations. The trained hybrid models are encoded in SBML and stored back in model databases, where they can be further analyzed as regular SBML models. The application of this approach is illustrated with three well-known case studies: the threonine synthesis model in Escherichia coli, the P58IPK signal transduction model, and the Yeast glycolytic oscillations model. The proposed framework is expected to greatly facilitate the widespread use of hybrid modeling techniques for systems biology applications.

Electricity load forecasting is a crucial undertaking within all the deregulated markets globally. In contemporary times, the transition from conventional electricity grids to Smart Grids constitutes an area where extensive research is conducted on a global scale. Among the research challenges, the investigation of Deep Transfer Learning (DTL) in the field of electricity load forecasting represents a fundamental effort that imparts generality to Artificial Intelligence applications, due to new capabilities, such as knowledge transfer and computational power reduction. In this paper a comprehensive study is conducted for day-ahead electricity load forecasting. For this purpose, three Sequence-to-Sequence (Seq2seq) Deep Learning (DL) models are used, namely the Multilayer Perceptron (MLP), the Convolutional Neural Network (CNN) and the Ensemble Learning Model (ELM), which is consisted of the weighted combination of the outputs of MLP and CNN models. Also, the study focuses on the development of different forecasting strategies based on DTL and emphasizing the way the datasets are trained and fine-tuning for higher forecasting accuracy. In order to implement the forecasting strategies using Deep Learning models, load datasets from three Greek islands, Rhodes, Lesvos, and Chios, are used. The main purpose is to apply DTL for day-ahead predictions (1-24 hours) for each month of the year for Chios dataset after training and fine-tuning the models using the datasets of the three islands in various combinations. After several trials, four DTL strategies are illustrated. In the first strategy (DTL Case 1), each of the three DL models is trained using only Lesvos dataset, while fine-tuning is performed on the dataset of Chios island, in order to create day-ahead predictions for Chios load. In the second strategy (DTL Case 2), data from both Lesvos and Rhodes concurrently are used for the DL model training period, and fine-tuning is performed on the data from Chios. The third DTL strategy (DTL Case 3) contains the training of the DL models using Lesvos dataset, and the testing period performed directly on the Chios dataset without fine-tuning. The fourth strategy is a Multi-task Deep Learning (MTDL) approach, which has been extensively studied in recent years. In MTDL, the three DL models are trained simultaneously on all three datasets and the final predictions are made on the unknown part of the dataset of Chios. In this paper, we explore the performance of DTL and compare the results with those produced with MTDL. The results demonstrated that DTL can be applied with high efficiency for day-ahead load forecasting. Specifically, the two cases with fine-tuning (DTL Case 1 and 2) outperformed MTDL in terms of load prediction accuracy. Regarding the DL models, all three exhibit very high prediction accuracy, especially in the two cases with fine-tuning. The ELM excels compared to the single models. More specifically, for conducting day-ahead predictions, it has been concluded that the MLP model presents best monthly forecasts with a MAPE of 6.24% and 6.01% for the first two cases, the CNN model presents best monthly forecasts with a MAPE of 5.57% and 5.60% respectively and the ELM model achieves best monthly forecasts with a MAPE of 5.29% and 5.31%, respectively, indicating the very high accuracy it can achieve.

The extensive utilization of drones has led to numerous scenarios that encompass both advantageous and perilous outcomes. By using deep learning techniques, it is aimed to reduce the dangerous effects of drone use through early detection of drones.The state of the art of the study is evaluation of deep learning approaches such as pre trained YOLOv8 drone detection for security issues. In this study, dataset collected by Mehdi Özel for a UAV competition are sourced from GitHub. These images are labeled using Roboflow, and the model is trained on Google Colab. To enhance model performance, dataset augmentation techniques including rotation and blurring are implemented. Following these steps, a significant precision value of 0.946, a notable recall value of 0.9605 and a considerable precision-recall curve value of 0.978 are achieved, surpassing many popular models such as Mask CNN, CNN, and YOLOv5.

The study aimed to investigate the impact of incorporating deep learning strategies in mathematics instruction on students' problem-solving abilities, critical thinking skills, academic achievement, and mathematical skills development. The study population consisted of 280 students, with a sample size of 162 participants. Employing a purely quantitative approach, data collection utilized questionnaires and achievement tests as instruments. Data analysis involved simple linear regression, descriptive analysis, and Pearson correlation. The results of the study revealed that there was no significant enhancement observed in students' problem-solving abilities and critical thinking skills when deep learning strategies were incorporated into mathematics instruction. Additionally, students' attitudes towards deep learning activities in mathematics showed no effect on their academic achievement. Furthermore, students' engagement levels in deep learning activities did not impact their development of mathematical skills. Lastly, the frequency of deep learning activities in the mathematics classroom did not show a significant relationship with students' academic performance. These findings suggest that the implementation of deep learning strategies in mathematics instruction did not yield notable improvements in the specified areas based on the study outcomes.

To identify objects in images, a complex set of skills is needed that includes understanding the context and being able to determine the borders of objects. In computer vision, this task is known as semantic segmentation and it involves categorizing each pixel in an image. It is crucial in many real-world situations: for autonomous vehicles, it enables the identification of objects in the surrounding area; in medical diagnosis, it enhances the ability to detect dangerous pathologies early, thereby reducing the risk of serious consequences. In this study, we compare the performance of various ensembles of convolutional and transformer neural networks. Ensembles can be created, e.g, by varying the loss function, data augmentation method or the learning rate strategy. Our proposed ensemble, which is based on the simple average rule, demonstrates exceptional performance on several datasets. All the resources used in this study are available online at the following GitHub repository: https://github.com/LorisNanni.

Polar Mesospheric Summer Echoes (PMSEs) are radar echoes that are observed in the mesosphere during the Arctic summer months in the polar regions. By studying PMSE, researchers can gain insights into physical and chemical processes that occur in the upper atmosphere specifically in the 80 to 90 km altitude range. In this paper, we employ fully convolutional networks such as UNET and UNET++ for the purposes of segmenting PMSE from the EISCAT VHF dataset. First, experiments are performed to find suitable weights and hyperparameters for UNET and UNET++. Second, different loss functions are tested to find one suitable for our task. Third, as the number of PMSE samples used is relatively small that can lead to poor generalization. To address this, image-level and object-level augmentation methods are employed. Four, we briefly explain our findings by employing layerwise relevance propagation.

In the field of low power wireless networks, one of the techniques on which many researchers are putting their efforts is related to positioning methodologies such as fingerprinting in dense urban areas. This paper presents an experimental study aimed at quantifying the mean location estimation error in densely urbanized areas.Using a dataset made available by the University of Antwerp, a neural network was implemented with the aim of providing the position of the end-devices. In this way it was possible to measure the mean location estimation error in an area with high urban density. The results obtained show an accuracy in the localization of the end-device of less than 150 meters.This result would make it possible to use the fingerprint instead of alternative, energy consuming, methodologies such as GPS in IoT (Internet of Things) applications where battery life is the primary requirement to be met.

Deep learning, specifically the supervised approach, has proved to be a breakthrough in depth prediction. However, the generalization ability of deep networks is still limited, and they cannot maintain a satisfactory performance on some inputs. Addressing a similar problem in the segmentation field, a scheme (f-BRS) has been proposed to refine predictions in the inference time. f-BRS adapts an intermediate activation function to each input by using user clicks as sparse labels. Given the similarity between user clicks and sparse depth maps, this paper aims at extending the application of f-BRS to depth prediction. Our experiments show that f-BRS, fused with a depth estimation baseline, is trapped in local optima, and fails to improve the network predictions. To resolve that, we propose a double-stage adaptive refinement scheme (DARS). In the first stage, a Delaunay-based correction module significantly improves the depth generated by a baseline network. In the second stage, a particle swarm optimizer (PSO) delineates the estimation through fine-tuning f-BRS parameters—that is, scales and biases. DARS is evaluated on an outdoor benchmark, KITTI, and an indoor benchmark, NYUv2 while for both the network is pre-trained on KITTI. The proposed scheme outperformed rival methods on both datasets.

Numerous approaches address over-fitting in neural networks: by imposing a penalty on the parameters of the network (L1, L2, etc); by changing the network stochastically (drop-out, Gaussian noise, etc.); or by transforming the input data (batch normalization, etc.). In contrast, we aim to ensure that a minimum amount of supporting evidence is present when fitting the model parameters to the training data. This, at the single neuron level, is equivalent to ensuring that both sides of the separating hyperplane (for a standard artificial neuron) have a minimum number of data points — noting that these points need not belong to the same class for the inner layers. We firstly benchmark the results of this approach on the standard Fashion-MINST dataset, comparing it to various regularization techniques. Interestingly, we note that by nudging each neuron to divide, at least in part, its input data, the resulting networks make use of each neuron, avoiding a hyperplane completely on one side of its input data (which is equivalent to a constant into the next layers). To illustrate this point, we study the prevalence of saturated nodes throughout training, showing that neurons are activated more frequently and earlier in training when using this regularization approach. A direct consequence of the improved neuron activation is that deep networks are now easier to train. This is crucially important when the network topology is not known a priori and fitting often remains stuck in a suboptimal local minima. We demonstrate this property by training a network of increasing depth (and constant width): most regularization approaches will result in increasingly frequent training failures (over different random seeds) whilst the proposed evidence-based regularization significantly outperforms in its ability to train deep networks.

Fracture analysis of reinforced concrete deep beam strengthened with carbon fiber-reinforced polymer (CFRP) plates was carried out. The present research aims to find out whether the crack propagation in a strengthened deep beam follows linear elastic fracture mechanics (LEFM) theory or nonlinear fracture mechanics theory. To do so, a new energy release rate based on nonlinear fracture mechanics theory was formulated on the finite element method and the discrete cohesive zone model (DCZM) was developed in deep beams. To validate and compare with numerical models, three deep beams with rectangular cross-sections were tested. The code results based on nonlinear fracture mechanics models were compared with experimental results and ABAQUS results carried out based on LEFM. The predicted values of initial stiffness, yielding point and failure load, energy absorption, and compressive strain in the concrete obtained by the proposed model were very close to the experimental results. However, the ABAQUS software results have greater differences with the experimental results. For example, the predicted failure load for the shear-strengthened deep beam using the proposed model has only 6.3% differences compared to the experimental result. However, the predicted failure load using ABAQUS software based on LEFM has greater differences (25.1%) compared to the experimental result.

GANs are well known for success in the realistic image gen-eration. However, they can be applied in tabular data generation as well.We will review and examine some recent papers about tabular GANs inaction. We will generate data to make train distribution bring closer tothe test. Then compare model performance trained on the initial traindataset, with trained on the train with GAN generated data, also wetrain the model by sampling train by adversarial training. We show thatusing GAN might be an option in case of uneven data distribution be-tween train and test data

The COVID-19 pandemic situation has created even more difficulties in the quick identification and screening of the COVID-19 patients for the medical specialists. Therefore, a significant study is necessary for detecting COVID-19 cases using an automated diagnosis method, which can aid in controlling the spreading of the virus. In this paper, the study suggests a Deep Convolutional Neural Network-based multi-classification approach (COV-MCNet) using eight different pre-trained architectures such as VGG16, VGG19, ResNet50V2, DenseNet201, InceptionV3, MobileNet, InceptionResNetV2, Xception which are trained and tested on the X-ray images of COVID-19, Normal, Viral Pneumonia, and Bacterial Pneumonia. The results from 3-class (Normal vs. COVID-19 vs. Viral Pneumonia) showed that only the ResNet50V2 model provides the highest classification performance (accuracy: 95.83%, precision: 96.12%, recall: 96.11%, F1-score: 96.11%, specificity: 97.84%) compared to rest of the models. The results from 4-class (Normal vs. COVID-19 vs. Viral Pneumonia vs. Bacterial Pneumonia) demonstrated that the pre-trained model DenseNet201 provides the highest classification performance (accuracy: 92.54%, precision: 93.05%, recall: 92.81%, F1-score: 92.83%, specificity: 97.47%). Notably, the ResNet50V2 (3-class) and DenseNet201 (4-class) models in the proposed COV-MCNet framework showed higher accuracy compared to the rest six models. This indicates that the designed system can produce promising results to detect the COVID-19 cases on the availability of more data. The proposed multi-classification network (COV-MCNet) significantly speeds up the existing radiology-based method, which will be helpful to the medical community and clinical specialists for early diagnosis of the COVID-19 cases during this pandemic.

The transportation industry is rapidly transitioning from Internal Combustion Engine (ICE) based vehicles to Electric Vehicles (EVs) to promote clean energy. However, large-scale adoption of EVs can compromise the reliability of the power grids by introducing large uncertainty in the demand. Demand response with a controlled charge scheduling strategy for EVs can mitigate such issues. In this paper, a deep reinforcement learning- based charge scheduling strategy is developed for individual EVs by considering user’s dynamic driving behavior and charging preferences. The temporal dynamics of user’s anxiety about charging the EV battery is rigorously addressed. A dynamic weight allocation technique is applied to continuously tune user’s priority for charging and cost-saving with respect to charging duration. The sequential charging control problem is formulated as a Markov decision process, and an episodic approach to the deep deterministic policy gradient (DDPG) algorithm with target policy smoothing and delayed policy update techniques is applied to develop the optimal charge scheduling strategy. A real-world dataset that captures user’s driving behavior, such as arrival time, departure time, and charging duration, is utilized in this study. The extensive simulation results reveal the effectiveness of the proposed algorithm in minimizing energy cost while satisfying user’s charging requirements.

This paper reviews non-intrusive load monitoring (NILM) approaches that employ deep neural networks to disaggregate appliances from low frequency data, i.e. data with sampling rates lower than the AC base frequency. We first review the many degrees of freedom of these approaches, what has already been done in literature, and compile the main characteristics of the reviewed publications in an extensive overview table. The second part of the paper discusses selected aspects of the literature and corresponding research gaps. In particular, we do a performance comparison with respect to reported MAE and F$_1$-scores and observe different recurring elements in the best performing approaches, namely data sampling intervals below 10\,s, a large field of view, the usage of GAN losses, multi-task learning, and post-processing. Subsequently, multiple input features, multi-task learning and related research gaps are discussed, the need for comparative studies is highlighted, and finally, missing elements for a successful deployment of NILM approaches based on deep neural networks are pointed out. We conclude the review with an outlook on possible future scenarios.

Breast cancer is the most prevalent type of disease among women. It has become one of the foremost causes of death among women globally. Early detection plays a significant role in administering personalized treatment and improving patient outcomes. Mammography procedures are often used to detect early-stage cancer cells. This traditional method of mammography while valuable has limitations of potential for false positives and negatives, patient discomfort, and radiation exposure. Therefore, there is a probe for more accurate techniques required in detecting breast cancer, leading to exploring the potential of machine learning in the classification of diagnostic images due to its efficiency and accuracy. This study conducted a comparative analysis of pre-trained CNNs (ResNet50 and VGG16) and Vision Transformers (ViT-Base and SWIN Transformer) with the inclusion of ViT-Base trained from scratch model architectures to effectively classify breast cancer mammographic images into benign and malignant cases. The Swin transformer exhibits superior performance with 99.8% accuracy and a precision of 99.8%. These findings demonstrate the efficiency of deep learning to accurately classify breast cancer mammographic images for the diagnosis of breast cancer, leading to improvement in patient outcomes.

Magnetic resonance imaging (MRI) is crucial for its superior soft tissue contrast and high spatial resolution. Integrating deep learning algorithms into MRI reconstruction enhances image quality and efficiency, but a comprehensive review of optimization-based deep learning models for MRI reconstruction has been missing. This study fills that gap by examining the latest optimization-based algorithms in deep learning for MRI reconstruction, including gradient descent algorithms, proximal gradient descent algorithms, ADMM, PDHG, and diffusion models combined with gradient descent. Learnable optimization algorithms (LOA) are highlighted for their ability to map optimization processes to structured neural networks, improving model interpretability and performance. The study demonstrates significant advancements in MRI reconstruction through deep learning, with successful clinical applications. These findings provide valuable insights and resources for researchers aiming to advance medical imaging using innovative deep learning techniques.

This study presents the Distilled MixUp Squeeze ResNet (DMSResNet), an optimized variant of the Residual Network architecture designed for high accuracy and computational efficiency. By integrating Squeeze-and-Excitation blocks, MixUp data augmentation, and knowledge distillation, it achieves competitive performance within strict parameter constraints and limited training resources. Evaluated on the CIFAR-10 dataset, the model attains a 96.56% accuracy with only 4.3 million total parameters. The results demonstrate that optimization of established architectures can yield performance comparable to newer models, especially in resource-limited scenarios. This work contributes to the development of efficient deep learning models for applications in constrained computational environments.

This case report presents a 63-year-old male patient who developed deep vein thrombosis (DVT) shortly after undergoing retroperitoneal laparoscopic nephroureterectomy for ureteral and renal pelvis tumors. Despite early postoperative mobilization and administration of low molecular weight heparin, the patient developed DVT, highlighting the limitations of current postoperative management strategies. The patient was promptly treated with the placement of an inferior vena cava filter and continued anticoagulation therapy, preventing the progression to pulmonary embolism and leading to a successful discharge.The report emphasizes the importance of vigilant postoperative monitoring and timely intervention in preventing complications such as DVT. It suggests that current strategies, including early mobilization and standard anticoagulation therapy, may not be sufficient to fully prevent DVT, necessitating further optimization of postoperative care. Regular monitoring of D-Dimer levels and ultrasound examinations, as well as appropriate limb protection measures during surgery, are recommended to detect and manage potential DVT early.This case contributes valuable insights into the early detection and management of postoperative DVT, providing useful references for similar cases and stressing the need for enhanced postoperative monitoring and timely intervention.

Market crises pose significant challenges for businesses, emphasizing the importance of effective crisis management strategies. Central to these strategies is the ability to identify and retain loyal customers, who often serve as the bedrock of stability during tumultuous times. This paper investigates the application of deep meta-learning analysis to predict loyal customers as a cornerstone of market crisis management. Drawing upon an extensive literature review, the study explores previous research on market crises, customer loyalty, and the evolution of deep learning and meta-learning in crisis management contexts. Methodologically, the paper outlines data collection, model development, and evaluation procedures tailored for deep meta-learning-based customer prediction. In results, the overall accuracy of model 85%, a precision 0.88, recall 0.82, and F1-score 0.85 were obtained. The analysis demonstrates the effectiveness of deep meta-learning models in accurately identifying loyal customers during market crises, offering insights into their performance and applicability compared to traditional methods. Practical implications include potential applications in crisis management for businesses and considerations for real-world implementation.

Deforestation through land-use conversion, illegal logging, and timber trafficking is believed to cause ~10% of the annual human carbon dioxide emissions at the global level. Given the large contribution, local and national policies have been set in place in the effort to reduce deforestation and support reforestation. However, accurate assessment of forest carbon stock is expensive and challenging in remote areas and on large geographical scales. To improve carbon stock monitoring and evaluation of fine-scale forest loss, we developed a rapid, automatic, scalable, and cost-efficient generalized deep learning framework that uses diverse remote sensing data and satellite imagery to derive aboveground carbon density from accurate estimates of tree canopy heights at fine-grained resolution (30x30 meters) in remote tropical rainforests. The remote sensing data is composed of Landsat-8, Sentinel-1, land cover, digital elevation model, and NASA CMS airborne LiDAR, as well as vegetation indices, texture metrics, and climatic data. Data sources were compiled into a data pipeline which produced >300 features and 2 million observations over forests in Indonesian Borneo. Using LiDAR swath data on canopy forest high for ~100,000 hectares in Indonesian Borneo to create a training and validation datasets, our neural network model produced aboveground carbon density estimates with R2 of 0.82, which is a significant improvement from comparable works by using Random Forest (R2 of 0.3-0.5). This deep learning framework can be used to facilitate further carbon stock modeling in other forest regions (e.g., Brazil) as well as for the general purpose of climate change mitigation.

Emotion detection holds significant importance in facilitating human-computer interaction, enhancing the depth of engagement. By integrating this capability, we pave the way for forthcoming AI technologies to possess a blend of cognitive and emotional understanding, bridging the divide between machine functionality and human emotional complexity. This progress has the potential to reshape how machines perceive and respond to human emotions, ushering in an era of empathetic and intuitive artificial systems. This paper introduces a novel approach to multimodal emotion recognition, seamlessly integrating speech and text modalities to accurately infer emotional states. Employing CNNs, we meticulously analyze speech using Mel spectrograms, while a BERT-based model processes the textual component, leveraging its bidirectional layers for profound semantic comprehension. The outputs from both modalities are combined using an attention-based fusion mechanism that optimally weighs their contributions. The proposed method undergoes meticulous testing on two distinct datasets: Carnegie Mellon University’s Multimodal Opinion Sentiment and Emotion Intensity (CMU-MOSEI) dataset and the Multimodal Emotion Lines Dataset (MELD). The results demonstrate superior efficacy compared to existing frameworks, achieving an accuracy of 88.4% and an F1-score of 87.9% on the CMU-MOSEI dataset, and a notable weighted accuracy (WA) of 67.81% and a weighted F1 (WF1) score of 66.32% on the MELD dataset. This comprehensive system offers precise emotion detection and introduces several significant advancements in the field.

The purpose of this paper is to propose a novel transfer learning regularization method based on knowledge distillation. Recently, transfer learning methods have been used in various fields. However, the problems such as knowledge loss still occur during the process of transfer learning to a new target dataset. To solve these problems, there are various regularization methods based on knowledge distillation techniques. In this paper, we propose a transfer learning regularization method based on feature map alignment used in the field of knowledge distillation. The proposed method is composed of two attention based submodules: self-pixel attention (SPA) and global channel attention (GCA). The self-pixel attention submodule utilizes both the feature maps of the source and target models, so that it provides an opportunity to jointly consider the features of the target and the knowledge of the source. The global channel attention submodule determines the importance of channels through all layers, unlike the existing methods that calculate these only within a single layer. Accordingly, transfer learning regularization is performed by considering both the internal of each single layer and the depth of the entire layer. Consequently, the proposed method using both of these submodules showed overall improved classification accuracy than the existing methods in classification experiments on commonly used dataset.

In recent years the field of computer science has experienced great changes, due to the incredible advances in the field of artificial intelligence. Deep Learning models are responsible for most of them since the biggest milestones occurred in 2012 when AlexNet won the image classification challenge called ImageNet. These models have demonstrated great performances in different types of complex tasks like image restoration, medical diagnosis or object recognition. Their disadvantages are related to their high data dependency, which forces experts in the field to follow a precise methodology to obtain accurate models. In this paper, we describe a complete workflow that begins with the management of the raw data until the in-depth interpretation of the performance of the models. This should be taken as a high-level consultation document describing good practices that should be applied. Apart from the step-by-step methodology, we present different use cases that correspond to the two main problems of the field: classification and regression.

Technological advances in DBS hardware and software represent a significant growth area in functional neurosurgery. These advancements have the potential to significantly enhance treatment outcomes and expand the scope of neurological disorders that can be effectively addressed through DBS interventions. The introduction of directional DBS marks a new phase in neuromodulation, offering extended benefits for patients with Parkinson's disease, more efficient optimization of clinical outcomes, and targeted treatment of refractory symptoms through data-driven approaches. Implementing a multimodal programming strategy, incorporating novel current fractionation technology and image-guided tools for lead localization and brain sensing, further reduces reliance on traditional trial-and-error programming methods, paving the way for a more predictive application of this therapy. These advancements are expected to drive the development of sophisticated closed-loop stimulation systems capable of integrating continuous data streams for improved patient care.

Neurodegenerative illnesses, such as Parkinson’s disease (PD), have a substantial impact on the overall well-being of those who are affected. This study investigates and contrasts the capabilities of convolutional neural networks (CNN) in detecting Parkinson’s disease (PD) by utilising hand-drawn images alongside wave and spiral images as input data. This study employs pre-trained CNN models, specifically MobileNet, ResNet50, EfficientNet-B1, and InceptionV3, to classify Parkinson’s disease (PD). The findings demonstrate that MobileNet surpasses other architectural designs, as evidenced by the F1-Score of the four classes: Spiral Normal (0.87), Spiral Parkinson (0.86), Wave Normal (0.97), and Wave Parkinson (0.97). MobileNet has also shown a remarkable accuracy of 0.92 in diagnosing Parkinson’s disease. The result demonstrates the efficacy of MobileNet in extracting features from images. The results of this study enhance the application of deep learning methods in the early detection of PD, as well as help indicate the effectiveness of patient therapy and exercise, promising better patient outcomes through timely intervention and treatment.

A remarkable strengths of deep neural networks lies in their ability to discover patterns and representations from complex data. Deep learning models can learn and refine features at different levels of abstraction, enabling them to handle complex tasks in computer vision recognition, natural language processing, and speech recognition. The ability to generalize from examples, coupled with the capacity to process vast amounts of data, empowers deep learning to achieve state-of-the-art results across a wide spectrum of applications. An argument for this ability is based on similarities between deep neural networks and the nervous system. In this report, we argue that despite the remarkable performance of deep learning, there are still gaps between deep learning and the nervous system that needs to be closed to enable deep learning doing tasks that currently only the nervous system can perform

Voice spoofing attempts to break into a specific automatic speaker verification (ASV) system by forging the user’s voice, and can be used through methods, such as text-to-speech (TTS), voice conversion (VC), and replay attacks. Recently, deep learning-based voice spoofing countermeasures have been developed. however, the problem with replay is that it is difficult to construct a large number of datasets because it requires a physical recording process. To overcome these problems, this study proposes a pre-training framework based on multi-order acoustic simulation for replay voice spoofing detection. Multi-order acoustic simulation utilizes existing clean signal and room impulse response (RIR) datasets to generate audios, which simulate the various acoustic configurations of the original and replayed audios. The acoustic configuration refers to factors, such as the microphone type, reverberation, time delay, and noise that may occur between a speaker and microphone during the recording process. We assume that a deep learning model trained on an audio that simulates the various acoustic configurations of the original and replayed audios can classify the acoustic configurations of the original and replay audios well. To validate this, we performed pre-training to classify the audio generated by the multi-order acoustic simulation into 3 classes: clean signal, audio simulating the acoustic configuration of the original audio, and audio simulating the acoustic configuration of the replay audio. We also set the weights of the pre-training model to the initial weights of the replay voice spoofing detection model using the existing replay voice spoofing dataset and then performed fine-tuning. To validate the effectiveness of the proposed method, we evaluated the performance of the conventional method without pre-training and proposed method using an objective metric, i.e., the accuracy. As a result, the conventional method achieved 92.94% accuracy and proposed method achieved 98.16% accuracy.

Estimating precipitation is of critical importance to climate systems and decision-making processes. This paper presents Espresso, a deep learning model designed for estimating precipitation from satellite observations on a global scale. Conventional methods, like ground-based radars, are limited in terms of spatial coverage. Satellite observations, on the other hand, allow global coverage. Combined with deep learning methods these observations offer the opportunity to address the challenge of estimating precicpation on a global scale. This research paper presents the development of a deep learning model using geostationary satellite data as input and generating instantaneous rainfall rates, calibrated using data from the Global Precipitation Measurement Core Observatory (GPMCO). The performance impact of various input data configurations on Espresso was investigated. These configurations include a sequence of four images from geostationary satellites and the optimal selection of channels. Additional descriptive features were explored to enhance the model’s robustness for global aplications. When evaluated against the GPMCO test set, Espresso demonstrated highly accurate precipitation estimation, especially within equatorial regions. A comparison against six other operational products using multiple metrics indicated its competitive performance. The model’s superior storm localization and intensity estimation were further confirmed through visual comparisons in case studies. Espresso has been incorporated as an operational product at Météo-France, delivering high-quality, real-time global precipitation estimates every 30 minutes.

Fruit quality is a critical factor in the produce industry, affecting producers, distributors, consumers, and the economy. High-quality fruits are more appealing, nutritious, and safe, boosting consumer satisfaction and revenue for producers. Artificial Intelligence can aid in assessing the quality of the fruit using images. This paper presents a general machine-learning model for assessing fruit quality using deep image features. The model leverages the learning capabilities of the recent successful networks for image classification called Vision Transformers (ViT). The ViT model is built and trained with a combination of various fruit datasets and learned to distinguish between good and rotten fruit images. The general model demonstrated impressive results in accurately identifying the quality of various fruits such as Apples (with a 99.50% accuracy), Cucumbers (99%), Grapes (100%), Kakis (99.50%), Oranges (99.50%), Papayas (98%), Peaches (98%), Tomatoes (99.50%), and Watermelons (98%). However, it showed slightly lower performance in identifying Guavas (97%), Lemons (97%), Limes (97.50%), mangoes (97.50%), Pears (97%), and Pomegranates (97%).

In Taiwan, peanuts are classified by hand. The kernel quality is judged based on the appearance of peanuts. The identification takes considerable time and labor. Fatigue can induce misrecogni-tion or grade cognition instability due. Peanut defects are generally divided into health, underde-velopment, insect bite, and rupture. This study employed two imaging methods push-boom FX10 and Snapshot. Deep learning was used to detect peanut defects. A push-broom instrument was used for testing. There were 1,560 peanuts, including 240 good and 240 bad ones. There were a total of 1,080 good peanuts and bad peanuts in the test set data, the band selection was used, and a CNN model was built. Based on the results from three methods, 3DCNN could classify with 97%accuracy. The Snapshot was used to achieve the real-time design of a lightweight CNN model. Finally, five bands were selected using PCA, and the screening could be accurate and efficient. The maximum overall accuracy was about 98.5%. Kappa was 97.3% with real-time recognition. This will be a major advantage in the future practical application and commercialization processes. This technique can reduce labor costs, achieve intelligent detection and intelligent grading, and bring breakthroughs to smart agriculture for other crops as well.

The paper is based on the research project P2Endure funded within the H2020-EE-2016-2017 call. The main aim of the research is to propose innovative solutions for deep renovation that are applicable and replicable for the widest range of building typologies, from public to residential and historic buildings. In this paper, the methodology adopted to perform the energy analyses and to verify the achievement of the expected targets in terms of savings is defined, based on a range of real case studies. The main aim is to promote a guideline for designers showing specific instruments to perform towards the attainment of the 60% threshold of the energy savings giving advice on how to further improve the energy performance of the buildings, basing on the overview of the energy analyses run for similar cases.

Modelling both long and short-term user interests from historical data is crucial for accurate recommendations. However, unifying these metrics across multiple application domains can be challenging, and existing approaches often rely on complex, intertwined models which can be difficult to interpret. To address this issue, we propose a lightweight, plug-and-play interest enhancement module that fuses interest vectors from two independent models. After analyzing the dataset, we identify deviations in the recommendation performance of long and short-term interest models. To compensate for these differences, we use feature enhancement and loss correction during training. In the fusion process, we explicitly split long-term interest features with longer duration into multiple local features. We then use a shared attention mechanism to fuse multiple local features with short-term interest features to obtain interaction features. To correct for bias between models, we introduce a comparison learning task that monitors the similarity between local features, short-term features, and interaction features. This adaptively reduces the distance between similar features. Our proposed module combines and compares multiple independent long-term and short-term interest models on multiple domain datasets. As a result, it not only accelerates the convergence of the models but also achieves outstanding performance in challenging recommendation scenarios.

Plant-based protein sources derived from microalgae offer promise as low-carbon emission and nutrient-dense sources for human consumption. The current management of microalgae cultivation relies heavily on manual microscopic examination in order to identify desired and competing species, as well as predators. In this study, we trained and tested a transfer learning model modified from EfficientNetV2 B3 on 434 and 161 prospectively acquired images of the preferred Nannochloropsis sp microalgae and competitor Spirulina, respectively, and achieved >98% classification for both species on tenfold cross-validation. The model was further enhanced with gradient-weighted class activation mapping, which allowed visualization of regions of the input images that were relevant to the classification, thereby improving its explainability. In this paper, we demonstrate that a simple deep transfer learning model can be used to identify microalgae species. The application addresses the practical need for robust automated monitoring of microalgae populations on industrial microalgae farms.

In this work we introduce an end-to-end multi-modal neural network to segment the Gross Tumor Volume (GTV) from 3D-CBCT’s during radiotherapy. We improve the tumor segmentation by using a U-net which takes additional information such as the tumor mask generated at the planning phase along with the CBCT volume. The mask contains relevant information about the tumor’s location and can guide the model to use this knowledge appropriately to give a better prediction. This technique could become an alternative to produce segmentation masks of GTV in CBCT automatically during radiotherapy as in the traditional RT-pipeline, they are not segmented. We have evaluated our model on a dataset of 82 patients who have undergone radiotherapy. We compare the results of registered target volumes from planning CT as mask seed with 2 different types of multi-modal architectures. Our model shows a DSC of 0.706±0.002 with Late Fusion and 0.702±0.015 with Early Fusion using the GTV Mask. The performance of the two models on this mask is similar, so we perform further experiments with different types of masks which suggest that Late Fusion model produces a better segmentation of the tumor than the Early Fusion model. We also provide an ablation study consisting of a single modality U-net and a metric based on the Planning CT mask registration. It indicates a clear advantage of using our model to produce segmentation for this type of imaging.

The emergence of unmanned aerial vehicles (UAVs) raised multiple concerns, given their potential for malicious misuse in unlawful acts Vision-based counter-UAV applications offer a reliable solution compared to acoustic and radio frequency-based solutions because of their high detection accuracy in diverse weather conditions. The existing solutions work well on trained datasets, but their accuracy is relatively low for real-time detection. In this paper, we model deep learning-empowered solutions to improve the multi-class UAV's classification performance using single-shot object detection algorithms YOLOv5 and YOLOv7. The transfer learning is employed for performance improvement and rapid training with improved results. We customized a multi-class dataset containing multi-rotor, fixed-wing, and single-rotor UAVs in challenging weather conditions. Experiments show that the integration of transfer learning has achieved good results, with an overall best average-classification precision of 94\%, an average recall of 93.1\%, a mAP$@$0.5 average of 95.3\%, and an average F1 score of 92.33\%. The dataset and code are available as an open source: https://github.com/ZeeshanKaleem/YOLOV5-Large-vs-YOLOV7.git

Hand gesture recognition has many valuable applications in engineering and health care. This study proposes a novel model which can accurately distinguish hand gestures using forearm muscles' surface electromyogram (sEMG) signals. A deep learning algorithm with hyper parameters impacting the final model’s accuracy and a convolutional neural network (CNN) were employed in the recognition stage. The number of convolutional layers, kernels per layer, and neurons in the dense layer were selected for optimization, while the remaining parameters, such as the learning rate, batch size, and number of epochs, were chosen based on trial and error and prior knowledge. The optimal values for the selected hyperparameters were obtained using a genetic algorithm to achieve maximum recognition accuracy. The UC2018 Dual-Myo database was used for training and testing the model based on EMG signals characterizing the activity of eight different hand gestures. The final structure of the model consisted of two convolutional layers with 131 and 28 kernels, a dense layer with 111 neurons, and a softmax layer with eight neurons. Upon optimizing the hyperparameters using the genetic algorithm, the accuracy of the proposed model increased from 91.86% to 96.4% at best and 95.3% on average in real-time applications and 99.6% in an offline mode. Future work is warranted towards improving the architecture and the computational cost.

Over the past decade, research in the field of Deep Learning has brought about novel improvements in image generation and feature learning; one such example being a Generative Adversarial Network. However, these improvements have been coupled with an increasing demand on mathematical literacy and previous knowledge in the field. Therefore, in this literature review, I seek to introduce Generative Adversarial Networks (GANs) to a broader audience by explaining their background and intuition at a more foundational level. I begin by discussing the mathematical background of this architecture, specifically topics in linear algebra and probability theory. I then proceed to introduce GANs in a more theoretical framework, along with some of the literature on GANs, including their architectural improvements and image-generation capabilities. Finally, I cover state-of-the-art image generation through style-based methods, as well as their implications on society.

This study evaluated the using of machine vision in combination with deep learning to identify weeds in real-time for wheat production system. PMAS-Arid Agriculture university research farm were selected for collection of images (6000 total images) of weeds and wheat crops under different weather condition. During growing season, the database was constructed to identify the weeds. For this study two framework were used TensorFlow and PyTorch under CNNs and Deep learning. Deep learning perfromed better with in PyTourch value as compared to another model in Tensorflow. comparing with other networks such as YOLOv4, we concluded that our network reaches a better result between speed and accuracy. Specifically, the maximum precision of weed and wheat plants were 0.89 and 0.91 respectively with 9.43 ms and 12.38 ms inference time per image (640 × 640) NVIDIA RTX2070 GPU.

To solve two key problems in the identification of people who are infected with COVID-19: the first problem is that the identification accuracy is not high enough. The second problem is that present identification method such as nucleic acid testing is expensive in many countries. Methods: So, I decided to design a fast identification method for COVID-19 patients which is based on deep learning. After the model (CoughNet) learns more than 6,000 cough spectrograms of both COVID-19 patients and normal people, the accuracy rate of identification of COVID-19 patients and normal people is higher than 99% in the test set. Structure: This paper is mainly divided into three parts: the first part introduces the main background and research status of the research; The second part introduces the research methods; The third part introduces the specific process of the experiment.

A hybrid smart process and material design system for nanoimprinting is proposed, which combines with a learning system based on experimental and numerical simulation results. Instead of carrying out extensive learning experiments for various condition, the simulation learning results are partially complimented where the results can be theoretically predicted by numerical simulation. In other word, the lacking data in experimental learning are complimented by simulation-based learning results. Therefore, the prediction of nanoimprint results under various conditions without experimental learning could be realized even for unknown materials. In this study, material and process design for a low-temperature nanoimprint process using glycerol-containing polyvinyl alcohol are demonstrated. Experimental results under limited conditions are learned to investigate optimum Glycerol concentrations and process temperatures. On the other hand, simulation-based learning is used to predict the dependence on press pressure and shape parameters. The prediction results for unknown Glycerol concentrations agreed well with the follow-up experiments.

The Covid-19 also known as the Coronavirus is a viral disease from the SARS-CoV-2 family of virus, as at December 2019 the first case of this virus infection was identified at Wuhan, China, this seemingly isolated case soon became a global pandemic, whose effect was felt globally which also had colossal effects on both health, economic and politics . As at the time of this research about 4.5 million people have died of the Coronavirus and over 215 million people already infected by it. This pandemic stood out not just for its scale but for how social media was a major contribution to its spread as well as to curbing it. The power of social media was used to spread misinformation as well as to spread awareness on the subject, with both having massive impact on the people. In this paper we will be running a sentimental analysis on twitter under the keyword “Covid-19 and Coronavirus”, twitter is a powerful social media tool that is known for its ability to keep trends in the form of tweets, we will be drawing correlations between the peaks of tweet with the peak of infection. We will also be analyzing to know what the impact of these tweets are having on the rate of the infection and vice versa. We will also be analyzing what people are tweeting most about, what are the talking points, comparing both real time and past tweets with real time infection and death rates using deep different learning methods to access what information can be derived from it.

Sentiment analysis is a process of extracting opinions into the positive, negative, or neutral categories from a pool of text using Natural Language Processing (NLP). In the recent era, our society is swiftly moving towards virtual platforms by joining virtual communities. Social media such as Facebook, Twitter, WhatsApp, etc are playing a very vital role in developing virtual communities. A pandemic situation like COVID-19 accelerated people's involvement in social sites to express their concerns or views regarding crucial issues. Mining public sentiment from these social sites especially from Twitter will help various organizations to understand the people's thoughts about the COVID-19 pandemic and to take necessary steps as well. To analyze the public sentiment from COVID-19 tweets is the main objective of our study. We proposed a deep learning architecture based on Bidirectional Gated Recurrent Unit (BiGRU) to accomplish our objective. We developed two different corpora from unlabelled and labeled COVID-19 tweets and use the unlabelled corpus to build an improved labeled corpus. Our proposed architecture draws a better accuracy of 87% on the improved labeled corpus for mining public sentiment from COVID-19 tweets.

Introduction: Dental segmentation in panoramic radiograph has become very relevant in dentistry, since it allows health professionals to carry out their assessments more clearly and helps them to define the best possible treatment plan for their patients. Objectives: In this work, a comparative study is carried out with four segmentation algorithms (U-Net, DCU-Net, DoubleU-Net and Nano-Net) that are prominent in the medical literature on segmentation and we evaluate their results with the current state of the art of dental segmentation in panoramic radiograph. Methods: These algorithms were tested with a dataset consisting of 1,500 images, considering experiment scenarios with and without augmentation data. Results: DoubleU-Net was the model that presented the best results among the analyzed models, reaching 96.591% accuracy and 92.886% Dice using the dataset with data augmentation. Another model that stood out was Nano-Net using the dataset without data augmentation; this model achieved results close to that of the literature with only 235 thousand trainable parameters, while the literature model (TSASNet) contains 78 million. Conclusions: The results obtained in this work are satisfactory and present paths for a better and more effective dental segmentation process.

Smartphones are prone to SMS phishing due to the rapid growth in the availability of smart mobile technologies driven by Internet connections. Also, detecting phishing SMS is a challenging task due to the unstructured nature of SMS text data with non-linear complex correlations. In this concern, considering the recent advancements in the domain of cybersecurity, we have proposed a hybrid deep learning framework that extracts robust features from SMS texts followed by an automatic detection of Phishing SMS. Due to combining the potential capability of individual models into one hybrid framework, it has outperformed various other individual machine learning and deep learning models. The proposed Phishing Detection framework is an effective hybrid combination of pretrained transformer model, MPNet (Masked and Permuted Language Modeling), with supervised ConvNets (CNN) and Bi-directional Gated Recurrent Units (GRU). It is intended to successfully detect unstructured short phishing text messages that contain complex patterns.

The usage and amount of information available on the internet increase over the past decade. This digitization leads to the need for automated answering system to extract fruitful information from redundant and transitional knowledge sources. Such systems are designed to cater the most prominent answer from this giant knowledge source to the user’s query using natural language understanding (NLU) and thus eminently depends on the Question-answering(QA) field. Question answering involves but not limited to the steps like mapping of user’s question to pertinent query, retrieval of relevant information, finding the best suitable answer from the retrieved information etc. The current improvement of deep learning models evince compelling performance improvement in all these tasks. In this review work, the research directions of QA field are analyzed based on the type of question, answer type, source of evidence-answer, and modeling approach. This detailing followed by open challenges of the field like automatic question generation, similarity detection and, low resource availability for a language. In the end, a survey of available datasets and evaluation measures is presented.

Beyond semantic segmentation,3D instance segmentation(a process to delineate objects of interest and also classifying the objects into a set of categories) is gaining more and more interest among researchers since numerous computer vision applications need accurate segmentation processes(autonomous driving, indoor navigation, and even virtual or augmented reality systems…) This paper gives an overview and a technical comparison of the existing deep learning architectures in handling unstructured Euclidean data for the rapidly developing 3D instance segmentation. First, the authors divide the 3D point clouds based instance segmentation techniques into two major categories which are proposal based methods and proposal free methods. Then, they also introduce and compare the most used datasets with regard to 3D instance segmentation. Furthermore, they compare and analyze these techniques performance (speed, accuracy, response to noise…). Finally, this paper provides a review of the possible future directions of deep learning for 3D sensor-based information and provides insight into the most promising areas for prospective research.

Autonomous image recognition has numerous potential applications in the field of planetary science and geology. For instance, having the ability to classify images of rocks would allow geologists to have immediate feedback without having to bring back samples to the laboratory. Also, planetary rovers could classify rocks in remote places and even in other planets without needing human intervention. Shu et al. classified 9 different types of rock images using a Support Vector Machine (SVM) with the image features extracted autonomously. Through this method, the authors achieved a test accuracy of 96.71%. In this research, Convolutional Neural Networks(CNN) have been used to classify the same set of rock images. Results show that a 3-layer network obtains an average accuracy of 99.60% across 10 trials on the test set. A version of Self-taught Learning was also implemented to prove the generalizability of the features extracted by the CNN. Finally, one model has been chosen to be deployed on a mobile device to demonstrate practicality and portability. The deployed model achieves a perfect classification accuracy on the test set, while taking only 0.068 seconds to make a prediction, equivalent to about 14 frames per second.

Semantic segmentation is a very popular topic in modern computer vision and it has applications to many fields. Researchers proposed a variety of architectures over time, but the most common ones exploit an encoder-decoder structure that aims to capture the semantics of the image and it low level features. The encoder uses convolutional layers, in general with a stride larger than one, to extract the features, while the decoder recreates the image by upsampling an using skip connections with the first layers. In this work, we use DeepLab as architecture to test the effectiveness of creating an ensemble of networks by randomly changing the activation functions inside the network multiple times. We also use different backbone networks in our DeepLab to validate our findings. We manage to reach a dice coefficient of 0.888, and a mean Intersection over Union (mIoU) of 0.825, in the competitive Kvasir-SEG dataset. Results in skin detection also confirm the performance of the proposed ensemble, which is ranked first with respect to other state-of-the-art approaches (including HardNet) in a large set of testing datasets. The developed code will be available at https://github.com/LorisNanni.

The betatron tune in the Large Hadron Collider (LHC) is measured using a Base-Band Tune (BBQ) system. The processing of these BBQ signals is often perturbed by 50 Hz noise harmonics present in the beam. This causes the tune measurement algorithm, currently based on peak detection, to provide incorrect tune estimates during the acceleration cycle with values that oscillate between neighbouring harmonics. The LHC tune feedback (QFB) cannot be used to its full extent in these conditions as it relies on stable and reliable tune estimates. In this work we propose new tune estimation algorithms, designed to mitigate this problem through different techniques. As ground-truth of the real tune measurement does not exist, we developed a surrogate model, which allowed us to perform a comparative analysis of a simple weighted moving average, Gaussian Processes and different deep learning techniques. The simulated dataset used to train the deep models was also improved using a variant of Generative Adversarial Networks (GANs) called SimGAN. In addition we demonstrate how these methods perform with respect to the present tune estimation algorithm.

Tremendous progress has been witnessed in artificial intelligence within the domain of neural network backed deep learning systems and its applications. As we approach the post Moore’s Law era, the limit of semiconductor fabrication technology along with a rapid increase in data generation rates have lead to an impending growing challenge of tackling newer and more modern machine learning problems. In parallel, quantum computing has exhibited rapid development in recent years. Due to the potential of a quantum speedup, quantum based learning applications have become an area of significant interest, in hopes that we can leverage quantum systems to solve classical problems. In this work, we propose a quantum deep learning architecture; we demonstrate our quantum neural network architecture on tasks ranging from binary and multi-class classification to generative modelling. Powered by a modified quantum differentiation function along with a hybrid quantum-classic design, our architecture encodes the data with a reduced number of qubits and generates a quantum circuit, loading it onto a quantum platform where the model learns the optimal states iteratively. We conduct intensive experiments on both the local computing environment and IBM-Q quantum platform. The evaluation results demonstrate that our architecture is able to outperform Tensorflow-Quantum by up to 12.51% and 11.71% for a comparable classic deep neural network on the task of classification trained with the same network settings. Furthermore, our GAN architecture runs the discriminator and the generator purely on quantum hardware and utilizes the swap test on qubits to calculate the values of loss functions. In comparing our quantum GAN, we note our architecture is able to achieve similar performance with 98.5% reduction on the parameter set when compared to classical GANs. With the same number of parameters, additionally, QuGAN outperforms other quantum based GANs in the literature for up to 125.0% in terms of similarity between generated distributions and original data sets.

Background: Albumin has a pivotal role in the homeostasis of osmotic pressure and is associated with cardiovascular, nephrotic, hepatic, and nutritional diseases. The detection of hypoalbuminemia is a cornerstone for diagnosis of hidden diseases and patient deterioration. We developed and validated a deep-learning-based model (DLM) for detection of hypoalbuminemia using electrocardiography (ECG). Methods: This historical cohort study included data from consecutive patients from two hospitals. The patient data in one hospital were divided into development (82,499 ECGs from 54,248 patients) and internal validation (20,664 ECGs from 20,664 patients) datasets, whereas the patient data in the other hospital were included in only an external validation (37,421 ECGs from 37,421 patients) dataset. An DLM was developed using a 12-lead ECG signal, age, and sex from the development dataset. The endpoint was hypoalbuminemia, defined by serum albumin concentration below 3.5 g/dL. Results: During the internal and external validations, the areas under the receiver operating characteristic curve of the DLM for the detection of hypoalbuminemia were 0.887 (0.877–0.897) and 0.888 (0.880–0.896), respectively. Among the 27,400 individuals without hypoalbuminemia at the initial laboratory exam, those identified by the DLM as higher-risk patients had a significantly larger change in developing hypoalbuminemia than those in the low-risk group (7.09% vs. 1.01%, p < 0.001) during 24 months. The sensitivity map showed that the DLM focused on the T wave and QRS complex for the detection of hypoalbuminemia. Conclusions: The DLM exhibited a high accuracy for hypoalbuminemia detection and prediction using 12-, 6-, and single-lead ECGs.

Classification of human actions from uni-modal and multi-modal datasets is an ongoing research problem in computer vision. This review is aimed to scope current literature on data-fusion and action-recognition techniques and to identify gaps and future research direction. Success in producing cost-effective and portable vision-based sensors has dramatically increased the number and size of datasets. The rise in number of action recognition datasets intersects with advances in deep-learning architectures and computational support, both of which offer significant research opportunities. Naturally, each action-data modality - such as RGB, depth, skeleton, and infrared - has distinct characteristics; therefore, it is important to exploit the value of each modality for better action recognition. In this article we will focus solely on areas such as data fusion and recognition techniques in the context of vision with a uni-modal and multi-modal perspective. We conclude by discussing research challenges, emerging trends, and possible future research directions.

An eruption of volcano is related to the past volcanic and tectonic seismicity. Recently, on November 29, 2020, a 1423 m Mount Ile Lewotolok in Lembata Island has erupted. In here, this paper aimed to assess the volcanic and tectonic seismicity as determinant factor and precursor of recent Mt Ile Lewotolok eruption. The assessment shows that Mt Ile Lewotolok volcanic activities were characterized by both tectonic and deep volcanic seismicity. Since 2010, mean tectonic quake magnitudes of M 4.133 (95%CI:M 3.205-5.062) have occurred at mean depth of 13.500 km (95%CI:8.201-18.799 km) within a distance of 3-4 km from the summit. Tectonic quake has occurred frequently in southwest of Ile Lewotolok and this has contributed to the past eruptions in 2012 and 2017. Recent eruption has been influenced by deep volcanic seismicity rather than local tectonic. Based on November 2020 record, mean of deep volcanic seismicity frequency was 2.190 events/day (95%CI:1.136-3.243 events/day km) that has outnumbered (t_test=2.665, P=0.013) the events of local tectonic quakes (mean 0.621 events/day; 95%CI:0.142-1.099events/day). Prior to the eruption there was significant increase of deep volcanic seismicity (P=0.023) while local tectonic quake was not showing an increasing trend (P=0.764). This result confirms that the deep volcanic seismicity frequency is a precursor that may trigger the eruption and deep volcanic seismicity data can be used as indicator of volcanic activities.

Mouthbrooding or oral incubation, the retention of early developmental stages inside of the mouth for an extended period of time, has evolved multiple times in bony fishes1,2. Though uncommon, this form of parental care has been documented and well-studied in several groups of freshwater fishes but is also known to occur in a small number of marine fishes, all inhabiting coastal waters1,2. A recent paper3, reported for the first time mouthbrooding in a deep-water fish species, the zeiform Parazen pacificus, which according to the authors “fills in a gap in the larval literature for this family of fishes and prompts further investigation into other novel reproductive modes of deep-sea fauna.”

The conventional machine learning (ML) algorithms are continuously advancing and evolving at a fast-paced by introducing the novel learning algorithms. ML models are continually improving using hybridization and ensemble techniques to empower computation, functionality, robustness, and accuracy aspects of modeling. Currently, numerous hybrid and ensemble ML models have been introduced. However, they have not been surveyed in a comprehensive manner. This paper presents the state of the art of novel ML models and their performance and application domains through a novel taxonomy.

This paper proposes a method for recognizing audio events in urban environments that combines handcrafted audio features with a deep learning architectural scheme (Convolutional Neural Networks, CNNs), which has been trained to distinguish between different audio context classes. The core idea is to use the CNNs as a method to extract context-aware deep audio features that can offer supplementary feature representations to any soundscape analysis classification task. Towards this end, the CNN is trained on a database of audio samples which are annotated in terms of their respective "scene" (e.g. train, street, park), and then it is combined with handcrafted audio features in an early fusion approach, in order to recognize the audio event of an unknown audio recording. Detailed experimentation proves that the proposed context-aware deep learning scheme, when combined with the typical handcrafted features, leads to a significant performance boosting in terms of classification accuracy. The main contribution of this work is the demonstration that transferring audio contextual knowledge using CNNs as feature extractors can significantly improve the performance of the audio classifier, without need for CNN training (a rather demanding process that requires huge datasets and complex data augmentation procedures).

An effective on-board cloud detection method in small satellites would greatly improve the downlink data transmission efficiency and reduce the memory cost. In this paper, an ensemble method combining a lightweight U-Net with wavelet image compression is proposed and evaluated. The red, green, blue and infrared waveband images from Landsat-8 dataset are trained and tested to estimate the performance of proposed method. The LeGall-5/3 wavelet transform is applied on the dataset to accelerate the neural network and improve the feasibility of on-board implement. The experiment results illustrate that the overall accuracy of the proposed model achieves 97.45% by utilizing only four bands. Tests on low coefficients of compressed dataset have shown that the overall accuracy of the proposed method is still higher than 95%, while its inference speed is accelerated to 0.055 second per million pixels and maximum memory cost reduces to 2Mb. By taking advantage of mature image compression system in small satellites, the proposed method provides a good possibility of on-board cloud detection based on deep learning.

This paper presents a new method : HIVEC to learn latent vector representations of graphs in a manner that captures the semantic dependencies of sub-structures. The representations can then be used in machine learning algorithms for tasks such as graph classification, clustering etcetera. The method proposed is unsupervised and uses the information of co-occurrence of sub-structures. It introduces a notion of hierarchical embeddings that allows us to avoid repetitive learning of sub-structures for every new graph. As an alternative to deep learning methods, the edit distance similarity between sub-structures is also used to learn vector representations. We compare the performance of these methods against previous work.

A novel indole, microindolinone A (1), was isolated from a deep-sea-derived actinomycete Microbacterium sp. MCCC 1A11207, together with 18 known compounds (2–19). By detailed analysis of the ¹H, ¹³C, HSQC, COSY, HMBC, HRESIMS, and CD data, the absolute configuration of 1 was elucidated as 5R-hydroxy-4,5,6,7-tetrahydroindole-4-one. Noteworthily, 1 is the second example of a saturated indole isolated from nature.

The COVID-19 pandemic has highlighted the need for a robust medical infrastructure and crisis management strategy as part of smart-city applications, with technology playing a crucial role. The Internet of Things (IoT) has emerged as a promising solution, leveraging sensor arrays, wireless communication networks, and artificial intelligence (AI)-driven decision-making. Advancements in Edge Computing (EC), Deep Learning (DL), and Deep Transfer Learning (DTL) have made IoT more effective in healthcare and pandemic-resilient infrastructure. When combined with medically oriented IoT setups, DL-architectures are suitable for integrating into pandemic-compliant medical infrastructure. The development of intelligent pandemic-compliant infrastructure requires combining IoT, edge and cloud computing, image processing, and AI tools to monitor adherence to social distancing norms, mask-wearing protocols, and contact tracing. The proliferation of 5G wireless communication has enabled ultra-wide broadband wireless communication, with high reliability and low latency, thereby enabling seamless medical support as part of smart-city applications. Such set-ups are designed to be ever-ready to deal with virus-triggered pandemic-like medical emergencies. This study presents the design of a pandemic-compliant mechanism leveraging IoT optimized for healthcare applications, edge- and cloud-computing frameworks, and a suite of DL-tools. The framework uses a composite attention-driven framework incorporating various DL-pre-trained models (DPTM) for protocol adherence and contact tracing. When connected to public networks, it can detect specific cyber-attacks. The results confirm the effectiveness of the proposed methodologies.

Deep Neutral Networks (DNNs) were initially proposed towards the midst of the 20th century, motivated by the neural structure and mechanism of the human brain. Thanks to major ad-vancements in computational resources, during the past decade DNN-based systems have demonstrated unprecedented performance in terms of accuracy and speed. However, recent work has shown that such models may not be sufficiently robust during the inference process. Furthermore, due to the data-driven learning nature of DNNs, designing interpretable and gen-eralizable networks is a major challenge, especially in critical applications, such as medical Computer Aided Diagnostics (CAD) and other medical imaging tasks, including classification, regression and reconstruction. Within this context, a line of physics-driven approaches for deep learning has recently emerged, aimed at improving the stability and generalization capacity of DNNs for medical imaging applications. In this paper, we review recent work focused on phys-ics-driven or prior-information learning for a variety of imaging modalities and medical applica-tions. We discuss how the inclusion of domain-related knowledge into the learning process and networks’ design supports their stability and generalization capability. In addition, we highlight current and future challenges within this scope.

Dynamic adaptive video streaming over HTTP (DASH) plays a crucial role in video transmission across networks. Traditional adaptive bitrate (ABR) algorithms adjust the quality of video segments based on network conditions and buffer occupancy. However, these algorithms rely on fixed rules within a complex environment, making it challenging to achieve optimal decisions considering the overall context. In this paper, we propose a novel Deep Reinforcement Learning-based approach for streaming DASH, focusing on maintaining consistent perceived video quality throughout the streaming session to enhance user experience. Our approach optimizes the Quality of Experience (QoE) by dynamically controlling the quality distance factor between consecutive video segments. We evaluate this approach through a simulation model that encompasses diverse wireless network environments and various video sequences. Additionally, we compare our proposed approach with state-of-the-art methods. The experimental results demonstrate significant improvements in QoE, ensuring users enjoy stable, high-quality video streaming sessions.

Deep learning has taken over - both in problems beyond the realm of traditional, hand-crafted machine learning paradigms as well as in capturing the imagination of the practitioner sitting on top of petabytes of data. While the public perception about the efficacy of deep neural architectures in complex pattern recognition tasks grows, sequentially up-to-date primers on the current state of affairs must follow. In this review, we seek to present a refresher of the many different stacked, connectionist networks that make up the deep learning architectures followed by automatic architecture optimization protocols using multi-agent approaches. Further, since guaranteeing system uptime is fast becoming an indispensable asset across multiple industrial modalities, we include an investigative section on testing neural networks for fault detection and subsequent mitigation. This is followed by an exploratory survey of several application areas where deep learning has emerged as a game-changing technology - be it anomalous behavior detection in financial applications or financial time-series forecasting, predictive and prescriptive analytics, medical imaging, natural language processing or power systems research. The thrust of this review is on outlining emerging areas of application-oriented research within the deep learning community as well as to provide a handy reference to researchers seeking to embrace deep learning in their work for what it is: statistical pattern recognizers with unparalleled hierarchical structure learning capacity with the ability to scale with information.

Background/Objectives Adult hearing-impaired patients qualifying for cochlear implants typically exhibit less than 60% sentence recognition in their best hearing aid condition, either in quiet or noisy environments, with speech and noise presented through a single speaker. This study examines the influence of deep neural network-based (DNN-based) based noise reduction on cochlear implant evaluation. Methods Speech perception was assessed using AzBio sentences in both quiet and noisy conditions (multi-talker babble) at 5 and 10 dB signal-to-noise ratio (SNR) through one loudspeaker. Sentence recognition scores were measured for 10 (5-bilateral, 5-unilateral) hearing-impaired patients using three hearing aid programs: calm situation, speech in noise, and spheric speech in loud noise (DNN-based noise reduction). Speech perception results were compared to bench analyses comprising the phase inversion technique, employed to predict SNR improvement, and the Hearing-Aid Speech Perception Index (HASPI v2), utilized to predict speech intelligibility. Results The spheric speech in loud noise program improved speech perception by 20 to 32% points as compared to the calm situation program. This improvement might make it difficult for some patients to achieve the

Short-term electrical demand forecasting is crucial for the effective operation of new electrical networks. This necessity arises from the requirement for more timely demand information compared to traditional networks, as well as the integration of new energy sources into the electrical grid. Most existing forecasting methodologies fail because they do not consider the nonlinearities of the system, and because they neglect to represent important external influences on the problem; in other words, they are not based on a holistic approach to electricity demand forecasting. In this study, we present a new approach to developing an electrical demand forecasting model, and consider a case study to evaluate the importance of several factors affecting electricity consumption and forecast accuracy. We use an ISO NE (Independent System Operator New England) dataset that represents the total electrical load of several New England cities from January 2017 to December 2019. This dataset includes 23 independent variables, such as meteorological data, economic indicators and market information. Our study presents a pipeline for building electrical demand forecast models using time series information as input. By carefully selecting variables and representing external factors, we demonstrate the feasibility of generating a more accurate forecasting model with reduced computational resources. Our results show that incorporating external factors can enhance model accuracy by up to 60%. To handle large input datasets, variable selection is crucial for reducing dimensionality while maintaining accuracy, enabling the effective application of deep learning models. Simulations reveal that deep learning models with no more than three intermediate layers and an optimal number of neurons perform efficiently. The CNN+LSTM composite model achieved the lowest error rate at 0.15%, surpassing previously studied models, including standalone CNN and LSTM approaches, which had error rates of 0.8% and 1.44%, respectively. The CNN's feature extraction abilities, combined with the LSTM's strength in handling time-series data, were instrumental in achieving superior performance across most scenarios.

Object detection often struggles with accurately identifying visually similar parts, a challenge commonly faced in industrial applications. To address this issue, we propose a clustering methodology based on the visual similarity of 3D object models. This approach is particularly effective when integrated with synthetic image generation, as both processes rely on 3D models. In a case study, we observed a 22 % increase in classification performance on a validation dataset when training an object detector on visually similar groups rather than on all classes, suggesting the potential of our method as a baseline for a multi-stage classification scheme.

Myelination, the process of forming a myelin sheath around nerve fibers, is critical not only during early development but also throughout adulthood. This article explores the multifaceted impacts of myelination on the adult brain, highlighting its importance in cognitive functions, neurological health, and overall brain efficiency. In adults, efficient myelination enhances information processing speed, supports memory consolidation, and contributes to cognitive reserve, potentially delaying the onset of neurodegenerative diseases. Myelination ensures synchronized neural activity and energy-efficient brain function, playing a key role in complex cognitive tasks and coordination. Despite a slower rate of plasticity compared to early development, the adult brain retains the ability to adapt and learn, with myelination facilitating these processes. This article also draws parallels between myelination in the brain and learning curves in deep learning Perceptrons, emphasizing how reduced stimulation and experience can significantly slow down cognitive and learning processes. Additionally, the clinical implications of myelination are discussed, including its role in multiple sclerosis and age-related cognitive decline, along with strategies to maintain and enhance myelin health through lifestyle choices and emerging pharmacological interventions. Understanding the ongoing significance of myelination in the adult brain, and its comparison to artificial neural networks, is crucial for developing interventions to support cognitive health and mitigate neurological disorders.

This study presents an advanced defect detection approach for solar cells using the YOLOv10 deep learning model. Leveraging a comprehensive dataset of 10,500 solar cell images, annotated with 12 distinct defect types, our model integrates Compact Inverted Blocks (CIB) and Partial Self-Attention (PSA) modules to enhance feature extraction and classification accuracy. Training on the Viking cluster with state-of-the-art GPUs, our model achieved remarkable results, including a mean Average Precision (mAP@0.5) of 98.5%. Detailed analysis of the model's performance revealed exceptional precision and recall rates for most defect classes, notably achieving 100% accuracy in detecting black core, corner, fragment, scratch, and short circuit defects. Even for challenging defect types such as thick line and star crack, the model maintained high performance with accuracies of 94% and 96%, respectively. The recall-confidence and precision-recall curves further demonstrate the model's robustness and reliability across varying confidence thresholds. This research not only advances the state of automated defect detection in photovoltaic manufacturing but also underscores the potential of YOLOv10 for real-time applications. Our findings suggest significant implications for improving the quality control process in solar cell production.

Living in extreme environments, marine organisms face constant exposure to a range of stressors, such as high radiation levels, fluctuations in temperature, and oxidative stress. Understanding extremophile fishes is crucial because it gives us valuable insights into the biochemical, physiological, and developmental processes that govern life, by observing how they operate under natural stressors. Among the most fascinating adaptations is the existence of specialised enzymes and compounds that function as potent antioxidants, successfully counteracting reactive oxygen species' deleterious effects. In this review, we analysed the findings from several studies on Antarctic and deep-sea fish species, while highlighting the environmental stressors effects toward the antioxidant system. The antioxidant defences of the considered extremophile fishes have been extensively studied, but there is still much to learn to fully understand this complex system, while the relative research is still ongoing. Consequently, we are properly anticipating further advancements over the next few years about our understanding of crucial physiological processes that support cell survival.

With advancements in digital technologies such as 5G communications, big data, and cloud computing, the components of network operation systems have become increasingly complex, significantly complicating system monitoring and maintenance. Correspondingly, automated log anomaly detection has become a crucial means to ensure stable network operation and protect them from malicious attacks or failures. Conventional machine learning and deep learning methods assume consistent distributions between the training and testing data, adhering to a closed-set recognition paradigm. Nevertheless, in realistic scenarios, systems may encounter new anomalies that were not present in the training data, especially in log anomaly detection. Inspired by evidential learning, we propose a novel anomaly detector called LogEDL, which supervises training of the model through an evidential loss function. Unlike traditional loss functions, the evidential loss function not only focuses on correct classification, but also quantifies the uncertainty of predictions. This enhances the robustness and accuracy of the model in handling anomaly detection tasks, while achieving functionality similar to open-set recognition. To evaluate the proposed LogEDL method, we conduct extensive experiments on three datasets, i.e., HDFS, BGL, and Thunderbird, to detect anomalous log sequences. Experimental results demonstrate that our proposed LogEDL achieves state-of-the-art performance in anomaly detection.

In directional forming processes such as rolling and extrusion, the microstructure is strongly elongated in the forming direction. This results in a preferred alignment of the crystal orientations of the initially randomly oriented grains. These preferred orientations – the texture – and elongations of the grains are the cause of the material anisotropy. This anisotropy leads to phenomena such as earing, which occur during further forming processes, e.g., during deep drawing of sheet metal. Considering anisotropic properties in numerical simulations allows to investigate the effects of texture-dependent defects in forming processes and the development of possible solutions. Purely phenomenological models for modelling anisotropy work by fitting material parameters or applying measured R-values in experiments which are applied to all elements of the part and remain constant over the duration of the simulation. In contrast, methods such as the Visco-Plastic Self-Consistent (VPSC) model, which has been used in this work, provide a deeper insight into the development of the material microstructure. By experimentally measuring the initial texture and using it as an initial condition for the simulations, it is possible to predict the evolution of the microstructure and the resulting effect on the mechanical properties during forming operations. The results of the simulations with VPSC show good qualitative agreement with tests typically used for rolled stock.

Analog neuromorphic prototyping is crucial for creating spiking neuron models that use memristive devices as synapses to design integrated circuits that leverage on-chip parallel deep neural networks. These models mimic how biological neurons in the brain communicate through electrical potentials. Doing so enables more powerful and efficient functionality than traditional artificial neural networks that run on von Neumann computers or graphic processing unit-based platforms. This technology can accelerate deep learning processing, aiming to exploit the brain’s unique features of asynchronous and event-driven processing by leveraging neuromorphic hardware’s inherent parallelism and analog computation capabilities. Therefore, this paper presents the design and implementation of a leaky integrate and fire neuron prototype implemented with commercially available components on a PCB board. Simulations conducted in LTSpice agree well with the electrical test measurements. The results demonstrate that this design can be employed to interconnect many boards to build layers of physical spiking neurons, interconnected with spike-timing-dependent plasticity as the primary learning algorithm, contributing to the realization of experiments in the early stage of adopting neuromorphic computing.

Sign language recognition technology can help people with hearing impairments to communicate with those who are hearing impaired. At present, with the rapid development of society, deep learning technology also provided certain technical support for sign language recognition work. In sign language recognition tasks, the use of traditional convolutional neural networks to extract spatio-temporal features from sign language videos suffers from insufficient feature extraction, resulting in low recognition rates. Nevertheless, video-based sign language datasets are very large and require a lot of computational resources for training and generalisation must be ensured, which poses a challenge for recognition. In this paper, we have presented a video-based sign language recognition method based on resnet and lstm. As the number of network layers increases, the ResNet network can effectively solve the granularity explosion problem and obtain better time series features. We use the Resnet convolutional network as the backbone model. At the initialisation stage, we obtain sign language features using ResNet; then, the learned feature space is used as the input of LSTM network to obtain long sequence features. The experimental results show that the accuracy of the above model is better than the mainstream model, and it can effectively extract the spatio-temporal features in sign language videos and improve the recognition rate of sign language actions.

Deep learning image reconstruction (DLIR) algorithms employs convolutional neural networks (CNNs) for CT image reconstruction to produce CT images with a very low noise level, even at low radiation dose. The aim of this study was to assess whether the DLIR algorithm reduces the CT effective dose (ED) and im-proves CT image quality in comparison with filtered back projection (FBP) and iterative reconstruction (IR) algorithms in intensive care unit (ICU) patients. We identified all consecutive patients referred to the ICU of a single hospital who underwent at least two consecutive chest and/or abdominal contrast-enhanced CT scans within a time period of 30 days using DLIR and subsequently FBP or IR algorithm (Advanced Mod-eled Iterative Reconstruction model-based algorithm [ADMIRE], or Adaptive Iterative Dose Reduction 3D [AIDR 3D]) for CT image reconstruction. The radiation ED, noise level, and signal-to-noise ratio (SNR) were compared between the different CT scanners. The non-parametric Wilcoxon test was used for statistical comparison. Statistical significance was set at p < 0.05. A total of 83 patients (mean age, 59 ± 15 years [standard deviation]; 56 men) were included. DLIR vs FBP reduced ED (18.45 ± 13.16 mSv vs 22.06 ± 9.55 mSv, P < 0.05), while DLIR vs FBP and vs ADMIRE and AIDR 3D IR algorithms reduced image noise (8.45 ± 3.24 vs 14.85 ± 2.73 vs 14.77 ± 32.77 and 11.17 ± 32.77\, P < 0.05) and increased SNR (11.53 ± 9.28 vs 3.99 ± 1.23 vs 5.84 ± 2.74 and 3.58 ± 2.74, P < 0.05). CT scanners employing DLIR reduced radiation ED and im-proved SNR compared to CT scanners using FBP, whereas CT scanners using DLIR improved SNR com-pared to CT scanners using FBP or IR algorithms in ICU patients.

We introduce a novel microscopic image dataset augmented with segmentation and detection labels specifically designed for microplastic analysis in sewage environments. Recognizing the increasing concern over microplastics — particles of synthetic polymers smaller than 5mm — and their detrimental effects on marine ecosystems and human health, our research focuses on enhancing detection and analytical methodologies through advanced computer vision and deep learning techniques. The dataset comprises high-resolution microscopic images of microplastics collected from sewage, meticulously labeled for both segmentation and detection tasks, aiming to facilitate accurate and efficient identification and quantification of microplastic pollution. In addition to dataset development, we present example deep learning models optimized for segmentation and detection of microplastics within complex sewage samples. The models demonstrate significant potential in automating the analysis of microplastic contamination, offering a scalable solution to environmental monitoring challenges. Furthermore, we ensure the accessibility and reproducibility of our research by making the dataset and model codes publicly available, accompanied by detailed documentation on GitHub and LabelBox. The dataset and example deep learning models are publicly available at the following GitHub link. https://anonymous.4open.science/r/Microplastics-in-Sewage-1BEF

Recent advancements in artificial intelligence (AI) and deep learning have revolutionized the field of protein engineering, particularly in the area of de novo protein design. This review article explores the impact of AI-driven approaches on protein design, with a specific focus on the role of structure prediction neural networks, such as AlphaFold and RoseTTAFold. The article discusses the paradigm shift brought about by these networks, which have enabled the design of proteins with unique structures and functions that are not found in nature. The review covers various aspects of AI-driven protein design, including the use of protein language models to harness evolutionary information, the development of de novo protein design workflows utilizing deep learning, and the application of generative models like RFdiffusion and RoseTTAFold All-Atom. The article also highlights the successes and applications of AI-driven protein design across diverse domains, such as enzyme engineering, antibody design, and vaccine development. Additionally, the review identifies current challenges and future directions in the field, emphasizing the need to address limitations in modeling conformational dynamics and designing proteins for in vivo functionality. The article concludes by underscoring the potential of AI-driven protein design to transform various aspects of science and technology, while also acknowledging the importance of interdisciplinary collaboration and the development of robust pipelines for the validation and optimization of designed proteins. Overall, this comprehensive review serves as a valuable resource for researchers and practitioners interested in understanding the current state and future prospects of AI-driven protein design.

Artificial intelligence (AI) refers to computer algorithms that replicate the cognitive function of humans. Machine learning is widely applicable using structured and unstructured data, while deep learning is derived from the neural networks of the human brain that process and interpret information. During the last decades, AI has been introduced in several aspects of healthcare. In this review, we aim to present the current application of AI in the neonatal intensive care unit. AI-based models have been applied to neurocritical care, including automated seizure detection algorithms and electroencephalogram-based hypoxic-ischemic encephalopathy severity grading systems. Moreover, AI models evaluating magnetic resonance imaging contributed to the progress of the evaluation of the neonatal developing brain and the understanding of how prenatal events affect both structural and functional network topologies. Furthermore, AI algorithms have been applied to predict the development of bronchopulmonary dysplasia and assess the extubation readiness of preterm neonates. Automated models have been also used for the detection of retinopathy of prematurity and the need for treatment. Among others, AI algorithms have been utilized for the detection of sepsis, the need for patent ductus arteriosus treatmnet, the evaluation of jaundice, and the detection of gastrointestinal morbidities. Finally, AI prediction models have been constructed for the evaluation of the neurodevelopmental outcome and the overall mortality of neonates. Although the application of AI in neonatology is encouraging, further research in AI models is warranted in the future including retraining clinical trials, validating the outcomes, and addressing serious ethics issues.

Objectives: The purpose of the study was to evaluate the effectiveness of dental caries segmentation on the panoramic radiographs taken from children in primary dentition, mixed dentition, and permanent dentition with an Artificial Intelligence (AI) models developed using the deep learning method. Methods: This study used 6075 panoramic radiographs taken from children aged between 4 and 14 to develop the AI model. The radiographs included in the study were divided into three groups as primary dentition (n: 1857), mixed dentition (n: 1406), and permanent dentition (n: 2812). U-Net model implemented with PyTorch library was used for segmentation of caries lesions. Confusion matrix was used to evaluation of model performance. Results: In the primary dentition group, the sensitivity, precision and F1 scores calculated using the confusion matrix were found to be 0.8525, 0.9128, and 0.8816, respectively. In the mixed dentition group, the sensitivity, precision, and F1 scores calculated using the confusion matrix were found as 0.7377, 0.9192, and 0.8185, respectively. In the permanent dentition group, the sensitivity, precision, and F1 scores calculated using the confusion matrix were found as 0.8271, 0.9125, and 0.8677, respectively. In the total group including primary, mixed, and permanent dentition, the sensitivity, precision and F1 score calculated using the confusion matrix were 0.8269, 0.9123 and 0.8675, respectively. Conclusion: Deep learning-based AI models are promising tools for the detection and diagnosis of caries in panoramic radiographs taken from children in different dentitions.

Agricultural crops productivity is heavily influenced by the strength of their seeds. Current methods for testing seed vigor, such as the germination rate and tetrazolium test, rely on time-consuming human visual inspections. Computer Vision, a crucial technology in smart production processes, offers an automated alternative and is increasingly applied in digital agriculture processes. Here we propose a method that combines YOLOv8 and morphological analysis of seeds using X-ray images to aid seed quality analysts in classifying seed lots based on their physiological potential. We examined the internal morphology of three seed lots and generated X-ray images to train a YOLOv8-based model with a post-segmentation module. This model assessed various parameters such as the total seed area, area filled by the embryo and endosperm, length, width, and calculated the percentage occupied by the embryo and endosperm in relation to the total seed area. Subsequently, seeds were classified into four categories based on the internal area occupied by the endosperm and embryo. Our findings reveal the robust performance of the YOLOv8 model in segmenting and classifying despite the challenges posed by a relatively small dataset. Notably, the proposed model achieved an impressive accuracy of up to 95.6% in identifying and segmenting the endosperm over 1,500 epochs with just 15 training images. The endosperm/seed area ratio, particularly within the 50-60% range covering over 50% of the samples, emerged as a significant metric for evaluating the viability of seed batches.

The combustion of fuel oil releases sulfur compounds, which react to form sulfur oxides that are harmful to the environment and pose a threat to human health. To mitigate these negative effects, it is crucial to pretreat fuel oil through desulfurization processes. One promising approach involves the use of deep eutectic solvents (DESs), carbon nanotubes (CNTs), and their composites (DESs/CNTs) to effectively reduce sulfur content in fuel oil.However, the conventional methods of DES-based extraction desulfurization and CNT-based adsorption desulfurization suffer from low desulfurization rates. Additionally, the composite material DESs/CNTs is in a semi-solidified form, resulting in high viscosity, complex operation procedures, and increased waste and environmental pollution. To address these challenges, this study focused on the preparation of CNTs/PEG-PVA hydrogel. This hydrogel not only maintains the high desulfurization rate observed in the DESs/CNTs composite material but also possesses several advantages. Firstly, it is convenient to use, offering ease of handling and application. Secondly, it is biodegradable, meaning it can break down naturally over time, reducing its impact on the environment. Lastly, the hydrogel is reusable, allowing for repeated use, thereby minimizing material waste. The desulfurization rate achieved with the CNTs/PEG-PVA hydrogel is remarkably high, reaching up to 99.8%. Moreover, the material utilization rate is close to 100%, meaning that almost all of the hydrogel is utilized in the desulfurization process, reducing unnecessary material consumption. In summary, the innovative CNTs/PEG-PVA hydrogel presents a highly efficient and environmentally friendly solution for desulfurization in fuel oil. Its superior desulfurization rate, alongside its convenience, biodegradability, and reusability, positions it as a promising alternative that not only ensures effective desulfurization but also contributes to sustainable practices in the energy industry.

Sustainable agriculture is at the forefront of modern farming practices, with a growing emphasis on optimizing crop production while minimizing environmental impact. The choice of fertilizer application technology plays a critical role in achieving these objectives. Soybean (Glycine max (L.) Merr.), a valuable crop globally, serves both as a source of human nutrition and livestock feed. To address the challenges of enhancing soybean production while minimizing ecological harm, this study evaluates and compares the performance of different fertilizer application technologies. In this study we combined remote and ground surveys to allow a comprehensive understanding of the impact of these technologies on soybean cultivation. The study focuses on deep placement of slow-release nitrogen fertilizers with coated urea at a 20 cm depth, conventional nitrogen fertilizer application, and manure organic fertilizer application. Remote surveys are conducted using advanced smart farming tools such as Unmanned Aerial Vehicle (UAV), ground surveys such as soil plant analysis development SPAD measurements, and soybean parameters to assess the impact of fertilizer application. This study revealed that deep placement fertilizer demonstrated better performance compared to other fertilizer application technologies for soybean cultivation. The findings from this study contribute to the advancement of sustainable agricultural practices and empower farmers with knowledge to make informed decisions for optimizing green soybean production.

PurposeThe purpose of this research was the development and evaluation of an intelligent assistive classification system for the recognition of endometriosis and the improvement of the accuracy of laparoscopic imaging in diagnosis. The research was developed with the use of Deep Learning approaches. Methods Data from 4448 laparoscopy images were used in a retrospective chart analysis. The data were divided into two folders, healthy and pathological including 2157 healthy and 2291 pathological images. Based on simple clinical and imaging information and criteria such as the diagnosis of endometriosis (included in an open-source dataset GLENDA of Kaggle repository), data mining algorithms were used to improve laparoscopic imaging accuracy. Results The final developed computer system based on the ResNet50 algorithm predicted the best outcome for all participants who had laparoscopic surgical therapy. The Keras tool was used and the generated code was implemented in Python programming language providing a mean accuracy of > 95 %. Conclusion The intelligent approach revealed better performance than the commonly used imaging criteria in predicting endometriosis improving the time and the total accuracy of diagnostic approaches.

Olive fruits at different ripening stages give rise to various table olive products and oil qualities. Therefore, developing an efficient method for recognizing and sorting olive fruits based on their ripening stages can greatly facilitate postharvest processing. This study introduces an automatic computer vision system that utilizes deep learning technology to classify the `Roghani` Iranian olive cultivar into five ripening stages using color images. The developed model employs convolutional neural networks (CNN) and transfer learning based on the Xception architecture and ImageNet weights as the base network. The model was fine-tuned by testing multiple configurations of well-known CNN layers. To minimize overfitting and enhance model generality, data augmentation techniques were employed. By considering different optimizers and two image sizes, four final candidate models were generated. These models were then compared in terms of loss and accuracy on the test dataset, classification performance (classification report and confusion matrix), and generality. All four candidates exhibited high accuracies ranging from 86.93% to 93.46% and comparable classification performance. In all models, at least one class was recognized with 100% accuracy. However, by taking into account the risk of overfitting, two models were discarded. Finally, a model with an image size of 224 × 224 and an SGD optimizer, which had a loss of 1.23 and an accuracy of 86.93%, was selected as the preferred option. The results of this study offer robust tools for automatic olive sorting systems, simplifying the differentiation of olives at various ripening levels for different post-harvest products.

The rate of accidents in Morocco is experiencing a significant increase. Automatic license plate detection and recognition (ALPR) is an essential road safety technology. It facilitates applications such as traffic control, law enforcement, and toll collection by allowing for the automated recognition of vehicles on the road. In this study, we incorporated ShuffleNet V2 into the end-to-end YOLOV5 object detection system. The objective was to develop a model capable of identifying Moroccan license plates with an accuracy of 87%. The proposed model is intended to attain a high processing performance of 60 frames per second (FPS) while maintaining a low weight of 1.3 megabytes (MB) and a parameter count of 0.44 million floating point operations (MGFLOP). Our model maintains superior performance and is highly compatible with embedded systems compared to other models utilized in the same context.

This paper presents ConF, a novel deep learning model designed for accurate and efficient prediction of non-coding RNA families. NcRNAs are essential functional RNA molecules involved in various cellular processes, including replication, transcription, and gene expression. Identifying ncRNA families is crucial for comprehensive RNA research, as ncRNAs within the same family often exhibit similar functionalities. Traditional experimental methods for identifying ncRNA fam-ilies are time-consuming and labor-intensive. Computational approaches relying on annotated secondary structure data face limitations in handling complex structures like pseudoknots and have restricted applicability, resulting in suboptimal prediction performance. To overcome these chal-lenges, ConF integrates mainstream techniques such as residual networks with dilated convolutions and cross multi-head attention mechanisms. By employing a combination of dual-layer convolu-tional networks and BiLSTM, ConF effectively captures intricate features embedded within RNA sequences. This feature extraction process leads to significantly improved prediction accuracy compared to existing methods. Experimental evaluations conducted on a ten-fold publicly available dataset demonstrate the superiority of ConF in terms of accuracy, sensitivity, and other perfor-mance metrics. Overall, ConF represents a promising solution for accurate and efficient ncRNA family prediction, addressing the limitations of traditional experimental and computational methods.