A real-time monitoring system for the operational status of fishing vessels is an essential element for the modernization of the fishing industry. The operational status of fishing vessels can be identified by using onboard cameras to detect the deck crew and the use of fishing nets. Due to the typically limited processing capacity of shipboard equipment and the significant memory consumption of detection models, however, general target detection models are unable to perform real-time image detection to identify the operational status of fishing vessels. In this paper, we propose a lightweight real-time deck crew and the use of fishing nets detection method, YOLOv5s-SGC. It is based on the YOLOv5s model, which uses surveillance cameras to obtain video of fishing vessels operating at sea and enhances the dataset. YOLOv5s-SGC replaces YOLOv5s’s backbone and the ordinary convolutional blocks in the feature fusion network with ShufflNetV2 and Ghost module.

Fluctuating stock prices make it difficult for investors to see investment opportunities. One tool that can help investors overcome this is forecasting techniques. Long Short-Term Memory (LSTM) is one of deep learning methods used in forecasting time series. The training and success of deep learning is strongly influenced by the selection of hyperparameters. This research uses a hybrid method between the Genetic Algorithm (GA) and LSTM to find a suitable model for predicting stock prices. GA is used in optimizing the architecture such as the number of epochs, window size, and the number of LSTM units in the hidden layer. Tuning optimizer is also carried out using several optimizers to achieve the best value. From method that has been applied, it shows that the method has a good level of accuracy with MAPE values below 10% in every optimizer used. The error rate generated is quite low, in case-1 with a minimum RMSE value of 93.03 and 94.40, & in case-2 with an RMSE value of 104.99 and 150.06 during training and testing. A fairly stable and small value is generated by setting it using the Adam Optimizer.

Over the last several decades, large wildfires are increasingly common across the United States causing disproportionate impact on forest health and function, human well-being, and economy. Here, we examine the severity of large wildfires across the Contiguous United States over the past decade (2011-2020) using a wide array of meteorological, vegetational, and topographical features in the Deep Neural Network model. A total of 4,538 wildfire incidents were used in the analysis covering 87,305 square miles of burned area. We observed the highest number of large wildfires in California, Texas, and Idaho, with lightning causing 43 % of these incidents. Importantly, results indicate that the severity of wildfire occurrences is highly correlated with the climatological forcings, land cover, location, and elevation of the ecosystem. Overall, results may serve useful guide in managing landscapes under changing climate and disturbance regimes.

The rapid development of Deep Learning brought novel methodologies for 3D Object Detection using LiDAR sensing technology. These improvements in precision and inference speed performances lead to notable high performance and real-time inference, which is especially important for self-driving purposes. However, the developments carried by these approaches overwhelm the research process in this area since new methods, new technologies, and software versions lead to different project necessities, specifications and requirements. Moreover, the improvements brought by the new methods may be due to improvements in newer versions of deep learning frameworks and not just the novelty and innovation of the model architecture. Thus, it became crucial to create a framework with the same software versions, specifications and requirements that accommodate all these methodologies and allow the easy introduction of new methods and models. A framework is proposed that abstracts the implementation, reusing and building of novel methods and models. The main idea is to facilitate the representation of state-of-the-art (SoA) approaches and simultaneously encourage the implementation of new approaches by reusing, improving and innovating modules in the proposed framework, which has the same software specifications to allow a fair comparison. This makes it possible to determine if the key innovation approach outperforms the current SoA by comparing models in a framework with the same software specifications and requirements.

Unmanned Aerial Vehicles (UAV) are increasingly being used in a variety of domains and precision agriculture is no exception. Precision agriculture is the future of agriculture and will play a key role in long-term sustainability of agricultural practices. This paper presents a survey of how image data collected using UAVs has been used in conjunction with ma-chine learning techniques to support precision agriculture. Numerous agricultural applications including classification of crop types and trees, crops detection, weed detection, cropland cover, and segmentation of farming fields are discussed. A variety of supervised, semi-supervised and unsupervised machine learning techniques for image-based preci-sion agriculture are compared. The survey showed that for traditional machine learning approaches, Random Forests performed better than Support Vector Machines (SVM) and K-Nearest Neighbor Algorithm (KNN) for crop/weed classification. And, while Convolutional Neural Networks (CNN) have been used extensively, the U-Net-based models out-performed conventional CNN models for classification and segmentation tasks. Among the Single Stage Detectors (SSD), YOLO series performed relatively well. Two-Stage Detectors like R-CNN, FPN, and Mask R-CNN generally tended to outperform SSDs. Vision Trans-formers (ViT) showed promising results amongst transformer-based models which did not generally perform better than CNNs. Finally, Generative Adversarial Networks (GANs) have been used to address the problem of smaller datasets and unbalanced data

As a new distributed machine learning (ML) approach, federated learning (FL) shows the great potential to preserve data privacy by enabling distributed data owners to collaboratively build a global model without sharing their raw data. However, the heterogeneity in terms of data distribution and hardware configurations make it hard to select participants from the thousands of nodes. In this paper, we propose a multi-objective node selection approach to improve time-to-accuracy performance while resisting malicious nodes. We firstly design a deep reinforcement learning assisted FL framework. Then the problem of multi-objective node selection under this framework is formulated as a Markov decision process (MDP), which aims to reduce the training time and improve model accuracy simultaneously. Finally, a deep Q-netwok (DQN) based algorithm is proposed to efficiently solve the optimal set of participants for each iteration. Simulation results show that the proposed method not only significantly improves the accuracy and training speed of FL, but has stronger robustness to resist malicious nodes.

Importance and need for cyber security have increased in folds since a decade. Indirectly, the country’s security depends on the country’s cyber-physical systems. Attackers are becoming more innovative, and attacks are becoming undetectable, causing huge risks to the systems. In this scenario, intelligent and evolving detection methods should be introduced to replace the basic and outworn ones. This article discusses about new-age intelligence and smart techniques dealing with artificial intelligence (AI) models. Artificial intelligence for cyber security is reviewed, and the performance of machine learning models (ML) and deep learning (DL) models are analysed. A real-time case study of stealthy local covert attacks with false data injection attacks is implemented on the DC-DC converter. A deep learning model is designed to mitigate cyber attacks, and its performance is evaluated.

For W-band long-range mm-wave wireless transmission systems, nonlinearity issues resulting from photoelectric devices, optical fiber, and wireless power amplifiers can be handled by deep learning equalization algorithms. Besides, the PS technique is considered an effective measure to further increase the capacity of the modulation-constraint channel. However, since the proba-bilistic distribution of m-QAM varies with the amplitude, there have been difficulties in learning valuable information from the minority class. This limits the gain of nonlinear equalization. To overcome the imbalanced machine learning problem, we propose a novel two-lane DNN (TLD) equalizer with the random over-sampling (ROS) technique in this paper. The combination of PS at the transmitter and ROS at the receiver improves the overall performance of the W-band wireless transmission system, and our 4.6-km ROF delivery experiment verifies its effectiveness for the W-band mm-wave PS-16QAM system. Based on our proposed equalization scheme, we realize single-channel 10-Gbaud W-band PS-16QAM wireless transmission over 100 m optical fiber link and 4.6-km wireless air-free distance. The results show that compared with the typical TLD without ROS, the TLD-ROS can improve the receiver sensitivity by 1dB. Furthermore, a reduction of 45.6% in complexity is realized, and training samples can be reduced by 15.5%. Considering the actual wireless physical layer with its requirements, there is much to be gained from the joint use of deep learning and balanced data pre-processing techniques.

Indoor localization is an active area of research dominated by traditional machine-learning techniques. Deep learning-based systems have shown unprecedented improvements and have accomplished exceptional results over the past decade, especially the Transformer network within natural language processing (NLP) and computer vision domains. We propose the hyper-class Transformer (HyTra), an encoder-only Transformer with multiple classification heads (one per class) and learnable embeddings, to investigate the effectiveness of Transformer-based models for received signal strength (RSS) based WiFi fingerprinting. HyTra leverages learnable embeddings and the self-attention mechanism to determine the relative position of the wireless access points (WAPs) within the high-dimensional embedding space, improving the prediction of user location. From an NLP perspective, we consider a fixed order sequence of all observed WAPs as a sentence and the captured RSS value(s) for every given WAP at a given reference point from a given user as words. We test our proposed network on public and private datasets of different sizes, proving that the quality of the learned embeddings and overall accuracy improves with increments in samples.

Raman-based Distributed Temperature Sensing (DTS) is a valuable tool for field testing and validating heat transfer models in Borehole Heat Exchanger (BHE) and Ground Source Heat Pump (GSHP) applications. However, temperature uncertainty is rarely reported in the literature. In this paper, a new calibration method is proposed for single-ended DTS configurations, along with a method to remove fictitious temperature drifts due to ambient air variations. The methods are implemented for a Distributed Thermal Response Test case study in an 800 m deep coaxial BHE. The results show that the calibration method and temperature drift correction are robust and give adequate results, with a temperature uncertainty of ± 0.58, 0.74, 1.1 and 1.7 K found at depths of 200, 300, 500 and 800 m, respectively. The temperature uncertainty increases non-linearly with depth and is dominated by the uncertainty in the calibrated parameters for depths larger than 200 m. The paper also offers insights into thermal features observed during the DTRT, including a heat flux inversion along the borehole depth and the slow temperature homogenization under circulation. Distributed temperature measurements and their uncertainties are useful tools in determining the strengths and drawbacks of ground heat transfer models.

Cyber-physical security is vital for protecting key computing infrastructure against cyber attacks. Individuals, corporations, and society can all suffer considerable digital asset losses due to cyber attacks, including Data loss, theft, financial loss, reputation harm, company interruption, infrastructure damage, ransomware attacks, and espionage. A cyber-physical attack harms both digital and physical assets. Cyber-physical system security is more challenging than software-level cyber security because it requires physical inspection and monitoring. This paper proposes an innovative and effective algorithm to strengthen Cyber-Physical Security (CPS) with minimal human intervention. It is a Human Activity Recognition (HAR)-based approach where a GoogleNet-BiLSTM network hybridization has been used to recognize suspicious activities in cyber-physical infrastructure perimeter. The proposed HAR-CPS algorithm classifies suspicious activities from real-time video surveillance with an average accuracy of 73.15%. It incorporates Machine Vision at the IoT Edge (Mez) technology to make the system latency tolerant. Dual-layer security has been ensured by operating the proposed algorithm and GoogleNet-BiLSTM hybrid network from a cloud server, which ensures the security of the proposed security system. The innovative optimization scheme makes it possible to strengthen cyber-physical security with $4.29 per month only.

Predicting Length of Stay (LoS) and understanding its underlying factors is essential to minimize the risk of hospital-acquired conditions, improve financial, operational, and clinical outcomes, and to better manage future pandemics. The purpose of this study is to forecast patients’ LoS using a deep learning model and analyze cohorts of risk factors minimizing or maximizing LoS. We employed various pre-processing techniques, SMOTE-N to balance data, and Tab-Transformer model to forecast LoS. Finally, Apriori algorithm was applied to analyze cohorts of risk factors influencing LoS at hospital. The Tab-Transformer outperformed the base Machine Learning models with an F1-score (.92), precision (.83), recall (.93), and accuracy (.73) for discharge dataset, and F1-score (.84), precision (.75), recall (.98), and accuracy (.77) for deceased dataset. The association mining algorithm was able to identify significant risk factors/indicators belonging to lab, X-Ray, and clinical data such as elevated LDH, and D-Dimer, lymphocytes count, and comorbidities such as hypertension and diabetes responsible for extending patients LoS. It also reveals what treatments has reduced the symptoms of COVID-19 patients leading to reduction in LoS particularly when no vaccines or medication such as Paxlovid were available.

Ultrasound (US) is commonly used for the diagnosis of liver masses. Ensemble learning has been widely used for image classification, but its methods have not been fully optimized. This study was performed to investigate the usefulness of ensemble learning and compare a number of ensemble learning techniques using multiple convolutional neural network (CNN)-trained models for image classification of liver masses in US images. The US imaging data set was classified into four categories: benign liver tumor (BLT, 6320 images), liver cyst (LCY, 2320 images), metastatic liver cancer (MLC, 9720 images), and primary liver cancer (PLC, 7840 images). In this study, 250 test images were randomly selected for each class, for a total of 1000 images, and the remaining images were used for training. Sixteen different CNNs were used for to train and test the US images. All four types of ensemble learning—soft voting (SV), stacking (ST), weighted average voting (WAV), and weighted hard voting (WHV)—showed greater accuracy than the single CNN. All four types also showed significantly better deep learning performance than ResNeXt101 alone. For image classification of liver masses using US images, ensemble learning improved the performance of deep learning over a single CNN.

Micro-expression (ME) is one of the key psychological stress reactions. It is a modest, spontaneous facial mechanism. ME has significant applicability in a variety of psychologically-related sectors because to its precision and unpredictability with regard to psychological manifestations. Nevertheless, the current Micro-expression recognition (MER) algorithms have poor accuracy and a limited quantity of ME data, and this study issue has not been thoroughly investigated. Therefore, we present an approach for deep learning based on a Spatio-temporal capsule network (STCP-Net). STCP-Net has four components: a jitter reduction module, a differential feature extraction module, an STCP module, and a fully linked layer. The first two modules are aimed to extract diversifying differential features more precisely and to limit the influence of head jitter. The STCP module is used to extract Spatio-temporal features layer by layer, taking the temporal and geographical connection between features into account. This research runs sufficient trials using the Leave One Subject Out (LOSO) methodology for cross-validation using the CASMEII dataset. The conclusion and analysis demonstrate that the algorithm is innovative and efficient.

A common challenge in real-world classification scenarios with sequentially appending target domain data is insufficient training datasets during the training phase. Therefore, conventional deep learning and transfer learning classifiers are not applicable especially when individual classes are not represented or are severely underrepresented at the outset. Domain Generalization approaches reach their limits when domain shifts become too large, making them occasionally unsuitable as well. In many (technical) domains, however, it is only the defect/ worn/ reject classes that are insufficiently represented, while the non-defect class is often available from the beginning. The proposed classification approach addresses such conditions and is based on a CNN encoder. Following a contrastive learning approach, it is trained with a modified triplet loss function using two datasets: Besides the non-defective target domain class (= 1st dataset), a state-of-the-art labeled source domain dataset that contains highly related classes (e.g., a related manufacturing error or wear defect) but originates from a (highly) different domain (e.g., different product, material, or appearance) (= 2nd dataset) is utilized. The approach learns the classification features from the source domain dataset while at the same time learning the differences between the source and the target domain in a single training step, aiming to transfer the relevant features to the target domain. The classifier becomes sensitive to the classification features and – by architecture – robust against the highly domain-specific context. The approach is benchmarked in a technical and a non-technical domain and shows convincing classification results. In particular, it is shown that the domain generalization capabilities and classification results are improved by the proposed architecture, allowing for larger domain shifts between source and target domains.

Breast cancer is among the most common and fatal diseases for women, and no permanent treatment has been discovered. Thus, early detection is a crucial step to control and cure breast cancer that can save the lives of millions of women. For example, in 2020, more than 65% of breast cancer patients were diagnosed in early-stage of cancer, from whom all survived cancer. Although early detection is the most effective approach for cancer treatment, breast cancer screening conducted by radiologists is very expensive and time-consuming. More importantly, conventional methods of analyzing breast cancer images suffer from high false detection rates. Different breast cancer imaging modalities are used to extract and analyze the key features affecting diagnosis and treatment of breast cancer. These imaging modalities can be divided into subgroups such as mammograms, ultrasound, magnetic resonance imaging, histopathological images, or any combination of them. Radiologists or pathologists analyze images produced by these methods manually that leads to increase the risk of wrong decisions for cancer detection. Thus, utilization of new automatic methods to analyze all kinds of breast screening images to assist radiologists to interpret images is required. Recently, artificial intelligence (AI) has been widely utilized to automatically improve the early detection and treatment of different types of cancer, specifically breast cancer, thereby enhancing the survival chance of patients. Advances in AI algorithms, such as deep learning, and availability of datasets obtained from various imaging modalities have opened an opportunity to surpass limitations of current breast cancer analysis methods. In this article, we first review breast cancer imaging modalities, and their strengths and limitations. Then, we explore and summarize the most recent studies that employed AI in breast cancer detection using various breast imaging modalities. In addition, we report available datasets on the breast cancer imaging modalities which are important in developing AI-based algorithms and training deep learning models. In conclusion, this review paper tries to provide a comprehensive resource to help researchers working in breast cancer imaging analysis.

Fresh potatoes were deep-fried in olive oil (OO) & extra virgin olive oil (EVOO) and their blends with 5%, 10% and 20% v/v sesame oil (SO). This is the first report on the use of sesame oil as natural source of antioxidant for olive oil deep-frying. Oil was evaluated for Peroxide Value (PV), Free Fatty Acids (FFA), K₂₃₂, K₂₇₀, Trolox Equivalent Antioxidant Capacity (TEAC) and Total Phenols (TP) until Total Polar Compounds (TPCs) reached 25%. Sesame lignan transformations were monitored through Reverse-phase HPLC. While TPCs in olive oils increased at a steady rate, the addition of 5%, 10% and 20% v/v SO created a time window lasting 1, 2 and 3 hours, respectively, where TPCs were constant. SO addition to OO increased the total frying time. Furthermore, the addition of SO reduced the peroxides formation rate for both OO and EVOO. EVOO was more resistant to oxidation than OO as measured by TPCs and TEAC, while frying time raised from 21.5 to 25.25 h when EVOO replaced OO. The increase in frying time for olive oil but not for extra virgin olive oil, after SO addition, is pointing out a niche market for extra virgin olive oil in deep-frying.

River flow prediction is a pivotal task in the field of water resource management during the era of rapid climate change. The highly dynamic and evolving nature of the climatic variables e.g., precipitation has a significant impact on the temporal distribution of the river discharge in recent days making the discharge forecasting even more complicated for diversified water-related issues e.g., flood prediction and irrigation planning. To predict the discharge, various physics-based numerical models are used using numerous hydrologic parameters. Extensive lab-based investigation and calibration are required to reduce the uncertainty involved in those parameters. However, in the age of data-driven predictions, several deep learning algorithms showed satisfactory performance in dealing with sequential data. In this research, Long Short-term Memory (LSTM) neural network regression model is trained using over 80 years of daily data to forecast the discharge time series up to 3 days ahead of time. The performance of the model is found satisfactory through the comparison of the predicted data with the observed data, visualization of the distribution of the errors and Root Mean Squared Error (RMSE) value of 0.09. Higher performance is achieved through the increase in the number of epochs and hyper parameter tuning. This model can be transferred to other locations with proper feature engineering and optimization to perform univariate predictive analysis and potentially be used to perform real-time river discharge prediction.

Early detection of wildfires has been limited using the sun-synchronous orbit satellites due to their low temporal resolution and wildfires’ fast spread in the early stage. NOAA’s geostationary weather satellites GOES-R can acquire images every 15 minutes at 2km spatial resolution, and have been used for early fire detection. However, advanced processing algorithms are needed to provide timely and reliable detection of wildfires. In this research, a deep learning framework, based on Gated Recurrent Units (GRU), is proposed to detect wildfires at early stage using GOES-R dense time series data. GRU model maintains good performance on temporal modelling while keep a simple architecture, makes it suitable to efficiently process time-series data. 36 different wildfires in North and South America under the coverage of GOES-R satellites are selected to assess the effectiveness of the GRU method. The detection times based on GOES-R are compared with VIIRS active fire products at 375m resolution in NASA’s Fire Information for Resource Management System (FIRMS). The results show that GRU-based GOES-R detections of the wildfires are earlier than that of the VIIRS active fire products in most of the study areas. Also, results from proposed method offer more precise location on the active fire at early stage than GOES-R Active Fire Product in mid-latitude and low-latitude regions.

Prediction of protein-ligand binding affinities is crucial for computational drug discovery. A number of deep learning approaches have been developed in recent years to improve the accuracy of such affinity prediction. While the predicting power of these systems have advanced to some degrees depending on the dataset used for model training and testing, the effects of the quality and quantity of the underlying data have not been thoroughly examined. In this study, we employed erroneous datasets and data subsets of different sizes, created from one of the largest databases of experimental binding affinities, to train and evaluate a deep learning system based on convolutional neural networks. Our results show that data quality and quantity do have significant impacts on the prediction performance of trained models. Depending on the variations in data quality and quantity, the performance discrepancies could be comparable to or even larger than those observed among different deep learning approaches. In particular, the presence of proteins during model training leads to a dramatic increase in prediction accuracy. This implies that continued accumulation of high-quality affinity data, especially for new protein targets, is indispensable for improving deep learning models to better predict protein-ligand binding affinities.

Smart devices such as smartphones, smartwatches, etc. are promising platforms that are being used for automatic recognition of human activities. However, it is difficult to accurately monitor complex human activities due to inter-class pattern similarity, which occurs when different human activities exhibit similar signal patterns or characteristics. Current smartphone-based recognition systems depend on the traditional sensors such as accelerometer and gyroscope, which are inbuilt in these devices. Therefore, apart from using information from the traditional sensors, these systems lack contextual information to support automatic activity recognition. In this article, we explore environment contexts such as illumination(light conditions) and noise level to support sensory data obtained from the traditional sensors using a hybrid of Convolutional Neural Networks and Long Short Time Memory(CNN_LSTM) learning models. The models performed sensor fusion by augmenting the low-level sensor signals with rich contextual data to improve recognition and generalisation ability of the proposed solution. Two sets of experiments were performed to validate the proposed solution. The first set of experiments used inertial sensing data whilst the second set of extensive experiments combined inertial signals with contextual information from environment sensing data. Obtained results demonstrate that contextual information such as environment noise level and illumination using hybrid deep learning models achieved better recognition accuracy than the traditional activity recognition models without contextual information.

Computer network attacks are evolving in parallel with the evolution of hardware and neural network architecture. Despite major advancements in Network Intrusion Detection System (NIDS) technology, most implementations still depend on signature-based intrusion detection systems, which can’t identify unknown attacks. Deep learning can help NIDS to detect novel threats since it has a strong generalization ability. The deep neural network’s architecture has a significant impact on the model’s results. We propose a genetic algorithm based model to find the optimal number of hidden layers and the number of neurons in each layer of the deep neural network (DNN) architecture for the network intrusion detection binary classification problem. Experimental results demonstrate that the proposed DNN architecture shows better performance than classical machine learning algorithms at a lower computational cost.

Deep eutectic solvents (DESs) are one of the most rapidly evolving types of solvents, appearing in a broad range of applications such as nanotechnology, electrochemistry, biomass transformation, pharmaceuticals, membrane technology, biocomposite development, modern 3D-printing, and many others. The range of their applicability continues to expand, which demands the development of new DESs with improved properties. To do so requires an understanding of the fundamental relationship between the structure and properties of DESs. Computer simulation and machine learning techniques provide a fruitful approach as they can provide predictions, reveal physical mechanisms and readily be linked to experiments. This review is devoted to the computational research of DESs and describes technical features of DES simulations and the corresponding perspectives on various DES applications. The aim is to demonstrate the current frontiers of computational research of DESs and discuss future perspectives.

Nowadays, an analysis of virtual media to predict society’s reaction to any events or processes is a task of great relevance. Especially it concerns meaningful information on healthcare problems. Internet sources contain a large amount of pharmacologically meaningful information useful for pharmacovigilance purposes and repurposing drug use. An analysis of such a scale of information demands developing the methods that require the creation of a corpus with labeled relations among entities. Before, there have been no such Russian language datasets. This paper considers the first Russian language dataset where labeled entity pairs are divided into multiple contexts within a single text (by used drugs, by different users, by the cases of use, etc.), and a method based on the XLM-RoBERTa language model, previously trained on medical texts to evaluate the state-of-the-art accuracy for the task of indication of the four types of relationships among entities: ADR–Drugname, Drugname–Diseasename, Drugname–SourceInfoDrug, Diseasename–Indication. As shown based on the presented dataset from the Russian Drug Review Corpus, the developed method achieves the F1-score of 81.2% (obtained using cross-validation and averaged for the four types of relationships), which is 7.8% higher than the basic classifiers.

The ACC feature when activated augments the engine performance in real-time. This article presents a novel methodology to predict the optimal adaptive cruise control set speed profile (ACCSSP) by considering all the effecting parameters. This paper investigates engine operating conditions (EOC) criteria to develop a predictive model of ACCSSP in real-time. We developed a deep learning (DL) model using the NARX method to predict engine operating point (EOP) mapping the vehicle-level vectors (VLV). We used real-world field data obtained from Cadillac test vehicles driven by activating the ACC feature for developing the DL model. We used a realistic set of assumptions to estimate the VLV for the future time steps for the range of allowable speed values and applied them at the input of the developed DL model to generate multiple sets of EOP’s. We imposed the defined EOC criteria on these EOPs, and the top three modes of speeds satisfying all the requirements are derived for each second. Thus three eligible speed values are estimated for each second, and an additional criterion is defined to generate a unique ACCSSP for future time steps. Performance comparison between predicted and constant ACCSSPs indicates that the predictive model outperforms constant ACCSSP.

The ever-burgeoning growth of autonomous unmanned aerial vehicles (UAVs) has demonstrated a promising platform for utilization in real-world applications. In particular, UAV equipped with a vision system could be leveraged for surveillance applications. This paper proposes a learning-based UAV system for achieving autonomous surveillance, in which the UAV can be of assistance in autonomously detecting, tracking, and following a target object without human intervention. Specifically, we adopted the YOLOv4-Tiny algorithm for semantic object detection and then consolidated it with a 3D object pose estimation method and Kalman Filter to enhance the perception performance. In addition, a back-end UAV path planning for surveillance maneuver is integrated to complete the fully autonomous system. The perception module is assessed on a quadrotor UAV, while the whole system is validated through flight experiments. The experiment results verified the robustness, effectiveness, and reliability of the autonomous object tracking UAV system in performing surveillance tasks. The source code is released to the research community for future reference.

Skin disease is a very common disease for humans. In the medical industry detecting skin disease and recognizing its type is a very challenging task. Due to the complexity of human skin texture and the visual closeness effect of the diseases, sometimes it is really difficult to detect the exact type. Therefore, it is necessary to detect and recognize the skin disease at its very first observation. In today's era, artificial intelligence (AI) is rapidly growing in medical fields. Different machine learning (ML) and deep learning(DL) algorithms are used for diagnostic purposes. These methods drastically improve the diagnosis process and also speed up the process. In this paper, a brief comparison between the machine learning process and the deep learning process was discussed. In both processes, three different and popular algorithms are used. For the machine Learning process Bagged Tree Ensemble, K-Nearest Neighbor (KNN), and Support Vector Machine(SVM) algorithms were used. For the deep learning process three pre-trained deep neural network models

Maritime ship monitoring plays an important role in maritime transportation. Fast and accurate detection of maritime ship is the key to maritime ship monitoring. The main sources of marine ship images are optical images and synthetic aperture radar (SAR) images. Different from natural images, SAR images are independent to daylight and weather conditions. Traditional ship detection methods of SAR images mainly depend on the statistical distribution of sea clutter, which leads to poor robustness. As a deep learning detector, RetinaNet can break this obstacle, and the problem of imbalance on feature level and objective level can be further solved by combining with Libra R-CNN algorithm. In this paper, we modify the feature fusion part of Libra RetinaNet by adding a bottom-up path augmentation structure to better preserve the low-level feature information, and we expand the dataset through style transfer. We evaluate our method on the publicly available SAR dataset of ship detection with complex backgrounds. The experimental results show that the improved Libra RetinaNet can effectively detect multi-scale ships through expansion of the dataset, with an average accuracy of 97.38%.

Neural decoding from spiking activity is an essential tool for understanding the information encoded in population neurons, especially in applications like brain-computer interface (BCI). Various quantitative methods have been proposed and have shown superiorities under different scenarios respectively. From the machine learning perspective, the decoding task is to map the high-dimensional spatial & temporal neuronal activity to the low-dimensional physical quantities (e.g., velocity, position). Because of the complex interactions and the abundant dynamics among neural circuits, good decoding algorithms usually have the capability of capturing flexible spatiotemporal structures embedded in the input feature space. Recently, the Transformer-based models are widely used in processing natural languages and images due to its superior performances in handling long-range and global dependencies. Hence, in this work we examine the potential applications of Transformers in neural decoding and introduce two Transformer-based models. Besides adapting the Transformer to neuronal data, we also propose a data augmentation method for overcoming the data shortage issue. We test our models on three experimental datasets and their performances are comparable to the previous state-of-the-art (SOTA) RNN-based methods. In addition, Transformer-based models show increased decoding performances when the input sequences are longer, while LSTM-based models deteriorate quickly. Our research suggests that Transformer-based models are important additions to the existing neural decoding solutions, especially for large datasets with long temporal dependencies.

The Unmanned Aerial Vehicles (UAVs), used in civilian applications such as emergency medical deliveries, precision agriculture, wireless communication provisioning, etc., face the challenge of limited flight time due to their reliance on the on-board battery. Therefore, developing efficient mechanisms for in-situ power transfer to recharge UAV batteries hold potential in extending their mission time. In this paper, we study the use of far-field wireless power transfer (WPT) technique from specialized, transmitter UAVs (tUAVs) carrying Multiple Input Multiple Output (MIMO) antennas for transferring wireless power to receiver UAVs (rUAVs) in a mission. The tUAVs can fly and adjust their distance to the rUAVs to maximize energy transfer. The use of MIMO antennas further boost the energy reception by narrowing the energy beam toward the rUAVs. The complexity of their dynamic operating environment increases with the growing number of tUAVs, and rUAVs with varying levels of energy consumption and residual power. We propose an intelligent trajectory selection algorithm for the tUAVs based on a deep reinforcement learning model called Proximal Policy Optimization (PPO) to optimize the energy transfer gain. Simulation results demonstrate that with the use of PPO, the system achieves a tenfold flight time extension compared to no wireless recharging. Further, PPO outperforms the benchmark movement strategies of ’Traveling Salesman Problem’ and ’Low Battery First’ when used by the tUAVs.

Physical inactivity increases the risk of many adverse health conditions, including the world’s major non-communicable diseases, such as coronary heart disease, type 2 diabetes, and breast and colon cancers, shortening life expectancy. There are minimal medical care and personal trainers’ methods to monitor a patient’s actual physical activity types. To improve activity monitoring, we propose an artificial-intelligence-based approach to classify the physical movement activity patterns. In more detail, we employ two deep learning (DL) methods, namely a deep feed-forward neural network (DNN) and a deep recurrent neural network (RNN) for this purpose. We evaluate the proposed models on two physical movement datasets collected from several volunteers who carried tri-axial accelerometer sensors. The first dataset is from the UCI machine learning repository, which contains 14 different activities-of-daily-life (ADL) and is collected from 16 volunteers who carried a single wrist-worn tri-axial accelerometer. The second dataset includes ten other ADLs and is gathered from 8 volunteers who placed the sensors on their hips. Our experiment results show that the RNN model provides the accuracy performance compared to the state-of-the-art methods in classifying the fundamental movement patterns with an overall accuracy of 84.89% and an overall F1-score of 82.56%. Our results indicate that the proposed method will provide the medical doctors and trainers a promising way to precisely track and understand a patient’s physical activities for better treatment.

We have an unprecedented ability to map the Earth’s surface as deep learning technologies are applied to an abundance of high-frequency Earth observation data. Simple, free, and effective methods are needed to enable a variety of stakeholders to use these tools to improve scientific knowledge and decision making. Here we present a trained U-Net model that can map and delineate ground mounted solar arrays using publicly available Sentinel-2 imagery, and that requires minimal data pre-processing and no feature engineering. By using label overloading and image augmentation during training, the model is robust to temporal and spatial variation in imagery. The trained model achieved a precision and recall of 91.5% each and an intersection over union of 84.3% on independent validation data from two distinct geographies. This generalizability in space and time makes the model useful for repeatedly mapping solar arrays. We use this model to delineate all ground mounted solar arrays in North Carolina and the Chesapeake Bay watershed to illustrate how these methods can be used to quickly and easily produce accurate maps of solar infrastructure.

Hypergravity accelerators are a type of large machinery used for gravity training or medical research. A failure of such large equipment can be a serious problem in terms of safety or costs. This paper proposes a prediction model that can proactively prevent failures that may occur in a hy-pergravity accelerator. The method proposed in this paper was to convert vibration signals to spectograms and perform classification training using a deep learning model. An experiment was conducted to evaluate the performance of the method proposed in this paper. A 4-channel accel-erometer was attached to the bearing housing, which is a rotor, and time-amplitude data were obtained from the measured values by sampling. The data were converted to a two-dimensional spectrogram, and classification training was performed using a deep learning model for four conditions of the equipment: Unbalance, Misalignment, Shaft Rubbing, and Normal. The ex-perimental results showed that the proposed method had a 99.5% F1-Score, which was up to 23% higher than the 76.25% for existing feature-based learning models.

This paper provides the state of the art of data science in economics. Through a novel taxonomy of applications and methods advances in data science are investigated. The data science advances are investigated in three individual classes of deep learning models, ensemble models, and hybrid models. Application domains include stock market, marketing, E-commerce, corporate banking, and cryptocurrency. Prisma method, a systematic literature review methodology is used to ensure the quality of the survey. The findings revealed that the trends are on advancement of hybrid models as more than 51% of the reviewed articles applied hybrid model. On the other hand, it is found that based on the RMSE accuracy metric, hybrid models had higher prediction accuracy than other algorithms. While it is expected the trends go toward the advancements of deep learning models.

The new barrier mode in Apache Spark allows embedding distributed deep learning training as a Spark stage to simplify the distributed training workflow. In Spark, a task in a stage doesn’t depend on any other tasks in the same stage, and hence it can be scheduled independently. However, several algorithms require more sophisticated inter-task communications, similar to the MPI paradigm. By combining distributed message passing (using asynchronous network IO), OpenJDK’s new auto-vectorization and Spark’s barrier execution mode, we can add non-map/reduce based algorithms, such as Cannon’s distributed matrix multiplication to Spark. We document an efficient distributed matrix multiplication using Cannon’s algorithm, which improves significantly on the performance of the existing MLlib implementation. Used within a barrier task, the algorithm described herein results in an up to 24% performance increase on a 10,000x10,000 square matrix with a significantly lower memory footprint. Applications of efficient matrix multiplication include, among others, accelerating the training and implementation of deep convolutional neural network based workloads, and thus such efficient algorithms can play a ground-breaking role in faster, more efficient execution of even the most complicated machine learning tasks

Corona Virus Infectious Disease (COVID-19) is newly emerging infectious disease. This disease is known to the globe in early 2019. Poor status of mental health is often caused by unemployment, ongoing socio-economic condition. Poor mental health may even accelerate the process of panic attack. It has been happening rapidly during COVID-19. It has a great effect on human health. This paper utilizes multiple related factors those have impact on causing panic attack. Recurrent Neural Network (RNN) based framework is utilized in this paper that assembles multiple RNN layers along with other parameters into a single platform. This method is implemented by capturing interfering factors and predicts panic attack tendency of people during COVID-19. Early prediction of panic attacks may assist in saving life from unwanted circumstances.

Big Data in medicine includes possibly fast processing of large data sets, both current and historical in purpose supporting the diagnosis and therapy of patients' diseases. Support systems for these activities may include pre-programmed rules based on data obtained from the interview medical and automatic analysis of test results diagnostic results will lead to classification of observations to a specific disease entity. The current revolution using Big Data significantly expands the role of computer science in achieving these goals, which is why we propose a Big Data computer data processing system using artificial intelligence to analyze and process medical images.

Photogrammetry involves aerial photography of the earth’s surface and subsequently processing the images to provide a more accurate depiction of the area (Orthophotography). It’s used by the Spanish Instituto Geográfico Nacional to update road cartography but requires a significant amount of manual labor due to the need to perform visual inspection of all tiled images. Deep Learning techniques (artificial neural networks with more than one hidden layer) can perform road detection but it is still unclear how to find the optimal network architecture. Our system applies grammar guided genetic programming to the search of deep neural network architectures. In this kind of evolutive algorithm all the population individuals (here candidate network architectures) are constrained to rules specified by a grammar that defines valid and useful structural patterns to guide the search process. Grammar used includes well-known complex structures (e.g. Inception-like modules) combined with a custom designed mutation operator (dynamically links the mutation probability to structural diversity). Pilot results show that the system is able to design models for road detection that obtain test accuracies similar to that reached by state of the art models when evaluated over a dataset from the Spanish National Aerial Orthophotography Plan.

Owing to the reduction of rupture instability and the avoidance of wrinkle defect, hydrodynamic deep drawing (HDD) process is gradually becoming attractive for fabricating lightweight and complicated products. Meanwhile, since metallic material presents anisotropic deformation behavior, it is necessary to select an appropriate constitutive model for the prediction of plastic deformation behavior of applied material with high precision. In the present research, several anisotropic yield criteria namely, Hill’48, Yld2000-2d and BBC2005 are implemented to investigate the effect of yield functions on the prediction accuracy of the critical process window and deformation behavior for HDD process of 2024 and 5754 aluminum alloys. Material constants in the yield criteria are determined by applying uniaxial and equi-biaxial tension tests and optimizing an error-function by using the Levenberg-Marquardt algorithm. Furthermore, the process window diagram is computed utilizing the stress analytical model combined material properties with workpiece geometrical features. Numerical simulation results of predicted material anisotropic parameters, process window and HDD deformation for aluminum alloys are compared with the experimental data. Through the comparison of diverse yield criteria based on materials anisotropic coefficients, critical process window prediction, earing profile, and thickness distribution, it is revealed that the Yld2000-2d and the BBC2005 yield criteria can offer more precise models of material behavior in planar anisotropy properties for HDD process of 2024 and 5754 aluminum alloys.

Deep learning has achieved lots of successes in many fields, but when trainable sample are extremely limited, deep learning often under or overfitting to few samples. Meta-learning was proposed to solve difficulties in few-shot learning and fast adaptive areas. Meta-learner learns to remember some common knowledge by training on large scale tasks sampled from a certain data distribution to equip generalization when facing unseen new tasks. Due to the limitation of samples, most approaches only use shallow neural network to avoid overfitting and reduce the difficulty of training process, that causes the waste of many extra information when adapting to unseen tasks. Euclidean space-based gradient descent also make meta-learner's update inaccurate. These issues cause many meta-learning model hard to extract feature from samples and update network parameters. In this paper, we propose a novel method by using multi-stage joint training approach to post the bottleneck during adapting process. To accelerate adapt procedure, we also constraint network to Stiefel manifold, thus meta-learner could perform more stable gradient descent in limited steps. Experiment on mini-ImageNet shows that our method reaches better accuracy under 5-way 1-shot and 5-way 5-shot conditions.

Electromyographic (EMG) signals provide information about a person's muscle activity. For hand movements, in particular, the execution of each gesture involves the activation of different combinations of the forearm muscles, which generate distinct electrical patterns. Conversely, the analysis of these muscle activation patterns, represented by EMG signals, allows recognizing which gesture is being performed. In this study, we aimed to implement an automatic identification system of hand or wrist gestures based on supervised Machine Learning (ML) techniques. We trained different computational models and determined which of these showed the best capacity to identify six hand or wrist gestures and generalize between different subjects. We used an open access database containing recordings of EMG signals from 36 subjects. Among the results obtained, we highlight the performance of the Random Forest model, with an accuracy of 95.39%, and the performance of a convolutional neural network with an accuracy of 94.77%.

In this study, by using Least Square Method, the dataset for the Gross Domestic Product per capita is modeled as a function satisfying the fractional differential equation. The function itself is assumed to be the finite summation of its previous values and the derivatives with unknown coefficients. Then, the prediction for the upcoming years is done by having an approach dividing the dataset into 4 regions corresponding to four different tasks. The mathematical model of the Gross Domestic Product (GDP) per capita of the countries (and union) which are Brazil, China, European Union (EU), India, Italy, Japan, UK, the USA, Spain, and Turkey is constructed with a new methodology called as the deep assessment method which comes from the expressing an arbitrary function modeling the dataset as the finite summation of its previous values and the derivatives with unknown coefficient. The method uses the fractional calculus properties combining with Least Square Method and is compared to Long short-term memory (LSTM) algorithm which is a special type of neural network used for time sequences in general.

Several researchers have contemplated deep learning-based post-filters to increase the quality of statistical parametric speech synthesis, which perform a mapping of the synthetic speech to the natural speech, considering the different parameters separately and trying to reduce the gap between them. The Long Short-term Memory (LSTM) Neural Networks have been applied successfully in this purpose, but there are still many aspects to improve in the results and in the process itself. In this paper, we introduce a new pre-training approach for the LSTM, with the objective of enhancing the quality of the synthesized speech, particularly in the spectrum, in a more efficient manner. Our approach begins with an auto-associative training of one LSTM network, which is used as an initialization for the post-filters. We show the advantages of this initialization for the enhancing of the Mel-Frequency Cepstral parameters of synthetic speech. Results show that the initialization succeeds in achieving better results in enhancing the statistical parametric speech spectrum in most cases when compared to the common random initialization approach of the networks.

Automated skin lesion analysis is one of the trending fields that has gained attention among the dermatologists and healthcare practitioners. Skin lesion restoration is an essential preprocessing step for lesion enhancements for accurate automated analysis and diagnosis. Digital hair removal is a non-invasive method for image enhancement by solving the hair-occlusion artefact in previously captured images. Several hair removal methods were proposed for skin delineation and removal. However, manual annotation is one of the main challenges that hinder the validation of these proposed methods on a large number of images or using benchmarking datasets for comparison purposes. In the presented work, we propose a realistic hair simulator based on context-aware image synthesis using image-to-image translation techniques via conditional adversarial generative networks for generation of different hair occlusions in skin images, along with the ground-truth mask for hair location. Besides, we explored using three loss functions including L1-norm, L2-norm and structural similarity index (SSIM) to maximise the synthesis quality. For quantitatively evaluate the realism of image synthesis, the t-SNE feature mapping and Bland-Altman test are employed as objective metrics. Experimental results show the superior performance of our proposed method compared to previous methods for hair synthesis with plausible colours and preserving the integrity of the lesion texture.

Background: Post-traumatic stress disorder (PTSD) is a common debilitating psychiatric condition for which pharmacological therapy is not always solvable. Various treatments have been suggested for these patients. Deep brain stimulation (DBS) is currently under investigation for patients affected by PTSD. 2) Methods: We review the neurocircuitry and up to date clinical concepts that may be of relevance for the implementation of DBS in posttraumatic stress disorder (PTSD). 3) Results: The role of DBS in treatment-refractory PTSD patients has been investigated relying on both preclinical and clinical studies. 4) Conclusions: DBS for PTSD is in its preliminary phases and likely to provide hope to patients with medical refractory PTSD following the results of randomized controlled studies.

With recent focus on deep neural network architectures for development of algorithms for computer-aided diagnosis (CAD), we provide a review of studies within the last 3 years (2015-2017) reported in selected top journals and conferences. 29 studies that met our inclusion criteria were reviewed to identify trends in this field and to inform future development. Studies have focused mostly on cancer-related diseases within internal medicine while diseases within gender-/age-focused fields like gynaecology/pediatrics have not received much focus. All reviewed studies employed image datasets, mostly sourced from publicly available databases (55.2%) and few based on data from human subjects (31%) and non-medical datasets (13.8%), while CNN architecture was employed in most (70%) of the studies. Confirmation of the effect of data manipulation on quality of output and adoption of multi-class rather than binary classification also require more focus. Future studies should leverage collaborations with medical experts to aid future with actual clinical testing with reporting based on some generally applicable index to enable comparison. Our next steps on plans for CAD development for osteoarthritis (OA), with plans to consider multi-class classification and comparison across deep learning approaches and unsupervised architectures were also highlighted.

: Cybersecurity has become a critical area in the digital field in recent years. The expansion of networks has revolutionised the way network structures are organised and managed. However, with increased connectivity and the growing complexity of modern networks, the threat of cyber-attacks has become more intense. As technology continues to advance, it brings both opportunities and challenges. One of the major challenges is the need to secure networks and sensitive data from various malicious activities. Traditional networks have evolved to include Software Defined Network (SDN), which offers a more flexible and programmable framework. Researchers should focus on detecting attacks in SDN because SDN networks are becoming more popular and attractive targets for clients due to their programmability and dynamic nature. The centralised controller, known as the backbone of an SDN, becomes a single point of failure and a potential target for attackers if not properly secured. Researchers need to emphasise the detection of attacks in SDN in order to mitigate these risks. By understanding the potential vulnerabilities and attack methods specific to SDN, researchers can develop effective detection approaches and propose countermeasures. This methodology helps protect the network from potential threats and minimises the impact of successful attacks. Therefore, this survey paper provides a review of Intrusion Detection Systems (IDSs) in Software-Defined Networks (SDN) to provide a thorough understanding of SDN security issues.

This paper presents an approach to the two-dimensional analysis of elastic isotropic deep beams using the finite difference method (FDM). Deep beams are subjected to in-plane loading and present a shear span to height ratio of less than 2.50; consequently, Euler-Bernoulli beam theory and Timoshenko beam theory do not apply. Deep beams analysis is generally conducted using numerical methods such as the finite element method and to a lesser extent the FDM; the strut-and-tie model and the stress field method are also widely utilized. Analytical approaches usually make use of the Airy stress function, where stresses are formulated in terms of the stress function; however, the exact solution of this function satisfying all of the boundary conditions can hardly be found, even for simple cases. In this paper, deep beams were analyzed using the FDM. The FDM is an approximate method for solving problems described with differential equations. The FDM does not involve solving differential equations; equations are formulated with values at selected nodes of the structure. Therefore, the deep beam was discretized with a two-dimensional grid, and additional nodes were introduced at the boundaries and at positions of discontinuity (openings, brutal change of material properties, non-uniform grid spacing), the number of additional nodes corresponding to the number of boundary conditions at the node of interest. The introduction of additional nodes allowed us to apply the governing equations at boundary nodes and satisfy the boundary and continuity conditions. An Airy stress function approach and a displacement potential function approach were considered in this study whereby strong formulations of equations (equilibrium, kinematic, and constitutive) were set. Stress and stability analyses were carried out with this model; furthermore, deep beams of varying stiffness, layered beams, and beams having openings were analyzed. For slender beams, the results obtained with the Airy stress function approach showed good agreement with those of the Euler-Bernoulli beam theory, and for deep beams, the shapes of stress distributions were in good agreement with a proper understanding of the behavior of structures. On the other hand, the displacement potential function approach delivered unsatisfactory results, probably due to the use of an inefficient equation solver; a more powerful tool will be needed in future research for this purpose.

Biodiversity, as an important component of ecosystems, faces unprecedented decline due to factors such as habitat loss. Many ecosystem services depend on biodiversity and influence crop yield and quality. In viticulture, the importance of groundcover management, as cover crops or spontaneous grassing, is particularly important for the conservation of biodiversity and consequently, the preservation of ecosystem services. The integration of remote sensing technologies and artificial intelligence in this context remains limited, especially concerning groundcover classification in viticulture. In response to this challenge, our research focuses on developing a methodology for the identification and classification of groundcovers within vineyards inter-row. Employing an Unmanned Aerial Vehicle, equipped with RGB cameras, and advanced deep learning models, we categorize groundcovers into nine classes of which seven distinct cover crop communities groups. Our study aims to establish a method for identifying vineyard inter-row groundcover, with the ultimate goal of creating a comprehensive groundcover map. The results demonstrate that using the UNet model with backbone architectures significantly enhances classification performance. Specifically, the model with an EfficientNetB0 backbone achieved an accuracy of 85.4%, a mean IoU of 59.8%, and a Jaccard score of 73.0%. This study validates the capabilities remote sensing and deep learning technologies in supporting the monitoring of biodiversity in vineyards, laying the groundwork for future researches to allow the model to monitor larger surface with time and cost advantages and expanding the dataset to encompass a broader range of vineyard types, soil conditions, and geographic.

There is considerable interest in cellulose solubilisation in deep eutectic solvents (DESs) as a new emerging class of green solvents. However, the experimental data have indicated that cellulose is almost insoluble in most of DESs. Carboxylic acid-based DESs are a category of DESs that exhibit a modest ability to solubilise cellulose. However, the molecular-level reasons why carboxylic acid-based DESs can dissolve cellulose are unknown. Such information is crucial in developing DESs with enhanced cellulose solubilising ability. Here, we use molecular dynamics simulations to explore the molecular mechanism of cellulose solubility in oxaline (choline chloride/oxalic acid) as a typical carboxylic acid-based DES and cellulose insolubility in reline (choline chloride/urea) as the most famous DES. It was revealed that oxalic acid possesses a greater ability to weaken the interaction between the chloride anion and the choline cation when compared to other urea in reline. Self-interactions of hydrogen bond donors are also highly attenuated in oxaline than in reline. These results explain the observed more significant melting point depression of oxaline in comparison to reline. We depict a molecular mechanism indicating how oxalic acid synergically works with chloride ion to dissolve cellulose and why urea cannot play such a role in reline.

The advent of the era of big data has led to the emergence of heterogeneous network data fusion as a prominent area of research. Heterogeneous network data is characterised by multi-modality, multi-source, and high dimensionality, which presents significant challenges for traditional data fusion methods. These methods often encounter difficulties in processing such data, including issues such as information redundancy, data inconsistency, and high computational complexity. This paper proposes a heterogeneous network data fusion model based on a deep neural network. The model employs the Multi-Layer Perceptron (MLP) as its fundamental framework, utilising the deep neural network to facilitate joint feature representation learning on data from disparate modalities. The Adaptive Feature Reconstruction Module enables the model to learn the interrelationships between different modalities and to balance the importance of different modal features in the fusion process in a dynamic manner. Furthermore, we introduce an innovative cross-modal attention mechanism, which is capable of effectively capturing the coupling relationship between deep features in heterogeneous data, thereby enhancing the expressiveness and data fusion efficacy of the model. The experimental results demonstrate that the proposed model markedly enhances the accuracy of classification and regression tasks in comparison to traditional methodologies.

Bosom disease remains a main source of death among women around the world. Early identification is significant for further developing endurance rates, accordingly making effective indicative strategies fundamental. This undertaking the malignant growth discovery utilizing mammography images (low dose x-ray), utilizing profound learning methods to robotize conclusions and upgrade exactness. The framework's center includes preprocessing the mammogram images, separating highlights, and arranging them utilizing Convolutional neural networks (CNNs) and other high-level profound learning models.

The agriculture industry is critical to the global economy, with product quality having a direct impact on marketability and waste management. Apples, one of the most extensively produced fruits, are affected by a variety of diseases that can reduce productivity and quality. Accurate diagnosis of diseases is critical, but traditional manual approaches are time-consuming, error-prone, and ineffective. Inadequate labeled data and a wide range of disease symptoms make it necessary to design an automated, robust, and accurate system. This article describes a hybrid model for Apple Fruit Disease Detection (HMAFDD) that combines the strengths of three pre-trained convolutional neural network (CNN) models: ResNet50, DenseNet121, and EfficientNetB0. It accomplish this by using multi-architecture feature extraction. The hybrid system, which combines both models, is able to recognize a wide range of features, from simple textures to complex patterns unique to a certain diseases. Grad-CAM, or gradient-weighted class activation mapping, creates heatmaps that highlight significant regions for prediction, which enhances the interpretability of the model.To increase robustness and accuracy, techniques such as spectral-shifted adversarial perturbation for data augmentation and spectrally-weighted global average pooling for feature aggregation are used. This technique provides 99.75% accuracy with minimal processing needs, making it acceptable for real-time applications in agricultural situations. This considerably improves apple disease management.

This study discusses the application of reinforcement learning (RL) in financial asset risk assessment, especially how to improve the ability of risk prediction and asset portfolio optimization through deep reinforcement learning (DRL) model. Traditional financial risk assessment methods are often unable to effectively cope with the dynamic changes and complexities of the financial market, but RL technology provides innovative solutions for modern financial risk management by means of real-time adjustment and adaptation mechanism. This paper experimentally verifies the advantages of DRL models in improving forecasting accuracy and risk management efficiency and discusses the potential impact of AI technology in the financial sector. The findings show that despite challenges such as data quality and model interpretation, the application of AI technology provides financial institutions with more precise and flexible risk management tools, driving further development of fintech.

The Association of Mutual Funds of India (AMFI) under the direction of the Securities and Exchange Board of India (SEBI) provided open access to various risk parameters with respect to MidCap and SmallCap funds on its website in February, 2024. This disclosure of month wise cross-sectional data depicting stress-test and liquidity analysis was hailed as a milestone as it provided a glimpse into inherent risk of mutual funds. The AMFI dataset consisted of 14 variables including stress test results, ownership concentration on both liability & asset sides, standard deviation on portfolio’s returns, portfolio beta, trailing 12-month PE ratio, portfolio turnover etc. Among these, the primary variable identified in grading mutual fund is the stress test parameter, expressed as number of days required to liquidate 50% and 25% of the portfolio respectively on a pro-rata basis under stress condition. A higher number of days to liquidate the portfolio under stress conditions represented higher risk for the investor. The objective of our paper is to test various neural network models which can help in predicting stress levels with the highest accuracy and specificity in MidCap and SmallCap mutual funds based on AMFI’s 14 parameters as predictors.Our study utilizes AMFI datasets available for MidCap and SmallCap mutual funds from February 2024 (inception period) to May 2024. To operationalize the analysis, the derived response variable was binned to represent high and low risk based on the time required for pro-rata liquidation of the portfolio. The study employed feed-forward neural networks, with the logistic regression as the step function, to model the relationships. Results suggest that the simpler neural network models with one hidden layer and two to three nodes show higher accuracy and specificity. Analysis also shows that more complex architectures do not always guarantee better performance as these models might suffer from higher bias. By leveraging advanced computational techniques, this study contributes to the ongoing discourse surrounding risk management and decision-making in the realm of mutual fund investments.

The prediction of stock prices is a challenging task, particularly for retail investors who may lack the resources and expertise to perform sophisticated quantitative trading. This study focuses on enhancing stock price prediction for retail investors by employing advanced machine learning techniques on data from the stock exchange market. We utilize a comprehensive methodology that includes data preprocessing to handle missing values and outliers, feature engineering, cross-validation, and parameter tuning. The techniques applied include Keras Deep Neural Networks (DNN), LightGBM, LSTM, GRU, and linear regression (LR). Our proposed ensemble model, which combines time series and deep learning models, demonstrates superior performance compared to individual models. This integration of methods leads to significant improvements in prediction accuracy, providing a robust solution for retail investors.

Accurate prediction of investment returns is crucial for financial decision-making. Traditional models often struggle with the complexity and variability inherent in financial data. This article presents an advanced predictive model integrating three deep neural networks, pretraining, and ensemble learning to enhance prediction accuracy. By leveraging a combination of DNNs for handling different complexities of data, and a CNN for efficient feature extraction, our approach significantly outperforms traditional methods. Our model incorporates sophisticated data handling techniques, including extensive preprocessing and data augmentation, and benefits from pretraining and ensemble methods to improve robustness and performance. This comprehensive framework offers a robust solution for investment return prediction, addressing the shortcomings of existing models and providing enhanced accuracy and reliability.

This study aims to enhance the detection and assessment of safety hazards in small-scale reservoir engineering using advanced image processing and deep learning techniques. Given the critical importance of small reservoirs in flood management, water supply, and ecological balance, the effective monitoring of their structural integrity is crucial. This paper developed a fully convolutional semantic segmentation method for hidden danger images of small reservoirs using an encoding-decoding structure, utilizing a deep learning framework of convolutional neural networks (CNNs) to process and analyze high-resolution images captured by UAVs. The method incorporated data augmentation and adaptive learning techniques to improve model accuracy under diverse environmental conditions. Finally, the quantification data of hidden dangers (length, width, area, etc.) was obtained by converting the image pixels to the actual size. Results demonstrate significant improvements in detecting structural deficiencies, such as cracks and seepage areas, with increased precision and recall rates compared to conventional methods, and the HHSN-25 network (Hidden Hazard Segmentation Network with 25 layers)proposed in this paper outperforms other methods. The main evaluation indicator mIoU of HHSN-25 is higher than other methods, reaching 87.00%, and Unet is 85.50%, Unet++is 85.55%. The proposed model achieves reliable real-time performance, allowing for early warning and effective management of potential risks. This study contributes to the development of more efficient monitoring systems for small-scale reservoirs, enhancing their safety and operational sustainability.

In the context of smart agriculture, accurately estimating plant leaf chemical parameters is crucial for optimizing crop management and improving agricultural yield. Hyperspectral imaging, with its ability to capture detailed spectral information across various wavelengths, has emerged as a powerful tool in this regard. However, the complex and high-dimensional nature of hyperspectral data poses significant challenges in extracting meaningful features for precise estimation. To address this challenge, this study proposes an end-to-end estimation network for multiple chemical parameters of Astragalus leaves based on attention mechanism (AM) and multivariate hyperspectral features (AM-MHENet). We leveraging HybridSN and multilayer perception (MLP) to extract prominent features from the hyperspectral data of Astragalus membranaceus var. mongholicus (AMM) leaves and stems, as well as the surface and deep soil surrounding AMM roots. This methodology allows us to capture the most significant characteristics present in these hyperspectral data with high precision. The AM is subsequently used to assign weights and integrate the hyperspectral features extracted from different parts of the AMM. The MLP is then employed to simultaneously estimate the chlorophyll content (CC) and nitrogen content (NC) of AMM leaves. Compared with estimation networks that utilize only hyperspectral data from AMM leaves as input, our proposed end-to-end AM-MHENet demonstrates superior estimation performance. Specifically, AM-MHENet achieves an R2 of 0.983 with an RMSE of 0.73 for the estimation of CC in AMM leaves. For NC estimation, AM-MHENet achieves an R2 value of 0.977 with an RMSE of 0.27. These results underscore AM-MHENet’s effectiveness in significantly enhancing the accuracy of both CC and NC estimation in AMM leaves. Moreover, these findings indirectly suggest a strong correlation between the development of AMM leaves and stems, as well as the surface and deep soil surrounding the roots of AMM, and directly highlight the ability of AM to effectively focus on the relevant spectral features within the hyperspectral data. This findings from this study could offer valuable insights into the simultaneous estimation of multiple chemical parameters in plants, thereby making a contribution to the existing body of research in this field.

Attention mechanisms have become a fundamental component of deep learning, including the field of computer vision. The key idea behind attention in computer vision is to help the model focus on the relevant spatial regions of the input image, rather than treating all regions equally. The traditional approaches to attention mechanisms in computer vision often suffer from distribution inconsistencies in the attention maps, resulting in sharp transitions that negatively affect model’s focus and lead to poor generalization on complex shapes. The problem of spatial incoherence is particularly pronounced in the task of semantic segmentation, where accurate pixel-level predictions require a detailed understanding of the spatial relationships within the image. In this paper, we propose an attention mechanism called Smooth Attention designed for convolutional neural networks to address the problem of spatial inconsistency in attention maps through multidimensional spatial smoothing. We conduct a series of experiments to evaluate the effectiveness of the proposed mechanism and demonstrate its superior performance compared to traditional methods.

Results: This case series study included 20 PD participants with STN-DBS, with a mean age of 61.2±10.1 years. There were no differences in equilibrium scores of SOT conditions between 60, 130, and 180 Hz frequencies (p> 0.05), except the SOT6 score (p= 0.003), where 60 Hz showed better equilibrium performance in SOT6, indicating an advantage in postural control when visual and somatosensory cues are disturbed, and vestibular cues are not.

In the recent years, with the advent of Artificial Intelligence (AI) traditional methods have seen a significant transformation in the agriculture sector, especially in soil management. Soil management involves practices that maintains and improves the physical, chemical as well as the biological properties of the soil. Soil health management is essential for both the environmental conservation and sustainable agriculture production, ensuring the soil productivity and functional aspects associated with the ecosystem. Soil health management is one of the most important aspects of agriculture and food production, hence preserving and enhancing the soil health is an essential factor for supporting agriculture. Integration of Artificial Intelligence (AI) technologies with soil health management offers the potential to enhance agricultural sustainability, productivity, adapt to the climatic changes and resource constraints. The study of AI tools that can help improve soil health management by providing more accurate and efficient monitoring, analysis and decision-making capabilities. This paper studies the potential AI technologies including machine learning, robotics, and remote sensing in enhancing soil health, raising crop yields, and lowering environmental concerns by examining previous research and case studies.

Recently, deep neural networks (DNNs) have been widely used in various fields such as autonomous vehicles and smart homes. Since these DNNs can be directly implemented on edge devices, they offer advantages such as real-time processing in low-power and low-bandwidth environments. However, deployment of DNNs in embedded systems, including edge devices, exposes them to threats such as fault injection attacks. In this paper, we introduce two methods to induce misclassifications by using clock and voltage glitch-based fault attacks in devices where DNN models are executed. As a result of experiments on a microcontroller implemented with a DNN for image classification, we show that glitch injection attacks can lead with high probability to serious misclassifications. Furthermore, we propose a countermeasure to glitch attacks on the softmax function, and confirm that this method is effective in preventing misclassifications.

Fire detection is crucial for ensuring human safety and minimizing property damage. By utilizing advanced technologies, we can identify fires early, before they escalate. Autonomous fire detection systems are particularly vital in high-risk areas with minimal human presence. To address this need, we propose an automated fire-detection model based on YOLOv8. We modified the architecture of YOLOv8 and achieved impressive performances. Through comprehensive analysis of different YOLO models and their predecessors, we identified and addressed substantial gaps in this field. Also, by augmenting the framework with additional capabilities, we surpassed previous models, enabling real-time fire detection. Our proposed model achieves impressive performances, with 98% accuracy in fire detection and 97.8% in smoke detection. Moreover, recall rates for fire and smoke are 97.1% and 97.4%, respectively, with a mean average precision (mAP) accuracy of 99.1% for both cases. Finally, we applied an explainable artificial intelligence technique to interpret our proposed model and its results. This study lays the foundation for future research aimed at enhancing fire detection efficiency. The reliability of our modified model presents promising opportunities for future advancements, ensuring increased effectiveness and precision in efficient fire detection.

This study introduces an advanced method for climate downscaling using a Super-Resolution Generative Adversarial Network (SRGAN), referred to as Elevation-integrated TEmperature and Precipitation SRGAN downscaling (E-TEPS), to enhance the spatial resolution of climate data, specifically 2m-temperature above the surface and total precipitation, from the high-resolution Euro-Mediterranean Center on Climate Change (CMCC) dataset over Italy. Traditional Numerical weather prediction (NWP) as well as global climate models (GCMs) generally lack the resolution required for local applications, and this research addresses that gap by refining GCM outputs to provide detailed local climate insights. E-TEPS incorporates elevation maps as auxiliary inputs, significantly improving the accuracy of downscaled temperature and precipitation data. Utilizing Google Cloud's computational resources, the model processes large climate datasets and runs complex training operations efficiently. The system is exceptionally fast, delivering all results in under 10 seconds. The model demonstrates superior performance over two interpolation methods (bicubic and linear), used as traditional downscaling methods, achieving lower Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE), higher Pearson Correlation for temperature, and higher agreement values for the precipitation. Visual comparisons confirm that E-TEPS generated high-resolution outputs exhibit fewer artifacts and better detail preservation, particularly in regions with complex terrain and variable climatic conditions. These findings underscore the potential of our model in delivering precise, high-resolution climate predictions, thereby enhancing the performance of early-warning systems and supporting more effective climate-related decision-making. This paper is the first part of a two-part series; the second part will demonstrate the integration of this downscaling method with the FourCastNet global forecasting model to enhance climate predictions at a regional level, particularly focusing on high-resolution forecasts for Italy.

The next generation network requires not only ultra high data rate, global coverage and connectivity, but also reducing network deployment costs and energy consumption. The emergence of reconfigurable intelligent surface (RIS) technology provides an effective way to improve efficiency and reduce cost, while the passive elements bring new challenges of channel estimation (CE) and beam tracking. For a RIS-aided multiple-input and single-output (MISO) system, in this paper, to achieve the channel state information (CSI), we propose a principle component analysis (PCA) based staged channel estimation method. Based on the estimated channel, we propose a deep learning (DL) based beam tracking scheme to realize low-complexity RIS reflection coefficient design, which effectively improves the signal-to-noise ratio (SNR) at the user side. Simulation results verify our proposed channel estimation scheme based on PCA and beam tracking scheme based on DNN for the semi-active RIS-aided MISO systems can obtain approximate performances with much lower computational complexity.

The present paper explored the possibility of discriminating between different malignant degrees of Chondrosarcoma (CS), a diffused bone tumor, and Enchondroma (EC), its benign version, through an approach based on the coupling between Confocal Raman Microscopy (CRM) and Machine Learning (ML) techniques. Recently, Raman Spectroscopy has proven to be a powerful tool for a complete grading of CS by distinguishing between grade 1, 2 and 3, and to distinguish CS from EC. In this paper, we tested some models, belonging either to ML or to the sub-type called Deep Learning (DL), showing excellent classification performances, especially the DL algorithms, with classification accuracy approaching to 100%, making the models promising for future implementations of Raman spectroscopy when applied to oncology diagnosis. In turn, we highlighted how some proper ML models result in worse classification performances but better resolution of specific chemical target compounds, possible candidate to become malignant markers, with relevant implication for a correct diagnosis.

In recent years, artificial intelligence technology has seen increasingly widespread application in the field of intelligent manufacturing, particularly with deep learning offering novel methods for recognizing geometric shapes with specific features. In traditional CNC machining, computer-aided manufacturing (CAM) typically generates G-code for specific machine tools based on existing models. However, the tool paths for most CNC machines consist of a series of collinear motion commands (G01), often resulting in discontinuities in the curvature of adjacent tool paths, leading to machining defects. To address these issues, this paper proposes a method for CNC system machining trajectory feature recognition and path optimization based on intelligent agents. This method employs intelligent agents to construct models, analyze the key geometric information in the G-code generated during CNC machining, and uses the MCRL deep learning model incorporating linear attention mechanisms and multiple neural networks for recognition and classification. Path optimization is then carried out using mean filtering, Bezier curve fitting, and an improved novel adaptive coati optimization algorithm (NACOA) according to the degree of unsmoothness of the path. Taking the optimization of a gear model’s process file as an example, the effectiveness of the proposed method is verified. The research results indicate that the CNC system machining trajectory feature recognition and path optimization method based on intelligent agents can significantly enhance the smoothness of CNC machining paths, reduce machining defects, and has substantial application value.

The popularity of handheld devices for point-of-care ultrasound (POCUS) has increased in recent years due to their portability and cost-effectiveness. However, POCUS has the drawback of lower imaging quality compared to conventional ultrasound, because of hardware limitations. Improving the quality of POCUS through post-image processing would therefore be beneficial, with deep learning approaches showing promise in this regard. This review investigates the state-of-the-art progress of image enhancement using deep learning suitable for POCUS applications. A systematic search was conducted from January 2024 to February 2024 on PubMed and Scopus. From the 457 articles that were found, the full text was retrieved for 69 articles. From this selection, 15 articles were identified addressing multiple quality enhancement aspects. A disparity in the baseline performance of the low-quality input images was seen across these studies, ranging between 8.65–29.24 dB for the Peak Signal-to-Noise Ratio (PSNR) and 0.03-0.71 for the Structural Similarity Index Measure (SSIM). In six studies, where both PSNR and the SSIM metrics were reported for the baseline and the generated images a mean difference of 6.60 (SD ± 2.99) and 0.28 (SD ± 0.15) was observed for the PSNR and SSIM, respectively. The reported performances demonstrate the potential of deep-learning-based image enhancement for POCUS. However, variability in the extent of performance gain across datasets and articles was notable, and the heterogeneity across articles makes quantifying the exact improvements challenging.

The proliferation of fake news on social media platforms poses significant challenges to information authenticity and public trust. In this study, the impact of various word embedding techniques on the performance of both machine learning and deep learning models for fake news detection is systematically evaluated. The comprehensive TruthSeeker dataset, which spans over a decade of labeled news articles and social media posts, is utilized. Three popular word embedding techniques are analyzed: TF-IDF, Word2Vec, and FastText. These embeddings were tested with multiple machine learning classifiers, including logistic regression, naive Bayes, K-nearest neighbors, support vector machines, multilayer perceptrons, decision trees, and random forests, as well as deep learning models, specifically convolutional neural networks (CNNs). Among machine learning models, the Support Vector Machine (SVM) with TF-IDF achieved the highest accuracy of 99.03%, precision of 98.92%, recall of 99.15%, and F1 score of 99.03%. The Multilayer Perceptron (MLP) also demonstrated strong performance with Word2Vec, reaching an accuracy of 95.24%, precision of 94.95%, recall of 95.50%, and F1 score of 95.07%. For deep learning models, CNNs combined with TF-IDF embeddings showed outstanding results, achieving an accuracy of 98.99%, precision of 98.56%, recall of 99.48%, and F1 score of 99.02%. These results highlight the critical role of word embeddings in enhancing model accuracy and reliability in misinformation detection.

Astronomy is entering an unprecedented era of Big Data science, driven by missions like the ESA's Gaia telescope, which aims to map the Milky Way in three dimensions. Gaia's vast dataset presents a monumental challenge for traditional analysis methods. The sheer scale of this data exceeds the capabilities of manual exploration, necessitating the utilization of advanced computational techniques. In response to this challenge, we developed a novel approach leveraging deep learning to estimate the metallicity of fundamental mode (ab-type) RR Lyrae stars from their light curves in the Gaia optical G-band. Our study explores applying deep learning techniques, particularly advanced neural network architectures, in predicting photometric metallicity from time-series data. Moreover, our deep learning models demonstrated notable predictive performance over experimentation and analysis. Through GRU-based models we achieved a low mean absolute error of 0.056 and a high $R^2$ regression performance of 0.9401 measured by cross-validation, showcasing the effectiveness of our approach in accurately estimating metallicity values. Our work underscores the importance of deep learning in astronomical research, particularly with large datasets from missions like Gaia. By harnessing the power of deep learning methods, we can provide precision in analyzing vast datasets, contributing to more precise and comprehensive insights into complex astronomical phenomena.

Deep inhalation breath hold (DIBH) has become a valuable technique in left breast cancer radiotherapy, offering the possibility to reduce radiation exposure to organs at risk (OARs) and minimize the risk of cardiac complications. The treatment method involves patients stopping breathing during irradiation to temporarily distance the heart from the radiation field, reducing potential cardiac and other complications. Research shows that DIBH reduces the radiation dose to the heart and lungs. However, the effectiveness of DIBH is determined by a variety of factors, including the patient's training, cooperation, anatomical features, age and choice of radiotherapy technique. Additionally, cardiovascular risk factors: diabetes, smoking, hypertension can affect the effectiveness and potential complications of DIBH. Moreover, lack of consent, fear, anxiety, depression and comorbid psychiatric disorders disqualify some patients from the DIBH treatment method. Careful patient selection, comprehensive training and optimization of treatment parameters are essential to maximize the benefits of DIBH while minimizing potential side effects. DIBH enhancement techniques, such as IMRT or VMAT, also have an important role to play. The purpose of this review article is to summarize the factors affecting the efficacy and side effects of DIBH in radiation therapy for left breast cancer, with the aim of optimizing its clinical application while minimizing side effects.

Land cover mapping, essential for understanding global land use patterns, relies on satellite imagery for monitoring changes, assessing ecosystem health, and supporting conservation ef- forts. However, significant challenges remain in managing large, complex satellite imagery datasets, acquiring specialized datasets due to high costs and labor intensity, and a lack of comparative studies for optimal deep learning model selection. Additionally, a scarcity of aerial datasets specifically tailored for agricultural areas exists. This study addresses these gaps by presenting a method for semantic segmentations of land covers in agricultural areas using satellite images and deep learning models with pre-trained backbones. We introduce an efficient methodology for preparing semantic segmentation datasets and contribute the "Land Cover Aerial Imagery" (LICAID) dataset for semantic segmentation. The study focuses on the Franciacorta area, Lombardy Region, leveraging the rich diversity of the dataset to effectively train and evaluate the models. We conduct a comparative study, employing cutting-edge deep learning-based segmentation models (U-Net, SegNet, DeepLabV3) with various pre-trained backbones (ResNet, Inception, DenseNet, EfficientNet) on our dataset acquired from Google Earth Pro. Through meticulous data acquisition, preprocessing, model selection, and evaluation, we demonstrate the effectiveness of these techniques in accurately delineating land cover classes. Integrating pre-trained feature extraction networks significantly improves performance across various metrics. Additionally, addressing challenges such as data availability, computational resources, and model interpret-ability is essential for advancing the field of remote sensing and supporting sustainable environmental stewardship worldwide.

In recent years, cryptocurrencies have received substantial attention from investors, researchers and the media due to their volatile behaviour and potential for high returns. This interest has led to an expanding body of research aimed at predicting cryptocurrency prices, which are notably influenced by a wide array of technical, sentimental, and legal factors. This paper reviews scholarly content from 2014 to 2024, employing a systematic approach to explore advanced quantitative methods for cryptocurrency price prediction. It encompasses a broad spectrum of predictive models, from early statistical analyses to sophisticated machine and deep learning algorithms. Notably, this review identifies and discusses the integration of emerging technologies such as Transformers and hybrid deep learning models, which offer new avenues for enhancing prediction accuracy and practical applicability in real-world scenarios. By thoroughly investigating various methodologies and parameters influencing cryptocurrency price predictions, including market sentiment, technical indicators, and blockchain features, this review highlights the field’s complexity and rapid evolution. The analysis identifies significant research gaps and under-explored areas, providing a foundational guideline for future studies. These guidelines aim to connect theoretical advancements with practical, profit-driven applications in cryptocurrency trading, ensuring that future research is both innovative and applicable.

In this work, we introduce the first pipeline that combines a refraction-aware Structure-from-Motion (SfM) method with a deep learning model specifically designed for airborne bathymetry. We accurately estimate the 3D positions of the submerged points by integrating refraction geometry within the SfM optimization problem. This way, no refraction correction as post-processing is required. Experiments with simulated data that approach the real-world capturing conditions demonstrate that SfM with refraction correction is extremely accurate, with submillimeter errors. We integrate our refraction-aware SfM within a deep learning framework that also takes into account radiometrical information, developing a combined spectral and geometry-based approach, with further improvements on accuracy and robustness to different seafloor types, both textured and textureless. We conducted experiments with real-world data at 2 locations in the southern Mediterranean Sea, with varying seafloor types which demonstrate the benefits of refraction correction for the deep learning framework. We open-source our refraction-aware SfM, providing researchers in airborne bathymetry with a practical tool to apply SfM in shallow water areas.

As modern systems are becoming more complex their control strategy cannot longer relay only on measurement information that usually comes from probes but also from mathematical models. Those systems models can lead to unbearable computation times due to their own complexity, turning the control process non-viable, which leads to the implementation of surrogate models that enable to achieve estimates within acceptable time to take decisions. A control trained with Deep Reinforcement Learning algorithm, using a Physics-Informed Neural Network to obtain the temperature map on the following time step, replaces the need of running Direct Numerical Simulations. On this work we considered an 1D heat conduction problem, which temperature distribution feeds a control system to activate a heat source aiming to obtain a constant, previously defined, temperature value. With this approach, control training becomes much faster without the need of performing numerical simulations or laboratory measurements, as well the control is taken based on Neural Network enabling its implementation on simple processors to edge computing.

Accurate segmentation of the Left Ventricle (LV) in Echocardiogram (Echo) images is essential for cardiovascular analysis. Conventional techniques are labor-intensive and exhibit inter-observer variability. Deep learning has emerged as a powerful tool for automated medical image segmentation, offering advantages in speed and potentially superior accuracy. This study explores the efficacy of employing the YOLO (You Only Look Once) segmentation model for automated LV segmentation in Echo images. YOLO, a cutting-edge object detection model, achieves exceptional speed-accuracy balance through its well-designed architecture. It utilizes efficient dilated convolutional layers and bottleneck blocks for feature extraction, while incorporating innovations like path aggregation and spatial attention mechanisms. These attributes make YOLO a compelling candidate for adaptation to LV segmentation in Echo images. We posit that by fine-tuning a pre-trained YOLO based model on a well-annotated Echo image dataset, we can leverage the model's strengths in real-time processing and precise object localization to achieve robust LV segmentation. The proposed approach entails fine-tuning a pre-trained YOLO model on a rigorously labeled Echo image dataset. Model performance has been evaluated using established metrics such as mean Average Precision (mAP) at an Intersection over Union (IoU) threshold of 50% (mAP50) with 98.31%and across a range of IoU thresholds from 50% to 95% (mAP50:95) with 75.27%. Successful implementation of YOLO for LV segmentation has the potential to significantly expedite and standardize Echo image analysis. This advancement could translate to improved clinical decision-making and enhanced patient care.

Recent advancements in computer vision, especially deep learning models, have shown considerable promise in tasks related to plant image object detection. However, the efficiency of these deep learning models heavily relies on input image quality, with low-resolution images significantly hindering model performance. Therefore, reconstructing high-quality images through specific techniques will help extract features from plant images, thus improve model performance. In this study, we explored the value of super-resolution technology for improving object detection model performance on plant images. Firstly, we built a comprehensive dataset comprising 1030 high-resolution plant images, named the PlantSR dataset. Subsequently, we developed a super-resolution model using the PlantSR dataset and benchmarked it against several state-or-the-art models designed for general image super-resolution tasks. Our proposed model demonstrated superior performance on the PlantSR dataset, indicating its efficacy in enhancing the super-resolution of plant images. Furthermore, we explored the effect of super-resolution on two specific object detection tasks: apple counting and soybean seed counting. By incorporating super-resolution as a pre-processing step, we observed a significant reduction in mean absolute error. Specifically, on the YOLOv7 model employed for apple counting, the mean absolute error decreased from 13.085 to 5.71. Similarly, on the P2PNet-Soy model utilized for soybean seed counting, the mean absolute error decreased from 19.159 to 15.085. These findings underscore the substantial potential of super-resolution technology in improving the performance of object detection models for accurately detecting and counting specific plants from images. The source codes and associated datasets are available at https://github.com/SkyCol/PlantSR.

This literature review delves into the realm of fault detection techniques for transmission lines, aiming to provide a comprehensive overview of recent advancements and key trends in the field. Employing a structured approach, the review synthesizes a plethora of research spanning from 2019 to 2024, sourced from diverse databases including IEEE Xplore, ScienceDirect, ResearchGate, Scopus, Litmaps, and Google Scholar. The methodology encompasses a systematic literature search protocol, stringent inclusion and exclusion criteria, meticulous data extraction, and a multi-dimensional analysis framework. The literature review uncovers a spectrum of fault detection methodologies, ranging from traditional signal processing techniques like Discrete Wavelet Transform and phase angle-based methods to cutting-edge deep learning algorithms such as Capsule Networks and Convolutional Neural Networks. Insights gleaned from the review underscore the critical importance of fault detection in maintaining the reliability, safety, and efficiency of power grids, highlighting its role as a frontline defense against widespread outages and equipment damage. Key findings from the review shed light on the efficacy of different fault detection approaches, showcasing their strengths and limitations across diverse system conditions. Furthermore, the review identifies common trends and challenges, including the need for real-world validation, scalability, adaptability to various network configurations, and cybersecurity considerations. This literature provides valuable insights and recommendations for future research endeavors in fault detection for transmission lines. By embracing advancements in both traditional and emerging techniques, researchers can continue to enhance the resilience and dependability of power transmission systems, ensuring their ability to withstand evolving challenges and safeguard critical infrastructure.

Breast cancer stands as a predominant health concern for women globally. As mammography is the primary screening tool for breast cancer detection, improving the detection of breast cancer at screening could save more lives. This mammography re-view paper provides a systematic review of computer-aided techniques during a specific time frame for the segmentation and classification of microcalcification, evaluating image processing, machine learning, and deep learning techniques. The systematic review is meticulously carried out, adhering closely to the preferred reporting items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. This article focuses on mammographic breast cancer detection approaches based on automated systems, discussed chronologically from 1970 through 2023. This article encompasses the breadth of artificial intelligence-based methods from the most primitive to the most sophisticated models. Image processing and machine learning-based methods are compre-hensively reviewed. The evaluation of a deep learning architecture based on self-extracted features for classification tasks demonstrated outclass performance. Large-scale datasets required for a broader and in-depth analysis of novel methods for breast cancer detection are also discussed in this article. This research work is aligned with the United Nations' sustainability de-velopment goals.

Recommendation systems are crucial in navigating the vast digital market. However, user data’s dynamic and non-stationary nature often hinders their efficacy. Traditional models struggle to adapt to the evolving preferences and behaviours inherent in user interaction data, posing a significant challenge for accurate prediction and personalisation. Addressing this, we propose a novel theoretical framework, the Non-stationary Transformer, designed to capture and leverage the temporal dynamics within data effectively. This approach enhances the traditional transformer architecture by introducing mechanisms accounting for non-stationary elements, offering a robust and adaptable solution for recommendation systems. Our experimental analysis, encompassing deep learning and reinforcement learning paradigms, demonstrates the framework’s superiority over benchmark models. The empirical results confirm the efficacy of our proposed framework, which not only provides significant performance enhancements, approximately 8% in Logloss reduction and up to 2% increase in F1 score but also underscores its potential applicability across accumulative reward scenarios. These findings advocate adopting Non-stationary Transformer models to tackle the complexities of today’s recommendation tasks.

Crop production in Afghanistan suffers from limited phosphorus (P) availability, which severely hinders national agriculture sustainability. This present study hypothesized that deep fertilizer placement could significantly enhance uptake of immobile P and thus tissue P accumulation and crop yield. A two-year pot experiment growing two maize (Zea mays) hybrid cultivars (Xida-789 and Xida-211) was therefore conducted to test these hypotheses under three contrasting fertilizer placement methods (broadcast, side band, and deep band). In doing so, P concentrations in both maize tissues and soils were compared at 45, 60, and 115 days after sowing (DAS) under nine combinations of nitrogen (N) and P fertilizer rates (kg ha-1: N112P45, N112P60, N112P75, N150P45, N150P60, N150P75, N187P45, N187P60, N187P75). Results have shown that deep band placement significantly increased P uptake efficiency, leading to greater P concentration and accumulation in maize tissues compared to other two fertilization methods. This improved P uptake was attributed to several factors associated with deep placement, including reduced P fixation, enhanced root access to P, and moisture availability for P uptake. Additionally, deep band placement combined with higher N application rates (N187 and N150) further enhanced plant P uptake by promoting P availability and utilization mechanisms. Deep band placement also resulted in significantly higher total soil P, Olsen-P, and P use efficiency than broadcast and side band methods, indicating a more efficient P fertilization strategy for maize that can improve growth and yield. This present study also found positive correlations between P concentration in plant organs and soil Olsen-P, highlighting the importance of adequate soil P levels for optimal plant growth. Overall, our results have showed that deep band fertilizer placement emerged as a superior strategy for enhancing P uptake efficiency, utilization, and maize productivity compared to broadcast and side band placement. This approach can potentially optimize P fertilizer utilization, improve maize growth and yield, while minimizing environmental P losses associated with broadcast applications.

The experience of ego dissolution and self-transcendence induced by psychedelic substances offers a profound challenge to our conventional understanding of the self and our relationship with the natural world. This paper delves into the philosophical underpinnings and implications of this phenomenon, exploring how the temporary breakdown of ego boundaries during a psychedelic experience may foster a sense of profound interconnectedness with nature. Drawing from Western and Eastern philosophical traditions, as well as recent insights from neuroscience and consciousness studies, we examine the potential of psychedelic-induced self-transcendence to catalyze a shift in our perspectives on the human-nature relationship. The paper investigates the philosophical implications of this experience, including its resonance with deep ecological philosophies and its potential to inspire more ecocentric ethical frameworks. Furthermore, we explore the possible practical implications of fostering a sense of ecological connectedness through psychedelics, such as promoting environmentally sustainable behaviors and addressing the pressing challenges of climate change and ecological degradation. Ultimately, this exploration highlights the value of engaging with altered states of consciousness as a means to reconsider our place within the intricate web of life on Earth.

Given the landscape of intricate and constantly evolving cyber threats, organizations demand refined strategies to deploy a Security Information and Event Management to support the management of a Cyber Security Operations Center. The dynamic nature of cyber threats complicates the efficient allocation of the location of network intrusion detection sensors, a critical component of a robust cybersecurity framework. Our research introduces an approach that integrates the precision of biomimetic optimization algorithms with the adaptability of Deep Q–Learning. By employing different biomimetic algorithms enhanced with deep learning, we aim to refine the deployment of sensors in network infrastructures, balancing the network security imperative against deployment costs. The results of computational tests demonstrate that the improved iterations leveraging Deep Q–Learning have outperformed their native counterparts. These findings underscore the importance of reinforcement learning, specifically through Deep Q–Learning, as a powerful tool to enhance the effectiveness of metaheuristics in addressing optimization challenges.

Microgrids (MGs) have evolved as critical components of modern energy distribution networks, providing increased dependability, efficiency, and sustainability. Effective control strategies are essential for optimising MG operation and maintaining stability in the face of changing environmental and load conditions. Traditional rule-based control systems are extensively used due to their interpretability and simplicity. However, these strategies frequently lack the flexibility for complex and changing system dynamics. This paper provides a novel method called hybrid intelligent control for adaptive MG that integrates basic rule-based-control and deep learning techniques, including gated recurrent units (GRU), basic recurrent neural network (RNN) and long short-term memory (LSTM). The main target of this hybrid approach is to improve MG management performance by combining the strengths of basic rule-based systems and deep learning techniques. These deep learning techniques readily enhance and adapt control decisions based on historical data and domain-specific rules, leading to increasing system efficiency, stability, and resilience in adaptive MG. Our results show that the proposed method optimises MG operation, especially under demanding conditions such as variable renewable energy supply and unanticipated load fluctuations. This study investigates special RNN architectures and hyper-parameter optimization techniques with the aim of predicting power consumption and generation within the adaptive MG system. Our promising results show the highest-performing models indicating high accuracy and efficiency in power prediction. The finest-performing model accomplishes an R2 value close to 1, representing a strong correlation between predicted and actual power values.

Facial acne is a prevalent dermatological condition regularly observed in the general population. This study introduces a novel deep learning model for facial acne segmentation utilizing a semi-supervised learning method known as bidirectional copy-paste, which synthesizes images by interchanging foreground and background parts between labeled and unlabeled images during the training phase. To overcome the lower performance observed in the labeled image training part compared to the previous methods, a new framework was devised to directly compute the training loss based on labeled images. The effectiveness of the proposed method was evaluated against previous semi-supervised learning methods using images cropped from facial images at acne sites. The proposed method achieved a Dice score of 0.5205 in experiments utilizing only 3% of labels, marking an improvement of 0.0151 to 0.0473 Dice score over previous methods. The proposed semi-supervised learning approach for facial acne segmentation demonstrates improvement in performance, offering a novel direction for future acne analysis.

The application of space robotic manipulators and heightened autonomy for ios represents a paramount pursuit for leading space agencies, given the substantial threat posed by space debris to operational satellites and forthcoming space endeavors. This work presents a guidance algorithm based on drl to solve for the space manipulator path-planning during the motion synchronization phase with the mission target. The goal is the trajectory generation and control of a spacecraft equipped with a 7-dof robotic manipulator, such that its end effector remains stationary with respect to the target point of capture. The ppo drl algorithm is used to optimize the manipulator’s guidance law, and the autonomous agent generates the desired joints rates of the robotic arm, which are then integrated and passed to a model-based feedback linearization controller. The agent is first trained to optimize its guidance policy and then tested extensively to validate the results against a simulated environment representing the motion synchronization scenario of an ios mission.

Despite their relatively low incidence globally, CNS tumors remain amongst the most lethal cancers, with only a few other malignancies surpassing them in 5-year mortality rates. Treatment decisions for brain tumors heavily rely on histopathological analysis, particularly intraoperatively, to guide surgical interventions and optimize patient outcomes. Frozen sectioning has emerged as a vital intraoperative technique, allowing for highly accurate, rapid analysis of tissue samples, although it poses challenges regarding interpretive errors and tissue distortion. Raman histology, based on Raman spectroscopy, has shown great promise in providing label-free, molecular information for accurate intraoperative diagnosis, aiding in tumor resection and identification of neurodegenerative disease. Techniques including Stimulated Raman Scattering (SRS), Coherent Anti-Stokes Raman Scattering (CARS), Surface-Enhanced Raman Scattering (SERS), and Tip-Enhanced Raman Scattering (TERS) have profoundly enhanced the speed and resolution of Raman imaging. Similarly, Confocal Laser Endomicroscopy (CLE) allows for real-time imaging and rapid intraoperative histologic evaluation of specimens. While CLE is primarily utilized in gastrointestinal procedures, its application in neurosurgery is promising, particularly in the context of gliomas and meningiomas. This review focuses on discussing the immense progress in intraoperative histology within neurosurgery and provides insight into the impact of these advancements on enhancing patient outcomes..

The term "code smell" refers to an indication of a problem with the quality of source code. Numerous studies have been conducted to identify problematic features in source code. Initially, the focus was on utilizing metric-based and heuristic-based approaches. In recent years, however, there has been a shift towards using machine learning and deep learning (DL) techniques for smell detection. Nevertheless, the current algorithms are still considered to be in the early stages of development. Recognizing the challenges associated with identifying smells using DL methods, both academics and software developers have made efforts to address these obstacles. This work involves constructing and evaluating new DL models for code smell detection. Two models are built upon stacked autoencoders, employing a hybrid architecture that combines bidirectional long short-term memory and convolutional neural network components.

The modern supply chain (SC) is growing in terms of data, devices, users, and stakeholders, which introduced new security challenges and threats, especially with the reliance on centralized servers or cloud platforms. In addition, increased trust among system participants exposes the SC to a higher risk of vulnerabilities which require strong security measures. This article proposes a hybrid security framework for SC systems, BC-DRLzSC, that integrates Blockchain (BC) and Deep Reinforcement Learning (DRL) designed to operate in a zero trust (ZT) environment. In particular, we propose a decentralized BC-based approach integrated with smart contracts to manage system participant registration and authentication and to control access to system resources. BC-DRLzSC adopts a ZT architecture to reinforce SC security, which can be achieved with an advocate to verify each entity’s trustworthiness before granting or retaining access to system resources. Incorporating the ZT architecture, with BC and DRL, can potentially and significantly bolster SC system security. DRL is employed to develop a proactive attack detection model that continuously monitors the incoming traffic from authenticated nodes within the network and predicts any malicious actions. Finally, we evaluate the performance of our proposed DRL solution using the NSL-KDD dataset.

Diabetic retinopathy (DR) is the primary factor leading to vision impairment and blindness in diabetic people. Uncontrolled diabetes can damage the retinal blood vessels. Initial detection and prompt medical intervention are vital in preventing progressive vision impairment. Today's growing medical field poses more significant workload and diagnostic demands on medical professionals. In the proposed study, a convolutional neural network (CNN) is employed to detect the stages of DR. This research is crucial for studying DR because of its innovative methodology incorporating two different public datasets. This strategy enhances the model's capacity to generalize unseen DR images, as each dataset encompasses unique demographics and clinical circumstances. The network can learn and capture complicated hierarchical image features with asymmetric weights. Each image undergoes preprocessing using contrast-limited adaptive histogram equalization (CLAHE) and the discrete wavelet transform (DWT). The model is trained and validated using the combined datasets of Dataset for Diabetic Retinopathy (DDR) and the Asia Pacific Tele-Ophthalmology Society (APTOS). The CNN model is tuned with different learning rates and optimizers. The CNN model with Adam optimizer achieved 70% accuracy and an area under curve (AUC) score of 0.82. The recommended study results may reduce diabetes-related vision impairment by early DR severity identification.

Prostate cancer is the second most frequent malignancy (after lung cancer) in men worldwide. Prostate tissue biopsies are usually graded using scores according to the Gleason grading system. This Gleason score is the most popular prognostic marker that reveals the potential aggressiveness of the disease. However, inter-observer variability in rating by human assessors is a major limiting factor. Such variability could lead to missing a severe case or suggesting unnecessary treatments. This study explores the discriminative ability of artificial intelligence (deep learning models) for Gleason score assessment. The study was designed to use whole-slide images of digitized H&E-stained biopsies of prostate tissues to automate the grading process and provide a remotely accessible clinical decision support system. Custom convolutional neural network architectures were trained on 10,616 images of prostate tissues. Gaussian filters were applied to pre-process the images and improve model performance. Transfer learning was applied to train eight machine learning architectures namely: Xception, VGG16 & 19, ResNet101, MobileNet, DenseNet121, EfficientNetB5 & B7. Efficient NetB7 had the best performance 85.2% compared with ground-truth classification by experts. Performance improves as more data is available. The model was deployed and hosted as a web application API on Google cloud service to ensure remote access. Tissue biopsy images can be uploaded and the corresponding Gleason score recovered immediately. This system reduces diagnostics turnaround time, increase throughput, and compensate for limited skills, especially in low resource settings.

Antimony (Sb) has gained significance as a critical raw material (CRM) within the European Union (EU) due to its strategic importance in various industrial sectors, particularly in the textile industry for flame retardants and as a component of Sb-based semiconductor materials. Moreover, Sb is emerging as a potential alternative for anodes used in lithium-ion batteries, a key element in the Energy transition. This study focused on exploring the feasibility of identifying and quantifying Sb mineralizations through the spectral signature of soils using reflectance spectroscopy, a non-invasive remote sensing technique, and by employing deep learning algorithms such as Convolutional Neural Networks (CNNs). Common signal preprocessing techniques were applied to the spectral data, and the soils were analyzed by Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Despite achieving high R-squared values, the study faces a significant challenge of generalization of the model to new data. Despite the limitations, this study provides valuable insights into potential strategies for future research in this field.

Educational institutions must identify students who are academically struggling to provide them with necessary support to improve their performance. In this context, recommendation systems powered by deep learning techniques are vital for detecting and categorizing such students. These systems help students plan their future by uncovering patterns in their historical academic data. This study introduces a new deep learning model designed to classify academically underperforming students in educational settings. The model incorporates a Gated Recurrent Neural Network (GRU) and includes specific neural network features like a dense layer, max-pooling layer, and the ADAM optimization algorithm. The model's training and evaluation were conducted using a dataset comprising 15,165 student assessment records from various academic institutions. The performance of the developed GRU model was benchmarked against other educational recommendation systems, including Recurrent Neural Network models, AdaBoost, and the Artificial Immune Recognition System v2. The proposed GRU model demonstrated remarkable accuracy, achieving an overall rate of 99.70%.

Object detection is complex and involved diverse requirements for different applications. In critical application such as visual impaired navigation guide and real-time surveillance for instance, identifying a particular object of instance is required and this involved a complex approach for realization. Literatures were reviewed, starting with the application of deep learning techniques for object detection; their gaps were identified and addressed using real-time object detection models literature review. From the review gaps were also detected and addressed using literature review on occlusion detection techniques. Object of instance detection in clustered scene with similar object of interest was identified as an open gap not addressed in the existing literature; even through it is vital for many applications. This research recommended an Occlusion Based Object of Instance Detection (OBOID) technique which used the spatial information to identify instant object of interest in clustered scene with many similarly object of interest. Limitation of the recommended OBOID is that it requires only system where position and distance of object is necessary to inform other decision.

In the age of social media, images come from unreliable sources that may change their content to fit a narrative. Automated detection of forged images is a complex task, especially considering recent technological advances in image manipulation software. However, in the existing literature, it is possible to identify two general approaches adopted by detection methods, active and passive. Active techniques preemptively act on an image by inserting structures before any manipulation is made that can verify its authenticity. In contrast, passive methods analyze the content of an image in search of traces of manipulation. In this way, this research proposes a novel approach to image manipulation detection, combining two passive methods through neural networks, creating a generalist approach capable of detecting with greater accuracy than the methods that compose it. Furthermore, this work used a combination of four datasets available in the literature for training and evaluation. After training, the merged approach obtained an accuracy of 89.59% in the set of validation images, significantly higher than the model trained with only unaltered images, which obtained 78.64%, and the two other models trained using images with a feature selection method applied to enhance inconsistencies that obtained 68.02% for Error-Level Analysis images and 50.70% for the method using Discrete Wavelet Transform. In addition to the performance improvement, the proposed approach’s accuracy variation was lower than the other models.

Deep eutectic solvents are a mixture of two or more components, that at a particular composition become liquids at room temperature. When the compounds that constitute the DES are primary metabolites, namely, amino acids, organic acids, sugars, or choline derivatives, the DES are so called natural deep eutectic solvents (NADES). NADES fully represent green chemistry principles. These solvents are highly welcome, as they are obtained from renewable resources, and gratifyingly biodegradable and biocompatible. They are an alternative to room temperature ionic liquids (RTILs). From the pharmaceutical industry point of view, they are highly desirable, but unfortunately have been rarely used despite their enormous potential. In this review, we will look at their impact on the asymmetric catalytic synthesis of key target molecules, via metal-based catalysis, biocatalysis and organocatalysis. In many cases the NADES that have been used are chiral and can even promote enantioselective reactions, this crucial and very exciting aspect will also be discussed and analysed.

A novel reinforcement learning deep deterministic policy gradient agent-based sliding mode control (DDPG-SMC) approach is proposed to suppress the chattering phenomenon in the attitude control for quadrotors, in the presence of external disturbances. First, the attitude dynamics model of the studied quadrotor is derived and the attitude control problem is described by formulas. Second, a sliding mode controller including its sliding mode surface and reaching law is selected for the nonlinear dynamic system, and the stability of the designed SMC system is supported by Lyapunov stability theorem. Third, a reinforcement learning (RL) agent based on deep deterministic policy gradient (DDPG) is trained to adjust the switching control gain adaptively. During the training process, the input signals of agent are the actual and desired attitude angles, and the output action is the time-varying control gain. Finally, the above trained agent is applied to the SMC as a parameter regulator, to implement the adaptive adjustment of the switching control gain related to reaching law, and the simulation results verify the robustness and effectiveness of the proposed DDPG-SMC method.

In areas where conventional fundus cameras are logistically unaffordable, the smartphone-based approach is considered as a promising future of glaucoma screening due to their affordability and ease-of-use. Optic disc localization is a critical stage during glaucoma screenings. For fundus images captured with standard fundus cameras, the majority of the models available out there can locate the optic disc with satisfactory performance. However, for images that are captured with smartphone-based fundus cameras, the inherent noise and lower quality of the fundus image makes it difficult for models to detect the optic disc region. In this study, we have proposed to utilize YOLOv8n for optic disc localization due to the model’s cutting-edge performance in diverse tasks and its lightweight nature. We have used a public dataset which has 2000 low quality fundus images that are captured using a smartphone-based fundoscopy device. From these, 60% of the data was used for fine-tuning the model, and 25% for testing. By using a confidence of 50% set as threshold, the model was able to detect the optic disc successfully on over 97% of test images with intersection over union of above 0.85. These results highlight the potential of the lightweight YOLOv8n model for deployment on resource-constrained environments, offering a promising performance on accurately localizing the OD and enhancing the feasibility of affordable glaucoma screening on smartphones.

The rapid growth of equipment tasks in 5G large-scale machine-type communication scenarios presents challenges due to user equipment’s (UE) limited computing power and power constraints. To address these limitations, task offloading to unmanned aerial vehicles (UAVs) has gained attention. However, traditional cloud computing centres far from the UE fail to meet latency requirements. Mobile Edge Computing (MEC) emerges as a solution by deploying computing servers at the edge of the cellular network to enhance service responsiveness. However, traditional MEC solutions need more flexibility. This paper explores the advantages of UAVs, including flexibility and rapid deployment, and considers UAV power constraints. It proposes a joint optimization approach for user offloading strategies and UAV trajectories to minimize the overall energy consumption of UE. By leveraging the benefits of UAVs and MEC, the proposed method aims to improve task execution efficiency in energy-constrained 5G machine-type communication scenarios.