The emergence of the Internet of Things (IoT) has led to the deployment of various types of sensors in many application fields, including environment monitoring, smart cities, health, industries, and others. The increasing number of connected devices has led to the creation of massive quantities of data that need to be analyzed. Typically, this data is ordered by time, as a time series. In this context, this paper presents a time series prediction model based on Recurrent Neural Networks in order to predict one step ahead. Result obtained through five Internet of Things monitoring datasets, showed that the Recurrent Neural Network obtained better performance that the prediction methods, ARIMA and SVM.

Accurate electricity price forecasting has become a substantial requirement since the liberalization of the electricity markets. Due to the challenging nature of the electricity prices, which includes high volatility, sharp price spikes and seasonality, various types of electricity price forecasting models still compete and can not outperform each other consistently. Neural Networks have been successfully used in machine learning problems and Recurrent Neural Networks (RNNs) have been proposed to address time-dependent learning problems. In particular, Long Short Term Memory and Gated Recurrent Units (GRU) are tailor-made for time series price estimation. In this paper, we propose to use Gated Recurrent Units as a new technique for electricity price forecasting. We have trained a variety of algorithms with rolling 3-year window and compared the results with the RNNs. In our experiments, 3-layered GRUs outperformed all other neural network structures and state of the art statistical techniques in a statistically significant manner in the Turkish day-ahead market.

Dynamic neural networks (DNN) are types of artificial neural networks (ANN) that are designed to work with sequential data where context in time is important. In contrast to traditional static neural networks that process data in a fixed order, dynamic neural networks use information about past inputs, which is important if dynamic of a certain process is emphasized. They are widely used in natural language processing, speech recognition and time series prediction. In industrial processes, their use is interesting for the prediction of difficult-to-measure process variables. In an industrial process of isomerization, it is crucial to measure the quality attributes affecting the octane number of gasoline. Process analyzers that are commonly used for this purpose are expensive and subject to failures, therefore, in order to achieve continuous production in case of malfunction, mathematical models for estimating the product quality attributes are imposed as a solution. In this paper, mathematical models were developed using dynamic recurrent neural networks (RNN), i.e., their subtype of a long short-term memory (LSTM) architecture. The results of the developed models were compared with the results of several types of other data-driven models developed for an isomerization process, such as multilayer perceptron (MLP) artificial neural networks, support vector machines (SVM) and dynamic polynomial models. The obtained results are satisfactory, which suggests a good possibility of application.

This article presents a forecast model that uses a hybrid architecture of recurrent neural networks (RNN) with surface neural networks (ANN), based on historical records of exported active energy (EAE) and weather data. Two types of models were developed: the first type includes six models that use EAE records and weather variables as inputs, while the second type includes eight models that use only weather variables. Different metrics were applied to assess the performance of these models, and the best model of each type was selected. Finally, a comparison of the performance between the selected models of both types is presented, and they are validated with real data provided by a solar plant, achieving acceptable levels of accuracy. The selected model of the first type has an RMSE of 0.19, MSE of 0.03, MAE of 0.09, a correlation coefficient of 0.96, and a determination coefficient of 0.93. The other selected model of the second type showed lower precision in the metrics (RMSE = 0.24, MSE = 0.06, MAE = 0.10, Corr. Coef. = 0.95, and Det. Coef. = 0.90). Both models demonstrated good performance and acceptable accuracy in forecasting the weekly photovoltaic energy production of the solar plant.

Electricity demand forecasting plays a significant role in energy markets. Accurate prediction of electricity demand is the key factor to optimize power generation, consumption, saving energy resources, and determining the energy prices. However, integrating energy mix scenarios, including solar and wind power which are highly non-linear and seasonal, into an existing grid increases uncertainty in generation, adds the challenges for precise forecast. To tackle these challenges, state-of-the-art methods and algorithms have been implemented in literature. We have developed Artificial Intelligence (AI) based deep learning models that can effectively handle the information of long time-series data. Based on the pattern of dataset, four different scenarios were developed and two best scenarios were selected for prediction. Dozens of models were developed and tested in deep AI networks. In the first scenario (Scenario1), data for weekdays excluding holidays was taken and in the second scenario (Scenario2) all the data in the basket was taken. Remaining two scenarios, weekends and holidays were tested and neglected because of their high prediction error. To find the optimal configuration, models were trained and tested within a large space of alternatives called hyper-parameters. In this study, an Aritificial Neural Network (ANN) based Feed-forward Neural Network (FNN) showed the minimum prediction error for Scenario1 while a Recurrent Neural Network (RNN) based Gated Recurrent Network (GRU) showed the minimum prediction error for Scenario2. While comparing the accuracy, the lowest MAPE of 2.47% was obtained from FNN for Scenario1. When evaluating the same testing dataset (non-holidays) of Scenario2, the RNN-GRU model achieved the lowest MAPE of 2.71%. Therefore, we can conclude that grouping of weekdays as Senario1 prepared by excluding the holidays provides better forecasting accuracy compared to the single group approach used in Scenario2, where all the dataset is considered together. However, Scenario2 is equally important to predict the demand for weekends and holidays.

Deep neural networks (DNNs) have made a huge impact in the field of machine learning by providing unbeatable humanlike performance to solve real-world problems such as image processing and natural language processing (NLP). Convolutional neural network (CNN) and recurrent neural network (RNN) are two typical architectures that are widely used to solve such problems. Time sequence-dependent problems are generally very challenging, and RNN architectures have made an enormous improvement in a wide range of machine learning problems with sequential input involved. In this paper, different types of RNN architectures are compared. Special focus is put on two well-known gated-RNN’s Long Term Short Memory (LSTM) and Gated Recurrent Unit (GRU). We evaluated these models on the task of force estimation system in pouring. In this study, four different models including multi-layers LSTM, multi-layers GRU, single-layer LSTM and single-layer GRU) were created and trained. The result suggests that multi-layer GRU outperformed other three models.

Online transactions are becoming more popular in present situation where the globe is facing an unknown disease COVID-19. Now authorities of Countries requested peoples to use cashless transaction as far as possible. Practically it is not always possible to use it in all transactions. Since number of such cashless transactions have been increasing during lockdown period due to COVID-19, fraudulent transactions are also increasing in a rapid way. Fraud can be analysed by viewing a series of customer transactions data that was done in his/her previous transactions. Normally banks or other transaction authorities warned their customers about the transaction If any deviation is noticed by them from available patterns. These authorities think that it is possibly of fraudulent transaction. For detection of fraud during COVID-19, banks and credit card companies are applying various methods such as data mining , decision tree, rule based mining, neural network, fuzzy clustering approach and machine learning methods. These approaches is try to find out normal usage pattern of customers based on their past activities. The objective of this paper is to find out such fraud transactions during such unmanageable situation.Digital payment schemes are often threatened by fraudulent activities. Detecting fraud transaction in during money transfer may save customers from financial loss. Mobile based money transactions are focused in this paper for fraud detection. A Deep Learning (DL) framework is suggested in this paper that monitors and detects fraudulent activities. Implementing and applying recurrent neural network on PaySim generated synthetic financial dataset, deceptive transactions are identified. The proposed method is capable to detect deceptive transactions with an accuracy of 99.87%, F1-Score of 0.99 and MSE of 0.01.

Multilabel learning goes beyond standard supervised learning models by associating a sample with more than one class label. Among the many techniques developed in the last decade to handle multilabel learning best approaches are those harnessing the power of ensembles and deep learners. This work proposes merging both methods by combining a set of gated recurrent units, temporal convolutional neural networks, and long short-term memory networks trained with variants of the Adam optimization approach. We examine many Adam variants, each fundamentally based on the difference between present and past gradients, with step size adjusted for each parameter. We also combine Incorporating Multiple Clustering Centers and a bootstrap-aggregated decision trees ensemble, which is shown to further boost classification performance. In addition, we provide an ablation study for assessing the performance improvement that each module of our ensemble produces. Multiple experiments on a large set of datasets representing a wide variety of multilabel tasks demonstrate the robustness of our best ensemble, which is shown to outperform the state-of-the-art. The MATLAB code for generating the best ensembles in the experimental section will be made available at https://github.com/LorisNanni.

When implementing SVMs, two major problems are encountered: (a) the number of local minima increases exponentially with the number of samples and (b) the quantity of required computer storage, required for a regular quadratic programming solver, increases by an exponential mag-nitude as the problem size expands. The Kernel-Adatron family of algorithms gaining attention lately which has allowed it to handle very large classification and regression problems. Howev-er, these methods treat different types of samples (Noise, border, and core) in the same manner, which causes searches in unpromising areas and increases the number of iterations. In this work, we introduce a hybrid method to overcome these shortcomings, namely Optimal Recurrent Neu-ral Network Density Based Support Vector Machine (Opt-RNN-DBSVM). This method consists of four steps: (a) characterization of different samples, (b) elimination of samples with a low probability of being a support vector, (c) construction of an appropriate recurrent neural network based on an original energy function, and (d) solution of the system of differential equations, managing the dynamics of the RNN, using the Euler-Cauchy method involving an optimal time step. The RNN remembers the regions explored during the search process thanks to its recurrent architecture. We demonstrated that RNN-SVM converges to feasible support vectors and Opt-RNN-DBSVM has a very low time complexity compared to RNN-SVM with constant time step, and KAs-SVM. Several experiments were performed on academic data sets. We used several classification performances measures to compare Opt-RNN-DBSVM to different classification methods and the results obtained show the good performance of the proposed method.

Arctic sea ice prediction holds significant importance for facilitating Arctic route planning, optimizing fisheries management, and advancing the field of sea ice dynamics research. While various deep learning models have been developed for sea ice prediction, they predominantly operate at the seasonal or sub-seasonal scale, often focusing on localized areas, and few cater to full-region daily scale prediction. This study introduces the use of spatiotemporal sequence data prediction models, namely, the convolutional LSTM (ConvLSTM) and predictive recurrent neural network (PredRNN), for the prediction of sea ice concentration (SIC). Our analysis reveals that, when solely utilizing SIC historical data as the input, the ConvLSTM model outperforms the PredRNN model in SIC prediction. To enhance the model's capacity to capture spatiotemporal relationships between multiple variables, we expanded the range of input data types to form the ConvLSTM-multi and PredRNN-multi models. Experimental findings demonstrate that the ConvLSTM-multi model excels in assimilating the influence of reanalysis data on sea ice within the sea ice edge region, thus exhibiting superior performance in predicting daily Arctic SIC over the subsequent 10 days. Furthermore, sensitivity tests on various model parameters highlight the substantial impact of sea surface temperature and prediction date on the accuracy of daily sea ice prediction. Additionally, meteorological and oceanographic parameters primarily affect the prediction accuracy of the thin ice region at the edge of the sea ice.

The environmental issues we are currently facing require long-term prospective efforts for sustainable growth. Renewable energy sources seem to be one of the most practical and efficient alternatives in this regard. Understanding a nation's pattern of energy use and renewable energy production is crucial for developing strategic plans. No previous study has been performed to explore the dynamics of power consumption with the change in renewable energy production on a country-wide scale. In contrast, a number of deep learning algorithms demonstrated acceptable performance while handling sequential data in the era of data-driven predictions. In this study, we developed a scheme to investigate and predict total power consumption and renewable energy production time series for eleven years of data using a Recurrent Neural Network (RNN). The dynamics of the interaction between the total annual power consumption and renewable energy production are investigated through extensive Exploratory Data Analysis (EDA) and a feature engineering framework. 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 0.084. Higher performance is achieved through the increase in the number of epochs and hyperparameter tuning. The proposed framework can be used and transferred to investigate the trend of renewable energy production and power consumption and predict the future scenarios for different communities. Incorporation of the cloud-based platform into the proposed pipeline may lead to real-time forecasting.

Many engineered approaches have been proposed over the years for solving the hard problem of performing indoor localization. However, specialising solutions for the edge cases remains challenging. Here we propose to build the solution with zero hand-engineered features, but having everything learned directly from data. We use a modality specific neural architecture for extracting preliminary features, which are then integrated with cross-modality neural network structures. We show that each modality-specific neural architecture branch is capable of estimating the location with good accuracy independently. But for better accuracy a cross-modality neural network fusing the features of those early modality-specific representations is a better proposition. Our multimodal neural network, MM-Loc, is effective because it allows the uniform flow of gradients during training across modalities. Because it is a data driven approach, complex features representations are learned rather than relying heavily on hand-engineered features.

Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been utilised to forecast the foreign exchange market (Forex). However, such models usually exhibit unstable behaviour, as data perturbations often downgrade the functionality of the entire network due to their monolithic architecture. Hence, this study proposes a novel neuroscience-informed modular network applied to the closing prices and the sentiments retrieved from Yahoo Finance and Twitter APIs, aiming to anticipate better price fluctuations in Euro to British Pound Sterling (EUR/GBP) rather than monolithic methods. The proposed model is based on a new modular CNN, replacing pooling layers with orthogonal kernel initialisation RNNs coupled with Monte Carlo Dropout (MCD), namely MCoRNNMCD. It combines two modules: i) a convolutional simple RNN and ii) a convolutional Gated Recurrent Unit (GRU), where orthogonality and MCD are added to reduce the overfitting, assessing each module's uncertainty. These parallel feature extraction modules concatenate their outputs to a final three-layer Artificial Neural Network (ANN) decision-making module. A comprehensive comparison viewing objective evaluation metrics such as the Mean Square Error (MSE) proved that the proposed MCoRNNMCD-ANN outperformed single CNN, LSTM, GRU, and the state-of-the-art hybrid BiCuDNNLSTM, CLSTM, CNN-LSTM, and LSTM-GRU in forecasting hourly EUR/GBP closing price fluctuations.

— In a smart healthcare system," Human Activity Recognition (HAR)" is considered as an efficient approach in pervasive computing from activity sensor readings. The "Ambient Assisted Living (AAL)" in the home or community helps the people to provide independent care and enhanced living quality. However, many AAL models are restricted to multiple factors that include both the computational cost and system complexity. Moreover, the HAR concept has more relevance because of its applications, such as content-based video search, sports play analysis, crowd behavior prediction systems, patient monitoring systems, and surveillance systems. This paper attempts to implement the HAR system using a popular deep learning algorithm, namely "Recurrent Neural Network (RNN)" with the activity data collected from smart activity sensors over time, and it is publicly available in the "UC Irvine Machine Learning Repository (UCI)". The proposed model involves three processes: (1) data collection, (b) optimal feature learning, and (c) activity recognition. The data gathered from the benchmark repository was initially subjected to optimal feature selection that helped to select the most significant features. The proposed optimal feature selection method is based on a new meta-heuristic algorithm called "Colliding Bodies Optimization (CBO)". An objective function derived from the recognition accuracy has been used for accomplishing the optimal feature selection. The proposed model on the concerned benchmark dataset outperformed the conventional models with enhanced performance.

Switched model predictive control (S-MPC) and recurrent neural network with long short-term memory (RNN-LSTM) are powerful control methods that have been extensively studied for energy management in microgrids (MGs). These methods are complementary in terms of constraint satisfaction, computational demand, adaptability, and comprehensibility, but typically one method is chosen over the other. The S-MPC method selects optimal models and control strategies dynamically based on the system’s operating mode and performance objectives. On the other hand, integration of auto-regressive (AR) with these powerful control methods improves prediction accuracy and system conditions’ adaptability. This paper compares the two approaches to control and proposes a novel algorithm called Switched Auto-regressive Neural Control (S-ANC) that combines their respective strengths. Using a control formulation equivalent to S-MPC and the same controller model for learning, the results indicate that pure RNN-LSTM cannot provide constraint satisfaction. The novel S-ANC algorithm can satisfy constraints and deliver comparable performance to MPC while enabling continuous learning. Results indicate that S-MPC optimization increases power flows within the MG, resulting in efficient utilization of energy resources. By merging the AR and LSTM, the model’s computational time decreased by nearly 47.2%. Also, this study evaluated our predictive model’s accuracy: (i) R-squared error is 0.951, indicating strong predictive ability, and (ii) mean absolute error (MAE) and mean square error (MSE) values of 0.571 indicate accurate predictions with minimal deviations from actual values.

We define the notions of strong and strict recurrency for actions of countable ordered groups on $\sigma$-finite non atomic measure spaces with quasi-invariant measures. We show that strong recurrency is equivalent to non existence of weakly wandering sets of positive measure. We also show that for certain p.m.p ergodic actions the system is not strictly recurrent, which shows that strong and strict recurrency are not equivalent.

Recurrent bacterial vaginosis (RBV) after antibiotic treatments has a relapse rate of 35% within 3 months and 60% within 12 months. Products containing polycarbophil (PLGG), that inhibits bacterial growth and mucoadhesive property, can impair biofilm formation. Here are shown the results of the POLARIS (Polybactum® to assess Recurrent Bacterial Vaginosis) study. The first phase was an interventional, open-label, non-controlled, and multicentre trial enrolling 56 women in Italy and Romania. The second phase was an observational 10-month follow-up without treatment conducted only in Romania. After 3 cycles with PLGG, only 8 BV recurrences out of 54 evaluable patients were identified (rate 14.81%) and for 26 out of 39 patients (66.67%) was evidenced positive effect on Lactobacilli in the vaginal secretions. In the follow-up 35 patients were observed after PLGG stopping treatment; 1 RBV (2.86%) at the 4th month and an additional 6 cases (17.14%) were evidenced at the end of the follow-up period. Therefore, no recurrence was evidenced in 12 subjects (34.28%) at 10th ± 2 months after the end of the PLGGtreatment. The use of PLGG vaginal ovules in the treatment of BV reduces the rate of relapses and improves the microbiological parameters (positive effect on Lactobacilli in 66.7 % of cases).

With an explosion of data in the nearly every domain, from physical and environmental sciences to econometrics and even social science, we have entered a golden era of time-domain data science. In addition to ever increasing data volumes and varieties of time-series, there's a plethora of techniques available to analyses them. Unsurprisingly, machine learning approaches are very popular. In environmental sciences (ranging from climate studies to space weather/solar physics), machine learning (ML )and deep learning (DL) approaches complement the traditional data assimilation approaches. In solar and astrophysics, multiwavelength time-series or lightcurves along with observations of other messengers like gravitational waves provide a treasure trove of time-domain data for ML/DL approaches for modeling and forecasting. In this regard, deep learning algorithms such as recurrent neural networks (RNNs) should prove extremely powerful for forecasting as it has in several other domains. Such methods agnostic of the physics can only predict transient flares that are represented in the original distribution trained on (eg. coloured noise fluctuations) and not special, isolated episodes (eg. a one-off outburst like a major coronal mass ejection) potentially distinguishing them. Therefore it is vital to test robustness of these techniques ; here we attempt to make a small contribution to this by exploring sensitivity of RNNs with some rudimentary experiments. We test the performance of a very successful class of RNNs, the Long Short Term Memory (LSTM) algorithms with simulated lightcurves. As an example, we focus on univariate forecasting of time-series (or lightcurves) characteristic certain variable astrophysical objects (eg. active galactic nuclei or AGN). Specifically, we explore the sensitivity of training and test losses to key parameters of the LSTM network and data characteristics namely gaps and complexity measured in terms of number of Fourier components. We find that typically, the performances of LSTMs are better for pink or flicker noise type sources. The key parameters on which performance is dependent are batch size for LSTM and the gap percentage of the lightcurves. While a batch size of $10-30$ seems optimal, the most optimal test and train losses are under $10 \%$ of missing data for both periodic and random gaps in pink noise. The performance is far worse for red noise. This compromises detectability of transients. Thus, we show that generative models (time-series simulations) are excellent guides for use of RNN-LSTMs in building forecasting systems

Abstract: Background: Surgery on varicose veins (crossectomy and stripping) leads to recurrence and has clinical and socio-economic repercussions. Their etiopathogenesis has yet to fully under-stood. Objective: Study the expression of endoglin and other molecules involved in the neovascu-larisation process in patients suffering from this disease. Methods: 43 patients that have undergone surgery for varicose veins (24 primary and 19 recurrent). They were identified on the venous wall (proximal -saphenofemoral junction- and distal), via real-time RT-PCR, and in serum, via ELISA: Endoglin (Eng), Vascular Endothelial Growth Factor (VEGF-A), its receptors 1 and 2 (VEGFR1 or FLT1), (VEGFR2 or FLK), and the Hypoxia-Inducible Factor (HIF-1A). All the patients signed a con-sent form. Results: The recurrent group recorded a higher expression of Eng, VEGF-A, VEGFR1 and VEGFR2 at the level of proximal venous wall compared to the primary group. HIF-1A did not record any differences. As regards the determination of the distal venous wall, no markers recorded differ-ences between the groups. Among the serum determinations, only sFLT1 recorded a significant drop among the patients with recurrent varicose veins. Conclusions: Patients with recurrent vari-cose veins record a higher expression of endoglin and other markers of angiogenesis in proximal veins. Endoglin in the blood (sEng) has not proven to be of any use in recurrent varicose veins.

Bacterial vaginosis (BV), the most common cause of vaginal discharge, is characterized by a shift in the vaginal microbiota from lactobacillus dominance to a diverse array of facultative and strict anaerobic bacteria which form a multi-species biofilm on vaginal epithelial cells. The rate of recurrence after therapy is high, often >60%. While the BV biofilm itself likely contributes to recurrent and/or refractory disease after treatment by reducing antimicrobial penetration, antimicrobial resistance in BV-associated bacteria including those, both within the biofilm and the vaginal canal, may be the result of independent, unrelated bacterial properties which are discussed in this paper. Our current recommendations for the treatment of refractory and recurrent BV are also provided.

An earthquake is a natural event by its general definition. This natural event is a disaster that causes significant damage, loss of life, and other economic effects that will damage the state. The possibility of predicting a natural event such as an earthquake will minimize the reasons mentioned. Data collection, data processing, and data evaluation were carried out in this study. Earthquake forecasting was performed using the data and the RNN (Recurrent Neural Network) method. The study was carried out on seismic data with a magnitude of 3.0 and above belonging to Düzce Province between 1990 and 2022. In order to increase the learning potential of the method, the b and d values of the earthquake were calculated and included in the data set, except for the earthquake magnitude. The determination of earthquakes in a specific time interval in regions of Turkey, the classification of earthquake-related seismic data using artificial neural networks, and the production of predictions for the future reveal the importance of this study.

In the post-PARP inhibitor era, potential changes in tumor biology after maintenance therapy have not been well investigated in recurrent ovarian cancer. We reported a case with alterations in the clinical and histological features of multiple relapsed disease associated with PARP inhibitor maintenance therapy. The patient with high-grade serous carcinoma exhibited BRCA wildtype and homologous recombination proficiency status, and suffered from three recurrences and surgeries accordingly. Olaparib maintenance had been used during the second-line therapy. We compared the differences in clinics and pathology among three recurrences and relapsed lesions. Disease-free survivals were dramatically decreased after the exposure to olaparib. At exploration of quaternary cytoreduction, the relapsed tumor was characterized by a carcinomatosis-like metastasis pattern and an easy tendency of bleeding. Tumor cytopathological changes and alterations were observed in both the tumoral and non-tumoral stroma, among relapsed tumor tissues derived from secondary, tertiary and quaternary cytoreduction. Histopathology indicated hemorrhage, necrosis, atypical tumor cells, massive angiogenesis, and decreased CD8+ tumor-infiltrating lymphocytes, particularly in the third relapsed disease. To our knowledge, this is the first report to show a unique metastatic pattern of angiogenic burst after PARP inhibitor maintenance therapy in ovarian cancer, which seemed to trigger invasive tumor growth and immune suppression. Further prospective studies and translational research focusing cytoreductive surgery after PARP inhibitor could progressively lead to an understanding of the biological behavior and metastatic patterns.

Water quality forecasting is increasingly significant for agricultural management and environmental protection. Enormous amounts of water quality data are increasingly being collected by advanced sensors, which leads to an interest in using data-driven models for predicting trends in water quality. However, the unpredictable background noises introduced during water quality monitoring seriously degrade the performance of those models. Meanwhile, artificial neural networks (ANN) with feed-forward architecture lack the capability of maintaining and utilizing the accumulated temporal information, which leads to biased predictions in processing time series data. Hence, we propose a water quality predictive model based on a combination of Kernal Principal Component Analysis (kPCA) and Recurrent Neural Network (RNN) to forecast the trend of dissolved oxygen. Water quality variables are reconstructed based on kPCA method, which aims to reduce the noise from the raw sensory data and preserve actionable information. With the RNN's recurrent connections, our model can make use of the previous information in predicting the trend in the future. Data collected from Burnett River, Australia was applied to evaluate our kPCA-RNN model. The kPCA-RNN model achieved R2 scores up to 0.908, 0.823 and 0.671 for predicting the concentration of dissolved oxygen in the upcoming 1, 2 and 3 hours, respectively. Compared to current data-driven methods like ANN and SVR, the predictive accuracy of the kPCA-RNN model was at least 8 %, 17 % and 21 % better than the comparative models in these 3 cases. The study demonstrates the effectiveness of the kPAC-RNN modeling technique in predicting water quality variables with noisy sensory data.

Thyroid surgery has been for years one of the most common elective procedures performed in general surgery. In the last 25 years, there have been significant advances in the diagnosis and treatment of thyroid disorders, and new technologies are being implemented. The aim of this study was to ana-lyze 25 years of experience in thyroid surgery in the Department of General, Minimally Invasive and Endocrine Surgery at the Medical University of Wroclaw in terms of demographic changes, indica-tions for surgical treatment, the type of thyroid surgery performed and complications. The impact of recent advances on changes in endocrine surgery was evaluated. For this purpose, clinical material from the years 1996-2020 was analyzed, with a total of 3748 patients (7285 RLN at risk of injury). Period I included the years: 1996-2003, period II: 2011-2015 and 2018-2020. Results: In the last 25 years, the percentage of patients operated on for thyroid cancer has increased threefold (p <0.00001); the extent of thyroid surgery has changed: in 1996-2003, subtotal thyroid operation was performed in more than 90% of patients vs. 1.7% of total thyroid gland surgery; 2011-2015 and 2018-2020 were more than 94% of total thyroid surgery vs. 4% of subtotal surgery (p<0.00001). The increase in the extent of surgery did not affect the percentage of total amount recurrent laryngeal nerve palsy (RLN palsy) (5.17% vs. 4.38% p = 0.1785), it did affect the percentage of transient RLN palsy in group II vs. group I (0.41% vs. 1.34%, p<0.00001), while the percentage of permanent RLN palsy in group I was statistically significantly higher than in group II (4.77% vs. 3.05%, p=0.0016). An increase in the percentage of postoperative clinical hypoparathyroidism in group II was observed: 4.84% vs. 8.93% (p<0.00001). Conclusions: Over 25 years, there has been a significant increase in the number of sur-geries performed for thyroid cancer, the range of surgeries from partial resections to total excision of the thyroid gland has changed, and the increased range of surgeries did not have a statistically signifi-cant effect on the number of vocal fold paralysis, but increased the percentage of postoperative hypo-parathyroidism. Intraoperative neuromonitoring had a significant impact on changing the scope of surgery and preventing complications.

The prediction of stock prices holds significant implications for researchers and investors in evaluating stock value and risk. In recent years, researchers have increasingly replaced traditional machine learning methods with deep learning approaches in this domain. However, the application of deep learning in forecasting stock prices is confronted with the challenge of overfitting. To address the issue of overfitting and enhance predictive accuracy, this study proposes a stock prediction model based on GRU (Gated Recurrent Unit) with reconstructed datasets. This model integrates data from other stocks within the same industry, thereby enriching the extracted features and mitigating the risk of overfitting. Additionally, an auxiliary module is employed to augment the volume of data through dataset reconstruction, thereby enhancing the model’s training comprehensiveness and generalization capabilities. Experimental results demonstrate a substantial improvement in prediction accuracy across various industries.

Objectives: Essential thrombocythemia (ET) is a chronic myeloproliferative neoplasm characterized by elevated platelet counts and an increased risk of thrombotic events, including ischemic strokes. Methods: We conducted a retrospective analysis of data from consecutive ischemic stroke patients with ET between March 2014 and February 2023. Results: This case series describes the clinical presentation, radiological features, and management of five patients with ET-associated ischemic strokes, all harboring the JAK2 mutation. The diverse radiological findings suggest that both large and small vessel diseases may be influenced by the prothrombotic state induced by ET. A significant elevation in platelet count was observed to correlate with the emergence of new acute infarctions in some cases. Conclusions: The study highlights combined use of antiplatelet and cytoreductive therapy in preventing secondary stroke events in patients with ET and JAK2 mutations. The heterogeneity of stroke patterns in this population necessitates a comprehensive understanding of the underlying pathophysiological mechanisms and tailored therapeutic approaches.

Video super-resolution, which utilizes the relevant information of several low-resolution frames to generate high-resolution images, is a challenging task. One possible solution called sliding window method tries to divide the generation of high-resolution video sequences into independent sub-tasks, and only adjacent low-resolution images are used to estimate the high-resolution version of the central low-resolution image. Another popular method named recurrent algorithm proposes to utilize not only the low-resolution images but also the generated high-resolution images of previous frames to generate the high-resolution image. However, both methods have some unavoidable disadvantages. The former one usually leads to bad temporal consistency and requires higher computational cost while the latter method always can not make full use of information contained by optical flow or any other calculated features. Thus more investigations need to be done to explore the balance between these two methods. In this work, a bidirectional frame recurrent video super-resolution method is proposed. To be specific, a reverse training is proposed that the generated high-resolution frame is also utilized to help estimate the high-resolution version of the former frame. With the contribution of reverse training and the forward training, the idea of bidirectional recurrent method not only guarantees the temporal consistency but also make full use of the adjacent information due to the bidirectional training operation while the computational cost is acceptable. Experimental results demonstrate that the bidirectional super-resolution framework gives remarkable performance that it solves the time-related problems when the generated high-resolution image is impressive compared with recurrent-based video super-resolution method.

This paper presents a quantitative approach to poetry, based on the use of several statistical measures (entropy, informational energy, N-gram, etc.) applied to a few characteristic English writings. We found that English language changes its entropy as time passes, and that entropy depends on the language used and on the author. In order to compare two similar texts, we were able to introduce a statistical method to asses the information entropy between two texts. We also introduced a method of computing the average information conveyed by a group of letters about the next letter in the text. We found a formula for computing the Shannon language entropy and we introduced the concept of N-gram informational energy of a poetry. We also constructed a neural network, which is able to generate Byron-type poetry and to analyze the information proximity to the genuine Byron poetry.

Buildings have started to play a critical role in the stability and resilience of modern smart grids, leading to a refocusing of large scale energy management strategies from the supply side to the consumer side. When the buildings integrate local renewable energy generation in the form of renewable energy resources they become prosumers and this reflects into additional complexity into the operation of the interconnected complex energy systems. A class of methods of modelling the energy consumption patterns of the building have recently emerged as black-box input-output approaches with the ability to capture underlying consumption trends. These make use and require large quantities of quality data produces by non-deterministic processes underlying the energy consumption. We present an application of a class of neural networks, namely deep learning techniques for time series sequence modelling with the goal of accurate and reliable building energy load forecasting. The Recurrent Neural Network implementation uses Long Short-Term Memory layers in increasing density of nodes to quantify prediction accuracy. The case study is illustrated on four university buildings from temperate climates over one year of operation using a reference benchmarking dataset that allows replicable results. The obtained results are discussed in terms of accuracy metrics and computational and network architecture aspects and are considered suitable for further used in future in situ energy management at the building and neighbourhood levels.

The work analyses the way in which symptoms and life themes manifest in middle-aged adults diagnosed with major recurrent depression. Specifically, the relationships between symptoms, life themes, and life themes - symptoms have been analyzed. For this purpose, Spearman correlation, and the methods of Latent Semantic Indexing (LSI) were used. Seven symptoms and twenty-six life themes were identified in the patients analyzed as well as similarities of symptoms/life themes (at patient level), the ranking of the importance of symptoms on each life theme (at the level of the group of patients), the rankings of the similarities of life themes in relation to different symptoms or various groups of symptoms (at the level of the group of patients), and dysfunctional cycles of symptoms and life themes (at patient level). The findings only refer to the patients analyzed. Although our findings cannot be generalized, there is a possibility that some of them may also be encountered in other patients. However, the design of the work can be used to initiate other similar studies.

Background: Patients with chronic obstructive pulmonary disease (COPD) often suffer from acute exacerbations. Our objective was to describe recurrent exacerbations in a GP based Swiss COPD cohort, and develop a statistical model for predicting exacerbation. Methods: In a questionnaire-based COPD cohort, demographic and medical data were recorded for 24 months. Data set was split into training (75%) and validation (25%) datasets. A negative binomial regression model was developed using the training dataset to predict the exacerbation rate within 1 year. An exacerbation prediction model was developed, and the overall performance was validated. A nomogram was created to facilitate the clinical use of the model. Results: Of the 229 COPD patients analyzed, 77% of patients had no exacerbation during the follow-up. The best subset in the training dataset found that lower forced expiratory volume, high scores on the MRC dyspnoea scale, exacerbation history, being on combination therapy of LABA+ICS or LAMA+LABA at baseline were associated with a higher rate of exacerbation. When validated, the area-under-curve (AUC) was 0.75 for one or more exacerbations. Calibration was accurate (predicted 0.34 exacerbations vs observed 0.28 exacerbations). Conclusion: Nomograms built from these models can assist clinicians in the decision-making process of their COPD care.

A theoretical model to translate the evolution over time, in early stages, of growth and accumulation of biofilm bacterial mass is introduced. The model implies the solution of a system of differential-difference master equations. The application of an algorithm like Miller´s tree term recurrence, already known for Bessel functions of first kind, allows an exact calculation of the solutions of such equations, for a wide range of parameters values and time. For biofilm model a five term recurrence is deduced and applied in a backwards computation. A suitable normalisation condition completes the reach of the solution.

Infrared thermography has already been proven to be a significant method in non-destructive evaluation since it gives information with immediacy, rapidity and low cost. However, the thorniest issue for wider application of IRT is the quantification. In this work, we proposed a specific depth quantifying technique by employing the Gated Recurrent Units (GRU) in composite material samples via pulsed thermography (PT). Carbon Fiber Reinforced Polymer(CFRP) embedded with flat bottom holes were designed via Finite Element Method (FEM) modeling in order to precisely control the depth and geometrics of the defects. The GRU model automatically quantified the depth of defects presented in the CFRP material. The proposed method evaluated the accuracy and performance of synthetic CFRP data from FEM for defect depth predictions.

The effects of four cycles of recurrent selection on simple sequence repeat (SSR) marker allele frequencies and population structure were examined in the Maksimir 3 Synthetic (M3S) maize population (Zea mays L.). Genotyping of 32 plants from each cycle of selection at 38 SSR loci revealed that the mean number of alleles per locus and the mean expected heterozygosity were preserved over cycles of selection, indicating maintenance of sufficient genetic variability in the population needed for future genetic gain. Waples test of selective neutrality revealed that genetic drift was the main force in changing allele frequencies in the population. The proportion of selectively nonneutral loci in single cycles of selection varied between 16% and 37%. Some nonneutral loci shared the same genomic locations with previously published QTLs controlling important agronomic traits. Between 5% and 29% of loci were found to be in significant Hardy-Weinberg (HW) disequilibrium with the majority showing an excess of homozygosity. Excess of homozygosity at several loci was highly consistent across cycle populations suggesting positive assortative mating as the possible cause of the observed HW disequilibrium. Linkage disequilibrium (LD) tests revealed that the M3S population was essentially in linkage equilibrium. The proportion of pairs of loci in significant LD varied across cycles of selection between 0.1% and 1.8% probably due to the effects of genetic drift and epistatic selection.

The aim of this study is to describe the cognitive and speech results obtained after growth hormone (GH) treatment and neurorehabilitation in a man that suffered a traumatic brain injury (TBI). 17 months after the accident, the patient was treated with growth hormone (GH), together with neurostimulation and speech therapy. At admission, the left vocal cord revealed paralyzed, in the paramedian position, a situation compatible with a recurrent nerve injury. Clinical and rehabilitation assessments revealed a prompt improvement in speech and cognitive functions, and following completion of treatment, endoscopic examination showed recovery of vocal cord mobility. These results, together with previous results from our group, indicate that GH treatment is safe and effective for helping neurorehabilitation in chronic speech impairment due to central laryngeal paralysis, as well as impaired cognitive functions.

Benign recurrent intrahepatic cholestasis (BRIC) is a rare genetic cause of cholestasis. It is considered as part of inherited intrahepatic cholestasis syndromes, such as progressive familial intrahepatic cholestasis (PFIC), and intrahepatic cholestasis of pregnancy. BRIC is presented in infancy or in early adulthood. It is characterized by exacerbations and remissions of jaundice with accompanying intense itching, lasting from weeks to years throughout lifetime. Normal gamma-glutamyl transferase (GGT) is a characteristic laboratory finding. Contrary to PFIC, which may progress to cirrhosis, BRIC does not progress to chronic liver disease or cirrhosis. However, incessant episodes of cholestasis result in marked reduction in quality of life and distinct mutations increase the risk of hepatobiliary malignancy. In intervals between the exacerbations, the histological findings of centrilobular cholestasis together with the abnormal laboratory parameters return to normal. In this context, liver biopsy might be avoided. In this review, we will focus on the genetic aspects of BRIC, its pathophysiology, clinical presentation, and prognosis of this autosomal recessive genetically determined cholestatic disorder. Moreover, triggering factors as well as treatment options will be further elucidated.

Rationale: Chronic Granulomatous Disease (CGD) is a rare primary immunodeficiency. Although most mutations are linked to the X chromosome, therefore affecting males, there are autosomal recessive forms that therefore affect both sexes. Clinically it is characterized by recurrent and severe bacterial and fungal infections, with formation of granulomas, due to the inability of phagocytes to generate reactive oxygen compounds, necessary for the intracellular death of phagocytosed microorganisms. Patient concerns: We reported a 12-year-old boy with pneumonia with pneumatocele of possible staphylococcal etiology at one year of age, bilateral orchitis secondary to Salmonella with hematogenous spread, osteomyelitis due to Salmonella, gastroenteritis due to Salmonella enteritidis, liver abscesses due to Staphycococcus aureus and lymphomatoid papulosis and pyodermititis. Diagnoses: In the biochemical diagnosis, the presence of yellow dye (NBT) and absence of dark bluish black dye (formazan) was evidenced. Regarding the genetic diagnosis, the presence of the mutation in exon 10 of CYBB (OMIM 306400) (Xp21.1) that encodes the gp91phox protein was demonstrated. Interventions and outcomes: Treatment was started with trimethoprim/sulfamethoxazole 400/80 mg every 24 hours via oral, and antifungal prophylaxis with itraconazole 200 mg every 24 hours orally is added, after which he does not present serious bacterial or fungal infections, not being able to administer immunomodulatory treatment with recombinant human Interferon gamma-1b. Lessons: The clinical finding of frequent severe infections associated with granulomas, CGD should be suspected. Treatment of this primary immunodeficiency is trimethoprim/sulfamethoxazole and itraconazole. Importantly, it indicated that the mutation in exon 10 of CYBB (Xp21.1) that encodes the gp91phox might influence the pathogenesis of CGD.

Autoinflammatory disorders encompass a wide range of conditions with systemic and neuro-logical symptoms, which can be acquired or inherited. These diseases are characterized by an abnormal response of the innate immune system, leading to an excessive inflammatory reac-tion. On the other hand, autoimmune diseases result from dysregulation of the adaptive im-mune response. Disease flares are characterized by systemic inflammation affecting the skin, muscles, joints, serosa, and eyes, accompanied by unexplained fever and elevated acute phase reactants. Autoinflammatory syndromes can present with various neurological manifestations, such as aseptic meningitis, meningoencephalitis, sensorineural hearing loss, and others. Early recognition of these manifestations by general neurologists can have a significant impact on the prognosis of patients. Timely and targeted therapy can prevent long-term disability by reducing chronic inflammation. This review provides an overview of recently reported neuroinflam-matory phenotypes, with a specific focus on genetic factors, clinical manifestations, and treatment options. General neurologists should have a good understanding of these important diseases.

We have developed deep learning models for automatic identification of the maternal heart rate (MHR) and, more generally, false signals (FSs) on fetal heart rate (FHR) recordings. The models can be used to preprocess FHR data prior to automated analysis or as a clinical alert system to assist the practitioner. Three models were developed and used to detect (i) FSs on the MHR channel (the FSMHR model), (ii) the MHR and FSs on the Doppler FHR sensor (the FSDop model), and (iii) FSs on the scalp ECG channel (the FSScalp model). The FSDop model was the most useful because FSs are far more frequent on the Doppler FHR channel. All three models were based on a multilayer, symmetric, GRU and were trained on data recorded during the first and second stages of delivery. The FSMHR and FSDop models were also trained on antepartum recordings. The training dataset contained 1030 expert-annotated periods (mean duration: 36 min) from 635 recordings. In an initial evaluation of routine clinical practice, 30 fully annotated recordings for each sensor type (mean duration: 5 h for MHR and Doppler sensors, and 3 h for the scalp ECG sensor) were analyzed. The sensitivity, positive predictive value (PPV) and accuracy were respectively 62.20%, 87.1% and 99.90% for the FSMHR model, 93.1%, 95.6% and 99.68% for the FSDop model, and 44.6%, 87.2% and 99.93% for the FSScalp model. We built a second test dataset with a more solid ground truth by selecting 45 periods (lasting 20 min, on average) on which the Doppler FHR and scalp ECG signals were recorded simultaneously. Using scalp ECG data, the experts estimated the true FHR value more reliably and thus annotated the Doppler FHR channel more precisely. The models achieved a sensitivity of 53.3%, a PPV of 62.4%, and an accuracy of 97.29%. In comparison, two experts (blinded to the scalp ECG data) respectively achieved a sensitivity of 15.7%, a PPV of 74.3%, and an accuracy of 96.91% and a sensitivity of 60.7%, a PPV of 83.5% and an accuracy of 98.24%. Hence, the models performed at expert level (better than one expert and worse than the other), although a well-trained expert with good knowledge of FSs could probably do better in some cases. The models and datasets have been included in the Fetal Heart Rate Morphological Analysis open-source MATLAB toolbox and can be used freely for research purposes.

Infrared thermography has already been proven to be a significant method in non-destructive evaluation since it gives information with immediacy, rapidity, and low cost. However, the thorniest issue for the wider application of IRT is quantification. In this work, we proposed a specific depth quantifying technique by employing the Gated Recurrent Units (GRU) in composite material samples via pulsed thermography (PT). Finite Element Method (FEM) modeling provides the economic examination of the response pulsed thermography. In this work, Carbon Fiber Reinforced Polymer (CFRP) specimens embedded with flat bottom holes are stimulated by a FEM modeling (COMSOL) with precisely controlled depth and geometrics of the defects. The GRU model automatically quantified the depth of defects presented in the stimulated CFRP material. The proposed method evaluated the accuracy and performance of synthetic CFRP data from FEM for defect depth predictions.

The onset of coronavirus disease 2019 (COVID-19), an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has sparked unprecedented change. Due to the public health guidelines imposed during the COVID-19 pandemic, there is no longer sufficient street traffic for remaining buskers to generate sufficient revenue, leading a majority of street musicians to pursue remote music production. However, real-time music production is notoriously difficult due to the excessively high latencies that current video call platforms such as Zoom and Google Meet harbor. In this paper, we propose an architecture for a platform with end-to-end, near-lossless audio transmission tailored specifically to online joint music production, called Latent Space. We discuss the usage of a recurrent autoencoder with sequence-aware encoding (RAES) and a 1D convolutional layer for audio compression, which we dub ClefNet, as well as propose a new evaluation metric for naive autoencoders (AEs), MSE-DTW loss, which combines the traditional mean square error (MSE) loss function with dynamic time warping (DTW) to prevent an increase in loss when the target sequence predicted by the AE is strictly a temporal variation of the source sequence. Moreover, we detail the logistics of a live system implementation which uses the Web Audio API to extract raw audio samples in real-time to feed into our client-side model before relaying the traffic using peer-to-peer WebRTC technology. The Latent Space platform can be accessed at https://latent-space.tech, and the code and data can be found under the MIT License at https://github.com/rvignav/ClefNet.

This study uses machine learning to predict the convergence results of the Duffing equation with and without damping. The Duffing equation represents a nonlinear second-order differential equation with interesting behavior in undamped free vibration and forced vibration with damping. Convergence alternates randomly between 1 and -1 in undamped free vibration, depending on initial conditions. For forced vibration with damping, multiple factors influence vibration patterns. We utilize the fourth-order Runge-Kutta method to collect convergence results for both conditions. Machine learning techniques, specifically the long short term memory (LSTM) and LSTM-Neural Network (LSTM-NN) method, are employed to predict these convergence values. The LSTM-NN model is a hybrid approach that combines the LSTM method with the addition of hidden layers of neurons. Both the LSTM and LSTM-NN models are thoroughly explored and analyzed in this research. The research process involves three stages: data preprocessing, training, and verification. The results show that the LSTM-NN model becomes more adept at predicting binary datasets, boasting an impressive accuracy of up to 98%. However, when it comes to predicting multiple solutions, the traditional LSTM method outperforms the LSTM-NN approach.

The intracranial pressure (ICP) signal, as monitored on patients in intensive care units, contains pulses of cardiac origin where P1 and P2 subpeaks can often be observed. When calculable, the ratio of their relative amplitudes is an indicator of the patient’s cerebral compliance. This characterization is particularly informative on the overall state of the cerebrospinal system. We developed a recurrent neural network-based framework for P2/P1 ratio computation that only takes a raw ICP signal as an input. Two tasks are performed, namely pulse classification and subpeak designation. Pulse classification was achieved with an area under the curve of 0.90 on a 4,344-pulse testing dataset, while the peak designation algorithm identified pulses with a P2/P1 ratio > 1 with a 97.3% accuracy. Although it still needs to be evaluated on a larger number of labeled recordings, our automated P2/P1 ratio calculation framework appears to be a promising tool for improving bedside monitoring devices.

We propose a chemical language processing model to predict polymers’ glass transition temperature (Tg) through a polymer language (SMILES, Simplified Molecular Input Line Entry System) embedding and recurrent neural network. This model only receives the SMILES strings of polymer’s repeat units as inputs and considers the SMILES strings as sequential data at the character level. Using this method, there is no need to calculate any additional molecular descriptors or fingerprints of polymers, and thereby, being very computationally efficient and simple. More importantly, it avoids the difficulties to generate molecular descriptors for repeat units containing polymerization point `*’. Results show that the trained model demonstrates reasonable prediction accuracy on unseen polymer’s Tg. Besides, this model is further applied for high-throughput screening on an unlabeled polymer database to identify high-temperature polymers that are desired for applications in extreme environments. Our work demonstrates that the SMILES strings of polymer’s repeat units can be used as an effective feature representation to develop a chemical language processing model for predictions of Tg. The framework of this model is general and can be used to construct structure-property relationships for other polymer’s properties.

Not only does this paper present a novel type-2 fuzzy system for identification and behavior prognostication of an experimental solar cell set and a wind turbine, but also it brings forward an exquisite technique to acquire an optimal number of membership functions and the corresponding rules. It proposes a seven-layered NCPRT2FS. For fuzzification in the first two layers, Gaussian type-2 fuzzy membership functions with uncertainty in the mean, are exploited. The third layer comprises rule definition and the forth one embeds fulfillment of type reduction. The three last remained layers are the ones in which resultant left–right firing points, two end-points and output all get assessed correspondingly. It should not be neglected off the nutshell that recurrent feedback at the fifth layer exerts delayed outputs ameliorating efficiency of the suggested NCPRT2FS. Later in the paper, a modern structural learning, established on type-2 fuzzy clustering, is held forth. An adaptively rated learning back-propagation algorithm is extended to adjust the parameters ensuring the convergence as well. Eventually, solar cell photo-voltaic and wind turbine are deemed as case studies. The experimental data are exploited and the consequent yields emerge so persuasive.

(1) Background: The aim of this study is to investigate the diagnostic-therapeutic problem of pelvic recurrence of rectum cancer, highlighting current surgical standards and the possible role of the minimally invasive approach. This retrospective analysis of our surgical case study wishes to enter this debate while suggesting a possible line of action, based on the site of recurrence; (2) Methods: We examined, retrospectively, all the patients diagnosed, between 2008 and 2018, with cancer of the rectum at the "Pietro Valdoni" Department of Surgery and monitored their follow-up for 5 years. The sample consisted of 368 patients with rectal neoplasm, 136 females and 232 males, with an average age of 65.8 (ranging from 37 to 86); (3) Results: In 103 of the cases, the neoplasm was located in the upper rectum (28%), in 119 cases in the middle rectum (32.3%), in 102 cases in the lower rectum (27.7%), in 31 cases (8.4%) at the level of the right/sigma junction and in 13 cases it was not possible to define the site with certainty (3.5%); (4) Conclusions: The discussion re-mains open as to which approach is best at surgical level, whether laparoscopic, robotic or open-air. Our experience informs us that the most dangerous site remains on the anastomotic site.

(Background) the aim was to determine related factors to recurrent vitreous hemorrhage (RVH) in a sample of proliferative diabetic retinopathy (PDR) patients. (Methods) A retrospective, review-based study. We studied 183 eyes from 121 type 2 diabetes patients with PDR. We recorded diabetes duration, history of hypertension, retinal photocoagulation status, the posterior vitreous status, the mean HbA1c, mean hemoglobin, the renal function, and the systemic complications related to diabetes. We also recorded the use of ranibizumab prior to vitrectomy and the following surgical variables: the application of segmentation and diathermy on fibrovascular proliferative tissue, the use of silicone oil, and the occurrence of surgical complications, to study which independent variables were significantly related to the presence of RVH. (Results) Duration of diabetes (P= 0.028), hemoglobin (P=0.02), status of the posterior vitreous (P=0.03), retinal photocoagulation (P=0.002) and use of segmentation surgery technique (P=0.003) have significant link to the presence of RVH. In addition, patients with diabetic polyneuropathy, myocardial infarction and ischemia in lower limbs had more vitreous hemorrhage events (p&lt;0.001). (Conclusions) Patients with PDR and with longer diabetes duration, anemia, attached posterior vitreous, deficient retinal photocoagulation, and previously cardiovascular events, were more prone to RVH.

This study proposes a mobile positioning method which adopts recurrent neural network algorithms to analyze the received signal strength indications from heterogeneous networks (e.g., cellular networks and Wi-Fi networks) for estimating the locations of mobile statioThis study proposes a mobile positioning method which adopts recurrent neural network algorithms to analyze the received signal strength indications from heterogeneous networks (e.g., cellular networks and Wi-Fi networks) for estimating the locations of mobile stations. The recurrent neural networks with multiple consecutive timestamps can be applied to extract the features of time series data for the improvement of location estimation. In practical experimental environments, there are 4,525 records, 59 different base stations, and 582 different Wi-Fi access points detected in Fuzhou University in China. The lower location errors can be obtained by the recurrent neural networks with multiple consecutive timestamps (e.g., 2 timestamps and 3 timestamps); the experimental results can be observed that the average error of location estimation was 9.19 meters by the proposed mobile positioning method with 2 timestamps.

Temporal lobe epilepsy is a common, chronic disorder with spontaneous seizures that is often refractory to drug therapy. A potential cause of temporal lobe epilepsy is primary brain injury, making prevention of epileptogenesis after the initial event an optimal method of treatment. Despite this, no preventive therapy for epilepsy is currently available. The purpose of this study was to evaluate the effects of anakinra, lamotrigine, and their combination on epileptogenesis using the rat lithium-pilocarpine model of temporal lobe epilepsy. The study showed that the treated and untreated animals showed no significant difference in the number and duration of seizures. However, the severity of seizures was significantly reduced after treatment. Anakinra and lamotrigine, alone or in combination, significantly reduced neuronal loss in the CA1 hippocampus compared to the control group. However, the drugs administered alone were found to be more effective for CA3 than their combination. The treatment alleviated the impairments in activity level, exploratory behavior and anxiety, but had a relatively weak effect on TLE-induced impairments in social behavior and memory. The efficacy of the combination treatment did not differ from that of anakinra and lamotrigine monotherapy. These findings suggest that anakinra and lamotrigine, alone or in combination, may have clinical utility in preventing epileptogenesis.

In recent years, unmanned aerial vehicles (UAVs) have become a valuable platform for many applications, including communication networks. UAV-enabled wireless communication faces challenges in complex urban and dynamic environments. UAVs can suffer from power limitations and path losses caused by non-line-of-sight connections, which may hamper better communication performance. To address these issues, reconfigurable intelligent surfaces (RIS) have been proposed as a helpful tool to enhance UAV communication networks. However, due to the high mobility of UAV, complex channel environments, and dynamic RIS configurations, it is challenging to estimate the link quality of ground users. In this paper, we propose a link quality estimation model using a gated recurrent unit (GRU) to assess the link quality of ground users for a multi-user RIS-assisted UAV-enabled wireless communication system. Our proposed framework uses a time series of user channel data and RIS phase shift information to estimate the quality of the link for each ground user. Simulation results showed that the proposed GRU model effectively and accurately estimates the link quality of the ground users in the RIS-assisted UAV-enabled wireless communication network.

ABSTRACT: Recurrent glioblastoma (rGBM) is a highly aggressive form of brain cancer that poses a significant challenge for treatment in neurooncology, and the survival status of patients after relapse usually means rapid deterioration, and also the leading cause of death among patients. In recent years, immunotherapy has emerged as a promising strategy for the treatment of recurrent glioblastoma by stimulate the body's immune system to recognize and attack cancer cells , which could be used as a in combination with other treatments such as surgery, radiation, and chemotherapy to improve outcomes for patients with recurrent glioblastoma, This therapy combines several key methods such as the use of monoclonal antibodies, chimeric antigen receptor T cell (CAR-T) therapy, checkpoint inhibitors, oncolytic viral therapy cancer vaccines, and combination strategies. In this review, we mainly document the latest immunotherapies for the treatment of glioblastoma and focus on the rGBM especially.

. Accurate coastal wave direction and speed forecasts are crucial in coastal and marine engineering, marine energy, maritime transport, fisheries, naval navigation, environmental research, and risk management. Approximately 269 km of Brunei Darussalam's coastline can generate between 15 and 126 GW of wave energy. As part of the preliminary feasibility study of wave energy harvesting in Brunei Darussalam and net zero commitment, in this study, we used two numerical methods, namely, finite difference and spectral element methods, for modeling and simulation of wave speed and direction. The mean error between numerical and analytical solutions was calculated in each simulation. Explanatory data analysis was used to provide insight into the study data. We then proposed wave direction and speed forecasting models using Long Short-Term Memory (LSTM) stacking on the data computed from the Acoustic Doppler Current Profiler (ADCP) sensor data. A univariate time series forecasting approach was adopted for this research. KerasTuner hyperparameter tuning API was used for tuning and optimizing hyperparameters, leading us to build models with the least training and test errors. Seven separate prediction experiments were conducted for wave speed and direction in degree and radian units for the next 1, 3, 6, 8, 10, 12, and 24 hours, respectively. Mean squared error (MSE) was used as a metric for both training and testing. The experimental results show that wave speed forecast has the lowest MSEs compared to direction, regardless of the unit of measure, but has a longer runtime. Moreover, the forecast of direction in the degree unit has the least errors compared to the radian unit; the running time of the latter is higher than that of the former. In the future, we intend to use advanced multivariate time series techniques to forecast wave speed and direction.

Machine learning (ML) is increasingly useful as data grows in volume and accessibility as it can perform tasks (e.g. categorisation, decision making, anomaly detection, etc.) through experience and without explicit instruction, even when the data are too vast, complex, highly variable, full of errors to be analysed in other ways , . Thus, ML is great for natural language, images, or other complex and messy data available in large and growing volumes. Selecting a ML algorithm depends on many factors as algorithms vary in supervision needed, tolerable error levels, and ability to account for order or temporal context, among many other things. Importantly, ML methods for explicitly ordered or time-dependent data struggle with errors or data asymmetry. Most data are at least implicitly ordered, potentially allowing a hidden `arrow of time’ to affect non-temporal ML performance. This research explores the interaction of ML and implicit order by training two ML algorithms on Twitter data before performing automatic classification tasks under conditions that balance volume and complexity of data. Results show that performance was affected, suggesting that researchers should carefully consider time when selecting appropriate ML algorithms, even when time is only implicitly included.

The vast majority of autonomous driving systems are limited to applications on roads with clear lane markings and are implemented using commercial grade sensing systems coupled with specialized graphic accelerator hardware. This research reviews an alternative approach for autonomously steering vehicles that eliminates the dependency on road markings and specialized hardware. A combination of machine vision, machine learning and artificial intelligence based on popular pre-trained Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) was used to drive a vehicle along roads lacking lane markings (unmarked roads). The team developed and tested this approach on the ACTor (Autonomous Camus TranspORt) vehicle - an autonomous vehicle development and research platform coupled with a low-cost webcam based sensing system and minimal computational resources. The proposed solution was evaluated on real-world roads and varying environmental conditions. It was found that this solution may be used to successfully navigate unmarked roads autonomously with acceptable road following behavior.

Wheat (Triticum aestivum L.) breeding programs can take over a decade to release a new variety. However, new methods of selection such as genomic selection (GS) must be integrated to decrease the time it takes to release new varieties to meet the demand of a growing population. The implementation of GS into breeding programs is still being explored, with many studies showing its potential to change wheat breeding through achieving higher genetic gain. In this review, we explore the integration of GS for a wheat breeding program by redesigning the traditional breeding pipeline to implement GS. We propose implementing a two-part breeding strategy by differentiating between population improvement and product development. The implementation of GS in the product development pipeline can be integrated into most stages and can predict within and across breeding cycles. Additionally, we explore optimizing the population improvement strategy through GS recurrent selection schemes to reduce crossing cycle time and significantly increase genetic gain. The recurrent selection schemes can be optimized for parental selection, maintenance of genetic variation, and optimal cross-prediction. Overall, we outline the ability to increase the genetic gain of a breeding program by implementing GS and a two-part breeding strategy.

INTRODUCTION The Anterior cruciate Ligament Reconstruction (ACLR) is a recurrent successfully surgery, however the literature shows from 3 to 15% of failure. Many recurrent ACL roptures are due to technical surgical mistakes, however, even when the neo-ligament is correctly located and well tensioned, a high rate of partial anterolateral rotatory instability can be clinically detected in those patients. We hypothesized that a recurrent ACL injury would often be associated with antero-lateral ligament lesion and so with an high grade pivoting. The purpose of our retrospective study is to demonstrate if, in the recurrent ACLR surgery, the Anterolateral ligament reconstruction (ALLR) is necessary due to restore a correct rotatory stability of the knee, to improve clinical outcomes and to determine a better return to sport activities. METHODS We included a total of 63 patients undergone to ACLR after a LCA recurrent lesion from January 2018 to January 2020 performed by the same surgeon. The pooled data passed through exclusion criteria and we created two different groups: group A with 11 patients (whom undergone to ACLR with bone-patellar-tendon-bone autograft) and group B with 12 patients (whom undergone to ACLR with bone-patellar-tendon-bone autograft + ALLR). A single operator (different from the surgeon) has clinically evaluated the patients with a 18.75 +/- 8.27 month follow-up. LYSHOLM score, IKDC and ACL-RSI scale clinical questionnaires were used to subjectively evaluation. Pivot-shift test and Rolimeter (Aircast Europa) arthrometric evaluation were used for an objective evaluation. All patients have done a knee XR due to confirm the correct bone tunnels direction. All data were meta-analyzed with the IBM-SPSS software (for MAC, version 22.0) and the results were published as media +/- standard deviation or Standard Error or percentage prevalence, when appropriate. To compare all the clinicals and scores data, we used T-Test for same data, Chi-Square for the percentage data. P-values under 0,05 were considered as significant. RESULTS Our study shows as in the recurrent ACLR surgery, the associated ACLR+ALLR is totally necessary and complementary due to statistically significantly reduction of the remaining anterolateral rotatory instability (p-value under 0.05) CONCLUSION Looking at the recent literature and the preliminary results of this study, the anatomical ALL reconstruction associated with the ACLR, should have been chosen as a better surgery practice in the revision surgery due to reduce anterolateral rotatory instability and the percentage of failure, to increase the odds of graft integrations, the clinical outcomes and the percentage of the return to the same level of sports. The indication of this surgery practice should have been custom-made, because it is mandatory to consider the functional necessary levels for the patient and because the rehabilitation needs time, dedication and commitment to achieve the best clinical and functional results.

SQL injection attacks are one of the most common types of attacks on web applications. These attacks exploit vulnerabilities in the application’s database access mechanisms, allowing attackers to execute unauthorized SQL queries. In this study, we propose an architecture for detecting SQL injection attacks using a recurrent neural network (RNN) autoencoder. The proposed architecture was trained on a publicly available dataset of SQL injection attacks. Then compared with several other machine learning models, including ANN, CNN, Decision Tree, Naïve Bayes, SVM, Random Forest, and Logistic Regression. The experimental result showed that the proposed approach achieved an accuracy of 94% and an F1 score of 92%, which demonstrate its effectiveness in detecting QL injection attacks with high accuracy in comparison with other models covered in the study.

This paper realizes early bearing fault warning through bearing fault time series prediction, and proposes a bearing fault time series prediction model based on optimized maximum correlation kurtosis deconvolution (MCKD) and long short-term memory (LSTM) recurrent neural network to ensure bearings operation reliability. The model is based on lifecycle vibration signal of the bearing, to begin, the cuckoo search (CS) is utilized to optimize the parameter filter length L and deconvolution period T of MCKD, taking into account the influence and periodicity of the bearing time series, the fault impact component of the optimized MCKD deconvolution time series is improved. Then select the LSTM learning rate α depending on deconvolution time series. Finally, the dataset obtained through various preprocessing approaches are used to train and predict the LSTM model. The average prediction accuracy of the optimized MCKD-LSTM model is 26 percent higher than that of the original time series, proving the efficiency of this method, and the prediction results track the real fault data well, according to the XI'AN JIAOTONG University XJTU-SY bearing dataset.

Several studies have analyzed human gait data obtained from inertial gyroscope and accelerometer sensors mounted on different parts of the body. In this article, we take a step further in gait analysis and provide a methodology for predicting the movements of the missing parts of the legs. In particular, we propose a method, called GaIn, to control non-invasive, robotic, prosthetic legs. GaIn can infer the movements of both missing shanks and feet for humans suffering from double trans-femoral amputation using biologically inspired recurrent neural networks. Predictions are performed for casual walking related activities such as walking, taking stairs, and running based on thigh movement. In our experimental tests, GaIn achieved a 4.55 degree prediction error for shank movements on average. However, a patient's intention to stand up and sit down cannot be inferred from thigh movements. In fact, intention causes thigh movements while the shanks and feet remain roughly still. The GaIn system can be triggered by thigh muscle activities measured with electromyography (EMG) sensors to make robotic prosthetic legs perform standing up and sitting down actions. The GaIn system has low prediction latency and is fast and computationally inexpensive to be deployed on mobile platforms and portable devices.

Accurate forecasting of wind power generation plays a key role in improving the operation and management of a power system network and thereby its reliability and security. However, predicting wind power is complex due to the existence of high non-linearity in wind speed that eventually relies on prevailing weather conditions. In this paper, a novel hybrid deep learning model is proposed to improve the prediction accuracy of very short-term wind power generation for the Bodangora Wind Farm located in New South Wales, Australia. The hybrid model consists of convolutional layers, gated recurrent unit (GRU) layers and a fully connected neural network. The convolutional layers have the ability to automatically learn complex features from raw data while the GRU layers are capable of directly learning multiple parallel sequences of input data. The data sets of five-minute intervals from the wind farm are used in case studies to demonstrate the effectiveness of the proposed model against other advanced existing models, including long short-term memory (LSTM), GRU, autoregressive integrated moving average (ARIMA) and support vector machine (SVM), which are tuned to optimise outcome. It is observed that the hybrid deep learning model exhibits superior performance over other forecasting models to improve the accuracy of wind power forecasting, numerically, up to 1.59 per cent in mean absolute error, 3.73 per cent in root mean square error and 8.13 per cent in mean absolute percentage error.

Celiac disease (CD) is an immune-mediated gluten-induced enteropathy that affects predisposed individuals of all ages. Many patients with CD do not report gastrointestinal symptoms making it difficult to reach an early diagnosis. On the other hand, CD is related to a wide spectrum of extra-intestinal manifestations, being dermatitis herpetiformis (DH) the best characterized. These associated conditions may be the clue for reaching the diagnosis of CD. Over the last years, there have been multiple reports of the association between CD and several cutaneous manifestations that may improve with a gluten-free diet (GFD). The presence of some of these skin diseases, even in absence of gastrointestinal symptoms, should give rise to an appropriate screening for CD. The aim of this paper is to describe the different cutaneous manifestations that have been associated to CD and the possible mechanisms involved.

We propose an advanced filtering scheme based on Recurrent Neural Networks (RNNs) and Deep Learning to enable efficient control strategies for Vertical Farming (VF) applications.  We demonstrate that the best RNN model incorporates five neuron layers, with the first and second containing ninety Long Short-Term Memory neurons.  The third layer implements one Gated Recurrent Units neuron.  The fourth segment incorporates one RNN network, while the output layer is designed by using a single neuron exhibiting a rectified linear activation function.  By utilizing this RNN digital filter, we introduce two variations: (1) A scaled RNN model to tune the filter to the signal of interest, and (2) A moving average filter to eliminate harmonic oscillations of the output waveforms. The RNN models are contrasted with conventional digital Butterworth, Chebyshev I, Chebyshev II, and Elliptic Infinite Impulse Response (IIR) configurations.  The RNN digital filtering schemes avoid introducing unwanted oscillations, which makes them more suitable for VF than their IIR counterparts.  Finally, by utilizing the advanced features of scaling of the RNN model, we demonstrate that the RNN digital filter can be pH selective, as opposed to conventional IIR filters.

An abdominal aortic aneurysm (AAA) is usually asymptomatic until rupture, which is associated with extremely high mortality. Consequently, early detection of AAAs is of paramount importance in reducing mortality; however, most AAAs are detected by medical imaging incidentally. The aim of this study was to investigate the feasibility of machine learning-based pulse wave (PW) analysis for the early detection of AAAs using a database of in silico PWs. PWs in the large systemic arteries were simulated using one-dimensional blood flow modelling. A database of in silico PWs representative of subjects (aged 55, 65 and 75 years) with different AAA sizes was created by varying the AAA-related parameters with major impacts on PWs – identified by parameter sensitivity analysis – in an existing database of in silico PWs representative of subjects without AAA. Then, a machine learning architecture for early detection of AAAs was proposed, which was trained and tested using the new in silico PW database. The parameter sensitivity analysis revealed that the AAA maximum diameter and stiffness of the large systemic arteries were the dominant AAA-related biophysical properties that significantly influence the PW. The simulated PW indexes extracted from the database showed that the PW was not only influenced by the presence of an AAA but was also significantly affected by multiple cardiovascular parameters that compromised the detection of AAAs by using individual PW indexes. Alternatively, the trained machine learning model performed well in classifying normal and AAA conditions using digital photoplethysmogram PWs from the database. These findings suggest that machine learning-based PW analysis is a promising approach for AAA screening using PW signals acquired by wearable devices.

The article analyzes the problems connected with ensuring the coordinated operation of slipway drives that arise during the launch of a ship. The dynamic model of load of the electric drive of the ship's cart is obtained taking into account the peculiarities of the construction of the ship-lifting complex, which allows to analyze the influence of external factors and random influences during the entire process of launching the ship. A linearized mathematical model of the dynamics of a complex vessel movement in the process of descent in the space of states is developed, which allows to identify the mode of operation of the multi-drive system, taking into account its structure. The analysis of efficiency of application of recurrent methods of identification (stochastic approximation and least squares) of the parameters of the linearized model in the space of states is carried out. A decision support system has been developed in the automated system of operational control by the module for estimating the situation and the control synthesis to ensure a coherent motion of a complex ship-carts object in a two-phase environment.

The Bayes filters, such as Kalman and particle filters, have been used in sensor fusion to integrate two sources of information and obtain the best estimate of the unknowns. Efficient integration of multiple sensors requires deep knowledge of their error sources and it is not trivial for complicated sensors, such as Inertial Measurement Unit (IMU). Therefore, IMU error modelling and efficient integration of IMU and Global Navigation Satellite System (GNSS) observations has remained a challenge. In this paper, we develop deep Kalman filter to model and remove IMU errors and consequently, improve the accuracy of IMU positioning. In other words, we add modelling step to the prediction and update steps of Kalman filter and the IMU error model is learned during integration. Therefore, our deep Kalman filter outperforms Kalman filter and reaches higher accuracy.

We present here a case of complex uterine anomaly – Obstructed HemiVagina with Ipsilateral Renal Agenesis (OHVIRA), also known as Herlyn-Werner-Wunderlich syndrome in a 13-year-old girl with the history of recurrent urinary tract infections (rUTI). In emergency room, trans-abdominal sonography revealed ovarian cyst and renal agenesis, without any suspicion of vaginal obstruction. This led to delay in the diagnosis of this uncommon anomaly. Finally, MRI findings confirmed the presence of OHVIRA syndrome. As the congenital anomalies of kidney and urinary tract (CAKUT) are in almost one third of cases associated with genital malformations, urologist should carefully search for patients with rUTI. The patient underwent simultaneous laparoscopy and vaginoscopy, what was in our opinion the most appropriate therapeutic decision. In this article we are also going to discuss the role of laparoscopy in the management of OHVIRA syndrome as well as other surgical techniques described in the literature.

An effective process monitoring strategy is a requirement for meeting the challenges posed by increasingly complex products and manufacturing processes. To address these needs, this study investigates a comprehensive scheme based on classical machine learning methods, deep learning algorithms, and feature extraction and selection techniques. In a first step, a novel deep learning architecture based on convolutional neural networks (CNN) and gated recurrent units (GRU) is introduced to predict the local weld quality based on mid-wave infrared (MWIR) and near-infrared (NIR) image data. The developed technology is used to discover critical welding defects including lack of fusion (false friends), sagging and lack of penetration, and geometric deviations of the weld seam. Additional work is conducted to investigate the significance of various geometrical, statistical, and spatio-temporal features extracted from the keyhole and weld pool regions. Furthermore, the performance of the proposed deep learning architecture is compared to that of classical supervised machine learning algorithms, such as multi-layer perceptron (MLP), logistic regression (LogReg), support vector machines (SVM), decision trees (DT), random forest (RF) and k-Nearest Neighbors (kNN). Optimal hyperparameters for each algorithm are determined by an extensive grid search. Ultimately, the three best classification models are combined into an ensemble classifier that yields the highest detection rates and achieves the most robust estimation of welding defects among all classifiers studied, which is validated on previously unknown welding trials.

Recent improvements in deepfake creation have made deepfake videos more realistic. Moreover, open-source software has made deepfake creation more accessible, which reduces the barrier to entry for deepfake creation. This could pose a threat to the people’s privacy. There is a potential danger if the deepfake creation techniques are used by people with an ulterior motive to produce deepfake videos of world leaders to disrupt the order of countries and the world. Therefore, research into the automatic detection of deepfaked media is essential for public security. In this work, we propose a deepfake detection method using upper body language analysis. Specifically, a many-to-one LSTM network was designed and trained as a classification model for deepfake detection. Different models were trained by varying the hyperparameters to build a final model with benchmark accuracy. We achieved 94.39% accuracy on the deepfake test set. The experimental results showed that upper body language can effectively detect deepfakes.

Cisplatin (CDDP) is a platinum-based drug effective against various cancers, including lung, bladder, prostate, ovarian, esophageal, stomach, and cervical cancer and malignant lymphoma. It plays a central role as the first choice in current anticancer therapy. However, CDDP causes serious side effects, such as neurological disorders, myelosuppression, and renal damage, because it is a small molecule indiscriminately distributed in normal tissues. Therefore, it is very important to prevent or attenuate CDDP toxicity. We hypothesized that liposomalization of CDDP could alleviate serious side effects. Therefore, this study aimed to evaluate the safety and efficacy of CDDP liposomes. A patient with multiple recurrent liver metastases from metastatic nasal carcinoma was administered CDDP liposomes with consent. Magnetic resonance imaging showed that the patient remained stable-diseased; however, no apparent side effects were observed, and blood draw data showed no worsening of renal function. Patients undergoing partial pancreatectomy and jejunoileal biliary anastomosis for biliary tract cancer who consented to receive CDDP liposomes demonstrated a partial response on angiographic computed tomography; however, they showed slight fatigue. To our knowledge, the present study is the first in Japan to suggest that liposomalization of CDDP may have anticancer effects while alleviating renal damage and bone marrow suppression.

Deep learning (DL) algorithms have recently emerged from machine learning and soft computing techniques. Since then, several deep learning (DL) algorithms have been recently introduced to scientific communities and are applied in various application domains. Today the usage of DL has become essential due to their intelligence, efficient learning, accuracy and robustness in model building. However, in the scientific literature, a comprehensive list of DL algorithms has not been introduced yet. This paper provides a list of the most popular DL algorithms, along with their applications domains.

The sustainability of the planet and its inhabitants is in dire danger and is among the highest priorities on global agendas such as the Sustainable Development Goals (SDGs) of the United Nations (UN). Solar energy -- among other clean, renewable, and sustainable energies -- is seen as essential for environmental, social, and economic sustainability. Predicting solar energy accurately is critical to increasing reliability and stability, and reducing the risks and costs of the energy systems and markets. Researchers have come a long way in developing cutting-edge solar energy forecasting methods. However, these methods are far from optimal in terms of their accuracies, generalizability, benchmarking, and other requirements. Particularly, no single method performs well across all climates and weathers due to the large variations in meteorological data. This paper proposes SENERGY (an acronym for Sustainable Energy), a novel deep learning-based auto-selective approach and tool that, instead of generalising a specific model for all climates, predicts the best performing deep learning model for GHI forecasting in terms of forecasting error. The approach is based on carefully devised deep learning methods and feature sets through an extensive analysis of deep learning forecasting and classification methods using ten meteorological datasets from three continents. We analyse the tool in great detail through a range of metrics and methods for performance analysis, visualization, and comparison of solar forecasting methods. SENERGY outperforms existing methods in all performance metrics including Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Forecast Skills (FS), Relative Forecasting Error, and the normalised versions of these metrics. The proposed auto-selective approach can be extended to other research problems such as wind energy forecasting and predict forecasting models based on different criteria (in addition to the minimum forecasting error used in this paper) such as the energy required or speed of model execution, different input features, different optimisations of the same models, or other user preferences.

This paper provides a review of an emerging field in the food processing sector, referring to efficient and safe food supply chains, ’from farm to fork’, as enabled by Artificial Intelligence (AI). Recent advances in machine and deep learning are used for effective food production, energy management and food labeling. Appropriate deep neural architectures are adopted and used for this purpose, including Fully Convolutional Networks, Long Short-Term Memories and Recurrent Neural Networks, Auto-Encoders and Attention mechanisms, Latent Variable extraction and clustering, as well as Domain Adaptation. Three experimental studies are presented, illustrating the ability of these AI methodologies to produce state-of-the-art performance in the whole food supply chain. In particular, these concern: (i) predicting plant growth and tomato yield in greenhouses, thus matching food production to market needs and reducing food waste or food unavailability; (ii) optimizing energy consumption across large networks of food retail refrigeration systems, through optimal selection of systems that can get shut-down and through prediction of the respective food de-freezing times, during peaks of power demand load; (iii) optical recognition and verification of food consumption expiry date in automatic inspection of retail packaged food, thus ensuring safety of food and people’s health.