Search

The inventory management of medical items in humanitarian operations is a challenging task due to the intermittent nature of their demand and long replenishment lead-times. An effective response to humanitarian emergencies is often associated with excess inventories, which leads to high costs. Henceforth, using accurate demand forecasts to control the inventories of these medical items is shown to be a significant lever to reduce inventory costs while keeping high service levels. This paper investigates the effectiveness of parametric and non-parametric demand forecasting methods that are commonly considered to deal with items characterised with intermittent demand patterns. Moreover, we propose a new hybrid deep learning method that combines the long short-term memory neural network and the support vector regression. Such combination enables to deal with complex and non-linear data. We conduct an empirical investigation by means of data related to 523 medical items managed in two warehouses of a major humanitarian organisation. The forecast accuracy and the inventory efficiency of the methods are analysed. The empirical results show the high performance of the deep learning methods. These insights are valuable not only in the context of medical items in humanitarian operations, but also for any items with intermittent demand patterns. [ABSTRACT FROM AUTHOR]

Heterogeneous ice nucleation (HIN) has applications in climate science, nanotechnology, and cryopreservation. Ice nucleation on the earth's surface or in the atmosphere usually occurs heterogeneously involving foreign substrates, known as ice nucleating particles (INPs). Experiments identify good INPs but lack sufficient microscopic resolution to answer the basic question: What makes a good INP? We employ molecular dynamics (MD) simulations in combination with machine learning (ML) to address this question. Often, the large amount of computational cost required to cross the nucleation barrier and observe HIN in MD simulations is a practical limitation. We use information obtained from short MD simulations of atomistic surface and water models to predict the likelihood of HIN. We consider 153 atomistic substrates with some surfaces differing in elemental composition and others only in terms of lattice parameters, surface morphology, or surface charges. A range of water features near the surface (local) are extracted from short MD simulations over a time interval (≤ 300 ns) where ice nucleation has not initiated. Three ML classification models, Random Forest (RF), support vector machine, and Gaussian process classification are considered, and the accuracies achieved by all three approaches lie within their statistical uncertainties. Including local water features is essential for accurate prediction. The accuracy of our best RF classification model obtained including both surface and local water features is 0.89 ± 0.05. A similar accuracy can be achieved including only local water features, suggesting that the important surface properties are largely captured by the local water features. Some important features identified by ML analysis are local icelike structures, water density and polarization profiles perpendicular to the surface, and the two-dimensional lattice match to ice. We expect that this work, with its strong focus on realistic surface models, will serve as a guide to the identification or design of substrates that can promote or discourage ice nucleation. [ABSTRACT FROM AUTHOR]

Incorporating decision-makers' risk preferences into data-driven newsvendor models and developing machine learning methods to solve the models are the challenging problems addressed in this study. To consider different distributions and decision-makers' different risk preferences for the two losses of the total cost newsvendor model, the symmetrical, the partial symmetrical and the asymmetrical CVaR criteria are introduced. The regularisation, the primal-dual approach and the kernels in support vector machines are used to transform the data-driven risk-averse newsvendor problems under the CVaR criterion into the convex quadratic programming problems with good theoretical properties. Computational experiments are conducted on a real-world dataset. The models under the partial symmetrical and the asymmetrical CVaR criteria obtained good performances, but that under the symmetrical CVaR criterion suffered the underfitting problem. Two factors including the degrees of risk aversion for the two losses in the total cost newsvendor model and the empirical errors of data-driven models affect order decisions. The degrees of risk aversion for the two losses have anti-directional effects on order quantities. The introduction of asymmetrical CVaR criterion paves a new way to reveal the effects of different risk references for different losses on order decisions, and has the potential to improve newsvendor decisions. [ABSTRACT FROM AUTHOR]

Efficient resource management methods are essential for spare parts used in the maintenance and repair of equipment. Forecasting plays a critical role in planning, especially under demand uncertainty. Existing works regarding spare parts with intermittent demand focus on the mere forecasting model while integrating the planning and forecasting models are not sufficiently investigated. We examine the interaction between two models to optimise planning and forecasting decisions and prevent sub-optimality. This paper presents two mathematical models, including a planning model that determines stock level, spare part order assignment to suppliers, equipment repair assignment, and the number of intervals over the planning horizon. The second model is the forecasting model by Support Vector Machine (SVM). Considering uncertainty, demand estimation is performed by piecewise linearisation considering the optimal number of intervals in the planning model used in forecasting. An interactive procedure is developed to optimise models. We use an empirical investigation from an oil company providing the spare part supply chain data. The analyses show that demand estimation by piecewise method and integrating the decisions optimises the cost, improves the forecasting accuracy, and planning performance. Moreover, we offer several insights to practitioners that shed light on spare part planning and forecasting decisions. [ABSTRACT FROM AUTHOR]

In forensic automatic speaker recognition (FASR), voice disguise is a most important disquiet in the field of multimedia applications. It is most important to classify the disguised voice in order to recognise the speaker's identity. Changing the pitch is one of the most popular types of voice disguises used by criminals. From the normal pitch to the change in raised pitch, lower pitch, fast pitch, and slow pitch, the most disguised type amongst the different types of disguised voice. The acoustic features in terms of mean and correlation coefficients for raised pitch and lower pitch have different coefficient values, similarly for fast and slow pitch. Using an innovative set of features, we have recognised the classification efficiency for raised pitch, lower pitch, fast speech, and slow speech disguised voice for female and male voice samples used in this proposed research work for multimedia applications. Mel-frequency cepstrum coefficient (MFCC), delta MFCC (ΔMFCC), and double delta MFCC (ΔΔMFCC) features are extracted from normal voice and entirely four varieties of disguised voice. The acoustic feature, correlation coefficients and mean are extracted from all voice samples using the three different MFCC feature extraction techniques in multimedia applications. Feature-based classifiers for example k-nearest neighbour (k-NN) and support vector machines (SVM) are used for classification and recognise the speaker's classification efficiency. The classification results show that applying these feature-based SVM classifiers increases the classification efficiency to 98.58%, compared to 94.23% when using feature-based k-NN classifiers. It is observed that for the standard normal voice, there is 100% classification efficiency for both SVM and k-NN classifiers. [ABSTRACT FROM AUTHOR]

This study develops an active security control strategy for Cyber-Physical Systems (CPSs) that are subject to attacks known as Denial-of-Service (DoS), which can target both channels from the controller to the actuator and from the controller to the sensor. Due to attack cost restrictions, the linked channels are subject to a limit on the number of continuous DoS attacks. A proactive security control method is then developed to combat two-channel DoS attacks, depending on a method for identifying IoT intrusions. Using the CICIDS dataset for attack detection, we examined the effectiveness of the Deep Convolutional Network Model (DCNM), a suggested deep learning model. The addressed CPS can be asymptotically stable against DoS assaults under the security controller's active security control technique without sacrificing control performance. Recent tests and simulations show how effective the security control strategy is active. The proposed model gives better trade-off compared to existing approaches like Deep Belief Networks (DBN), Recurrent Neural Networks (RNN), Support Vector Machines (SVM), Supervised Neural Networks (SNN) and Feed Forward Neural Networks (FNN). The proposed model gives 99.3%, 99.5%, 99.5%, 99.6%, 99%, 98.9%, 99% accuracy with normal attack detection, botnet attack detection, Brute force attack detection, DoS attack detection, Infiltration attack detection, Portscan attack detection and web attack detection respectively. [ABSTRACT FROM AUTHOR]

The major goal of this research is to develop a Deep Learning (DL) based automatic identification and classification of white blood cells (WBCs) with high accuracy and efficiency. The first phase of research is pre-processing and is accomplished by the Improved Median Wiener Filter (IMWF), which effectively eliminates the noises. The image is resized into a standard image size before filtering. The segmentation process takes place using Color Balancing Binary Threshold (CBBT) algorithm to divide the WBCs and another non-relevant background to improve the classification performance. The features like shape, texture and color of the WBCs are extracted from the segmented images. Finally, the classification takes place, and this is processed by a fine optimal deep convolution residual network (Fine Optimal DCRNet). In addition, the bionic model is introduced to improve classification accuracy. The dataset used in this research is BCCD and LISC datasets. The performance of the proposed model is validated using existing methods of Support Vector Machine (SVM), K-Nearest Neighbor (KNN), VGG-16, VGG-19, ResNet-50, DensetNet-121, DensetNet-169, Inception-V3, InceptionResNet-V2, Xception, MobileNet-224, Mobile NasNet, Tree, Naive Bayes, Ensemble active contour model, k-means clustering and handcraft and deep learned features-scale-invariant feature transform (HCDL-SIFT) in terms of Accuracy, Precision, Recall, Specificity, F-score, Relative Distance Error (RDE), Over-Segmentation Rate (OSR), Under-Segmentation Rate (USR) and Overall Error Rate (OER). For the LISC dataset, the detection model attains an outcome of 99%, 98%, 98%, 99%, 98%, 1.143, 0.0125, 0.056 and 0.125, respectively. For the BCCD dataset, apart from RDE, OSR, USR and OER metrics, the performance is evaluated as 98%, 96%, 98%, 99% and 97%. [ABSTRACT FROM AUTHOR]

Research on the visual system and its development is not only crucial for understanding how we see and interpret the environment from basic visual processing to complex perception but also aids in the development of technologies like virtual reality, augmented reality, and computer vision. In our previous paper, we have proposed a novel biological motion direction detection mechanism. Based on this mechanism, the authors of this paper proposed a learning artificial visual system (AVS) and use it to achieve motion direction detection. The learning AVS consists of two layers: the local visual feature detective neuron layer and the global feature detective neuron layer. For local motion direction detection, we use dendritic neurons to implement local motion direction-detective neurons and use them to extract local motion direction information. Global feature detective neurons are simply implemented by summing the local motion direction-detective neurons in each direction of possible motion. We trained the learning AVS by eliminating unnecessary branches or synapses and keeping only those that are strongly connected to the function of motion direction detection. Computer simulations demonstrate that AVS can learn the correct motion direction of objects with different shapes, size and positions. We compare the performance of AVS with support vector machine (SVM) and traditional convolutional neural networks (CNNs)—LeNet-5 and EfficientNet-B0 (EfN)—in motion direction detection and find that our learning AVS has better recognition accuracy, noise immunity and less demands on size of dataset and learning costs. [ABSTRACT FROM AUTHOR]

The importance of technology and managerial risk management in banks has increased due to the financial crisis. Banks are the most affected since there are so many of them with poor financial standing. Due to this problem, an unstable and inefficient financial system causes economic stagnation in both the banking sector and overall economy. Data envelopment analysis (DEA) has been used to examine decision-making units (DMUs) performance to enhance efficiency. Currently, with the rapid growth of big data, adding more DMUs will likely require a large amount of memory and CPU time on the computer system, which will be the biggest challenge. As a result, machine learning (ML) approaches have been used to analyze financial institution performance, but many of them have variances in predictions or model stability, making measuring bank efficiency extremely difficult. For this, ensemble learning is commonly used to evaluate the performance of financial institutions in this context. This paper presents a robust super learner ensemble technique for assessing bank efficiency, with four machine learning models serving as base learners. These models are the support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF), and AdaBoost classifier (ADA) which represent the base learners and their results utilized to train the meta-learner. The super learner (SL) approach is an extension of the stacking technique, which generates an ensemble based on cross-validation. One important benefit of this cross-validation theory-based technique is that it can overcome the overfitting issue that plagues most other ensemble approaches. When SL and base learners were compared for their forecasting abilities using different statistical standards, the results showed that the SL is superior to the base learners, where different variable combinations were used. The SL had accuracy (ACC) of 0.8636–0.9545 and F1-score (F1) of 0.9143–0.9714, while the basic learners had ACC of 0.5909–0.8182 and F1 of 0.6897–0.9143. So, SL is highly recommended for improving the accuracy of financial data forecasts, even with limited financial data. [ABSTRACT FROM AUTHOR]

Asthma is a common disease. The clinical diagnosis is usually confirmed on a pulmonary function test, which is not always readily accessible. We aimed to develop a computationally lightweight handcrafted machine learning model for asthma detection based on cough sounds recorded using mobile phones. Toward this aim, we proposed a novel feature extractor based on a one-dimensional version of the published attractive-and-repulsive center-symmetric local binary pattern (1D-ARCSLBP), which we tested on a new cough sound dataset. We prospectively recorded cough sounds from 511 asthmatics and 815 non-asthmatic subjects (comprising mostly healthy volunteers), which yielded 1875 one-second cough sound segments for analysis. Our model comprised four steps: (i) preprocessing, in which speech signals and stop times (silent zones between coughs) were removed, leaving behind analyzable cough sound segments; (ii) feature extraction, in which tunable q-factor wavelet transformation was used to perform multilevel signal decomposition into wavelet subbands, allowing 1D-ARCSLBP to extract local low- and high-level features; (iii) feature selection, in which neighborhood component analysis was used to select the most discriminative features; and (iv) classification, in which a standard shallow cubic support vector machine was deployed to calculate binary classification results (asthma versus non-asthma) using tenfold and leave-one-subject-out cross-validations. Our model attained 98.24% and 96.91% accuracy rates with tenfold and leave-one-subject-out cross-validation strategies, respectively, and obtained a low-time complexity. The excellent results confirmed the feature extraction capability of 1D-ARCSLBP and the feasibility of the model being developed into a real-world application for asthma screening. [ABSTRACT FROM AUTHOR]

Optical character recognition is a way of converting scanned images of printed or handwritten documents into machine-encoded text, making it easier to store, browse, retrieve, and process electronic data. In this research, a Printed Amharic Characters Recognition dataset is prepared to train and test a model. Images in the dataset only contain 231 basic Amharic characters that are normalized to 32 × 32 pixels. In this work, a hybrid model of the two super classifiers is developed: the convolutional neural network (CNN) and the support vector machine (SVM). In this novel hybrid CNN-SVM model, CNN works as an automatic feature extractor from the raw images, and then, the extracted feature vectors are given as input to SVM for classification and recognition. A 99.84% accuracy was achieved on the own-prepared dataset and 95.59% accuracy on the benchmark Amharic Optical Character Recognition Database in classifying the testing dataset images. The proposed hybrid CNN-SVM model gave better results than the CNN with a fully connected layer. Moreover, the proposed model outperforms previously existing works attempted by others to recognize printed Amharic characters on the same and different datasets. [ABSTRACT FROM AUTHOR]

ABSTRACT This study employs advanced data mining techniques to investigate the DASS‐42 questionnaire, a widely used psychological assessment tool. Administered to 680 students at Necmettin Erbakan University's Ahmet Kelesoglu Faculty of Education, the DASS‐42 comprises three distinct subscales—depression, anxiety and stress—each consisting of 14 items. Departing from traditional statistical methodologies, the study harnesses the power of the WEKA data mining program to analyse the dataset. Employing Naive Bayes (NB), Artificial Neural Network (ANN), Logistic Regression (LR), Support Vector Machine (SVM) and Random Forest (RF) algorithms, the research unveils novel insights. The ANN method emerges as a standout performer, achieving remarkable distinctiveness scores for all subscales: depression (99.26%), anxiety (98.67%) and stress (97.35%). The study highlights the potential of data mining in enhancing psychological assessment and showcases the ANN's prowess in capturing intricate patterns within complex psychological dimensions. By charting a course beyond conventional statistical methods, this research pioneers a new frontier for employing data mining within the realm of social sciences. As a result of the study, it is recommended that teacher candidates in the teacher education process should have knowledge about depression, anxiety and stress, and relevant courses on these topics should be added to the curriculum of teacher education programs. [ABSTRACT FROM AUTHOR]

This article explores the importance of calibration in prediction models when applied to new populations. It compares different calibration approaches and introduces Elkan's general updating approach (GUA), which accounts for differences in base rates between populations without re-estimation. The article explains how calibrated probability can be expressed as a function of base rates, which are estimated using proportions of subjects with a certain outcome in training and calibration data. The text discusses two approaches for estimating base probability in a training data set: Elkan's GUA calculates the average of the target variable, while Baker's GUA suggests using the average of the conditional probability. Both approaches assume that the only difference between populations is the change in base rate and require a calibration data set. A simulation study comparing different calibration methods for machine learning models found that beta calibration performed the best, while isotonic calibration and IVAP calibration were similar. Elkan calibration and Baker calibration also showed similar performance. The article provides detailed information on the simulation scenarios and the performance of each calibration method. [Extracted from the article]

Breast carcinoma or breast cancer is the most common type of cancer in women. It is very dangerous, and early detection is important for proper diagnosis and reduction of mortality rate. Cancer starts in a benign state, and if there is no proper treatment, it becomes malignant. Computer-aided diagnosis (CAD) techniques can be used for early detection of cancers. These techniques can significantly and effectively help in the diagnosis of cancer. Deep learning is the most popular technique used for accurate data analysis. It is also a powerful tool for categorizing histopathological images of breast cancer and analyzing breast cell shapes and densities. This paper proposes a hybrid EfficientNet with an SVM to categorize breast mass density with consideration of tumor types. (i.e., malignant or benign). In this hybrid, EfficientNet, the mammography images are applied to EfficientNet, and then the classification is carried out using a Support Vector Machine (SVM). The proposed model achieved accuracy of 94.48%, precision of 94.79%, specificity of 99.21%, Precision of 94.79%, FPR of 0.79%, score of 94.45%, MCC of 93.78%, and Kappa of 74.76%. The model outperformed state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

In the digitalized world, efficient big data clustering is necessary for massive data generation. The clustering algorithm plays an important role in resolving the computational complexity. The big data arriving from various sources are being processed using the MapReduce framework (MRF) by the knowledge of the clustering algorithms. Moreover, the clustering algorithm is useful for mining the significant information from the dataset. Generally, there are several difficulties in applying the clustering approach to big data as its new challenges are based on computation cost and reasonable time. Hence, this research introduced the Competitive Jaya Leader Harris Hawks Optimization assisted Entropy Weighted Power K-Means Clustering (CJayaLHHO_EWPKMC) for big data clustering. In addition, the overall processing of the devised method for big data clustering is carried out in the MapReduce (MR) framework. In mapper, the feature selection is done using Support vector Machine-Recursive Feature Elimination (SVM-RFE) assisted Jaya Leader Harris Hawks Optimization (JayaLHHO). In the reducer, the big data clustering is established using the EWPKMC method, wherein the weight of EWPKMC is modified with the CJayaLHHO algorithm such that the clustering outcome is attained. The proposed method is scalable, simple, cost-effective, and able to integrate with other technologies. The experimental result portrays that the developed method attained a superior presentation than the conventional methods based on the clustering accuracy is 0.937, the Jaccard coefficient is 0.913, and the rand coefficient is 0.912. [ABSTRACT FROM AUTHOR]

In order to be more efficient in the optimization of concrete mixture, a new intelligent optimization model of concrete mixture was established by using particle swarm optimization on the basis of the original optimization model, and the specific performance was fitted and predicted by using the support vector machine. Using the optimization model established in this work, the test results of 1,443 groups of mixed ratios show that the prediction accuracy of the new model is better than that of the traditional model. The average relative errors of the 7‐ and 28‐day compressive strengths and initial collapse were 10.42%, 7.43%, and 17.14%, respectively. This method is used in a practical project where a set of C25, C30, C35, and C40 concrete mix ratios are designed. The prediction and actual error of 7‐day intensity and 28‐day intensity are −14.4% to −6.1% and −17.6% to 0.6%, respectively, while the predicted value and actual slump error are 15%. The results show that the error between the measured value and the design value is basically within 15%, which meets the design requirements. The method can reduce the time and cost of concrete production, with the cost per cubic meter after optimization is 3.5–8 CHY lower than the concrete mix calculated by the conventional method. The relative decrease ranges from 1.53% to 3.22%. Finally, a practical quality and cost control system (compos) is established to make the strategy easier to apply. [ABSTRACT FROM AUTHOR]

Infiltration parameters and roughness are essential parameters in the design and management of furrow irrigation. Accurate determination of these parameters is essential for achieving reliable irrigation techniques and improving furrow irrigation performance. Therefore, this study proposed a method combining least squares support vector machines (LSSVM) and genetic algorithm (GA) to invert infiltration parameters and roughness. During the parameters inversion, it is necessary to know the simulated values of advance and recession time to construct the objective function, but the relationship between the furrow irrigation advance, recession time and soil infiltration parameters and roughness is implicit in the zero-inertia model, lacking a direct solution. Manually inputting parameters into WinSRFR for validation would severely hinder the efficiency of parameter estimation. Therefore, this study utilized LSSVM to map the complex nonlinear relationship the furrow irrigation advance, recession times and soil infiltration parameters and roughness. The advance time LSSVM regression model (regularization constant 60, RBF kernel width 35) and recession time LSSVM regression model (regularization constant 70, RBF kernel width 20) were constructed. Results showed that the root mean square error (RMSE) for advance and recession time in training samples ranged from 0.84 to 1.75 min and 2.18 to 3.16 min, respectively. For testing samples, the RMSE values ranged from 0.85 to 1.77 min and 2.41 to 3.62 min, demonstrating high prediction accuracy. This approach reduces the computational burden of manual input in WinSRFR by using LSSVM regression models to simulate advance/recession time, thus facilitating parameter estimation with GA. The key to implementing the LSSVM-GA method lies in the data transfer and iteration between LSSVM and GA when calculating the fitness function. To further verify the reliability of the proposed method, furrow irrigation experiments were conducted on closed-end furrows in cornfields in Shuangzhao Village, Xianyang, Shaanxi Province, China. Two typical furrow lengths were selected, each with two test plots to observe the actual advance and recession times. Each scheme was repeated three times, and the final experimental data were averaged. Multiple nonlinear regression (MNR) and Merriam-Keller post-irrigation volume balance analysis (MK-PIVB) from WinSRFR were also used for comparison. Results indicated that for the advance process simulation, the parameters estimated by MK-PIVB and LSSVM-GA provided similar curves, with RMSE ranging from 1.30 to 2.94 min and coefficient of determination (R²) from 0.97 to 0.99. In contrast, MNR performed slightly worse, particularly for F1, with an RMSE of 5.52 min. Further analysis of the recession process showed that MNR's parameter estimates were not ideal, with RMSE ranging from 4.77 to 6.00 min and R² below 0.60. MK-PIVB's estimates were satisfactory only for F2, but were inaccurate for the other three furrows with RMSE of 4.75, 4.13, and 4.18 minutes and R² of 0.49, 0.70, and 0.54, respectively. The LSSVM-GA method provided improved parameter estimates for both processes, with RMSEs ranging from 1.06 to 2.12 minutes for advance and 2.28 to 3.11 minutes for recession. The LSSVM-GA method effectively combined nonlinear mapping and heuristic algorithms, overcoming the limitations of the other two methods, such as the need for predefined regression functions, reliance on manual experience and judgment, and high computational cost. [ABSTRACT FROM AUTHOR]

Aim Methods Results Conclusion Length of stay (LOS) is an outcome measure and is assumed to be related to quality. The objective of this study is to examine the quality of birth care and risk factors associated with LOS after birth.A nationwide population‐based Turkish Demographic and Health Survey (TDHS) was used for the year 2018. A total of 1849 women ages 15–49 were included. Explanatory factor analysis and machine learning predictors such as Random Forest, Support Vector Machine, Neural Network, k‐Nearest Neighbor, and Naïve Bayes were used to identify the quality of birth care and risk factors associated with LOS after birth.As a result of the explanatory factor analysis, factor structures of quality of birth care, antenatal check‐ups and supplements, and risk factors associated with birth were obtained using the Categorical Component Analysis method. The type of delivery, place of delivery, age, and type of place, which are under the quality of birth care, and risk factors associated with birth factors were found to be the variables that had the highest impact on LOS estimation. Random forest (Accuracy = 0.5789), support vector machine (radial) (Accuracy = 0.5766), and neural network (Accuracy = 0.5750) models outperformed, respectively.Type of delivery which is an indicator of quality of birth care is a strong predictor of LOS after birth according to the Random Forest model. We demonstrated that machine learning techniques offer precise LOS prediction after birth. Further studies assessing the effect of quality of birth care on predicting LOS at birth would be beneficial. [ABSTRACT FROM AUTHOR]

Feature extraction is a core aspect in hyperspectral image classification, which can extract key information closely related to ground cover from complex scene, thus improving classification accuracy. Therefore, designing an effective feature extraction network is a hotspot and a challenge in the current research. In this paper, a feature extraction framework based on 3-D homogeneous attribute decomposition (3D-HAD) is proposed for HSI classification, which consists of the following key technologies. First, the principal component analysis algorithm is applied to the raw HSI to extract the principal components (PCs), and the raw HSI is clustered into many 3D superpixel blocks according to the first three PCs-based over-segmentation strategy. Then, a superpixel intrinsic attribute decomposition (SIAD) is designed to capture reflectance feature and suppress shading feature. Meanwhile, a metric entropy is introduced into the decomposition process to overcome the spectral-spatial weak assumption among pixels. Next, superpixel-guided recursive filtering is employed to preserve global details of HSI to enhance accuracy in HSI classification. Finally, the support vector machine classifier is used to obtain classification results of HSI. Experiments performed on several real hyperspectral data sets with limited training samples indicate that the proposed 3D-HAD method outperforms the classic, advance, and deep learning classification methods. [ABSTRACT FROM AUTHOR]

Due to the joint influence of human activities and climate change, lakes experience significant shrinkage or expansion, posing enormous challenges for water resource management and the ecological balance of lake ecosystems. As an important region abundant in water resources in China, the balance of water resources in the middle and lower Yangtze River Basin (MLY) is crucial for region sustainable development. However, to date, few studies have quantitatively analysed the lake area during heat waves in the MLY. To fill this gap, this study selected the summer of 2022, which encountered unprecedented heat waves, to analyse the changes of lake area in the MLY. Specifically, Sentinel-1 images were used as data sources, and the Random Forest (RF) model was applied to extract the monthly area of 62 lakes (area >10 km2) in the MLY from 2019 to 2022. The results showed that: (1) Compared to the Support Vector Machine (SVM) model and the Otsu method, the RF model had the highest accuracy, with its accuracy and Kappa coefficient in the testing samples of 97.14% and 0.94, respectively. During the period from June to September in 2022, the total area of 62 lakes decreased by 3475.50 km2. (2) Compared to the same period in the summer of 2019–2021, 23 lakes (37.10%) were observed to experience shrinkage. Among them, the area of Poyang Lake and Dongting Lake decreased to the level of the previous winter dry seasons in August 2022. (3) Temperature was the most important meteorological factor that caused changes in the lake area in the MLY, compared to precipitation, evaporation, and runoff. This research can help understand the impact of climate change on lakes and provide references for decision-making in the MLY to formulate the strategies of water resource management under the changing climate. [ABSTRACT FROM AUTHOR]

Study Design. Retrospective study. Objectives. The objective of this investigation was to formulate and internally verify a customized machine learning (ML) framework for forecasting cerebrospinal fluid leakage (CSFL) in lumbar fusion surgery. This was accomplished by integrating imaging parameters and using the SHapley Additive exPlanation (SHAP) technique to elucidate the interpretability of the model. Summary of Background Data. Given the increasing incidence and surgical volume of spinal degeneration worldwide, accurate predictions of postoperative complications are urgently needed. SHAP-based interpretable ML models have not been used for CSFL risk factor analysis in lumbar fusion surgery. Methods. Clinical and imaging data were retrospectively collected from 3505 patients who underwent lumbar fusion surgery. Six distinct machine learning models were formulated: extreme gradient boosting (XGBoost), decision tree (DT), random forest (RF), support vector machine (SVM), Gaussian naive Bayes (GaussianNB), and K-nearest neighbors (KNN) models. Evaluation of model performance on the test dataset was performed using performance metrics, and the analysis was executed through the SHAP framework. Results. CSFL was detected in 95 (2.71%) of 3505 patients. Notably, the XGBoost model exhibited outstanding accuracy in forecasting CSFLs, with high precision (0.9815), recall (0.6667), accuracy (0.8182), F1 score (0.7347), and AUC (0.7343). In addition, through SHAP analysis, significant predictors of CSFL were identified, including ligamentum flavum thickness, zygapophysial joint degeneration grade, central spinal stenosis grade, decompression segment count, decompression mode, intervertebral height difference, Cobb angle, intervertebral height index difference, operation mode, lumbar segment lordosis angle difference, Meyerding grade of lumbar spondylolisthesis, and revision surgery. Conclusions. The combination of the XGBoost model with the SHAP is an effective tool for predicting the risk of CSFL during lumbar fusion surgery. Its implementation could aid clinicians in making informed decisions, potentially enhancing patient outcomes and lowering healthcare expenses. This study advocates for the adoption of this approach in clinical settings to enhance the evaluation of CSFL risk among patients undergoing lumbar fusion. [ABSTRACT FROM AUTHOR]

Bipolar disorder (BD) is associated with increased morbidity/mortality. Adverse outcome prediction might help with the management of patients with BD. We systematically reviewed the performance of machine learning (ML) studies in predicting adverse outcomes (relapse or recurrence, hospital admission, and suicide-related events) in patients with BD. Demographic, clinical, and neuroimaging-related poor outcome predictors were also reviewed. Three databases (PubMed, Scopus, and Web of Science) were explored from inception to July 2023. Eighteen studies, accounting for >30,000 patients, were included. Support vector machine, decision trees, random forest, and logistic regression were the most frequently used ML algorithms. ML models' area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, and specificity ranged from 0.71 to 0.98, 72.7–92.8 %, and 59.0–95.2 % for relapse/recurrence prediction (4 studies (3 on relapses and 1 on recurrences). The corresponding values were 0.78–0.88, 21.4–100 %, and 77.0–99.7 % for hospital admissions (3 studies, 21,266 patients), and 0.71–0.99, 44.4–97.9 %, and 38.9–95.0 % for suicide-related events (10 studies, 5558 patients). Also, one study addressed a combination of the interest outcomes. Adverse outcome predictors included early onset BD, BD type I, comorbid psychiatric or substance use disorder, circadian rhythm disruption, hospitalization characteristics, and neuroimaging parameters, including increased dynamic amplitude of low-frequency fluctuation, decreased frontolimbic functional connectivity and aberrant dynamic functional connectivity in corticostriatal circuitry. ML models can predict adverse outcomes of BD with relatively acceptable performance measures. Future studies with larger samples and nested cross-validation validation should be conducted to reach more reliable results. • Adverse outcome prediction aids in the management of patients with BD. • We reviewed studies assessing ML models in predicting relapses/recurrences, hospitalizations, and suicide-related events in BD. • ML predicts adverse outcomes of BD with various performance measures across different models. • Future research is needed to integrate this knowledge into the clinical management of in BD. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the application value of support vector machine (SVM) model based on diffusion-weighted imaging (DWI), dynamic contrast-enhanced (DCE) and amide proton transfer- weighted (APTW) imaging in predicting isocitrate dehydrogenase 1(IDH-1) mutation and Ki-67 expression in glioma. Methods: The DWI, DCE and APTW images of 309 patients with glioma confirmed by pathology were retrospectively analyzed and divided into the IDH-1 group (IDH-1(+) group and IDH-1(-) group) and Ki-67 group (low expression group (Ki-67 ≤ 10%) and high expression group (Ki-67 > 10%)). All cases were divided into the training set, and validation set according to the ratio of 7:3. The training set was used to select features and establish machine learning models. The SVM model was established with the data after feature selection. Four single sequence models and one combined model were established in IDH-1 group and Ki-67 group. The receiver operator characteristic (ROC) curve was used to evaluate the diagnostic performance of the model. Validation set data was used for further validation. Results: Both in the IDH-1 group and Ki-67 group, the combined model had better predictive efficiency than single sequence model, although the single sequence model had a better predictive efficiency. In the Ki-67 group, the combined model was built from six selected radiomics features, and the AUC were 0.965 and 0.931 in the training and validation sets, respectively. In the IDH-1 group, the combined model was built from four selected radiomics features, and the AUC were 0.997 and 0.967 in the training and validation sets, respectively. Conclusion: The radiomics model established by DWI, DCE and APTW images could be used to detect IDH-1 mutation and Ki-67 expression in glioma patients before surgery. The prediction performance of the radiomics model based on the combination sequence was better than that of the single sequence model. [ABSTRACT FROM AUTHOR]

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

Affected by the uncertainty of external environmental factors, wind power generation has significant characteristics of randomness and non-stationarity. Accurately predicting wind power is a necessary technical means to improve the efficiency of wind energy utilization and reduce wind abandonment. Integrating large-scale wind power into the grid has put higher requirements for wind power prediction accuracy. Hence, this study proposes a combined prediction model based on wavelet denoising (WD), hybrid enhanced seagull optimization algorithm (HESOA) and support vector machine (SVM) to predict short-term wind power. The WD technology obtains effective signals from original wind power data, providing a reasonable basis for prediction research. The HESOA is proposed by introducing the Tent mapping, individual self-learning factor and Gaussian population strategy into the original seagull optimization algorithm to optimize SVM's prediction ability. The case study results reflect that the prediction method is feasible and universal in suppressing wind power generation volatility in spring and winter. The RMSE of short-term wind power prediction is controlled below 65 kW, and the value of R2 is above 98%. This study improves the accuracy of wind power prediction and helps to strengthen the renewable energy consumption level and reliability of wind power grid-connected generation systems. [ABSTRACT FROM AUTHOR]

Compound channels play an important role in hydraulic and hydrological engineering. Many rivers and streams have naturally compound channels, which are important for both flood control and river ecosystems. When predicting discharge, interval predictions are usually more realistic and reliable than traditional point predictions made using ML models. In this study, the authors used three well-known machine learning techniques (Adaptive Neuro Fuzzy Inference System (ANFIS), Support Vector Machine (SVM) and Artificial Neural Networks (ANN)) to predict compound channel discharge. To accurately assess uncertainty, two resampling techniques were employed: Bootstrap (B) and Jackknife after Bootstrap (JB). The proposed approach reflects the impacts of statistical uncertainties using prediction confidence intervals and the reliability of point prediction. The JB resampling technique has proven to offer superior accuracy in terms of point prediction and prediction intervals when compared to Bootstrap. Bootstrapping did not yield superior outcomes compared with the standalone ML models. The JB-ANN and JB-ANFIS models outperformed the JB-SVM and Bootstrap ML models in the testing phase, as shown by various statistical measures. This study suggests that the JB-ANN approach is a reliable and robust tool for accurately predicting compound discharges. The findings reveal that when the models have low accuracy in point prediction, the level of uncertainty in predicting the interval also increases. The innovation of this research is enhancing the examination of uncertainties in compound channel discharge by incorporating two resampling techniques to analyze three different ML models. [ABSTRACT FROM AUTHOR]

The rapid urbanization in Pakistan frequently leads to urban waterlogging due to storms. The event often leads to significant harm to the environment, people, and urban economies. Early identification of rainstorm events and urban waterlogging disasters is essential in reducing associated damages. Twitter (X), a widely used global microblogging platform, offers a large amount of real-time tweets that can be used for immediate monitoring purposes. This study introduces a method for recognizing microblogs with information about urban rainstorms and waterlogging and uses blog posts to assess the waterlogging risk. In light of the preliminary examination of microblog content, we determine the efficacy of cluster and support vector machine methods for classification. In addition to text vector attributes, we incorporate sentiment aspects to improve the precision and clarity of our results. We also constructed a lexicon for waterlogging severity to evaluate the risk of waterlogging based on the content of Tweets. Afterward, we generate a risk map using ArcGIS, with findings suggesting that SVM is suitable for detecting rainstorms and waterlogging events in real time. The waterlogging location aligns with the findings of the hazard assessment. The proposed risk assessment method can be a precise tool for promptly addressing emergencies. [ABSTRACT FROM AUTHOR]

A large-scale integration of photovoltaic (PV) systems can degrade the stability of the power grid. Therefore, it is important to accurately predict the short-term output power generated from PV systems to achieve better grid power distribution and allocation. For this reason, a short-term PV power prediction model that uses the data collected from temperature sensors, irradiance sensors, and other relevant sensors was proposed, in which an improved dragonfly algorithm (IDA) was applied to optimize the support vector machine (SVM). First, the output power curves of PV systems under sunny, cloudy, and rainy conditions were analyzed to determine the input variables of the prediction model, which included temperature, relative humidity, and solar radiation intensity. Second, the original dragonfly algorithm in the optimization process was improved, and then, this IDA was utilized to optimize the parameters of SVM, enhancing the predictive capability of the model. Finally, the IDA-optimized SVM (IDA-SVM) model was applied to predict the PV output power. Test performance results demonstrated that the average absolute percentage errors of IDA-SVM were 2.42, 5.96, and 7.44% for sunny, cloudy, and rainy days, respectively, outperforming other comparative models. The performance results showed that the proposed model can not only improve the stability of PV integration, but also effectively increase the penetration rate of PV energy and enhance the reliability of power system operation. [ABSTRACT FROM AUTHOR]

Background: Hospitalizations for exacerbations of congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD) and diabetic ketoacidosis (DKA) are costly in the United States. The purpose of this study was to predict in-hospital charges for each condition using machine learning (ML) models. Results: We conducted a retrospective cohort study on national discharge records of hospitalized adult patients from January 1st, 2016, to December 31st, 2019. We constructed six ML models (linear regression, ridge regression, support vector machine, random forest, gradient boosting and extreme gradient boosting) to predict total in-hospital cost for admission for each condition. Our models had good predictive performance, with testing R-squared values of 0.701-0.750 (mean of 0.713) for CHF; 0.694-0.724 (mean 0.709) for COPD; and 0.615-0.729 (mean 0.694) for DKA. We identified important key features driving costs, including patient age, length of stay, number of procedures, and elective/nonelective admission. Conclusions: ML methods may be used to accurately predict costs and identify drivers of high cost for COPD exacerbations, CHF exacerbations and DKA. Overall, our findings may inform future studies that seek to decrease the underlying high patient costs for these conditions. [ABSTRACT FROM AUTHOR]

Background: Anomalous origin of coronary artery is a common coronary artery anatomy anomaly. The anomalous origin of the coronary artery may lead to problems such as narrowing of the coronary arteries at the beginning of the coronary arteries and abnormal alignment, which may lead to myocardial ischemia due to the compression of the coronary arteries. Clinical symptoms include chest tightness and dyspnea, with angina pectoris as a common symptom that can be life-threatening. Timely and accurate diagnosis of anomalous coronary artery origin is of great importance. Coronary computed tomography angiography (CCTA) can provide detailed information on the characteristics of coronary arteries. Therefore, we combined CCTA and artificial intelligence (AI) technology to analyze the CCTA image features and clinical features of patients with anomalous origin of the right coronary artery to predict angina pectoris and the relevance of different features to angina pectoris. Methods: In this retrospective analysis, we compiled data on 15 characteristics from 126 patients diagnosed with anomalous right coronary artery origins. The dataset encompassed both CCTA imaging attributes, such as the positioning of the right coronary artery orifices and the alignment of coronary arteries, and clinical parameters including gender and age. To identify the most salient features, we employed the Chi-square feature selection method, which filters features based on their statistical significance. We then focused on features yielding a Chi-square score exceeding a threshold of 1, thereby narrowing down the selection to seven key variables, including cardiac function and gender. Subsequently, we evaluated seven classifiers known for their efficacy in classification tasks. Through rigorous training and testing, we conducted a comparative analysis to identify the top three classifiers with the highest accuracy rates. Results: The top three classifiers in this study are Support Vector Machine (SVM), Ensemble Learning (EL), and Kernel Approximation Classifier. Among the SVM, EL and Kernel Approximation Classifier-based classifiers, the best performance is achieved for linear SVM, optimizable Ensembles Learning and SVM kernel, respectively. And the corresponding accuracy is 75.7%, 75.7%, and 73.0%, respectively. The AUC values are 0.77, 0.80, and 0.75, respectively. Conclusions: Machine learning (ML) models can predict angina pectoris caused by the origin anomalous of the right coronary artery, providing valuable auxiliary diagnostic information for clinicians and serving as a warning to clinicians. It is hoped that timely intervention and treatment can be realized to avoid serious consequences such as myocardial infarction. [ABSTRACT FROM AUTHOR]

Background: Breast cancer poses a significant health risk to women worldwide, with approximately 30% being diagnosed annually in the United States. The identification of cancerous mammary tissues from non-cancerous ones during surgery is crucial for the complete removal of tumors. Results: Our study innovatively utilized machine learning techniques (Random Forest (RF), Support Vector Machine (SVM), and Convolutional Neural Network (CNN)) alongside Raman spectroscopy to streamline and hasten the differentiation of normal and late-stage cancerous mammary tissues in mice. The classification accuracy rates achieved by these models were 94.47% for RF, 96.76% for SVM, and 97.58% for CNN, respectively. To our best knowledge, this study was the first effort in comparing the effectiveness of these three machine-learning techniques in classifying breast cancer tissues based on their Raman spectra. Moreover, we innovatively identified specific spectral peaks that contribute to the molecular characteristics of the murine cancerous and non-cancerous tissues. Conclusions: Consequently, our integrated approach of machine learning and Raman spectroscopy presents a non-invasive, swift diagnostic tool for breast cancer, offering promising applications in intraoperative settings. [ABSTRACT FROM AUTHOR]

This study investigated whether machine learning (ML) has better predictive accuracy than logistic regression analysis (LR) for gait independence at discharge in subacute stroke patients (n = 843) who could not walk independently at admission. We developed prediction models using LR and five ML algorithms—specifically, the decision tree (DT), support vector machine, artificial neural network, ensemble learning, and k-nearest neighbor methods. Functional Independence Measure sub-items were used to evaluate the ability to walk independently. Model predictive accuracies were evaluated using areas under receiver operating characteristic curves (AUCs) as well as accuracy, precision, recall, F1 score, and specificity. The AUC for DT (0.812) was significantly lower than those for the other algorithms (p < 0.01); however, the AUC for LR (0.895) did not differ significantly from those for the other models (0.893–0.903). Other performance metrics showed no substantial differences between LR and ML algorithms. In conclusion, the DT algorithm had significantly low predictive accuracy, and LR showed no significant difference in predictive accuracy compared with the other ML algorithms. As its predictive accuracy is similar to that of ML, LR can continue to be used for predicting the prognosis of gait independence, with additional advantages of being easily understandable and manually computable. [ABSTRACT FROM AUTHOR]

Non-alcoholic Fatty Liver Disease (NAFLD), noted for its widespread prevalence among adults, has become the leading chronic liver condition globally. Simultaneously, the annual disease burden, particularly liver cirrhosis caused by NAFLD, has increased significantly. Neutrophil Extracellular Traps (NETs) play a crucial role in the progression of this disease and are key to the pathogenesis of NAFLD. However, research into the specific roles of NETs-related genes in NAFLD is still a field requiring thorough investigation. Utilizing techniques like AddModuleScore, ssGSEA, and WGCNA, our team conducted gene screening to identify the genes linked to NETs in both single-cell and bulk transcriptomics. Using algorithms including Random Forest, Support Vector Machine, Least Absolute Shrinkage, and Selection Operator, we identified ZFP36L2 and PHLDA1 as key hub genes. The pivotal role of these genes in NAFLD diagnosis was confirmed using the training dataset GSE164760. This study identified 116 genes linked to NETs across single-cell and bulk transcriptomic analyses. These genes demonstrated enrichment in immune and metabolic pathways. Additionally, two NETs-related hub genes, PHLDA1 and ZFP36L2, were selected through machine learning for integration into a prognostic model. These hub genes play roles in inflammatory and metabolic processes. scRNA-seq results showed variations in cellular communication among cells with different expression patterns of these key genes. In conclusion, this study explored the molecular characteristics of NETs-associated genes in NAFLD. It identified two potential biomarkers and analyzed their roles in the hepatic microenvironment. These discoveries could aid in NAFLD diagnosis and management, with the ultimate goal of enhancing patient outcomes. [ABSTRACT FROM AUTHOR]

Background: The prognosis, recurrence rates, and secondary prevention strategies varied significantly among different subtypes of acute ischemic stroke (AIS). Machine learning (ML) techniques can uncover intricate, non-linear relationships within medical data, enabling the identification of factors associated with etiological classification. However, there is currently a lack of research utilizing ML algorithms for predicting AIS etiology. Objective: We aimed to use interpretable ML algorithms to develop AIS etiology prediction models, identify critical factors in etiology classification, and enhance existing clinical categorization. Methods: This study involved patients with the Third China National Stroke Registry (CNSR-III). Nine models, which included Natural Gradient Boosting (NGBoost), Categorical Boosting (CatBoost), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Light Gradient Boosting Machine (LGBM), Gradient Boosting Decision Tree (GBDT), Adaptive Boosting (AdaBoost), Support Vector Machine (SVM), and logistic regression (LR), were employed to predict large artery atherosclerosis (LAA), small vessel occlusion (SVO), and cardioembolism (CE) using an 80:20 randomly split training and test set. We designed an SFS-XGB with 10-fold cross-validation for feature selection. The primary evaluation metrics for the models included the area under the receiver operating characteristic curve (AUC) for discrimination and the Brier score (or calibration plots) for calibration. Results: A total of 5,213 patients were included, comprising 2,471 (47.4%) with LAA, 2,153 (41.3%) with SVO, and 589 (11.3%) with CE. In both LAA and SVO models, the AUC values of the ML models were significantly higher than that of the LR model (P < 0.001). The optimal model for predicting SVO (AUC [RF model] = 0.932) outperformed the optimal LAA model (AUC [NGB model] = 0.917) and the optimal CE model (AUC [LGBM model] = 0.846). Each model displayed relatively satisfactory calibration. Further analysis showed that the optimal CE model could identify potential CE patients in the undetermined etiology (SUE) group, accounting for 1,900 out of 4,156 (45.7%). Conclusions: The ML algorithm effectively classified patients with LAA, SVO, and CE, demonstrating superior classification performance compared to the LR model. The optimal ML model can identify potential CE patients among SUE patients. These newly identified predictive factors may complement the existing etiological classification system, enabling clinicians to promptly categorize stroke patients' etiology and initiate optimal strategies for secondary prevention. [ABSTRACT FROM AUTHOR]

In this work, a novel framework for non-destructive, non-intrusive and completely automatic concealed weapon detection under human clothing for entry control and security check applications using passive terahertz (THz) images has been given. The technique models the problem of concealed weapon detection as that of binary segmentation, where the pixels corresponding to concealed weapons are foreground while the remaining are background. The core idea of the proposed framework is to first oversegment the THz image into superpixels and subsequently multiple handcrafted features are extracted from each superpixel. Finally, a machine learning-based classifier is used to perform binary classification thereby classifying each superpixel as foreground or background such that each pixel in a foreground superpixel is a foreground pixel and vice-versa for background. It is worth mentioning that this technique does not require a perfect segmentation to extract features from, rather an oversegmentation suffices. Furthermore, both global features like those based on image saliency and local features like those based on intensity are extracted considering the image properties. The detailed experiments and comparative analysis confirm that the proposed technique efficiently detects concealed weapons, exhibiting superior performance compared to state-of-the-art binary segmentation methods for the purpose. [ABSTRACT FROM AUTHOR]

Recently, renewable energy sources integrated with microgrid (MG) networks have provided safe, secure, and reliable power supply to both utility and industrial purposes. Power quality disturbances (PQDs) seriously affect the performance of MG networks and reduce the lifecycle of numerous sensitive devices in MG networks. Hence, this paper presents a new approach to detect and classify the PQDs using discrete wavelet transform, multiresolution analysis, and optimized‐kernel support vector machine. The obtained unique features from DWT‐MRA are fed to train the well‐known intelligent classifiers. In the optimized‐kernel SVM model, computing power is enhanced for classifying multiple PQ events based on the local density and leave‐one‐out (LOO) algorithm. To get higher separation in feature space, the kernel width of each sample is estimated based on the local density. By using the LOO method, an improved grid search strategy is implemented to get the penalty parameter to achieve satisfactory results. Moreover, a typical MG network is simulated in MATLAB software considering the validation of the proposed technique to address the power quality issues in MG networks, and the results of the proposed method are compared with other conventional ML classifiers. The simulation results confirm that the proposed method is more effective and accurate than other intelligent classifiers. [ABSTRACT FROM AUTHOR]

Purpose: The objective of this research is to explore the applicability of machine learning and fully homomorphic encryption (FHE) in the private pathological assessment, with a focus on the inference phase of support vector machines (SVM) for the classification of confidential medical data. Methods: A framework is introduced that utilizes the Cheon-Kim-Kim-Song (CKKS) FHE scheme, facilitating the execution of SVM inference on encrypted datasets. This framework ensures the privacy of patient data and negates the necessity of decryption during the analytical process. Additionally, an efficient feature extraction technique is presented for the transformation of medical imagery into vectorial representations. Results: The system's evaluation across various datasets substantiates its practicality and efficacy. The proposed method delivers classification accuracy and performance on par with traditional, non-encrypted SVM inference, while upholding a 128-bit security level against established cryptographic attacks targeting the CKKS scheme. The secure inference process is executed within a temporal span of mere seconds. Conclusion: The findings of this study underscore the viability of FHE in enhancing the security and efficiency of bioinformatics analyses, potentially benefiting fields such as cardiology, oncology, and medical imagery. The implications of this research are significant for the future of privacy-preserving machine learning, promoting progress in diagnostic procedures, tailored medical treatments, and clinical investigations. [ABSTRACT FROM AUTHOR]

Objective Methods Results Significance Functional seizures (FS) look like epileptic seizures but are characterized by a lack of epileptic activity in the brain. Approximately one in five referrals to epilepsy clinics are diagnosed with this condition. FS are diagnosed by recording a seizure using video‐electroencephalography (EEG), from which an expert inspects the semiology and the EEG. However, this method can be expensive and inaccessible and can present significant patient burden. No single biomarker has been found to diagnose FS. However, the current limitations in FS diagnosis could be improved with machine learning to classify signal features extracted from EEG, thus providing a potentially very useful aid to clinicians.The current study has investigated the use of seizure‐free EEG signals with machine learning to identify subjects with FS from those with epilepsy. The dataset included interictal and preictal EEG recordings from 48 subjects with FS (mean age = 34.76 ± 10.55 years, 14 males) and 29 subjects with epilepsy (mean age = 38.95 ± 13.93 years, 18 males) from which various statistical, temporal, and spectral features from the five EEG frequency bands were extracted then analyzed with threshold accuracy, five machine learning classifiers, and two feature importance approaches.The highest classification accuracy reported from thresholding was 60.67%. However, the temporal features were the best performing, with the highest balanced accuracy reported by the machine learning models: 95.71% with all frequency bands combined and a support vector machine classifier.Machine learning was much more effective than using individual features and could be a powerful aid in FS diagnosis. Furthermore, combining the frequency bands improved the accuracy of the classifiers in most cases, and the lowest performing EEG bands were consistently delta and gamma. [ABSTRACT FROM AUTHOR]

Summary: Indoor location based on Wi‐Fi fingerprint has attracted extensive attention recently, since it only requires a mobile device and existing network; it does not require additional infrastructure and hardware. However, localization using only received signal strength fingerprint is susceptible to dynamic environments and device heterogeneity. Localization using signal strength differences fingerprint and hyperbolic location fingerprint which extracted from received signal strength fingerprint is a powerful measure to overcome the device heterogeneity. Existing fusion‐based methods do not fully use of the mutual fusion function of multiple fingerprints, which leads to low positioning accuracy. In order to further improve the positioning accuracy, we propose a multifingerprint multiclassifier and two‐layer fusion weight positioning model. The proposed approach first creates a multiple fingerprints group by gleaning signal strength differences fingerprint and hyperbolic location fingerprint from received signal strength fingerprint. In order to deal with the accuracy decline caused by a single classifier in complex scenes, three typical classifiers, random forest (RF), K‐nearest neighbor (KNN), and support vector machine (SVM), are adopted, and a fusion weight selection algorithm is proposed to improve the localization accuracy by choosing fusion weights intelligently from the two‐layer fusion weights. Experiment results show that the proposed scheme is effective to resolve device heterogeneity, and the position accuracy is improved about 16.9% and 5.8%, respectively, compared with single fingerprint and classifier schemes. The results show that, compared with the single fingerprint scheme, the position accuracy is improved at least 16.9% and 28.4% for the homogeneous and heterogeneous equipments, respectively; as for the proposed multiclassifier fusion method, it achieves at least 5.8% and 8.7% accuracy improvement for the cases with homogeneous and heterogeneous equipments, respectively, compared with other traditional classifiers such as RF, KNN, and SVM. [ABSTRACT FROM AUTHOR]

Brain tumor is the most serious and deadly disease, and it is formed due to abnormal cell production. There are two different sorts of tumors including benign (non-cancerous) and malignant (cancerous), and the third level of a brain tumor is cancerous, which is a highly deadly form of cancer. Detecting brain tumors early is crucial for accurate diagnosis and efficient treatment planning. Although numerous techniques are established, the prediction techniques are time-consuming, have higher computational complexity, prone to overfitting, and have limited accuracy. Thus we proposed an Attention Bi-directional Gated Recurrent Unit Ensemble Support Vector Machine (ABES model) for detecting brain tumors. Here, the Bi-GRU layer learns the most important context information for every image. The attention layer also uses the attention mechanism to assign weights to each Bi-GRU layer output. The ensemble SVM classifier performs the categorization of brain tumors. Then, the training set is used to train the Attention BiGRU model and Ensemble SVM classifiers. The ABES classifier weights are optimized by utilizing the improved whale optimization algorithm. To demonstrate the efficiency of our proposed method, we compare it to four existing methods including Fully Automatic Heterogeneous Segmentation-based Support Vector Machine, Whale Harris Hawks Optimization-based Deep Learning, Machine learning-based back propagation neural networks, and Deep Convolutional Neural Network. Also, the ABES model is validated using the MRI dataset and the Figshare brain tumor dataset. Accuracy, precision, specificity, and sensitivity are the performance metrics used to evaluate classification performance, which achieves 97.2% accuracy, 98.61% Precision, 96.48% Specificity, and 97.34% Sensitivity, respectively. [ABSTRACT FROM AUTHOR]

Background: Early detection of T790M mutation in exon 20 of epidermal growth factor receptor (EGFR) in non-small cell lung cancer (NSCLC) patients with brain metastasis is crucial for optimizing treatment strategies. In this study, we developed radiomics models to distinguish NSCLC patients with T790M-positive mutations from those with T790M-negative mutations using multisequence MR images of brain metastasis despite an imbalanced dataset. Various resampling techniques and classifiers were employed to identify the most effective strategy. Methods: Radiomic analyses were conducted on a dataset comprising 125 patients, consisting of 18 with EGFR T790M-positive mutations and 107 with T790M-negative mutations. Seventeen first- and second-order statistical features were selected from CET1WI, T2WI, T2FLAIR, and DWI images. Four classifiers (logistic regression, support vector machine, random forest [RF], and extreme gradient boosting [XGBoost]) were evaluated under 13 different resampling conditions. Results: The area under the curve (AUC) value achieved was 0.89, using the SVM-SMOTE oversampling method in combination with the XGBoost classifier. This performance was measured against the AUC reported in the literature, serving as an upper-bound reference. Additionally, comparable results were observed with other oversampling methods paired with RF or XGBoost classifiers. Conclusions: Our study demonstrates that, even when dealing with an imbalanced EGFR T790M dataset, reasonable predictive outcomes can be achieved by employing an appropriate combination of resampling techniques and classifiers. This approach has significant potential for enhancing T790M mutation detection in NSCLC patients with brain metastasis. [ABSTRACT FROM AUTHOR]

Introduction: Pre-existent pools of coronavirus-specific or cross-reactive T cells were shown to shape the development of cellular and humoral immune responses after primary mRNA vaccination against SARS-CoV-2. However, the cellular determinants of responses to booster vaccination remain incompletely understood. Therefore, we phenotypically and functionally characterized spike antigen-specific T helper (Th) cells in healthy, immunocompetent individuals and correlated the results with cellular and humoral immune responses to BNT162b2 booster vaccination over a six-month period. Methods: Blood of 30 healthy healthcare workers was collected before, 1, 3, and 6 months after their 3rd BNT162b2 vaccination. Whole blood was stimulated with spike peptides and analyzed using flow cytometry, a 13-plex cytokine assay, and n Counter-based transcriptomics. Results: Spike-specific IgG levels at 1 month after booster vaccination correlated with pre-existing CD154+CD69+IFN-γ+CD4+ effector memory cells as well as spike-induced IL-2 and IL-17A secretion. Early post-booster (1-month) spike IgG levels (r=0.49), spike-induced IL‑2 (r=0.58), and spike-induced IFN‑γ release (r=0.43) correlated moderately with their respective long-term (6-month) responses. Sustained robust IgG responses were significantly associated with S-specific (CD69+±CD154+±IFN-γ+) Th-cell frequencies before booster vaccination (p=0.038), especially double/triple-positive type-1 Th cells. Furthermore, spike IgG levels, spike-induced IL‑2 release, and spike-induced IFN‑γ release after 6 months were significantly associated with increased IL‑2 & IL‑4, IP‑10 & MCP1, and IFN‑γ & IP‑10 levels at 1 month post-booster, respectively. On the transcriptional level, induction of pathways associated with both T-cell proliferation and antigen presentation was indicative of sustained spike-induced cytokine release and spike-specific IgG production 6 months post-booster. Using support vector machine models, pre-booster spike-specific T-cell frequencies and early post-booster cytokine responses predicted sustained (6-month) responses with F1 scores of 0.80-1.00.Discussion: In summary, spike-specific Th cells and T-cellular cytokine signatures present before BNT162b2 booster vaccination shape sustained adaptive cellular and humoral responses post-booster. Functional T-cell assays might facilitate early identification of potential non-responders. [ABSTRACT FROM AUTHOR]

EEG-based classification from human states is still challenging in human-computer interaction (HCI). Since it reflects brain activity directly, electroencephalography (EEG) has significant advantages in emotion classification research. This study recorded EEG signals from 12 participants while they perceived emotional audio-visual stimulation movie clips for four minutes. The six emotions studied were anger, excitement, fear, happiness, sadness, and a neutral state. We also perform raw data preprocessing to obtain clean data, extracting the power spectrum using Fast Fourier Transform (FFT) to generate feature vectors. In addition, we conduct extensive experiments to validate the classification of human states using subject-independent machine-learning techniques. As a result, the LSTM network achieved the highest classification accuracy of 81.46% for six emotional states, while the SVM network achieved only 68.64%. In addition, we achieved 82.89% accuracy in the Bi-LSTM network with two layers when applying the deep learning methods to different layers. In conclusion, extensive experiments were conducted on our collected dataset. Experimental results indicate that the LSTM network, a time-sequence-related model, has superior classification results of 82.89% than other methods. It also shows that the long duration of EEG signals is crucial for detecting the emotional state changes of various subject types, including individual subjects and cross-subjects. In future research, we will investigate multiple evaluation experiments utilizing deep learning models or propose novel EEG-based emotion classification features to improve the accuracy of emotion classification. [ABSTRACT FROM AUTHOR]

Successful identification of estrum or other stages in a cycling bitch often requires a combination of methods, including assessment of its behavior, exfoliative vaginal cytology, vaginoscopy, and hormonal assays. Vaginoscopy is a handy and inexpensive tool for the assessment of the breeding period. The present study introduces an innovative method for identifying the stages in the estrous cycle of female canines. With a dataset of 210 vaginoscopic images covering four reproductive stages, this approach extracts deep features using the inception v3 and Residual Networks (ResNet) 152 models. Binary gray wolf optimization (BGWO) is applied for feature optimization, and classification is performed with the extreme gradient boosting (XGBoost) algorithm. Both models are compared with the support vector machine (SVM) with the Gaussian and linear kernel, k-nearest neighbor (KNN), and convolutional neural network (CNN), based on performance metrics such as accuracy, specificity, F1 score, sensitivity, precision, matthew correlation coefficient (MCC), and runtime. The outcomes demonstrate the superiority of the deep model of ResNet 152 with XGBoost classifier, achieving an average model accuracy of 90.37%. The method gave a specific accuracy of 90.91%, 96.38%, 88.37%, and 88.24% in predicting the proestrus, estrus, diestrus, and anestrus stages, respectively. When performing deep feature analysis using inception v3 with the same classifiers, the model achieved an accuracy of 89.41%, which is comparable to the results obtained with the ResNet model. The proposed model offers a reliable system for identifying the optimal mating period, providing breeders and veterinarians with an efficient tool to enhance the success of their breeding programs. [ABSTRACT FROM AUTHOR]

Accurate inflow forecasting is an essential non-engineering strategy to guarantee flood management and boost the effectiveness of the water supply. As inflow is the primary reservoir input, precise inflow forecasting may also offer appropriate reservoir design and management assistance. This study aims to generalize the machine learning model using the support vector machine (SVM), which is support vector regression (SVR), to predict the discharges of the Euphrates River upstream of the Haditha Dam reservoir in Anbar province West of Iraq. Time series data were collected for the period (1986-2024) for the river's daily, monthly, and seasonal flow. Different kernel functions of SVR were applied in this study. The kernels are linear, Quadratic, and Gaussian (RBF). The results showed that the daily time scale is better than the monthly and seasonal performance. In contrast, the linear kernel outperformed the other SVR kernel with a time delay of one day based on the value of the coefficient of determination (R2 = 0.95) and the root mean square error (RMSE = 53.29) m3/sec for predicting daily river flow. The results showed that the proposed machine learning model performed well in predicting the daily flow of the Euphrates River upstream of the Haditha Dam reservoir; this indicates that the model might effectively forecast flows, which helps improve water resource management and dam operations. [ABSTRACT FROM AUTHOR]

Enumeration of Campylobacter from environmental waters can be difficult due to its low concentrations, which can still pose a significant health risk. Spectrophotometry is an approach commonly used for fast detection of water-borne pollutants in water samples, but it has not been used for pathogen detection, which is commonly done through a laborious and time-consuming culture or qPCR Most Probable Number enumeration methods (i.e., MPN-PCR approaches). In this study, we proposed a new method, MPN-Spectro-ML, that can provide rapid evidence of Campylobacter detection and, hence, water concentrations. After an initial incubation, the samples were analysed using a spectrophotometer, and the spectrum data were used to train three machine learning (ML) models (i.e., supported vector machine ‐ SVM, logistic regression–LR, and random forest–RF). The trained models were used to predict the presence of Campylobacter in the enriched water samples and estimate the most probable number (MPN). Over 100 stormwater, river, and creek samples (including both fresh and brackish water) from rural and urban catchments were collected to test the accuracy of the MPN-Spectro-ML method under various scenarios and compared to a previously standardised MPN-PCR method. Differences in the spectrum were found between positive and negative control samples, with two distinctive absorbance peaks between 540-542nm and 575-576nm for positive samples. Further, the three ML models had similar performance irrespective of the scenario tested with average prediction accuracy (ACC) and false negative rates at 0.763 and 13.8%, respectively. However, the predicted MPN of Campylobacter from the new method varied from the traditional MPN-PCR method, with a maximum Nash-Sutcliffe coefficient of 0.44 for the urban catchment dataset. Nevertheless, the MPN values based on these two methods were still comparable, considering the confidence intervals and large uncertainties associated with MPN estimation. The study reveals the potential of this novel approach for providing interim evidence of the presence and levels of Campylobacter within environmental water bodies. This, in turn, decreases the time from risk detection to management for the benefit of public health. [ABSTRACT FROM AUTHOR]

The declining fertility rate and increasing marriage age among girls pose challenges for policymakers, leading to issues such as population decline, higher social and economic costs, and reduced labor productivity. Using machine learning (ML) techniques to predict the desire to have children can offer a promising solution to address these challenges. Therefore, this study aimed to predict the childbearing tendency in women on the verge of marriage using ML techniques. Data from 252 participants (203 expressing a "desire to have children" and 49 indicating "reluctance to have children") in Abadan, and Khorramshahr cities (Khuzestan Province, Iran) was analyzed. Seven ML algorithms, including multilayer perceptron (MLP), support vector machine (SVM), logistic regression (LR), random forest (RF), J48 decision tree, Naive Bayes (NB), and K-nearest neighbors (KNN), were employed. The performance of these algorithms was assessed using metrics derived from the confusion matrix. The RF algorithm showed superior performance, with the highest sensitivity (99.5%), specificity (95.6%), and receiver operating characteristic curve (90.1%) values. Meanwhile, MLP emerged as the top-performing algorithm, showcasing the best overall performance in accuracy (77.75%) and precision (81.8%) compared to other algorithms. Factors such as age of marriage, place of residence, and strength of the family center with the birth of a child were the most effective predictors of a woman's desire to have children. Conversely, the number of daughters, the wife's ethnicity, and the spouse's ownership of assets such as cars and houses were among the least important factors in predicting this desire. ML algorithms exhibit excellent predictive capabilities for childbearing tendencies in women on the verge of marriage, highlighting their remarkable effectiveness. This capacity to offer accurate prognoses holds significant promise for advancing research in this field. [ABSTRACT FROM AUTHOR]

In intensive care unit (ICU) patients undergoing mechanical ventilation (MV), the occurrence of difficult weaning contributes to increased ventilator-related complications, prolonged hospitalization duration, and a significant rise in healthcare costs. Therefore, early identification of influencing factors and prediction of patients at risk of difficult weaning can facilitate early intervention and preventive measures. This study aimed to strengthen airway management for ICU patients by constructing a risk prediction model with comprehensive and individualized offline programs based on machine learning techniques. This study involved the collection of data from 487 patients undergoing MV in the ICU, with a total of 36 variables recorded. The dataset was divided into a training set (70% of the data) and a test set (30% of the data). Five machine learning models, namely logistic regression, random forest, support vector machine, light gradient boosting machine, and extreme gradient boosting, were compared to predict the risk of difficult weaning in ICU patients with MV. Significant influencing factors were identified based on the results of these models, and a risk prediction model for ICU patients with MV was established. When evaluating the models using AUC (Area under the Curve of ROC) and Accuracy as performance metrics, the Random Forest algorithm exhibited the best performance among the five machine learning algorithms. The area under the operating characteristic curve for the subjects was 0.805, with an accuracy of 0.748, recall (0.888), specificity (0.767) and F1 score (0.825). This study successfully developed a risk prediction model for ICU patients with MV using a machine learning algorithm. The Random Forest algorithm demonstrated the highest prediction performance. These findings can assist clinicians in accurately assessing the risk of difficult weaning in patients and formulating effective individualized treatment plans. Ultimately, this can help reduce the risk of difficult weaning and improve the quality of life for patients. [ABSTRACT FROM AUTHOR]

AbstractObjectivesPatients and methodsResultsConclusionNeuropsychological tests (NPTs) are standard tools for assessing cognitive function. These tools can evaluate the cognitive status of a subject, which can be time-consuming and expensive for interpretation. Therefore, this paper aimed to optimize the systematic NPTs by machine learning and develop new classification models for differentiating healthy controls (HC), mild cognitive impairment, and Alzheimer’s disease dementia (ADD) among groups of subjects.A total dataset of 14,926 subjects was obtained from the formal 46 NPTs based on the Seoul Neuropsychological Screening Battery (SNSB). The statistical values of the dataset included an age of 70.18 ± 7.13 with an education level of 8.18 ± 5.50 and a diagnosis group of three; HC, MCI, and ADD. The dataset was preprocessed and classified in two- and three-way machine-learning classification from scikit-learn (www.scikit-learn.org) to differentiate between HC versus MCI, HC versus ADD, HC versus Cognitive Impairment (CI) (MCI + ADD), and HC versus MCI versus ADD. We compared the performance of seven machine learning algorithms, including Naïve Bayes (NB), random forest (RF), decision tree (DT), k-nearest neighbors (KNN), support vector machine (SVM), AdaBoost, and linear discriminant analysis (LDA). The accuracy, sensitivity, specificity, positive predicted value (PPV), negative predictive value (NPV), area under the curve (AUC), confusion matrixes, and receiver operating characteristic (ROC) were obtained from each model based on the test dataset.The trained models based on 29 best-selected NPT features were evaluated, the model with the RF algorithm yielded the best accuracy, sensitivity, specificity, PPV, NPV, and AUC in all four models: HC versus MCI was 98%, 98%, 97%, 98%, 97%, and 99%; HC versus ADD was 98%, 99%, 96%, 97%, 98%, and 99%; HC versus CI was 97%, 99%, 92%, 97%, 97%, and 99% and HC versus MCI versus ADD was 97%, 96%, 98%, 97%, 98%, and 99%, respectively, in predicting of cognitive impairment among subjects.According to the results, the RF algorithm was the best classification model for both two- and three-way classification among the seven algorithms trained on an imbalanced NPTs SNSB dataset. The trained models proved useful for diagnosing MCI and ADD in patients with normal NPTs. These models can optimize cognitive evaluation, enhance diagnostic accuracy, and reduce missed diagnoses. [ABSTRACT FROM AUTHOR]

Metabolic syndrome (MetS) is a cluster of interconnected metabolic risk factors, including abdominal obesity, high blood pressure, and elevated fasting blood glucose levels, that result in an increased risk of heart disease and stroke. In this research, we aim to identify the risk factors that have an impact on MetS in the Bangladeshi population. Subsequently, we intend to construct predictive machine learning (ML) models and ultimately, assess the accuracy and reliability of these models. In this particular study, we utilized the ATP III criteria as the basis for evaluating various health parameters from a dataset comprising 8185 participants in Bangladesh. After employing multiple ML algorithms, we identified that 27.8% of the population exhibited a prevalence of MetS. The prevalence of MetS was higher among females, accounting for 58.3% of the cases, compared to males with a prevalence of 41.7%. Initially, we identified the crucial variables using Chi-Square and Random Forest techniques. Subsequently, the obtained optimal variables are employed to train various models including Decision Trees, Random Forests, Support Vector Machines, Extreme Gradient Boosting, K-nearest neighbors, and Logistic Regression. Particularly we employed the ATP III criteria, which utilizes the Waist-to-Height Ratio (WHtR) as an anthropometric index for diagnosing abdominal obesity. Our analysis indicated that Age, SBP, WHtR, FBG, WC, DBP, marital status, HC, TGs, and smoking emerged as the most significant factors when using Chi-Square and Random Forest analyses. However, further investigation is necessary to evaluate its precision as a classification tool and to improve the accuracy of all classifiers for MetS prediction. [ABSTRACT FROM AUTHOR]