Utilizing Support Vector Machine Research Articles

TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients

Federated learning is a distributed collaborative machine learning paradigm that has gained strong momentum in recent years. In federated learning, a central server periodically coordinates models with clients and aggregates the models trained locally by clients without necessitating access to local data. Despite its potential, the implementation of federated learning continues to encounter several challenges, predominantly the slow convergence that is largely due to data heterogeneity. The slow convergence becomes particularly problematic in cross-device federated learning scenarios where clients may be strongly limited by computing power and storage space, and hence counteracting methods that induce additional computation or memory cost on the client side such as auxiliary objective terms and larger training iterations can be impractical. In this paper, we propose a novel federated aggregation strategy, TurboSVM-FL, that poses no additional computation burden on the client side and can significantly accelerate convergence for federated classification task, especially when clients are "lazy" and train their models solely for few epochs for next global aggregation. TurboSVM-FL extensively utilizes support vector machine to conduct selective aggregation and max-margin spread-out regularization on class embeddings. We evaluate TurboSVM-FL on multiple datasets including FEMNIST, CelebA, and Shakespeare using user-independent validation with non-iid data distribution. Our results show that TurboSVM-FL can significantly outperform existing popular algorithms on convergence rate and reduce communication rounds while delivering better test metrics including accuracy, F1 score, and MCC.

A two-stage framework for Arabic social media text misinformation detection combining data augmentation and AraBERT

Misinformation can profoundly impact the reputation of an entity, and eliminating its spread has become a critical concern across various applications. Social media, often a primary source of information, can significantly influence individuals’ perspectives through content from less credible sources. The utilization of machine-learning (ML) algorithms can facilitate automated, large-scale analysis of textual content, contributing to the rapid and efficient processing of extensive datasets for informed decision-making. Since the performance of ML models is highly affected by the size of the training data, many research papers have presented different approaches to solve the problem of limited dataset size. The data augmentation (DA) approach is one of these strategies, aiming to enhance ML model performance by increasing the amount of training data. DA generates new instances by applying different transformations to the original data instances. While many DA techniques have been investigated for various languages, such as English, achieving an enhancement of the classification model’s performance on the new augmented dataset compared to the original dataset, there is a lack of studies on the Arabic language due to its unique characteristics. This paper introduces a novel two-stage framework designed for the automated identification of misinformation in Arabic textual content. The first stage aims to identify the optimal representation of features before feeding them to the ML model. Diverse representations of tweet content are explored, including N-grams, content-based features, and source-based features. The second stage focuses on investigating the DA effect through the back-translation technique applied to the original training data. Back-translation entails translating sentences from the target language (in this case, Arabic) into another language and then back to Arabic. As a result of this procedure, new examples for training are created by introducing variances in the text. The study utilizes support vector machine (SVM), naive Bayes, logistic regression (LR), and random forest (RF) as baseline algorithms. Additionally, AraBERT transformer pre-trained language models are used to relate the instance’s label and feature representation of the input. Experimental outcomes demonstrate that misinformation detection, coupled with data augmentation, enhances accuracy by a noteworthy margin 5 to 12% compared to baseline machine-learning algorithms and pre-trained models. Remarkably, the results show the superiority of the N-grams approach over traditional state-of-the-art feature representations concerning accuracy, recall, precision, and F-measure metrics. This suggests a promising avenue for improving the efficacy of misinformation detection mechanisms in the realm of Arabic text analysis.

Decision-making for sustainable urban transportation: A statistical exploration of innovative mobility solutions and reduced emissions

This research aims to investigate sustainable urban transportation decision-making through innovative mobility solutions, aiming to curtail emissions. To this end, this research presents a smart stochastic architecture to optimize renewable energy-dependent systems, addressing challenges in networked microgrids. By integrating electric vehicle (EV) charging demands and harnessing the unpredictability of renewable energy resources into the urban networks, the research introduces a vehicle-to-grid layout and a novel machine learning-enabled probabilistic Technique. These advancements cover the way for stable, sustainable, and efficient urban transportation systems with reduced environmental impact. A smart stochastic architecture to optimize the operation and management of such systems that rely heavily on renewable energy is proposed in this study, taking into account the poor reliability and complicated energy management in networked microgrids (MGs). The stochastic and variable charging EVs and also the unpredictability of renewable energy resources (RERs) are incorporated in the suggested layout. A vehicle to grid layout, which has become suitable for the MG cost function, is being used for mitigating adverse impacts of vehicles on MGs. In order to support the use of RERs, this study proposes a new machine learning-enabled probabilistic method utilizing support vector machine and point estimation methods for keeping the system in a stable state taking into account their random behavior. Furthermore, this paper develops a novel optimization algorithm using modified particle swarm optimization to determine what is optimal. IEEE systems evaluate the suitability of the suggested smart layout.

Integrating Image Processing and Machine Learning for the Non-Destructive Assessment of RC Beams Damage

Non-destructive testing (NDT) is a crucial method for detecting damages in concrete structures. Structural damage can lead to functional changes, necessitating a range of damage detection techniques. Non-destructive methods enable the pinpointing of the location of the damage without causing harm to the structure, thus saving both time and money. Damaged structures exhibit alterations in their static and dynamic properties, primarily stemming from a reduction in stiffness. Monitoring these changes allows for the determination of the failure location and severity, facilitating timely repairs and reinforcement before further deterioration occurs. A systematic approach to damage detection and assessment is pivotal for fortifying structures and preventing structural collapse, which can result in both financial and human losses. In this study, we employ image processing to categorize damaged beams based on their crack growth and propagation patterns. We also utilize support vector machine (SVM) and k-nearest neighbor (KNN) methods to detect the type, location, and extent of failures in reinforced concrete beams. To provide context and relevance for the laboratory specimens, we will compare our findings to the results from controlled experiments in a controlled laboratory setting.

Diagnosis of attention deficit hyperactivity disorder: A deep learning approach

<abstract> <p>In recent years, there has been significant interest in the analysis and classification of brain dis-orders using electroencephalography (EEG). We presented machine learning and deep learning (DL) frameworks that integrate an EEG-based brain network with various DL models to diagnose attention deficit hyperactivity disorder (ADHD). By incorporating an objective biomarker into the diagnostic process, the accuracy and effectiveness of diagnosis could be enhanced. We used public EEG datasets from 61 ADHD youngsters and 60 normally developing children. The raw EEG data underwent preprocessing, including the application of filters in clinically relevant frequency bands and notch filters. From the preprocessed EEG segments, statistical features (e.g., standard deviation, kurtosis) and spectral features (e.g., entropy) were extracted. Principal component analysis (PCA) and chi-square with PCA were used as feature selection methods to obtain the most useful features and keep them. The machine learning models achieved the highest accuracy result of 94.86% by utilizing support vector machines (SVM) with PCA features. Furthermore, integrating models combining a convolutional neural network (CNN) with bidirectional long short-term memory (BiLSTM) networks, and gated recurrent unit-Transformer (GRU-Transformer block) with Chi-square and PCA features achieved accuracies of 94.50% and 95.59%, respectively. The suggested framework demonstrated a wide range of applicability in addressing the identification of ADHD. To evaluate the performance of the proposed models, comparisons were made with existing models, and the proposed system exhibited superior performance. We enhanced EEG-based analysis and categorization of ADHD by demonstrating the capabilities of advanced artificial intelligence models in enhancing diagnostic accuracy and efficacy.</p> </abstract>

EXplainable artificial intelligence for automatic defect detection in additively manufactured parts using CT scan analysis

AbstractAdditive Manufacturing (AM) and in particular has gained significant attention due to its capability to produce complex geometries using various materials, resulting in cost and mass reduction per part. However, metal AM parts often contain internal defects inherent to the manufacturing process. Non-Destructive Testing (NDT), particularly Computed Tomography (CT), is commonly employed for defect analysis. Today adopted standard inspection techniques are costly and time-consuming, therefore an automatic approach is needed. This paper presents a novel eXplainable Artificial Intelligence (XAI) methodology for defect detection and characterization. To classify pixel data from CT images as pores or inclusions, the proposed method utilizes Support Vector Machine (SVM), a supervised machine learning algorithm, trained with an Area Under the Curve (AUC) of 0.94. Density-Based Spatial Clustering with the Application of Noise (DBSCAN) is subsequently applied to cluster the identified pixels into separate defects, and finally, a convex hull is employed to characterize the identified clusters based on their size and shape. The effectiveness of the methodology is evaluated on Ti6Al4V specimens, comparing the results obtained from manual inspection and the ML-based approach with the guidance of a domain expert. This work establishes a foundation for automated defect detection, highlighting the crucial role of XAI in ensuring trust in NDT, thereby offering new possibilities for the evaluation of AM components.

A comprehensive radiopathological nomogram for the prediction of pathological staging in gastric cancer using CT-derived and WSI-based features

ObjectiveThis study aims to develop and validate an innovative radiopathomics model that combines radiomics and pathomics features to effectively differentiate between stages I-II and stage III gastric cancer (pathological staging). MethodsOur study included 200 patients with well-defined stages of gastric cancer divided into a training cohort (n = 140) and a test cohort (n = 60). Radiomics features were extracted from contrast-enhanced CT images using PyRadiomics, while pathomics features were obtained from whole slide images of pathological specimens through a fine-tuned deep learning model (ResNet-18). After rigorous feature dimensionality reduction and selection, we constructed radiomics models (SVM_rad, LR_rad, and MLP_rad) and pathomics models (SVM_path, LR_path, and MLP_path) utilizing support vector machine (SVM), logistic regression (LR), and multilayer perceptron (MLP) algorithms. The optimal radiomics and pathomics models were chosen based on comprehensive evaluation criteria such as ROC curves, Hosmer‒Lemeshow tests, and calibration curve tests. Feature patterns extracted from the best-performing radiomics model (MLP_rad) and pathomics model (SVM_rad) were integrated to create a powerful radiopathomics nomogram. ResultsFrom a pool of 1834 radiomics features extracted from CT images, 14 were selected to construct radiomics models. Among these, the MLP_rad model exhibited the most robust predictive performance (AUC, training cohort: 0.843; test cohort: 0.797). Likewise, 10 pathomics features were chosen from 512 extracted from whole slide images to build pathomics models, with the SVM_path model demonstrating the highest predictive efficiency (AUC, training cohort: 0.937; test cohort: 0.792). The combined radiopathomics nomogram model exhibited optimal discriminative ability (AUC, training cohort: 0.951; test cohort: 0.837), as confirmed by decision curve analysis (DCA), which indicated superior clinical effectiveness. ConclusionThis study presents a cutting-edge radiopathomics nomogram model designed to predict pathological staging in gastric cancer, distinguishing between stages I-II and stage III. Our research leverages preoperative CT images and histopathological slides to forecast gastric cancer staging accurately, potentially facilitating the estimation of staging before radical gastric cancer surgery in the future.

Modeling and detection of low-altitude flight conflict network based on SVM

With the continuous increase in low altitude flight density, the issue of low altitude navigation safety has attracted widespread attention. Due to the complex low altitude environment, low altitude flight is more susceptible to ground obstacles and weather effects than commercial aviation. In order to ensure the flight safety of helicopters in low altitude airspace, this paper proposes an improved support vector machine based flight conflict detection model. By modeling the conflict network in low altitude flight areas and utilizing Support Vector Machine (SVM) classification features, the safety discrimination of low altitude flight was achieved, ultimately achieving the safety of aircraft in low altitude flight. This article adopts a protected area model that considers the shape of the aircraft as a conflict zone. In order to reduce the complexity of the conflict detection model, an improved ID3 decision tree algorithm and random forest are used to reduce the complexity of the classifier. The study solved the saturation problem of S-type functions in conflict detection models by using more sensitive functions for probability mapping. And use intelligent optimization algorithms to pre train key parameters, achieving efficient conflict detection suitable for low altitude flight.

Enhancing the Recognition of Speakers in Different Distances using Voice Features

This paper proposes a method to enhance speaker recognition at varying distances by adjusting the reference voice based on voice features. Speaker recognition is the process of identifying an individual based on their voice. It involves analyzing and comparing various acoustic features of a person's voice with the reference voice in the database. Conventional speaker recognition techniques have limitations of reduced accuracy when speakers are from varying distances. In this work, we found that high-frequency signals tend to decline faster than low-frequency ones with respect to speaker distance. Based on this, we propose a method that utilizes support vector machines (SVM) to classify speaker distance using sound features, such as the amplitude sum of high-frequency signals and the dynamic range. Once the speaker distance is determined, the reference signal in the database is adjusted according to the distance before being used for speaker recognition. Mel-Frequency Cepstral Coefficients (MFCC) and Dynamic Time Warping (DTW) were employed as the recognition algorithm. Experiments were conducted with speakers placed at three distances, 0.1, 1, and 2.5 meters from the microphone. The experimental results reveal that signals with the frequency of 4 kHz and above experience a faster decline in amplitude than lower ones with increasing distance. The recognition results also demonstrate a significant improvement in accuracy.

Ancient Glass Products Identification and Composition Analysis Based on Support Vector Machine

Studying the chemical composition and weathering-related properties of ancient glass holds immense significance as it served as a vital trading commodity along the early Silk Road in China. The texture transformations of ancient glass are closely linked to its chemical composition and storage conditions. Throughout the weathering process, internal elements continually interact with the surrounding environment, leading to alterations in their chemical proportions. To address this, this article has developed a classification method utilizing support vector machines for identifying glass types and analyzing their compositions. This article research provides archaeologists with a convenient tool for investigating ancient glasses, significantly enhancing the identification process.

Enhancing the livelihoods of small-scale paddy farmers in sri lanka through computational approaches for inclusive and equitable agriculture

Paddy cultivation, also known as rice cultivation, contributes significantly to Sri Lanka's food security and economic prosperity. Nevertheless, a number of challenges exist in the practice, such as insufficient furrow irrigation techniques, inefficient fertilizer subsidy distribution, a lack of yield prediction models, and limited access to research findings. In order to address these concerns, this research proposes an integrated system with four components. An application for furrow irrigation planning is included in the first component, utilizing support vector machine algorithms and random forest algorithms to optimize irrigation paths. Second, a priority-based fertilizer subsidy system is introduced, employing Gradient Boosting to allocate subsidies efficiently based on farmers' needs. The third component focuses on improving the dissemination of research findings by employing text summaries in order to make the findings more accessible to small-scale farmers. In the fourth component, a yield prediction model is implemented using a Random Forest algorithm that takes into account climatic and soil variables in order to forecast expected yields. The integrated system aims to enhance the efficiency, productivity, and profitability of paddy farming in Sri Lanka, providing practical solutions to challenges faced by farmers. By adopting these innovations, farmers can make informed decisions and optimize their agricultural practices, leading to sustainable rice production and economic growth.

Spatio-temporal features for fast early warning of unplanned self-extubation in ICU

Patients’ behaviors in the Intensive Care Units (ICU) have garnered research attention, particularly regarding the impact of Unplanned Extubation (UEX). However, there is currently no existing report on methods for early warning of UEX action in RGB video. Applying traditional human action recognition algorithms to UEX in the complex ICU environment proves challenging. To address the above issue, we propose a novel feature for early warning of UEX action in patients using RGB videos. Firstly, we employ the YOLOv3 detection method to extract the region of interest (ROI), which corresponds to the region where the patient is located. Subsequently, we develop a spatio-temporal (ST) feature for human action tracking by using the L-K optical flow algorithm. This ST feature encompasses optical flow corner number, trajectory distance, and wavelet transform features. Finally, we utilize support vector machine (SVM) for patient action classification and early warning. Experimental results on the ICU monitoring dataset demonstrate the superior performance of the proposed feature in UEX prediction.

Ensemble learning models for the prediction of the weekly peak of PM2.5 concentration in Algiers, Algeria

Introduction: This paper focuses on the prediction of weekly peak levels of Particulate Matter with an aerodynamic diameter of less than 2.5 µm (PM2.5 ), using various Machine Learning (ML) models. The study compares ML models to deep learning models and emphasizes the explain ability of ML models for PM2.5 prediction.
 Materials and methods: We examine different combinations of features and time window dimensions to evaluate the performance of ML models. It utilizes Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), Decision Tree (DT), and five Ensemble Models (EL) including AdaBoost, XGBoost, LightGBM, CatBoost, and Random Forest (RF). The dataset includes three years of daily measurements of weather parameters and PM2.5.
 Results: Lagged values of PM2.5 improves prediction performance, particularly when the lagged value window size spans seven days or multiples thereof. This confirms that road traffic, which exhibits a weekly seasonality, is the primary source of PM2.5 in Algiers. Interestingly, including lagged values of weather parameters decreases prediction performance, even when chosen based on their correlation with PM2.5. The AdaBoost model performs the best, achieving a Root Mean Squared Error (RMSE) of 2.899 µg/m³ and an R2 value of 0.96.
 Conclusion: EL models, specifically AdaBoost, exhibit strong performance in predicting PM 2.5 levels. They not only provide accurate predictions but also allow analysis of feature importance. Lagged values of PM2.5 have a greater impact on predictions compared to weather parameters. Surprisingly, including weather parameters hampers prediction performance. Therefore, the utilization of ensemble learning models offers valuable insights into feature significance.

Occluded Face Recognition Using Machine Learning

A fascinating and intense issue nowadays is the difficult problem of identifying human faces when they are obscured by various obstructions. Due to a sizable occluded zone and a dearth of un-occluded portions, the capacity of most identification algorithms goes off dramatically. We describe a pre-processing approach for occluded faces that is effective and features-based to the fullest extent. Pose, lighting, age, emotions, and hair covering the face are among the 40 different occlusions that are included in the dataset. Edge-preserving enhancement and contrast measurement-correction for the occluded face pictures are part of the two simultaneous procedures used in the pre-processing step. Utilizing Gabor coefficients, Linear Binary Patterns based on Haar Wavelet components, and Histogram of Gaussian features, the finest textural characteristics are retrieved. The other set of characteristics used to represent the entire occluded face region is composed of wavelet components, color histograms, and statistical global features based on first order. In order to validate utilizing support vector machines, the study takes into account 100 celebrities from the CelebA dataset with posture and other occlusions. Compared to other well- known systems, the suggested efficient pre-processing and optimal features-based face recognition system exhibited increased classification accuracy in experiment analysis.

Identification of biomarkers associated with oxidative stress-related genes in postmenopausal osteoporosis.

Osteoporosis (OP) is a prevalent metabolic disease, with aging and menopause being the major risk factors. Studies have shown that nearly one-third of postmenopausal women suffer from osteoporosis. However, there is a scarcity of research on antioxidant systems for the prevention and treatment of postmenopausal osteoporosis (PM-OP). To address this gap, we performed differential analysis using Limma to identify differentially expressed genes (DEGs) in PM-OP samples. We employed weighted correlation network analysis (WGCNA) to identify oxidative stress (OS)-related genes (OSRGs) highly correlated with PM-OP. The intersection of key modular genes and DEGs yielded differentially expressed OSRGs (DE-OSRGs) specific to PM-OP. We conducted GO and KEGG functional enrichment analyses on these genes. Additionally, we constructed a protein-protein interaction (PPI) network and utilized support vector machine recursive feature elimination (SVM-RFE) and random forest (RF) algorithms to identify signature genes. The diagnostic value of the signature genes was evaluated and validated using ROC curves. GSEA enrichment analysis was employed to explore the potential mechanisms associated with the signature genes. Finally, we constructed a regulatory network involving TF-miRNA-mRNA interactions for the signature genes and verified the biological roles of FOXO3 and DDIT3 in PM-OP and healthy groups using quantitative real-time PCR (qRT-PCR). Our analysis revealed 20 DE-OSRGs specific to PM-OP, obtained by intersecting modular and differential genes. The PPI network identified central genes (DDIT3, MAPK8, CDK2, SIRT1, and FOXO3) with more than 3 nodes. Through integration with machine learning algorithms, we identified DDIT3 and FOXO3 as signature genes. The ROC curve analysis indicated that the AUC value was greater than 0.7, suggesting the potential diagnostic value of these signature genes. Furthermore, GSEA results revealed their involvement in pathways related to the regulation of neutrophil activation, oxidative phosphorylation, MAPK signaling, mitochondrial matrix, and phagocytosis. Lastly, we constructed a regulatory network comprising 27 nodes (22 TFs, 3 miRNAs, and 2 mRNAs) and 28 edges. Additionally, qRT-PCR confirmed the significant up-regulation of FOXO3 and DDIT3 expressions in the PM-OP group compared to the healthy control group. In summary, this study employed bioinformatics analysis to identify OS-related biomarkers (DDIT3 and FOXO3) in PM-OP, providing new biological targets for clinical treatment.

A hybrid strategy to extract metadata from scholarly articles by utilizing support vector machine and heuristics

A hybrid strategy to extract metadata from scholarly articles by utilizing support vector machine and heuristics

Impact of DEMs for Improvement Sentinel 2 Lithological Mapping Utilizing Support Vector Machine: A Case Study of Mineralized Fe-Ti-Rich Gabbroic Rocks from the South Eastern Desert of Egypt

Fused remote sensing datasets have greatly contributed to enhancing lithological targets and providing significant information for mineral exploration. For instance, multispectral datasets can discriminate rock units through their unique spectral signatures. Digital Elevation Models (DEMs) could be an effective tool boosting lithological discrimination based mainly on their topographic variations. Consequently, the current study applied the power of the support vector machine (SVM) algorithm and the integrated Phased Array L-type band Synthetic Aperture Radar (PALSAR) DEM and multispectral Sentinel 2 data to: 1—Create an updated lithological map of a poorly mapped arid terrain (Khashir-Ras El-Kharit district, Eastern Desert of Egypt), and 2—Compare and assess the distribution of ferrogabbros with the aim of improving the localization of these rock bodies and investigating their mineral content. Our results enhanced the lithological mapping and revealed a better generalization of mineralized ferrogabbros when the input was a fused DEM with Sentinel 2 compared to the salt and pepper effect observed when adopting the Sentinel 2 solely as the input data. Additionally, the current research strongly recommends detailed exploration programs for Fe-Ti oxide ores within the gabbroic rocks delineated through the resultant thematic map. The ferrogabbros were subjected to a comprehensive analysis involving whole rock geochemistry, XRD, EDX, and light-reflecting investigations, which revealed the existence of magnetite, titanomagnetite, and hematite. The titanomagnetite crystals display subhedral morphology and exhibit fine- to medium-grained surfaces with irregular shapes. X-ray diffraction (XRD) analysis revealed the presence of magnetite and hematite in the concentrated Fe-Ti oxides. Additionally, the whole rock geochemistry highlighted the origin of the ferrogabbros and their proposed tectonic setting as an alkaline to calc-alkaline magma type that developed in a continental arc setting.

Transferable structural, static and dynamic functional MRI features for predicting conversion from mild cognitive impairment to Alzheimer’s disease in Shanghai Memory Study and ADNI cohorts

AbstractBackgroundMild cognitive impairment (MCI) is a critical prodromal stage of Alzheimer’s disease (AD) with high risk of conversion. It is of great challenge to construct reliable biomarkers for predicting conversion from MCI to AD, while the underlying mechanism is still not fully explored. Inter‐dataset generalizability is a prerequisite for clinical use of biomarkers and always a shortage of neuroimaging‐based studies.MethodIn this study, we propose a novel framework by integrating structural MRI (sMRI) and both static and dynamic resting‐state functional MRI (fMRI) measurements to investigate the differences between MCI converters (MCI_C) and non‐converters (MCI_NC), and then utilized support vector machine (SVM) to construct the prediction models based on selected features. A total of 186 MCI patients with both MRI and three‐year outcome data were selected from two independent cohorts: Shanghai Memory Study (SMS) cohort for selection of MRI predictors and internal cross‐validation, and ADNI cohort for external validation on the generalizability of these MRI predictors.ResultIn comparison with MCI_NC, the MRI converters were mainly characterized by alterations of medial temporal lobe (MTL) with atrophy extending to lateral temporal and regional hyperactivity and instability, posterior parietal cortex (PPC) with atrophy and inter‐regional hypo‐connectivity and connectional instability, and occipital cortex with functional instability. All of the imaging‐based prediction models achieved an AUC above 0.7 and ACC above 70% in both SMS and ADNI cohorts. The combination of static and dynamic fMRI features resulted in overall good performance as relative to static or dynamic fMRI solely, supporting the contribution of dynamic features into the fMRI model. In both cohorts, the best imaging model was the multi‐modality MRI model which provided excellent performance with AUC above 0.85 and average ACC/sensitivity/specificity around 80%.ConclusionThis inter‐cohort validation study provides a new insight into the mechanisms of MCI conversion and paves a way for eventual clinical use of MRI biomarkers.

Automated detection of inorganic powders in X-ray images of airport luggage

At the checkpoint, the detection of illicit inorganic powders in passenger luggage using conventional X-ray can be challenging. An algorithm is presented for the automated detection of inorganic powder-like substances from complex X-ray images of highly cluttered passenger bags using computer vision. The proposed method utilizes support vector machine (SVM) classifiers built from local binary patterns (LBP) texture features. When tested on a dataset created in-house, the algorithm achieves a detection precision of 97% and a false positive rate of 3%. This is the first study performed on a realistic dataset, including different amounts and shapes of powders and electronic clutter, and where the success of the automated method is compared with inter-observer variability.

A novel and efficient Wavelet Scattering Transform approach for primitive-stage dyslexia-detection using electroencephalogram signals

Dyslexia is a neurological disorder affecting reading and writing abilities. Early intervention is important for affected individuals’ social and academic development. The accuracy and objectivity limitations of traditional dyslexia detection systems based on behavioral symptoms and standard tests can pose challenges to the early detection of the condition. In response, an electroencephalogram (EEG) based detection method has been proposed to aid medical professionals in addressing these limitations. A comparison is made between the Wavelet Scattering Transform (WST) approach and three other approaches, namely Spectral Statistical Features (SSF), Connectivity Features with Autoencoders (CFA), and Hybrid Features (HF), using two datasets. These two datasets were chosen for various reasons, including the fact that they were collected during different tasks and from different countries. Another significant factor is that the age range of the participants was 7 and 12 years old, marking the beginning of their educational journey. This age range is ideal for detecting dyslexia in its primitive stages, making these datasets a perfect fit for this research. The performance evaluation of the approaches involved utilizing Support Vector Machine (SVM) classifiers with three non-linear kernels and k-fold cross-validation implementation. The findings suggest that the other three approaches could not achieve more than 80% accuracy, and their accuracy results were inconsistent with each dataset. In contrast, the WST approach achieved a high accuracy rate, with an average accuracy of 96.96% and 97.12% for dataset 1 and dataset 2, respectively, using the Radial Basis Function (RBF) kernel. The accuracy of WST features is further improved to 98.72% and 98.67% through the Majority Voting method. These findings demonstrate the effectiveness and generalization of the WST approach.