Search

Social desirability bias (SDB) is a common threat to the validity of conclusions from responses to a scale or survey. There is a wide range of person-fit statistics in the literature that can be employed to detect SDB. In addition, machine learning classifiers, such as logistic regression and random forest, have the potential to distinguish between biased and unbiased responses. This study proposes a new application of these classifiers to detect SDB by considering several person-fit indices as features or predictors in the machine learning methods. The results of a Monte Carlo simulation study showed that for a single feature, applying person-fit indices directly and logistic regression led to similar classification results. However, the random forest classifier improved the classification of biased and unbiased responses substantially. Classification was improved in both logistic regression and random forest by considering multiple features simultaneously. Moreover, cross-validation indicated stable area under the curves (AUCs) across machine learning classifiers. A didactical illustration of applying random forest to detect SDB is presented. [ABSTRACT FROM AUTHOR]

As the Internet of Things (IoT) landscape rapidly evolves, robust network security measures are imperative. In particular, Intrusion Detection Systems play a very important role in the preservation of an IoT environment from malicious activities. This paper provides a comprehensive performance comparison of various machine learning classifiers, including K-Nearest Neighbors, Gradient Boosting, XGBoost, Support Vector Machines, Random Forests, Decision Trees, and Extremely Randomized Trees, for intrusion detection in IoT networks. Comparative analysis shows that although all models did very well, the ensemble methods—GB, XGBoost, RF, and ERT—constantly performed better than others in F1-Score, recall, accuracy, and precision. Among them, ERT is turned out to be the most effective model for real-time attack detection on IoT devices, with an accuracy of 99.7% besides excellent precision and recall. XGBoost and RF also turn out to have high reliability and accuracy with F1-Scores of 0.95. These findings further underscore that ensemble methods outperform in intrusion detection for IoT networks and, thus, offer important insights to improve security within networks and protect critical IoT-based infrastructures from a variety of threats.

The performance functions of large-scale complex structures are often implicit and nonlinear, which leads to problems such as high computational costs and low computational accuracy in reliability calculation. In this regard, a dynamic machine learning classifier surrogate model based on Monte Carlo simulation (DMLC-MCS) is proposed. The training sample points are generated by the Markov chain Monte Carlo (MCMC) method and numerical analysis to create the training sample dataset, and the surrogate model based on machine learning classifiers (MLCs) are used to reconstruct the limit state function (LSF). Then, samples are extracted by MCS technique, and the LSF values are predicted by the trained surrogate model. An iterative process is proposed around the most probable point (MPP), and the failure probability obtained by the MCS technique is taken as the convergence condition. If the convergence condition is not satisfied, the MPP information is added to the original sample set to refine the surrogate model. Compared with the traditional reliability method, the proposed method significantly reduces the computational cost on the premise of ensuring high accuracy. In addition, the method is easy to combine with numerical analysis and is proven to be applicable for reliability analysis of real-world complex engineering problems. [ABSTRACT FROM AUTHOR]

Engine fault diagnosis is a critical task in automotive aftermarket management. Developing appropriate fault-labeled datasets can be challenging due to nonlinearity variations and divergence in feature distribution among different engine kinds or operating scenarios. To solve this task, this study experimentally measures audio emission signals from compression ignition engines in different vehicles, simulating injector failures, intake hose failures, and absence of failures. Based on these faults, a hybrid approach is applied to classify different conditions that help the planning and decision-making of the automobile industry. The proposed hybrid approach combines the wavelet packet transform (WPT), Markov blanket feature selection, random convolutional kernel transform (ROCKET), tree-structured Parzen estimator (TPE) for hyperparameters tuning, and ten machine learning (ML) classifiers, such as ridge regression, quadratic discriminant analysis (QDA), naive Bayes, k-nearest neighbors (k-NN), support vector machine (SVM), multilayer perceptron (MLP), random forest (RF), extra trees (ET), gradient boosting machine (GBM), and LightGBM. The audio data are broken down into sub-time series with various frequencies and resolutions using the WPT. These data are subsequently utilized as input for obtaining an informative feature subset using a Markov blanket-based selection method. This feature subset is then fed into the ROCKET method, which is paired with ML classifiers, and tuned using Optuna using the TPE approach. The generalization performance applying the proposed hybrid approach outperforms other standard ML classifiers. [ABSTRACT FROM AUTHOR]

Background: Diabetic retinopathy (DR) is an eye disease developed due to long-term diabetes mellitus, which affects retinal damage. The treatment at the right time supports people in retaining vision, and the early detection of DR is the only solution to prevent blindness. Objective: The development of DR shows few symptoms in the early stage of progression; it is difficult to identify the disease to give treatment from the beginning. Manual diagnosis of DR on fundus images is time-consuming, costly, and liable to be misdiagnosed when compared to computer-aided diagnosis systems. Methods: In this work, we proposed a deep convolutional neural network for the recognition and classification of diabetic retinopathy lesions to identify the severity of the disease. The performance evaluation of the proposed model was tested with other machine learning classifiers such as K-nearest neighbor (KNN), Naïve Bayes (NB), logistic regression (LR), support vector machine (SVM), decision tree (DT), and random forest (RF). Results: Our proposed model achieves 98.5% accuracy for the recognition and classification of the severity level of DR stages such as no DR, mild DR, moderate DR, severe DR, and proliferative DR. Conclusion: The training and testing of our model are carried out on images from the Kaggle APTOS dataset, and this work can act as a base for the autonomous screening of DR. [ABSTRACT FROM AUTHOR]

Brain tumors are frequently classified with high accuracy using convolutional neural networks (CNNs) to better comprehend the spatial connections among pixels in complex pictures. Due to their tiny receptive fields, the majority of deep convolutional neural network (DCNN)-based techniques overfit and are unable to extract global context information from more significant regions. While dilated convolution retains data resolution at the output layer and increases the receptive field without adding computation, stacking several dilated convolutions has the drawback of producing a grid effect. This research suggests a dilated parallel deep convolutional neural network (PDCNN) architecture that preserves a wide receptive field in order to handle gridding artifacts and extract both coarse and fine features from the images. This article applies multiple preprocessing strategies to the input MRI images used to train the model. By contrasting various dilation rates, the global path uses a low dilation rate (2,1,1), while the local path uses a high dilation rate (4,2,1) for decremental even numbers to tackle gridding artifacts and to extract both coarse and fine features from the two parallel paths. Using three different types of MRI datasets, the suggested dilated PDCNN with the average ensemble method performs best. The accuracy achieved for the multiclass Kaggle dataset-III, Figshare dataset-II, and binary tumor identification dataset-I is 98.35%, 98.13%, and 98.67%, respectively. In comparison to state-of-the-art techniques, the suggested structure improves results by extracting both fine and coarse features, making it efficient. [ABSTRACT FROM AUTHOR]

Monitoring river water quality is crucial for safeguarding public health, protecting ecosystems, and ensuring economic sustainability. It helps detect contaminants, ensures drinking water safety, and facilitates early intervention for environmental protection and legal compliance. The objective of this study is to evaluate multiple machine learning algorithms to analyze water quality parameters in computing water quality index (WQI) and classification thereof, aiming to devise a reliable method for forecasting water quality with high accuracy. In this study, fourteen machine learning classifiers applied include Support Vector Machine (SVM), Random Forest (RF), Logistic Regression (LR), Decision Tree (DT), Multilayer Perceptron (MLP), K-Nearest Neighbor (KNN), Naïve Bayes, Gradient boosting, AdaBoost, Bagging, Extra Trees, Quadratic Discriminant Analysis (QDA), XGBoost, and CATBoost. A total of 1096 sample data was used where each data consists of nineteen analytical water quality parameters. To assess the performance of various classifiers, several evaluation techniques were utilized including confusion matrices, classification reports detailing precision and accuracy ratios, and Receiver Operating Characteristic (ROC) curves. The study also utilizes explainable AI (LIME and SHAP) to provide clear insights into the decision-making processes used to classify river water quality. The results indicated that all ML models demonstrate satisfactory performance in predicting WQI. Among the classifiers used, Gradient Boosting achieves the highest Accuracy (99.64 %), Precision (0.95), Recall (0.96), and F1-Score (0.95), indicating its superior ability to correctly classify instances and suggesting a balanced performance across different evaluation metrics. The analysis presented in this article holds the promise of providing accurate water quality data to researchers, thereby enhancing monitoring effectiveness through the application of machine learning techniques.

Abstract This study explores the practical applications of artificial intelligence (AI) in medical imaging, focusing on machine learning classifiers and deep learning models. The aim is to improve detection processes and diagnose diseases effectively. The study emphasizes the importance of teamwork in harnessing AI’s full potential for image analysis. Collaboration between doctors and AI experts is crucial for developing AI tools that bridge the gap between concepts and practical applications. The study demonstrates the effectiveness of machine learning classifiers, such as forest algorithms and deep learning models, in image analysis. These techniques enhance accuracy and expedite image analysis, aiding in the development of accurate medications. The study evidenced that technologically assisted medical image analysis significantly improves efficiency and accuracy across various imaging modalities, including X-ray, ultrasound, CT scans, MRI, etc. The outcomes were supported by the reduced diagnosis time. The exploration also helps us to understand the ethical considerations related to the privacy and security of data, bias, and fairness in algorithms, as well as the role of medical consultation in ensuring responsible AI use in healthcare.

The ubiquitous adoption of Android devices has unfortunately brought a surge in malware threats, compromising user data, privacy concerns, and financial and device integrity, to name a few. To combat this, numerous efforts have explored automated botnet detection mechanisms, with anomaly-based approaches leveraging machine learning (ML) gaining attraction due to their signature-agnostic nature. However, the problem lies in devising accurate ML models which capture the ever evolving landscape of malwares by effectively leveraging all the possible features from Android application packages (APKs).This paper delved into this domain by proposing, implementing, and evaluating an image-based Android malware detection (AMD) framework that harnessed the power of feature hybridization. The core idea of this framework was the conversion of text-based data extracted from Android APKs into grayscale images. The novelty aspect of this work lied in the unique image feature extraction strategies and their subsequent hybridization to achieve accurate malware classification using ML models. More specifically, four distinct feature extraction methodologies, namely, Texture and histogram of oriented gradients (HOG) from spatial domain, and discrete wavelet transform (DWT) and Gabor from the frequency domain were employed to hybridize the features for improved malware identification. To this end, three image-based datasets, namely, Dex, Manifest, and Composite, derived from the information security centre of excellence (ISCX) Android Malware dataset, were leveraged to evaluate the optimal data source for botnet classification. Popular ML classifiers, including naive Bayes (NB), multilayer perceptron (MLP), support vector machine (SVM), and random forest (RF), were employed for the classification task. The experimental results demonstrated the efficacy of the proposed framework, achieving a peak classification accuracy of 93.03% and recall of 97.1% for the RF classifier using the Manifest dataset and a combination of Texture and HOG features. These findings validate the proof-of-concept and provide valuable insights for researchers exploring ML/deep learning (DL) approaches in the domain of AMD.

Abstract This research addresses the accuracy issues in IoT-based human activity recognition (HAR) applications, essential for health monitoring, elderly care, gait analysis, security, and Industry 5.0. This study uses 12 machine learning approaches, split equally between support vector machine (SVM) and k-nearest neighbor (k-NN) models. Data from 102 individuals, aged 18–43, were used to train and test these models. The researchers aimed to detect twelve daily activities, such as sitting, walking, and cycling. Results showed k-NN models achieved slightly higher accuracy (97.08%) compared to SVM models (95.88%), though SVM had faster processing times. The improved machine learning approaches proved effective in accurately classifying daily activities, with k-NN models outperforming SVM models marginally. The paper provides significant contributions to the field of HAR by enhancing the performance of SVM and k-NN classifiers, optimizing them for higher accuracy and faster processing. Through robust testing with samples of real-world data, the study provides a detailed comparative analysis that highlights strengths and weaknesses of each classifier model, specifically within IoT-based systems. This work not only advances the theoretical understanding and practical applications of HAR systems in areas, such as healthcare and industrial automation, but also sets the stage for future research that could explore hybrid models or further enhancements, consequently improving the efficiency and functionality of IoT devices based on activity recognition.

Brain tumors are frequently classified with high accuracy using convolutional neural networks (CNNs) to better comprehend the spatial connections among pixels in complex pictures. Due to their tiny receptive fields, the majority of deep convolutional neural network (DCNN)-based techniques overfit and are unable to extract global context information from more significant regions. While dilated convolution retains data resolution at the output layer and increases the receptive field without adding computation, stacking several dilated convolutions has the drawback of producing a grid effect. This research suggests a dilated parallel deep convolutional neural network (PDCNN) architecture that preserves a wide receptive field in order to handle gridding artifacts and extract both coarse and fine features from the images. This article applies multiple preprocessing strategies to the input MRI images used to train the model. By contrasting various dilation rates, the global path uses a low dilation rate (2,1,1), while the local path uses a high dilation rate (4,2,1) for decremental even numbers to tackle gridding artifacts and to extract both coarse and fine features from the two parallel paths. Using three different types of MRI datasets, the suggested dilated PDCNN with the average ensemble method performs best. The accuracy achieved for the multiclass Kaggle dataset-III, Figshare dataset-II, and binary tumor identification dataset-I is 98.35%, 98.13%, and 98.67%, respectively. In comparison to state-of-the-art techniques, the suggested structure improves results by extracting both fine and coarse features, making it efficient.

This study explores the practical applications of artificial intelligence (AI) in medical imaging, focusing on machine learning classifiers and deep learning models. The aim is to improve detection processes and diagnose diseases effectively. The study emphasizes the importance of teamwork in harnessing AI's full potential for image analysis. Collaboration between doctors and AI experts is crucial for developing AI tools that bridge the gap between concepts and practical applications. The study demonstrates the effectiveness of machine learning classifiers, such as forest algorithms and deep learning models, in image analysis. These techniques enhance accuracy and expedite image analysis, aiding in the development of accurate medications. The study evidenced that technologically assisted medical image analysis significantly improves efficiency and accuracy across various imaging modalities, including X-ray, ultrasound, CT scans, MRI, etc. The outcomes were supported by the reduced diagnosis time. The exploration also helps us to understand the ethical considerations related to the privacy and security of data, bias, and fairness in algorithms, as well as the role of medical consultation in ensuring responsible AI use in healthcare. [ABSTRACT FROM AUTHOR]

The accurate detection of ancient artifacts is very crucial in recognizing and tracking the origin of these relics. The methodologies used in engraving characters onto these objects are different from the ones used in the modern era, prompting the need to develop tools that are accurately tailored to detect these characters. The challenge encountered in developing an object character recognition model for this purpose is the lack of sufficient data needed to train these models. In this work, we propose Styleformer-ART to augment the ancient artifact character images. To show the performance of Styleformer-ART, we compared Styleformer-ART with different state-of-the-art data augmentation techniques. To make a conclusion on the best augmentation method for this special dataset, we evaluated all the augmentation methods employed in this work using the Frétchet inception distance (FID) score between the reference images and the generated images. The methods were also evaluated on the recognition accuracy of a CNN model. The Styleformer-ART model achieved the best FID score of 210.72, and Styleformer-ART-generated images achieved a recognition accuracy with the CNN model of 84%, which is better than all the other reviewed image-generation models. [ABSTRACT FROM AUTHOR]

The ubiquitous adoption of Android devices has unfortunately brought a surge in malware threats, compromising user data, privacy concerns, and financial and device integrity, to name a few. To combat this, numerous efforts have explored automated botnet detection mechanisms, with anomaly-based approaches leveraging machine learning (ML) gaining attraction due to their signature-agnostic nature. However, the problem lies in devising accurate ML models which capture the ever evolving landscape of malwares by effectively leveraging all the possible features from Android application packages (APKs).This paper delved into this domain by proposing, implementing, and evaluating an image-based Android malware detection (AMD) framework that harnessed the power of feature hybridization. The core idea of this framework was the conversion of text-based data extracted from Android APKs into grayscale images. The novelty aspect of this work lied in the unique image feature extraction strategies and their subsequent hybridization to achieve accurate malware classification using ML models. More specifically, four distinct feature extraction methodologies, namely, Texture and histogram of oriented gradients (HOG) from spatial domain, and discrete wavelet transform (DWT) and Gabor from the frequency domain were employed to hybridize the features for improved malware identification. To this end, three image-based datasets, namely, Dex, Manifest, and Composite, derived from the information security centre of excellence (ISCX) Android Malware dataset, were leveraged to evaluate the optimal data source for botnet classification. Popular ML classifiers, including naive Bayes (NB), multilayer perceptron (MLP), support vector machine (SVM), and random forest (RF), were employed for the classification task. The experimental results demonstrated the efficacy of the proposed framework, achieving a peak classification accuracy of 93.03% and recall of 97.1% for the RF classifier using the Manifest dataset and a combination of Texture and HOG features. These findings validate the proof-of-concept and provide valuable insights for researchers exploring ML/deep learning (DL) approaches in the domain of AMD. [ABSTRACT FROM AUTHOR]

The emergence of COVID-19 in late 2019 in Wuhan, China, has led to a global health crisis that has claimed many lives worldwide. A thorough understanding of the available COVID-19 datasets can enable healthcare professionals to identify cases at an early stage. This study presents an innovative pipeline-based framework for predicting survival and mortality in patients with COVID-19 by leveraging the Mexican COVID-19 patient dataset (COVID-19-MPD dataset). Preprocessing plays a pivotal role in ensuring that the framework delivers high-quality outcomes. We deploy various machine learning models with optimized hyperparameters within the framework. Through consistent experimental conditions and dataset utilization, we conducted multiple experiments employing diverse preprocessing techniques and models to maximize the area under the receiver operating characteristic curve (AUC) for COVID-19 prediction. Given the considerable dimensions of the dataset, feature selection is crucial for identifying factors influencing COVID-19 mortality or survival. We employ feature dimension reduction methods, such as principal component analysis and independent component analysis, in addition to feature selection techniques such as maximum relevance minimum redundancy and permutation feature importance. Impactful features related to patient outcomes can significantly aid experts in disease management by enhancing treatment efficacy and control measures. Following various experiments with standardized data and AUC assessment using the k-nearest neighbor algorithm with four components, the proposed framework achieves optimal results, attaining an AUC of 100%. Given its effectiveness in COVID-19 prediction, this framework has the potential for integration into medical decision support systems. [ABSTRACT FROM AUTHOR]

Breathing pattern assessment holds critical importance in clinical practice for detecting respiratory dysfunctions and their impact on health and wellbeing. This systematic literature review investigates the efficacy of inertial sensors in assessing adult human breathing patterns, exploring various methodologies, challenges, and limitations. Utilizing the PSALSAR framework, incorporating the PICOC method and PRISMA statement for comprehensive research, 22 publications were scrutinized from the Scopus, Web of Science, and PubMed databases. A diverse range of sensor fusion methods, data signal analysis techniques, and classifier performances were investigated. Notably, Madgwick's algorithm and the Principal Component Analysis showed superior performance in tracking respiratory movements. Classifiers like Long Short-Term Memory Recurrent Neural Networks exhibited high accuracy in detecting breathing events. Motion artifacts, limited sample sizes, and physiological variability posed challenges, highlighting the need for further research. Optimal sensor configurations were explored, suggesting improvements with multiple sensors, especially in different body postures. In conclusion, this systematic literature review elucidates methods, challenges, and potential future developments in using inertial sensors for assessing adult human breathing patterns. Overcoming the challenges related to sensor placement, motion artifacts, and algorithm development is essential for progress. Future research should focus on extending sensor applications to clinical settings and diverse populations, enhancing respiratory health management. [ABSTRACT FROM AUTHOR]

Abstract An increasingly strong connection between artificial intelligence and medicine has enabled the development of predictive models capable of supporting physicians’ decision-making. Artificial intelligence encompasses much more than machine learning, which nevertheless is its most cited and used sub-branch in the last decade. Since most clinical problems can be modeled through machine learning classifiers, it is essential to discuss their main elements. This review aims to give primary educational insights on the most accessible and widely employed classifiers in radiology field, distinguishing between “shallow” learning (i.e., traditional machine learning) algorithms, including support vector machines, random forest and XGBoost, and “deep” learning architectures including convolutional neural networks and vision transformers. In addition, the paper outlines the key steps for classifiers training and highlights the differences between the most common algorithms and architectures. Although the choice of an algorithm depends on the task and dataset dealing with, general guidelines for classifier selection are proposed in relation to task analysis, dataset size, explainability requirements, and available computing resources. Considering the enormous interest in these innovative models and architectures, the problem of machine learning algorithms interpretability is finally discussed, providing a future perspective on trustworthy artificial intelligence. Relevance statement The growing synergy between artificial intelligence and medicine fosters predictive models aiding physicians. Machine learning classifiers, from shallow learning to deep learning, are offering crucial insights for the development of clinical decision support systems in healthcare. Explainability is a key feature of models that leads systems toward integration into clinical practice. Key points • Training a shallow classifier requires extracting disease-related features from region of interests (e.g., radiomics). • Deep classifiers implement automatic feature extraction and classification. • The classifier selection is based on data and computational resources availability, task, and explanation needs. Graphical Abstract

Electric motors play a fundamental role in various industries, and their relevance is strengthened in the context of the energy transition. Having efficient tools and techniques to detect and diagnose faults in electrical machines is crucial, as is providing early alerts to facilitate prompt decision-making. This study proposes indicators based on the magnitude of the space vector stator current for detecting and diagnosing incipient inter-turn short circuits (ITSCs) in induction motors (IMs). The effectiveness of these indicators was evaluated using four machine learning methods previously documented in the literature: random forests (RFs), support vector machines (SVMs), the k-nearest neighbor (kNN), and feedforward and recurrent neural networks (FNNs and RNNs). This assessment was conducted using experimental data. The results were compared with indicators based on discrete wavelet transform (DWT), demonstrating the viability of the proposed approach, which opens up a way of detecting incipient ITSCs in three-phase IMs. Furthermore, utilizing features derived from the magnitude of the spatial vector led to the successful identification of the phase affected by the fault. [ABSTRACT FROM AUTHOR]

Machine learning techniques have been widely used in the oil and gas industry to improve the qualitative and quantitative characterization of subsurface reservoirs. Because rock properties are strongly influenced by lithological and sedimentological information, lithofacies classification is an important step in 3D reservoir modeling. The aim of this study is to use supervised classification algorithms to predict the spatial distribution pattern of lithofacies classes using borehole and seismic data. In this study, lithofacies classes are distributed away from the wells using a machine-learning classifier. Seismic data attributes extracted from well locations are utilized as training data features in various supervised classification algorithms. Machine learning classifiers trained and evaluated for lithofacies classification include K-nearest neighbors, support vector machine, Gaussian naive Bayes, decision tree, Gradient Boosting, and Random Forests. A number of parameters are optimally determined in order to achieve the highest value of classification accuracy in the model. Comparing machine learning classifiers based on evaluation metrics reveals that ensemble-based decision tree approaches such as Random Forests and Gradient Boosting are the most effective for supervised classification. The results are validated using testing data and have an 80% classification accuracy. The predicted volume of lithofacies classes contributes to improved 3D reservoir modeling for the pre-salt carbonate reservoir. [ABSTRACT FROM AUTHOR]

We investigated the molecular underpinnings of variation in immune responses to the live attenuated typhoid vaccine (Ty21a) by analyzing the baseline immunological profile. We utilized gene expression datasets obtained from the Gene Expression Omnibus (GEO) database (accession number: GSE100665) before and after immunization. We then employed two distinct computational approaches to identify potential baseline biomarkers associated with responsiveness to the Ty21a vaccine. The first pipeline (knowledge- based) involved the retrieval of differentially expressed genes (DEGs), functional enrichment analysis, protein-protein interaction network construction, and topological network analysis of post-immunization datasets before gauging their pre-vaccination expression levels. The second pipeline utilized an unsupervised machine learning algorithm for data-driven feature selection on pre-immunization datasets. Supervised machine-learning classifiers were employed to computationally validate the identified biomarkers. Baseline activation of NKIRAS2 (a negative regulator of NF-kB signalling) and SRC (an adaptor for immune receptor activation) was negatively associated with Ty21a vaccine responsiveness, whereas LOC100134365 exhibited a positive association. The Stochastic Gradient Descent (SGD) algorithm accurately distinguished vaccine responders and non-responders, with 88.8%, 70.3%, and 85.1% accuracy for the three identified genes, respectively. This dual-pronged novel analytical approach provides a comprehensive comparison between knowledge-based and data-driven methods for the prediction of baseline biomarkers associated with Ty21a vaccine responsiveness. The identified genes shed light on the intricate molecular mechanisms that influence vaccine efficacy from the host perspective while pushing the needle further towards the need for development of precise enteric vaccines and on the importance of pre-immunization screening. • Baseline immunological profiles are determiner of post-immunization immune responses. • Immunization with Typhoid vaccine (Ty21a) have shown sub-optimal vaccine efficacies; reasons remain elusive. • Over expression NKIRAS2 , inhibitor of NF-kB signalling and SRC linked to poor vaccine outcome. • LOC100134365 positive marker of Ty21a vaccine induced protection; needs further exploration. • Both knowledge-based and data-driven pipeline can reveal novel insights from gene expression datasets. [ABSTRACT FROM AUTHOR]

COVID-19 is an extremely contagious disease that transmits from person to person by contaminated droplets or virus containing airborne particles. It causes severe damage to a patient's lungs by forming patchy pulmonary lesions and consolidations, which are apparent in chest radiographs such as CXR (X-ray) and CT (computed tomography) images. Therefore, CXR and CT are considered crucial sources of information for early detection of COVID-19 infection. Manual inspection of this information requires expertise, high manpower and substantial amount of time. In order to tackle these issues, deep learning techniques can be utilized in the field of COVID-19 detection. This paper aims at analyzing the performance of different convolutional neural network models in COVID-19 detection using CXR and CT images. These models employ transfer learning and are formed by combining four well-known convolution bases with five distinct machine learning classifiers. All the models are comprehensively trained and tested on CXR and CT datasets each and are thoroughly compared with one another in terms of various evaluation metrics. Amongst these models, the best classification accuracy of 91.18% is provided by the Inception V3 with a neural network classifier on CXR images. Moreover, to assess the improvement of a COVID-19 detection method due to using different techniques, a comparative study of these transfer learning based models with other existing frameworks is also provided. [ABSTRACT FROM AUTHOR]

An increasingly strong connection between artificial intelligence and medicine has enabled the development of predictive models capable of supporting physicians' decision-making. Artificial intelligence encompasses much more than machine learning, which nevertheless is its most cited and used sub-branch in the last decade. Since most clinical problems can be modeled through machine learning classifiers, it is essential to discuss their main elements. This review aims to give primary educational insights on the most accessible and widely employed classifiers in radiology field, distinguishing between "shallow" learning (i.e., traditional machine learning) algorithms, including support vector machines, random forest and XGBoost, and "deep" learning architectures including convolutional neural networks and vision transformers. In addition, the paper outlines the key steps for classifiers training and highlights the differences between the most common algorithms and architectures. Although the choice of an algorithm depends on the task and dataset dealing with, general guidelines for classifier selection are proposed in relation to task analysis, dataset size, explainability requirements, and available computing resources. Considering the enormous interest in these innovative models and architectures, the problem of machine learning algorithms interpretability is finally discussed, providing a future perspective on trustworthy artificial intelligence. Relevance statement The growing synergy between artificial intelligence and medicine fosters predictive models aiding physicians. Machine learning classifiers, from shallow learning to deep learning, are offering crucial insights for the development of clinical decision support systems in healthcare. Explainability is a key feature of models that leads systems toward integration into clinical practice. Key points • Training a shallow classifier requires extracting disease-related features from region of interests (e.g., radiomics). • Deep classifiers implement automatic feature extraction and classification. • The classifier selection is based on data and computational resources availability, task, and explanation needs. [ABSTRACT FROM AUTHOR]

Anorexia nervosa is associated with impaired cognitive flexibility and central coherence, i.e., the ability to provide an overview of complex information. Therefore, the aim of the present study was to evaluate EEG features elicited from patients with anorexia nervosa and healthy controls during mental tasks (valid and invalid Aristotelian syllogisms and paradoxes). Particularly, we examined the combination of the most significant syllogisms with selected features (relative power of the time–frequency domain and wavelet-estimated EEG-specific waves, Higuchi fractal dimension (HFD), and information-oriented approximate entropy (AppEn)). We found that alpha, beta, gamma, theta waves, and AppEn are the most suitable measures, which, when combined with specific syllogisms, form a powerful tool for efficiently classifying healthy subjects and patients with AN. We assessed the performance of triadic combinations of "feature–classifier–syllogism" via machine learning techniques in correctly classifying new subjects in these two groups. The following triads attain the best classifications: (a) "AppEn-invalid-ensemble BT classifier" (accuracy 83.3%), (b) "Higuchi FD-valid-linear discriminant" (accuracy 75%), (c) "alpha amplitude-valid-SVM" (accuracy 83.3%), (d) "alpha RP-paradox-ensemble BT" (accuracy 85%), (e) "beta RP-valid-ensemble" (accuracy 85%), (f) "gamma RP-valid-SVM" (accuracy 85%), and (g) "theta RP-valid-KNN" (accuracy 80%). Our findings suggest that anorexia nervosa has a specific information-processing style across reasoning tasks in the brain as measured via EEG activity. Our findings also contribute to further supporting the view that entropy-oriented, i.e., information-based features (the AppEn measure used in this study) are promising diagnostic tools (biomarkers) in clinical applications related to medical classification problems. Furthermore, the main EEG-specific frequency waves are extremely enhanced and become powerful classification tools when combined with Aristotle's syllogisms. [ABSTRACT FROM AUTHOR]

The escalating evolution of aquaculture has wielded a profound and far-reaching impact on regional sustainable development, ecological equilibrium, and food security. Currently, most aquaculture mapping efforts mainly focus on coastal aquaculture ponds rather than diverse inland aquaculture areas. Recognizing all types of aquaculture areas and accurately classifying different types of aquaculture areas remains a challenge. Here, on the basis of the Google Earth Engine (GEE) and the time-series Sentinel-1 and -2 data, we developed a novel hierarchical framework extraction method for mapping fine inland aquaculture areas (aquaculture ponds + rice-crawfish fields) by employing distinct phenological disparities within two temporal windows (T1 and T2) in Qianjiang, so-called "Home of Chinese Crawfish". Simultaneously, we evaluated the classification performance of four distinct machine learning classifiers, namely Random Forest (RF), Support Vector Machine (SVM), Classification and Regression Trees (CART), and Gradient Boosting (GTB), as well as 11 feature combinations. Following an exhaustive comparative analysis, we selected the optimal machine learning classifier (i.e., the RF classifier) and the optimal feature combination (i.e., feature combination after an automated feature selection method) to classify the aquaculture areas with high accuracy. The results underscore the robustness of the proposed methodology, achieving an outstanding overall accuracy of 93.8%, with an F1 score of 0.94 for aquaculture. The result indicates that an area of 214.6 ± 10.5 km2 of rice-crawfish fields, constituting approximately 83% of the entire aquaculture area in Qianjiang, followed by aquaculture ponds (44.3 ± 10.7 km2, 17%). The proposed hierarchical framework, based on significant phenological characteristics of varied aquaculture types, provides a new approach to monitoring inland freshwater aquaculture in China and other regions of the world. [ABSTRACT FROM AUTHOR]

In electric vehicle applications, early detection of demagnetization faults is crucial for ensuring the smooth operation of Permanent Magnet Synchronous Motors (PMSM). Indeed, an efficient maintenance operation can be carried out with the assistance of defect identification and classification. In this paper, Machine Learning Classifiers (MLC) based demagnetization fault binary classification for PMSM using motor current signal spectral analysis are presented. Threephase current signals are obtained by building a Finite Elements (FE) model with predefined demagnetization faults in order to obtain currents data. The Power Spectral Density (PSD) is used to extract the Amplitude of Sideband Components (ASBCs) from the frequency pattern. In order to classify the demagnetization fault state, different MLC are finally trained and evaluated by using the extracted feature set. The MLC exhibits really encouraging outcomes for accuracy and other recognized performance metrics. The suggested approach outperforms prior research studies with an accuracy that is marginally higher than 99%. [ABSTRACT FROM AUTHOR]

Speech Emotion Recognition (SER) is the process of recognizing and classifying emotions expressed through speech. SER greatly facilitates personalized and empathetic interactions, enhances user experiences, enables sentiment analysis, and finds applications in psychology, healthcare, entertainment, and gaming industries. However, accurately detecting and classifying emotions is a highly challenging task for machines due to the complexity and multifaceted nature of emotions. This work gives a comparative analysis of two approaches for emotion recognition based on original and augmented speech signals. The first approach involves extracting 39 Mel Frequency Cepstrum Coefficients (MFCC) features, while the second approach involves using MFCC spectrograms and extracting features using deep learning models such as MobileNet V2, VGG16, Inception V3, VGG19 and ResNet 50. These features are then tested on Machine learning classifiers such as SVM, Linear SVM, Naive Bayes, k-Nearest Neighbours, Logistic Regression and Random Forest. From the experiments, it is observed that the SVM classifier works best with all the feature extraction techniques Furthermore, to enhance the results, ensembling techniques involving CatBoost, and the Voting classifier along with SVM were utilized, resulting in improved test accuracies of 97.04% on the RAVDESS dataset, 93.24% on the SAVEE dataset, and 99.83% on the TESS dataset, respectively. It is worth noting that both approaches are computationally efficient as they required no training time. [ABSTRACT FROM AUTHOR]

Brain hemorrhages have become increasingly common and can be fatal if left untreated. Current methods for monitoring the progression of the disorder that rely on MRI and PET scans are inconvenient and costly for patients. This has spurred research toward portable and cost-effective techniques for predicting the current stage and malignancy of the hemorrhages. In this study, simulated S-parameter data obtained from a two-antenna system placed over the head is used in conjunction with machine learning to detect the dielectric changes in the brain caused by hemorrhage non-invasively. Several machine learning classifiers are used to analyze the data, and their performance metrics are compared to determine the optimal classifier for this case. The study revealed that Decision Tree, KNN, and Random Forest classifiers are better than SVM and MLP classifiers in terms of accuracy, precision, and recall in predicting Brain hemorrhage at the most probable locations. Contrary to conventional microwave imaging systems requiring several antennas for brain hemorrhage detection, this study demonstrates that integrating machine learning with microwave sensors enables accurate solutions with a reduced antenna count. The results present a transformative strategy for monitoring systems in clinics, where a simple, safe, and low-cost microwave antenna-based system can be intelligently integrated with machine learning to diagnose the presence of Brain hemorrhage.

Software fault prediction is a critical aspect of software engineering aimed at improving software quality and reliability. However, it faces significant challenges, including the class imbalance issue in fault data and the need for robust predictive models that generalize well across different projects. In this research, we delve into these challenges and investigate the impact of class imbalance and model generalization on software fault prediction using cross-project analysis. Our study addresses three primary research questions: Firstly, we examine the critical issue of class imbalance in fault prediction, which poses a significant hurdle to accurate model performance. Through extensive experimentation with various classifiers on diverse datasets from different software projects, we highlight the variations in classifier performance and the necessity of addressing class imbalance for reliable predictions. Secondly, we evaluate the reliability of cross-project prediction, aiming to understand how effectively predictive models trained on one project can generalize to predict faults in other projects. We demonstrate the importance of training with datasets sharing similar characteristics with the target project for achieving reliable cross-project prediction. Thirdly, we analyze the impact of increasing training samples from different projects on prediction accuracy, emphasizing the benefits of utilizing cross-project analysis to enhance predictive model performance. In addition to addressing these research questions, we provide a comprehensive comparison of classifier performance metrics, including accuracy, precision, recall, and F1 Score. Our findings not only shed light on the challenges and opportunities in software fault prediction but also emphasize the importance of considering class imbalance and model generalization for developing robust and reliable fault prediction models. This research contributes to advancing the field by providing insights into effective modeling approaches and highlighting the motivation behind addressing these challenges.

This study aimed to investigate the application of label propagation techniques to propagate labels among photoplethysmogram (PPG) signals, particularly in imbalanced class scenarios and limited data availability scenarios, where clean PPG samples are significantly outnumbered by artifact-contaminated samples. We investigated a dataset comprising PPG recordings from 1571 patients, wherein approximately 82% of the samples were identified as clean, while the remaining 18% were contaminated by artifacts. Our research compares the performance of supervised classifiers, such as conventional classifiers and neural networks (Multi-Layer Perceptron (MLP), Transformers, Fully Convolutional Network (FCN)), with the semi-supervised Label Propagation (LP) algorithm for artifact classification in PPG signals. The results indicate that the LP algorithm achieves a precision of 91%, a recall of 90%, and an F1 score of 90% for the “artifacts” class, showcasing its effectiveness in annotating a medical dataset, even in cases where clean samples are rare. Although the K-Nearest Neighbors (KNN) supervised model demonstrated good results with a precision of 89%, a recall of 95%, and an F1 score of 92%, the semi-supervised algorithm excels in artifact detection. In the case of imbalanced and limited pediatric intensive care environment data, the semi-supervised LP algorithm is promising for artifact detection in PPG signals. The results of this study are important for improving the accuracy of PPG-based health monitoring, particularly in situations in which motion artifacts pose challenges to data interpretation.

Anomaly detection in power systems is crucial for operational reliability and safety, often addressed through binary classification in existing research. However, a research gap exists in multiclass classification for enhanced reliability. To bridge this gap, this study employs four machine learning (ML) classifiers: Random Forest (RF), Decision Tree, Naive Bayes (NB), and Support Vector Machine (SVM) using comprehensive testing on a dataset comprising sixteen indices and their pair combinations (totaling 136 pairs). These classifiers, trained on a dataset derived from simulating a test system with hybrid DGs, exhibit superior anomaly detection, especially with the $\frac {dv}{dq}\& \frac {dv}{dp}$ pair. Among them, RF and DT classifier achieves precision, recall, and F score of unity and outperforming NB and SVM. The performance of the proposed RF and DT classifiers with $\frac {dv}{dq}\& \frac {dv}{dp}$ pair is compared with existing research papers in terms of accuracy and data division. The comparison shows that the proposed RF and DT classifiers with $\frac {dv}{dq}\& \frac {dv}{dp}$ pair achieve 100% accuracy even with 50% data division, whereas other techniques fail to achieve it even at 20% for testing and 80% for training. The study underscores the critical role of pair selection and classifier combinations in effective anomaly detection, facilitating the implementation of robust mitigating strategies for power system stability.

Accurate classification of software requirements, distinguishing between functional and non-functional aspects, is crucial for developing reliable and efficient software systems. However, existing methods often struggle with insufficient semantic understanding and managing diverse software requirements. In this study, we introduce an innovative framework named EnsCL-CatBoost (Ensembled Contrastive Learning with CatBoost) to address these challenges by enhancing classification accuracy and robustness. Our method uses a weighted ensemble of Doc2Vec, Word2Vec, and FastText, leveraging their strengths for richer semantic representation. Unlike conventional strategies, we incorporate contrastive learning with the InfoNCE loss function, boosting discriminative power by clustering similar samples and distancing dissimilar ones, thus enhancing robustness and model generalization. To tackle class imbalance, we integrate SMOTE-Tomek links into the embedding process, achieving balanced class distribution before classification. Additionally, our evaluation extends beyond test datasets to new, unlabeled datasets, demonstrating practical applicability in real-world scenarios. We compare our framework’s performance with five machine learning classifiers which we trained using traditional embedding techniques. The results show that our method significantly outperforms others, achieving 94% accuracy. Our research transparently presents tool-based results, highlighting the transformative potential of automation in software requirement classification and setting a new benchmark for practical deployment in diverse environments, paving the way for future research in the field.

The automatic identification of plant species using unmanned aerial vehicles (UAVs) is a valuable tool for ecological research. However, challenges such as reduced spatial resolution due to high-altitude operations, image degradation from camera optics and sensor limitations, and information loss caused by terrain shadows hinder the accurate classification of plant species from UAV imagery. This study addresses these issues by proposing a novel image preprocessing pipeline and evaluating its impact on model performance. Our approach improves image quality through a multi-step pipeline that includes Enhanced Super-Resolution Generative Adversarial Networks (ESRGAN) for resolution enhancement, Contrast-Limited Adaptive Histogram Equalization (CLAHE) for contrast improvement, and white balance adjustments for accurate color representation. These preprocessing steps ensure high-quality input data, leading to better model performance. For feature extraction and classification, we employ a pre-trained VGG-16 deep convolutional neural network, followed by machine learning classifiers, including Support Vector Machine (SVM), random forest (RF), and Extreme Gradient Boosting (XGBoost). This hybrid approach, combining deep learning for feature extraction with machine learning for classification, not only enhances classification accuracy but also reduces computational resource requirements compared to relying solely on deep learning models. Notably, the VGG-16 + SVM model achieved an outstanding accuracy of 97.88% on a dataset preprocessed with ESRGAN and white balance adjustments, with a precision of 97.9%, a recall of 97.8%, and an F1 score of 0.978. Through a comprehensive comparative study, we demonstrate that the proposed framework, utilizing VGG-16 for feature extraction, SVM for classification, and preprocessed images with ESRGAN and white balance adjustments, achieves superior performance in plant species identification from UAV imagery.

Abstract Background and objective Diabetes is a life-threatening chronic disease with a growing global prevalence, necessitating early diagnosis and treatment to prevent severe complications. Machine learning has emerged as a promising approach for diabetes diagnosis, but challenges such as limited labeled data, frequent missing values, and dataset imbalance hinder the development of accurate prediction models. Therefore, a novel framework is required to address these challenges and improve performance. Methods In this study, we propose an innovative pipeline-based multi-classification framework to predict diabetes in three classes: diabetic, non-diabetic, and prediabetes, using the imbalanced Iraqi Patient Dataset of Diabetes. Our framework incorporates various pre-processing techniques, including duplicate sample removal, attribute conversion, missing value imputation, data normalization and standardization, feature selection, and k-fold cross-validation. Furthermore, we implement multiple machine learning models, such as k-NN, SVM, DT, RF, AdaBoost, and GNB, and introduce a weighted ensemble approach based on the Area Under the Receiver Operating Characteristic Curve (AUC) to address dataset imbalance. Performance optimization is achieved through grid search and Bayesian optimization for hyper-parameter tuning. Results Our proposed model outperforms other machine learning models, including k-NN, SVM, DT, RF, AdaBoost, and GNB, in predicting diabetes. The model achieves high average accuracy, precision, recall, F1-score, and AUC values of 0.9887, 0.9861, 0.9792, 0.9851, and 0.999, respectively. Conclusion Our pipeline-based multi-classification framework demonstrates promising results in accurately predicting diabetes using an imbalanced dataset of Iraqi diabetic patients. The proposed framework addresses the challenges associated with limited labeled data, missing values, and dataset imbalance, leading to improved prediction performance. This study highlights the potential of machine learning techniques in diabetes diagnosis and management, and the proposed framework can serve as a valuable tool for accurate prediction and improved patient care. Further research can build upon our work to refine and optimize the framework and explore its applicability in diverse datasets and populations.

Background: Recurrent and unpredictable seizures characterize epilepsy, a neurological disorder affecting millions worldwide. Epilepsy diagnosis is crucial for timely treatment and better outcomes. Electroencephalography (EEG) timeseries data analysis is essential for epilepsy diagnosis and surveillance. Complex signal processing methods used in traditional EEG analysis are computationally demanding and difficult to generalize across patients. Researchers are using machine learning to improve epilepsy detection, particularly visual feature extraction from EEG time-series data. Objective: This study examines the application of a Gramian Angular Summation Field (GASF) approach for the analysis of EEG signals. Additionally, it explores the utilization of image features, specifically the Scale-Invariant Feature Transform (SIFT) and Oriented FAST and Rotated BRIEF (ORB) techniques, for the purpose of epilepsy detection in EEG data. Methods: The proposed methodology encompasses the transformation of EEG signals into images based on GASF, followed by the extraction of features utilizing SIFT and ORB techniques, and ultimately, the selection of relevant features. A state-of-the-art machine learning classifier is employed to classify GASF images into two categories: normal EEG patterns and focal EEG patterns. Bern-Barcelona EEG recordings were used to test the proposed method. Results: This method classifies EEG signals with 96% accuracy using SIFT features and 94% using ORB features. The Random Forest (RF) classifier surpasses state-of-the-art approaches in precision, recall, F1-score, specificity, and Area Under Curve (AUC). The Receiver Operating Characteristic (ROC) curve shows that Random Forest outperforms Support Vector Machine (SVM) and k-Nearest Neighbors (k-NN) classifiers. Significance: The suggested method has many advantages over time-series EEG data analysis and machine learning classifiers used in epilepsy detection studies. A novel image-based preprocessing pipeline using GASF for robust image synthesis and SIFT and ORB for feature extraction is presented here. The study found that the suggested method can accurately discriminate between normal and focal EEG signals, improving patient outcomes through early and accurate epilepsy diagnosis. [ABSTRACT FROM AUTHOR]