Search Results (4,120)

The early detection of faults in power transformers is crucial for ensuring operational reliability and minimizing system disruptions. This study introduces a novel machine learning framework that integrates Particle Swarm Optimization (PSO) and Dwarf Mongoose Optimization (DMO) algorithms for feature selection and hyperparameter tuning, combined with advanced classifiers such as Decision Trees (DT), Random Forests (RF), and Support Vector Machines (SVM). A 5-fold cross-validation approach was employed to ensure a robust performance evaluation. Feature extraction was performed using both Discrete Wavelet Decomposition (DWD) and Matching Pursuit (MP), providing a comprehensive representation of the dataset comprising 2400 samples and 41 extracted features. Experimental validation demonstrated the efficacy of the proposed framework. The PSO-optimized RF model achieved the highest accuracy of 97.71%, with a precision of 98.02% and an F1 score of 98.63%, followed by the PSO-DT model with a 95.00% accuracy. Similarly, the DMO-optimized RF model recorded an accuracy of 98.33%, with a precision of 98.80% and an F1 score of 99.04%, outperforming other DMO-based classifiers. This novel framework demonstrates significant advancements in transformer protection by enabling accurate and early fault detection, thereby enhancing the reliability and safety of power systems. Full article

Background: Intracranial atherosclerotic stenosis (ICAS) is a leading cause of ischemic stroke, particularly in the anterior circulation. Understanding the underlying stroke mechanisms is essential for guiding personalized treatment strategies. This study proposes an integrated framework that combines CT perfusion imaging, vascular anatomical features, computational fluid dynamics (CFD), and machine learning to classify stroke mechanisms based on the Chinese Ischemic Stroke Subclassification (CISS) system. Methods: A retrospective analysis was conducted on 118 patients with intracranial atherosclerotic stenosis. Key indicators were selected using one-way ANOVA with nested cross-validation and visualized through correlation heatmaps. Optimal thresholds were identified using decision trees. The classification performance of six machine learning models was evaluated using ROC and PR curves. Results: Time to Maximum (Tmax) > 4.0 s, wall shear stress ratio (WSSR), pressure ratio, and percent area stenosis were identified as the most predictive indicators. Thresholds such as Tmax > 4.0 s = 134.0 mL and WSSR = 86.51 effectively distinguished stroke subtypes. The Logistic Regression model demonstrated the best performance (AUC = 0.91, AP = 0.85), followed by Naive Bayes models. Conclusions: This multimodal approach effectively differentiates stroke mechanisms in anterior circulation ICAS and holds promise for supporting more precise diagnosis and personalized treatment in clinical practice. Full article

Background: Glioblastoma multiforme (GBM) is an aggressive brain tumor with a poor prognosis. Traditional diagnosis relies on invasive biopsies, which pose surgical risks. Advances in artificial intelligence (AI) and machine learning (ML) have improved non-invasive GBM diagnosis using magnetic resonance imaging (MRI), offering potential advantages in accuracy and efficiency. Objective: This review aims to identify the methodologies and technologies employed in AI-based GBM diagnostics. It further evaluates the performance of AI models using standard metrics, highlighting both their strengths and limitations. Methodology: In accordance with the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, a systematic review was conducted across major academic databases. A total of 104 articles were identified in the initial search, and 15 studies were selected for final analysis after applying inclusion and exclusion criteria. Outcomes: The  included studies indicated  that the signal T1-weighted imaging (T1WI) is the most frequently used MRI modality in AI-based GBM diagnostics. Multimodal approaches integrating T1WI with diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) have demonstrated improved classification performance. Additionally, AI models have shown potential in surpassing conventional diagnostic methods, enabling automated tumor classification and enhancing prognostic predictions. Full article

Primary angle-closure glaucoma (PACG), characterized by angle closure (AC) with insidious and irreversible progression, requires precise assessment of AC mechanisms for accurate diagnosis and treatment. This study developed an artificial intelligence system, ACM-Assessor, to evaluate AC mechanisms in ultrasound biomicroscopy (UBM) images. A dataset of 8482 UBM images from 1160 patients was retrospectively collected. ACM-Assessor comprises models for pixel-to-physical spacing conversion, anterior chamber angle boundary segmentation, and scleral spur localization, along with three binary classification models to assess pupillary block (PB), thick peripheral iris (TPI), and anteriorly located ciliary body (ALCB). The integrated assessment model classifies AC mechanisms into pure PB, pure non-PB, multiple mechanisms (MM), and others. ACM-Assessor’s evaluation encompassed external testing (2266 images), human–machine competition and assisting beginners’ assessment (an independent test set of 436 images). ACM-Assessor achieved accuracies of 0.924 (PB), 0.925 (TPI), 0.947 (ALCB), and 0.839 (integrated assessment). In man–machine comparisons, the system’s accuracy was comparable to experts (p > 0.05). With model assistance, beginners’ accuracy improved by 0.117 for binary classification and 0.219 for integrated assessment. ACM-Assessor demonstrates expert-level accuracy and enhances beginners’ learning in UBM analysis. Full article

Necrotizing enterocolitis remains one of the most severe gastrointestinal diseases in neonates, particularly affecting preterm infants. It is characterized by intestinal inflammation and necrosis, with significant morbidity and mortality despite advancements in neonatal care. Recent advancements in artificial intelligence (AI) and machine learning (ML) have shown potential in improving NEC prediction, early diagnosis, and management. A systematic search was conducted across multiple databases to explore the application of AI and ML in predicting NEC risk, diagnosing the condition at early stages, and optimizing treatment strategies.AI-based models demonstrated enhanced accuracy in NEC risk stratification compared to traditional clinical approaches. Machine learning algorithms identified novel biomarkers associated with disease onset and severity. Additionally, deep learning applied to medical imaging improved NEC diagnosis by detecting abnormalities earlier than conventional methods. The integration of AI and ML in NEC research provides promising insights into patient-specific risk assessment. However, challenges such as data heterogeneity, model interpretability, and the need for large-scale validation studies remain. Future research should focus on translating AI-driven findings into clinical practice, ensuring ethical considerations and regulatory compliance. Full article

Dairy herds around the world are undergoing several changes. Herd sizes are increasing, as are both milk yield and quality. The implementation of new technologies in various domains of dairy production is leading to an increase in the quantity of data available. This, in turn, creates a need to extract useful information from these data to improve production efficiency. This paper presents the findings of a preliminary study that utilizes a machine learning (ML) approach to assess the accuracy of somatic cell count (SCC) and neutrophils + lymphocytes count/mL (PLCC) in identifying cows at risk of developing intramammary infection (IMI) due to major pathogens. These pathogens (MajPs) include S. aureus, S. agalactiae, S. uberis, and S. dysgalactiae. This study identified these pathogens either by real-time PCR (qPCR) methods or by conventional bacteriology, following the cows’ calving process. This study encompassed a total of 424 cows and 1696 quarter milk samples. A comparison of the two methods revealed significant disparities in the prevalence of MajPs, with the qPCR method demonstrating a higher prevalence than conventional bacteriology. However, the prevalence of negative results was comparable, with both methods yielding approximately 71.0% and 72.1%, respectively. The comprehensive results of this study substantiated that all the cellular markers exhibited the most accurate when MajP IMI was diagnosed using quarter milk samples, but this result is mainly due to the very high specificity. The cellular markers exhibited nearly equivalent performance, irrespective of the ML algorithm employed. The findings indicate that approaches based on SCC or PLCC may be useful for identifying healthy cows or quarters. However, it is essential to confirm all “non-negative” results through subsequent analysis within 7–15 days to ensure accuracy. However, further studies are necessary to enhance diagnostic accuracy. Full article

Machine learning (ML) techniques have attracted many microwave researchers and engineers for their potential to improve performance in microwave- and millimeter-wave-based medical applications. This paper reviews ML algorithms, data acquisition, training techniques, and applications that have emerged in recent years. It also reviews state-of-the-art ML techniques applied for the detection of various organ diseases with microwave signals, achieving more successful results than using traditional methods alone, such as a higher diagnosis accuracy or spatial resolution and significantly improved efficiency. Challenges and the outlook of using ML in future microwave medical applications are also discussed. Full article

As electric vehicles (EVs) are growing, the fault diagnosis in their drive motor becomes more important to have optimal performance and safety. Traditional fault detection methods suffer mainly from high false positive and false negative rates, computational complexity, and lack of transparency in decision-making methods. In addition, existing models are also heavy and inefficient. A lightweight framework for fault diagnosis in EV drive motors is presented with the aid of Recursive Feature Elimination with Cross-Validation (RFE-CV), parameter optimization, and in-depth preprocessing. We further optimize the models and their combination to a hybrid Soft Voting Classifier. These techniques were applied to a dataset of 40,040 data entries that had been simulated by a Variable Frequency Drive (VFD) model. We evaluated eight machine learning models, and our proposed Soft Voting Classifier has the highest test accuracy of 94.52% and a Kappa score of 0.9210 on diagnostic performance. Also, the model has minimal memory usage and low inference latency. In addition, Local Interpretable Model-Agnostic Explanations (LIME) were used to improve transparency and gain an understanding of decisions made through the Soft Voting Classifier. Also, the framework was validated by an additional real-world dataset, thereby further confirming its robustness and consistency in performance for different conditions, which indicates the generalizability of the framework in real-world applications. RFE-CV is found to be very effective in feature selection and helps to construct a lightweight and cost-effective ensemble voting model for enhancing fault diagnosis for EV Drive Motors, overcoming its unsatisfactory transparency, accuracy, and computational efficiency. Finally, it contributes to the development of safer and more reliable EV systems through the development of models supervised on fewer features to give the computing time that is a little lighter without compromising its diagnostic performance. Full article

Precision rolling bearings serve as critical components in a range of diverse industrial applications, where their continuous health monitoring is essential for preventing costly downtime and catastrophic failures. Early-stage bearing defects present significant diagnostic challenges, as they manifest as weak, nonlinear, and non-stationary transient features embedded within high-amplitude random noise. While entropy-based methods have evolved substantially since Shannon’s pioneering work—from approximate entropy to multiscale variants—existing approaches continue to face limitations in their computational efficiency and information preservation. This paper introduces the Adaptive Composite Multiscale Slope Entropy (ACMSlE) framework, which overcomes these constraints through two innovative mechanisms: a time-window shifting strategy, generating overlapping coarse-grained sequences that preserve critical signal information traditionally lost in non-overlapping segmentation, and an adaptive scale optimization algorithm that dynamically selects discriminative scales through entropy variation coefficients. In a comparative analysis against recent innovations, our integrated fault diagnosis framework—combining Fast Ensemble Empirical Mode Decomposition (FEEMD) preprocessing with Particle Swarm Optimization-Extreme Learning Machine (PSO-ELM) classification—achieves 98.7% diagnostic accuracy across multiple bearing defect types and operating conditions. Comprehensive validation through a multidimensional stability analysis, complexity discrimination testing, and data sensitivity analysis confirms this framework’s robust fault separation capability. Full article

Poultry body temperature is closely related to their metabolism and vital activities, which can indicate their physiological status and health. Therefore, monitoring these temperature changes by analyzing thermal images can help in the early and accurate diagnosis of their diseases using a non-destructive method. On the other hand, it is very important to state which part of the bird has the greatest effect on the diagnosis of the disease. This not only speeds up the diagnosis process but also determines an important index for animal pathologists. In this study, an intelligent algorithm was presented with the aim of early diagnosis and classification of two diseases, Avian influenza and Newcastle disease, in the early hours of disease transmission. For this purpose, three different models were developed based on thermal images, including: original images, images with background removal, and images with the head and legs of the chicken separated by the YOLO-v8 model. Then, the features extracted from the thermal images, including texture and color, were evaluated in all three models with a support vector machine (SVM) classifier. Also, the most important and effective features of thermal images for the diagnosis of two diseases, Avian influenza and Newcastle disease, were introduced to other researchers by the Relief feature selection algorithm. The classification results of the original images, images without background and images of the head and legs of chickens for Avian influenza were 75.89, 83.93, and 92.48%, respectively, and for Newcastle disease were 83.04, 91.52, and 94.20% respectively. The model developed for early diagnosis of the disease showed the ability to diagnose the two diseases at 8 h after disease infection with an accuracy of more than 90%. The results show that the contribution of texture-related features is greater than other features extracted from thermal images in the diagnosis of poultry diseases. Also, focusing on the head and feet areas by the YOLO-v8 algorithm will increase the classification accuracy, which allows for more accurate diagnosis in real time and in the early stages of the disease. Full article

Early fault detection in rotating machinery is crucial for optimizing maintenance and minimizing downtime costs, especially in medium-to-large-scale industrial applications. This study presents a multibody model developed in the Simulink^® Simscape environment to simulate the dynamic behavior of medium-sized spherical bearings. The model includes descriptions of the six degrees of freedoms of each subcomponent, and was validated by comparison with experimental measurements acquired on a test rig capable of applying heavy radial loads. The results show a good fit between experimental and simulated signals in terms of identifying characteristic fault frequencies, which highlights the model’s ability to reproduce vibrations induced by localized defects on the inner and outer races. Amplitude differences can be attributed to simplifications such as neglected housing compliancies and lubrication effects, and do not alter the model’s effectiveness in detecting fault signatures. In conclusion, the developed model represents a promising tool for generating useful datasets for training diagnostic and prognostic algorithms, thereby contributing to the improvement of predictive maintenance strategies in industrial settings. Despite some amplitude discrepancies, the model proves useful for generating fault data and supporting condition monitoring strategies for industrial machinery. Full article

Background and Objectives: Ovarian cancer surgery requires multiple radical resections with a high risk of complications. The aim of this single-centre, retrospective study was to determine the best method for predicting Clavien–Dindo grade ≥ III complications using machine learning techniques. Material and Methods: The study included 179 patients who underwent surgery at the gynaecological oncology department of Antalya Training and Research Hospital between January 2015 and December 2020. The data were randomly split into training set n = 134 (75%) and test set n = 45 (25%). We used 49 predictors to develop the best algorithm. Mean absolute error, root mean squared error, correlation coefficients, Mathew’s correlation coefficient, and F1 score were used to determine the best performing algorithm. Cohens’ kappa value was evaluated to analyse the consistency of the model with real data. The relationship between these predicted values and the actual values were then summarised using a confusion matrix. True positive (TP) rate, False positive (FP) rate, precision, recall, and Area under the curve (AUC) values were evaluated to demonstrate clinical usability and classification skills. Results: 139 patients (77.65%) had no morbidity or grade I-II CDC morbidity, while 40 patients (22.35%) had grade III or higher CDC morbidity. BayesNet was found to be the most effective prediction model. No dominant parameter was observed in the Bayesian net importance matrix plot. The true positive (TP) rate was 76%, false positive (FP) rate was 15.6%, recall rate (sensitivity) was 76.9%, and overall accuracy was 82.2% A receiver operating characteristic (ROC) analysis was performed to estimate CDC grade ≥ III. AUC was 0.863 with a statistical significance of p < 0.001, indicating a high degree of accuracy. Conclusions: The Bayesian network model achieved the highest accuracy compared to all other models in predicting CDC Grade ≥ III complications following epithelial ovarian cancer surgery. Full article

This study explores an innovative approach to early Parkinson’s disease (PD) detection by analyzing speech data collected using a mixed reality (MR) system. A total of 57 Polish participants, including PD patients and healthy controls, performed five speech tasks while using an MR head-mounted display (HMD). Speech data were recorded and analyzed to extract acoustic and linguistic features, which were then evaluated using machine learning models, including logistic regression, support vector machines (SVMs), random forests, AdaBoost, and XGBoost. The XGBoost model achieved the best performance, with an F1-score of 0.90 ± 0.05 in the story-retelling task. Key features such as MFCCs (mel-frequency cepstral coefficients), spectral characteristics, RASTA-filtered auditory spectrum, and local shimmer were identified as significant in detecting PD-related speech alterations. Additionally, state-of-the-art deep learning models (wav2vec2, HuBERT, and WavLM) were fine-tuned for PD detection. HuBERT achieved the highest performance, with an F1-score of 0.94 ± 0.04 in the diadochokinetic task, demonstrating the potential of deep learning to capture complex speech patterns linked to neurodegenerative diseases. This study highlights the effectiveness of combining MR technology for speech data collection with advanced machine learning (ML) and deep learning (DL) techniques, offering a non-invasive and high-precision approach to PD diagnosis. The findings hold promise for broader clinical applications, advancing the diagnostic landscape for neurodegenerative disorders. Full article

Artificial intelligence (AI) has emerged as a promising adjunct in the diagnosis and management of infective endocarditis (IE), a disease characterized by diagnostic complexity and significant morbidity. Machine learning (ML) models such as SABIER and SYSUPMIE have demonstrated strong predictive accuracy for early IE diagnosis, embolic risk stratification, and postoperative mortality, surpassing traditional clinical scoring systems. In imaging, AI-enhanced echocardiography and advanced modalities like FDG-PET/CT offer improved sensitivity, specificity, and reduced inter-observer variability, potentially transforming clinical decision making. Additionally, AI-powered microbiological techniques, including MALDI-TOF mass spectrometry combined with ML and neural network-based metagenomic classifiers, show promise in rapidly identifying pathogens and predicting antimicrobial resistance. Despite encouraging early results, widespread adoption faces barriers, including data limitations, interpretability issues, ethical concerns, and the need for robust validation. Future directions include leveraging generative AI as clinical consultative tools, provided their capabilities and limitations are carefully managed. Ultimately, collaborative efforts addressing these challenges could transform IE care, enhancing diagnostic accuracy, clinical outcomes, and patient safety. Full article

Background: Breast cancer is the second leading cause of cancer-related mortality among women, accounting for 12% of cases. Early diagnosis, based on the identification of radiological features, such as masses and microcalcifications in mammograms, is crucial for reducing mortality rates. However, manual interpretation by radiologists is complex and subject to variability, emphasizing the need for automated diagnostic tools to enhance accuracy and efficiency. This study compares a radiomics workflow based on machine learning (ML) with a deep learning (DL) approach for classifying breast lesions as benign or malignant. Methods: matRadiomics was used to extract radiomics features from mammographic images of 1219 patients from the CBIS-DDSM public database, including 581 cases of microcalcifications and 638 of masses. Among the ML models, a linear discriminant analysis (LDA) demonstrated the best performance for both lesion types. External validation was conducted on a private dataset of 222 images to evaluate generalizability to an independent cohort. Additionally, a deep learning approach based on the EfficientNetB6 model was employed for comparison. Results: The LDA model achieved a mean validation AUC of 68.28% for microcalcifications and 61.53% for masses. In the external validation, AUC values of 66.9% and 61.5% were obtained, respectively. In contrast, the EfficientNetB6 model demonstrated superior performance, achieving an AUC of 81.52% for microcalcifications and 76.24% for masses, highlighting the potential of DL for improved diagnostic accuracy. Conclusions: This study underscores the limitations of ML-based radiomics in breast cancer diagnosis. Deep learning proves to be a more effective approach, offering enhanced accuracy and supporting clinicians in improving patient management. Full article