Search

• Two-phase flow regime identification using ML with enhanced interpretability. • Evaluation of various dimensionality reduction techniques to optimize feature representation. • Kernel Fisher discriminant analysis is the best performing dimensionality reduction technique. • AutoML can impartially select the most effective ML classifier. • Our workflow can develop a virtual flow regime map to automate flow regime recognition. Accurate identification of flow regimes is paramount in several industries, especially in chemical and hydrocarbon sectors. This paper describes a comprehensive data-driven workflow for flow regime identification. The workflow encompasses: i) the collection of dynamic pressure signals using an experimentally verified numerical two-phase flow model for three different flow regimes: stratified, slug and annular flow, ii) feature extraction from pressure signals using Discrete Wavelet Transformation (DWT), iii) Evaluation and testing of 12 different Dimensionality Reduction (DR) techniques, iv) the application of an AutoML framework for automated Machine Learning classifier selection among K-Nearest Neighbors, Artificial Neural Networks, Support Vector Machines, Gradient Boosting, Random Forest, and Logistic Regression, with hyper-parameter tuning. Kernel Fisher Discriminant Analysis (KFDA) is the best DR technique, exhibiting superior goodness of clustering, while KNN proved to be the top classifier with an accuracy of 92.5 % and excellent repeatability. The combination of DWT, KFDA and KNN was used to produce a virtual flow regime map. The proposed workflow represents a significant step forward in automating flow regime identification and enhancing the interpretability of ML classifiers, allowing its application to opaque pipes fitted with pressure sensors for achieving flow assurance and automatic monitoring of two-phase flow in various process industries. [ABSTRACT FROM AUTHOR]

Purpose: Bearings play a critical role in the reliable operation of induction machines, and their failure can lead to significant operational challenges and downtime. Detecting and diagnosing these defects is imperative to ensure the longevity of induction machines and preventing costly downtime. The purpose of this paper is to develop a novel approach for diagnosis of bearing faults in induction machine. Design/methodology/approach: To identify the different fault states of the bearing with accurately and efficiently in this paper, the original bearing vibration signal is first decomposed into several intrinsic mode functions (IMFs) using variational mode decomposition (VMD). The IMFs that contain more noise information are selected using the Pearson correlation coefficient. Subsequently, discrete wavelet transform (DWT) is used to filter the noisy IMFs. Second, the composite multiscale weighted permutation entropy (CMWPE) of each component is calculated to form the features vector. Finally, the features vector is reduced using the locality-sensitive discriminant analysis algorithm, to be fed into the support vector machine model for training and classification. Findings: The obtained results showed the ability of the VMD_DWT algorithm to reduce the noise of raw vibration signals. It also demonstrated that the proposed method can effectively extract different fault features from vibration signals. Originality/value: This study suggested a new VMD_DWT method to reduce the noise of the bearing vibration signal. The proposed approach for bearing fault diagnosis of induction machine based on VMD-DWT and CMWPE is highly effective. Its effectiveness has been verified using experimental data. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this paper is to achieve high accuracy in forecasting generation reliability by accurately evaluating the reliability of power systems. This study uses the RTS-79 reliability test system to measure the method's effectiveness, using mean absolute percentage error as the performance metrics. Accurate reliability predictions can inform critical decisions related to system design, expansion and maintenance, making this study relevant to power system planning and management. Design/methodology/approach: This paper proposes a novel approach that uses a radial basis kernel function-based support vector regression method to accurately evaluate the reliability of power systems. The approach selects relevant system features and computes loss of load expectation (LOLE) and expected energy not supplied (EENS) using the analytical unit additional algorithm. The proposed method is evaluated under two scenarios, with changes applied to the load demand side or both the generation system and load profile. Findings: The proposed method predicts LOLE and EENS with high accuracy, especially in the first scenario. The results demonstrate the method's effectiveness in forecasting generation reliability. Accurate reliability predictions can inform critical decisions related to system design, expansion and maintenance. Therefore, the findings of this study have significant implications for power system planning and management. Originality/value: What sets this approach apart is the extraction of several features from both the generation and load sides of the power system, representing a unique contribution to the field. [ABSTRACT FROM AUTHOR]

A patent application titled "Systems And Methods For Analyzing Brain Activity And Applications Thereof" has been published by the US Patent and Trademark Office. The invention aims to improve the resolution of electroencephalography (EEG) by continuously recording brain activity and projecting it onto wavelet packet atoms. The system also monitors physiological and environmental factors, uses machine learning to determine personalized mental states, and provides feedback through visual or audible signals. This invention has potential applications in healthcare and neuroscience research. [Extracted from the article]

A new study conducted at the Affiliated Hospital of Zunyi Medical University in China has developed a high-performance prediction model for the recurrence of ischemic stroke (IS) using deep learning techniques. The model, called the ischemic stroke multi-group learning (ISGL) model, combines magnetic resonance imaging (MRI) data and biochemical test data to predict the risk of stroke recurrence. The ISGL model was found to have an accuracy of 95%, specificity of 96%, sensitivity of 94%, and an area under the curve of 95%. The study also identified risk factors such as low-density lipoprotein, smoking, and heart disease history that positively correlated with stroke recurrence, while total cholesterol, high-density lipoprotein, and diabetes were negatively correlated. This research has implications for early intervention and secondary prevention of stroke. [Extracted from the article]

• An approach is proposed for estimating monotonic change point in multistage process. • The quality of the process at each stage is characterized by simple linear profiles. • Two SVM algorithms are proposed for identifying the occurrence and type of change. • MLE is proposed for estimating both single and multiple monotonic change points. • The performance of the proposed approach is evaluated under various scenarios. In this paper, a hybrid method is proposed for estimating monotonic change points in multistage processes where, at each stage, the quality of the process is represented by simple linear profile models. In the case of monotonic changes, the change type is not known a priori, and the only assumption is that the changes are non-decreasing or non-increasing in nature. In the proposed method, a support vector machine (SVM) algorithm is first developed to identify the parameters experiencing a change in each process stage. Then, the second SVM algorithm is applied to identify the type of change occurring in the related parameters. Finally, considering the identified parameters and change types, the maximum likelihood estimator (MLE) is proposed to estimate the change points of the process under different scenarios, including single and multiple change points, upward and downward step changes, increasing and decreasing linear trends, shift in both parameters of the profiles, shift in different stages of the process, and weak and strong autocorrelation coefficients. The performance of the proposed method is evaluated through extensive simulation experiments. The results indicate that the proposed method recognizes the patterns and estimates the monotonic change points accurately. [ABSTRACT FROM AUTHOR]

Researchers from Zagazig University in Egypt have developed a new framework for the early detection of Alzheimer's disease (AD) using artificial intelligence (AI) and magnetic resonance imaging (MRI) scans. The framework, called SECNN-RF, combines a Squeeze-and-Excitation Convolutional Neural Network (SECNN) with a Random Forest classifier to accurately categorize features extracted from MRI scans. The SECNN-RF achieved high accuracy (99.89%) and offers explainable analysis, making it more transparent and interpretable. The researchers also explored the use of other machine learning algorithms, such as Decision Tree and Support Vector Machine, but found that Random Forest achieved the highest accuracy. [Extracted from the article]

A study conducted by the School of Engineering and Sciences explores the use of machine learning for estimating indoor occupancy. The study compares supervised and semi-supervised machine learning approaches in three university classrooms. The researchers found that semi-supervised learning achieved superior performance compared to supervised models, demonstrating its effectiveness in estimating occupancy while reducing the need for extensive data collection and labeling. This research has implications for space management, energy efficiency, and crowd control, particularly in the context of the COVID-19 pandemic. [Extracted from the article]

A recent report from the Malaviya National Institute of Technology in Rajasthan, India, discusses a new approach to detecting schizophrenia using graph-based features from EEG signals. The researchers utilized graph signal processing to capture the functional connectivity between different brain regions, allowing for a more nuanced understanding of neural network abnormalities associated with schizophrenia. The study achieved a classification accuracy of 97.22% using support vector machine classifiers. This research provides valuable insights into the early detection of schizophrenia and its significance in medical contexts. [Extracted from the article]

This article discusses the use of in silico prediction methods to understand the binding of peptides to human leucocyte antigen class II (HLAII) molecules. Traditional structure-based methods have faced challenges in accuracy due to the dynamic nature of peptide-HLAII interactions. In recent years, machine learning techniques, such as support vector machines and neural networks, have been used to predict peptide-HLAII binding affinities. The article benchmarks the use of AlphaLink2 and AlphaFold2 Multimer in predicting correct peptide binding motifs, which could have implications for vaccine development and protein deimmunization. However, it is important to note that this preprint has not been peer-reviewed. [Extracted from the article]

Researchers at the University Medical Center Utrecht in the Netherlands have conducted a study on vessel disease, specifically focusing on lesion-symptom mapping (LSM) in cerebral small vessel disease. LSM is used to understand how lesions in the brain impact cognitive performance. The researchers compared the performance of neural networks (NNs) with support vector regression (SVR), a current gold standard method. While the CNN with explainable AI (XAI) did not outperform SVR, it was able to detect relevant lesion locations, showing potential for LSM. This research provides insights into the use of machine learning techniques in understanding vessel disease. [Extracted from the article]

In partial label learning (PLL) problem, each training sample corresponds to a group of candidate labels, in which only one label is the ground-truth label (correct label). Almost all the existing PLL algorithms attempt to eliminate the ambiguity of the candidate label sets by treating all labels indiscriminately. However, this kind of approach lacks the consideration of the complexity between the labels and the instances in the training process. Encouraged by the extensive researches of the self-paced learning (SPL) and transfer learning (TL) in various fields, this paper introduces SPL and TL together to address the PLL problem and proposes a new SPL framework, which is called self-paced method for transfer partial label learning (SPTPLL). The proposed model utilizes transfer learning model to share the parameters and regularization terms of the Support Vector Machine (SVM), which can transfer knowledge from the source task to the target task. Additionally, we implement the self-paced learning scheme by choosing a suitable self-paced function to enhance the robustness of the proposed model. In the process of learning iteration, the priority of training examples with their candidate labels is ranked through self-paced learning to control the learning process. Finally, we demonstrate the superior performance of the proposed method through a large number of experiments compared with state-of-the-art baseline methods. [ABSTRACT FROM AUTHOR]

A recent report from the Chinese University of Hong Kong discusses the use of machine learning in predicting heart disease. The researchers highlight the advancements in machine learning and deep learning that have improved the accuracy of heart disease prediction by utilizing various patient data, such as electrocardiograms, historical diagnoses, and genomic information. However, challenges still remain, including algorithmic safety, ethical considerations, and the need for clinical validation. The researchers suggest that future research should focus on integrating multimodal data, enhancing model interpretability, and addressing privacy concerns to further improve predictive accuracy and clinical utility. [Extracted from the article]

A recent study conducted by researchers at Nanjing Drum Tower Hospital Clinical College of Nanjing Medical University explored the value of multiple preprocessors and classifiers in constructing models for predicting microsatellite instability (MSI) status in colorectal cancer (CRC). The study aimed to accurately and non-invasively predict MSI status, which is crucial for cost savings and guiding clinical treatment strategies. The researchers used a retrospective study of 307 CRC patients and performed feature extraction and model construction using various preprocessors and classifiers. The logistic regression model based on the quantile preprocessor exhibited excellent performance and repeatability, suggesting it may improve the model's predictive capabilities for MSI status in CRC. The study also identified tumor location, N stage clinical status, and hypertension as independent clinical predictors of MSI status. The combined model showed excellent predictive performance and clinical applicability. [Extracted from the article]

A recent report discusses research findings on pneumonia and anastomotic leak after transthoracic esophagectomy for cancer. The study aimed to identify predictive risk factors and procedure-related failure-to-rescue (FTR) using a large dataset. The researchers analyzed 1,355 patients and found that factors associated with pneumonia were more numerous than those associated with anastomotic leak. However, the rates of failure-to-rescue did not significantly differ among different operation types, suggesting that minimally-invasive procedures can be safely implemented with experience. Machine-learning models were unable to accurately predict complications or failure-to-rescue. [Extracted from the article]

A recent study conducted at the University of Wisconsin Madison explores the impact of the tumor microenvironment on cancer growth and development. The study focuses on early breast cancer patients and examines the collagen fibers surrounding the tumor tissue to identify potential biomarkers for diagnosis and prognosis. The researchers propose a novel support vector machine (SVM)-based predictive algorithm to analyze the complex relationship between the fibers and the tumor status. The study offers reproducible implementations of the algorithms in the R package mildsvm. This research has been peer-reviewed and published in the Annals of Applied Statistics. [Extracted from the article]

The article focuses on the development of an AI-based diagnostic model for pertussis pneumonia by researchers at Chongqing University. Topics include the use of machine learning algorithms combined with biochemical blood tests, the comparison of diagnostic accuracy among different algorithms, and the identification of key clinical and biochemical features associated with pertussis, with the XGBoost model showing the highest performance in terms of diagnostic accuracy.

A study conducted at Dortmund Hospital in Germany has developed an artificial-intelligent prediction model for the occurrence of cerebral vasospasms, a dangerous complication of a ruptured cerebral aneurysm. The existing scales used to predict the outcome and occurrence of vasospasms are not reliable. The study used an algorithm based on machine learning and achieved an accuracy of 61%-86% in predicting symptomatic vasospasms. The researchers believe that this algorithm can help identify patients at high risk for vasospasms and allow for early intervention to prevent permanent damage. [Extracted from the article]

Researchers from Qingdao Agricultural University in China have proposed a novel method for food safety testing based on infrared spectroscopy. The method, called negentropy-sorted kernel independent component analysis (NS-kICA), was used to analyze Fourier transform infrared spectroscopy data and rank the components based on non-Gaussian independent components. The researchers found that NS-kICA outperformed conventional feature selection methods, and the support vector machine model was the most effective for recognition performance. These findings are significant for food safety testing and can contribute to achieving high, stable, and efficient results. [Extracted from the article]

A recent study explored the role of DNA methylation in lung adenocarcinoma (LUAD), the most common subtype of lung cancer. The study identified differentially methylated sites (DMSs) and found that hypermethylation in certain regions of the DNA was associated with critical developmental processes and disrupted cellular functions, potentially contributing to tumorigenesis. The study also demonstrated the diagnostic potential of these hypermethylated sites and identified distinct methylation subgroups within LUAD tumors. These findings contribute to our understanding of LUAD and highlight the clinical utility of DNA methylation as a diagnostic tool. However, it is important to note that this study has not yet undergone peer review. [Extracted from the article]

A recent study conducted at the University of Utah has found that ataxic breathing (AB), a known effect of opioids, can be used to detect opioid-induced respiratory depression (OIRD) earlier in normal volunteers. The researchers used a support vector machine learning approach to quantify AB in volunteers receiving increasing doses of remifentanil. They found that AB was frequently present even when traditional measures of OIRD, such as reduced mental alertness and bradypnea, were not detected. These findings suggest that AB could be a useful additional measure in recognizing OIRD and further trials are needed to explore its clinical utility with other opioids. [Extracted from the article]

Researchers at Vilnius Gediminas Technical University in Lithuania have conducted a study on the detection of pathological gaits using electromyography (EMG) signals. The aim of the study was to determine the most appropriate advanced methods for distinguishing the gait of healthy children from those with cerebral palsy. The study used an EMG database of 22 children and employed three classification algorithms: support vector machine (SVM), k-nearest neighbors (KNNs), and decision tree (DT). The highest classification accuracy was achieved using the KNN algorithm, indicating that muscle coactivation can be a reliable parameter in gait classification tasks. [Extracted from the article]

The article reports on a new atrial fibrillation detection method developed by Foshan University, which combines advanced techniques like CEPNCC-BiLSTM and photoelectric pulse wave technology to enhance detection accuracy and reduce false positives. Topics discussed include the limitations of traditional detection methods, the improved efficiency and sensitivity of the new method, and the introduction of the ET-score evaluation system for a comprehensive performance assessment.

A researcher at the Georgia Institute of Technology has published a study on support vector machines (SVM) in the field of magnetic resonance elastography (MRE). MRE is an imaging technique that measures tissue stiffness, but poor image acquisition can occur if there is insufficient contact between the MRE actuator and the patient. The researcher developed a sensor-embedded tissue phantom setup paired with an SVM classifier to automate the MRE actuator testing process. The study found that this setup showed potential for improving the MRE actuator development process by automatically detecting contact-related issues before clinical testing. [Extracted from the article]

This article discusses the importance of predicting the quaternary structure and function of proteins in various applications such as pharmaceuticals, drug design, and medical or engineering fields. With the increasing number of protein sequences available, it is crucial to quickly understand the quaternary structural characteristics of proteins for genomics research. The study presents a method using amino acid composition and a support vector machine with a sequential optimization technique to discriminate between hetero-oligomers and non-hetero-oligomers. It is important to note that this preprint has not undergone peer review. [Extracted from the article]

A study conducted in Warangal, India, by researchers from the School of Engineering has developed a hybrid deep learning method for accurately classifying breast lesions as benign or malignant. The method combines a modified Lipschitz-based image augmentation technique, k-fold cross-validation, and a hybrid convolutional neural network (CNN) approach. The deep features extracted from the CNNs are further reduced using a recursive feature elimination technique and then fed into machine learning classifiers. The results show that the proposed model outperforms existing techniques, achieving high accuracy, sensitivity, specificity, precision, and F1 score. This research has been peer-reviewed and provides valuable insights into breast cancer classification. [Extracted from the article]

A recent study conducted by researchers at Southeast University focused on predicting cervical cancer using a machine learning-based approach. Cervical cancer is a leading cause of premature death among women, with the majority of cases occurring in underdeveloped countries. The study utilized early screening and risk trends from individual health records to develop a predictive model. By employing machine learning classifiers and feature selection techniques, the researchers achieved an accuracy rate of 99.19%. The study highlights the effectiveness of the ensemble machine learning classifier in handling classification problems and addressing the issue of high-class imbalance. [Extracted from the article]

Researchers at the University of Ha'il in Saudi Arabia have used artificial intelligence (AI) to model the solubility of the medication Alendronate. By inputting temperature and pressure parameters, the researchers evaluated three different models, with the AdaBoost model showing the highest performance. The research concluded that using this ensemble method, the root mean square error rate was 4.30E-02 and the mean absolute error was 2.96E-02. This study demonstrates the potential of AI in optimizing drug solubility and improving therapeutic effectiveness. [Extracted from the article]

Researchers at Gachon University in Seongnam si, South Korea have developed a novel hybrid approach called Neuro-Health Guardian for precision brain stroke prediction and healthcare analytics. The approach integrates machine learning techniques such as Logistic Regression, Naive Bayes, K-Nearest Neighbors, and Support Vector Machine (SV) to create an ensemble model that accurately predicts stroke risk. The study emphasizes the importance of early stroke detection and intervention to prevent worsening of the condition. The research offers valuable insights for the development of predictive models related to stroke prevention and treatment. [Extracted from the article]

A recent study conducted at the Indian Institute of Technology Kanpur has proposed a new approach to feature extraction in motor imagery-based brain-computer interfaces (BCIs). The traditional method, known as common spatial pattern (CSP), often lacks robustness against variations in electroencephalogram (EEG) signals. The researchers introduced a novel method called SCSP-3, which combines spectral information obtained from power-spectrum estimation with temporal variations. The results showed that SCSP-3 outperformed traditional CSP in terms of accuracy and generalizability, making it a promising alternative for consistent performance across different subject categories. [Extracted from the article]

A recent study conducted by researchers at Motilal Nehru National Institute of Technology Allahabad in Uttar Pradesh, India, focused on using machine learning-based classification models to identify autism spectrum disorder (ASD) using neuroimaging data. The study introduced a novel correlation-based functional connectivity method to extract improved Region of Interest (ROI) coupling features from the Autism Brain Imaging Data Exchange (ABIDE) dataset. The researchers also employed graph embedding domain adaptation (GEDA) to mitigate dataset heterogeneity and enhance classification accuracy. The proposed model achieved an accuracy of 78.1% and AUROC of 83.9% in identifying ASD patients, showing a substantial improvement compared to previous models. The study concluded that there are disruptions in brain connectivity between anterior and posterior brain regions in individuals with ASD. [Extracted from the article]

Researchers from Universitas Muslim Indonesia conducted a study to analyze the performance of three classification models (Decision Tree Classifier, Support Vector Machine, and Naive Bayes Classifier) in predicting lung cancer using the "Lung Cancer Prediction" dataset. The study found that all three models demonstrated good potential in lung cancer prediction, with the Decision Tree Classifier and Naive Bayes Classifier showing similar high performance and the Support Vector Machine exhibiting competitive performance. However, the selection of the best model should consider relevant evaluation metrics and the limitations of each model. Further evaluation and analysis are needed to improve the accuracy and consistency of lung cancer prediction. [Extracted from the article]

A new report discusses research findings on the integration of quantum computing and machine learning in healthcare. The study, conducted by researchers at Yonsei University in Seoul, South Korea, focused on predicting mortality in patients with early-onset colorectal cancer. The researchers compared the performance of quantum support vector machine (QSVM) models with classical support vector machine (SVM) models and found that the QSVM models outperformed the conventional SVM models in terms of predictive accuracy. The study highlights the potential of quantum computing to improve predictive modeling in healthcare, particularly for rare diseases with limited available data. [Extracted from the article]

This article discusses the role of exosomal microRNAs (exomiRs) in cancer metastasis. ExomiRs are small molecules that are transported between cells and play a crucial role in intercellular communication. In cancer, exomiRs influence tumor progression by regulating processes such as proliferation, angiogenesis, and metastasis. The study aims to characterize the role of exomiRs in influencing the pre-metastatic niche (PMN) in various types of cancer. The analysis reveals that exomiRs are associated with specific cancer-related processes and tumor suppressor genes. Additionally, exomiR targets are significantly associated with patient survival and can be used to classify responders and non-responders to chemotherapy. Overall, this study highlights the diagnostic and therapeutic potential of exomiRs in cancer treatment. [Extracted from the article]

A recent report from the Research Institute Children's Cancer Center Hamburg in Germany discusses the characterization of the immune and stromal microenvironment in pediatric low-grade gliomas. The study analyzed gene-expression data from 400 low-grade gliomas and 193 high-grade gliomas to identify different microenvironment cell populations. The researchers found that low-grade gliomas can be divided into three distinct immunologic tumor microenvironments, which may have implications for future immunotherapeutic research. The study also revealed the negative prognostic relevance of vascular markers in low-grade gliomas. [Extracted from the article]

A study conducted by researchers at Peking University in Beijing, China, aimed to establish a noninvasive model for screening metabolic dysfunction-associated steatotic liver disease (MASLD) in children aged 6-14 years. MASLD is becoming increasingly prevalent among children, especially in regions with high obesity rates. The study identified the waist-to-height ratio (WHtR) as the optimal noninvasive index for predicting MASLD, with a cutoff value of >= 0.48. The combination of WHtR and lipid accumulation product (LAP) was recommended for individual MASLD screening. The study also proposed three MASLD screening recommendations based on obesity prevalence in different regions. [Extracted from the article]

A new report from the Rochester Institute of Technology discusses the use of Back Support Industrial Exoskeletons (BSIEs) to reduce the risk of low-back injury during repetitive trunk flexion tasks. The researchers conducted a study with 14 participants who performed trunk bending tasks with and without fatigue, while their muscle activity and whole-body stability were measured. Machine learning algorithms were used to predict perceived back fatigue based on sensor data. The study found that the XGBoost algorithm performed the best in predicting back fatigue, and the findings can be useful in developing fatigue detection products for industrial scenarios. [Extracted from the article]

A recent study conducted by researchers at Islamic Azad University in Tehran, Iran, focused on the classification of breast cancer masses using non-linear quadratic support vector machine (SVM) and compared it with self-organizing neural network (SONN). The study emphasized the importance of early detection in breast cancer and aimed to develop a method for easier and faster diagnosis. The results showed that the SVM method achieved an overall classification accuracy of 98.2%, indicating its optimal performance in breast cancer data classification. This research provides valuable insights into the potential of machine learning techniques in improving breast cancer diagnosis and treatment. [Extracted from the article]

A study conducted by researchers from Eindhoven University of Technology proposes the use of mathematical algorithms to predict hypertensive disorders in pregnancy, including preeclampsia, eclampsia, and chronic hypertension. The study compares different modeling methodologies, including random forest, decision tree, support vector machine, and the Takagi-Sugeno (T-S) and fuzzy c-means (FCM) methods. The T-S and FCM method achieved the highest predictive ability, with a success rate of 90%. The research concludes that these algorithms can be effective tools for predicting hypertensive disorders in pregnancy. [Extracted from the article]

Researchers at the University of Michigan have conducted a study on sepsis, a potentially life-threatening condition caused by bloodstream infections. The study focused on predicting the risk of poor sepsis-related outcomes using minimal variables. The researchers used various machine learning techniques and physiological signals data, such as electronic health records, electrocardiograms, and arterial line signals, to make predictions. They found that structuring the waveform features as a tensor, a mathematical object, allowed for continuous real-time predictions and preserved important structural and temporal information. The study highlights the potential benefits of using these methods for sepsis prediction. [Extracted from the article]

A recent study conducted by the Higher Institute of Engineering and Technology explores the use of a wearable textile bra-tenna system for breast cancer detection. The researchers investigate the effectiveness of different polarization signals combined with machine learning techniques to enhance detection accuracy. The system utilizes machine learning algorithms to predict breast cancer in its early stages, achieving high accuracy rates. The study concludes that the Support Vector Machine algorithm performs exceptionally well in both tumor and non-tumor detection scenarios. The research provides valuable insights for doctors in effectively investigating breast cancer. [Extracted from the article]