Search

In human neuroscience, machine learning can help reveal lower-dimensional neural representations relevant to subjects behavior. However, state-of-the-art models typically require large datasets to train, and so are prone to overfitting on human neuroimaging data that often possess few samples but many input dimensions. Here, we capitalized on the fact that the features we seek in human neuroscience are precisely those relevant to subjects behavior rather than noise or other irrelevant factors. We thus developed a Task-Relevant Autoencoder via Classifier Enhancement (TRACE) designed to identify behaviorally-relevant target neural patterns. We benchmarked TRACE against a standard autoencoder and other models for two severely truncated machine learning datasets (to match the data typically available in functional magnetic resonance imaging [fMRI] data for an individual subject), then evaluated all models on fMRI data from 59 subjects who observed animals and objects. TRACE outperformed alternative models nearly unilaterally, showing up to 12% increased classification accuracy and up to 56% improvement in discovering cleaner, task-relevant representations. These results showcase TRACEs potential for a wide variety of data related to human behavior.

Biomarkers that aid in early detection of neurodegeneration are needed to enable early symptomatic treatment and enable identification of people who may benefit from neuroprotective interventions. Increasing evidence suggests that sleep biomarkers may be useful, given the bi-directional relationship between sleep and neurodegeneration and the prominence of sleep disturbances and altered sleep architectural characteristics in several neurodegenerative disorders. This study aimed to demonstrate that sleep can accurately characterize specific neurodegenerative disorders (NDD). A four-class machine-learning algorithm was trained using age and nine sleep biomarkers from patients with clinically-diagnosed manifest and prodromal NDDs, including Alzheimers disease dementia (AD = 27), Lewy body dementia (LBD = 18), and isolated REM sleep behavior disorder (iRBD = 15), as well as a control group (CG = 58). The algorithm was validated in a total of 381 recordings, which included the training data set plus an additional AD = 10, iRBD = 18, Parkinson disease without dementia (PD = 29), mild cognitive impairment (MCI = 78) and CG = 128. Test-retest consistency was then assessed in LBD = 10, AD = 9, and CG = 46. The agreement between the NDD profiles and their respective clinical diagnoses exceeded 75% for the AD, LBD, and CG, and improved when NDD participants classified Likely Normal with NDD indications consistent with their clinical diagnosis were considered. Profiles for iRBD, PD and MCI participants were consistent with the heterogeneity of disease severities, with the majority of overt disagreements explained by normal sleep characterization in 27% of iRBD, 21% of PD, and 26% of MCI participants. For test-retest assignments, the same or similar NDD profiles were obtained for 88% of LBD, 86% in AD, and 98% of CG participants. The potential utility for NDD subtyping based on sleep biomarkers demonstrates promise and requires further prospective development and validation in larger NDD cohorts.

Traditional methods for identifying hit molecules from a large collection of potential drug-like candidates rely on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have a significant limitation in that they require exceptional computing capabilities for even relatively small collections of molecules. Increasingly large and complex state-of-the-art deep learning approaches have gained popularity with the promise to improve the productivity of drug design, notorious for its numerous failures. However, as deep learning models increase in their size and complexity, their acceleration at the hardware level becomes more challenging. Hyperdimensional Computing (HDC) has recently gained attention in the computer hardware community due to its algorithmic simplicity relative to deep learning approaches. The HDC learning paradigm, which represents data with high-dimension binary vectors, allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas (computer vision, bioinformatics, mass spectrometery, remote sensing, edge devices, etc.). To the best of our knowledge, our work is the first to consider HDC for the task of fast and efficient screening of modern drug-like compound libraries. We also propose the first HDC graph-based encoding methods for molecular data, demonstrating consistent and substantial improvement over previous work. We compare our approaches to alternative approaches on the well-studied MoleculeNet dataset and the recently proposed LIT-PCBA dataset derived from high quality PubChem assays. We demonstrate our methods on multiple target hardware platforms, including Graphics Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs), showing at least an order of magnitude improvement in energy efficiency versus even our smallest neural network baseline model with a single hidden layer. Our work thus motivates further investigation into molecular representation learning to develop ultra-efficient pre-screening tools. We make our code publicly available at https://github.com/LLNL/hdbind .

The rapidly evolving field of radiomics has shown that radiomic features are able to capture characteristics of both tumor and normal tissue that can be used to make accurate and clinically relevant predictions. In the present study we sought to determine if radiomic features can characterize the adverse effects caused by normal tissue injury as well as identify if human embryonic stem cell (hESC) derived extracellular vesicle (EV) treatment can resolve certain adverse complications. A cohort of 72 mice (n = 12 per treatment group) were exposed to X-ray radiation to the whole lung (3 × 8 Gy) or to the apex of the right lung (3 × 12 Gy), immediately followed by retro-orbital injection of EVs. Cone-Beam Computed Tomography images were acquired before and 2 weeks after treatment. In total, 851 radiomic features were extracted from the whole lungs and

Preterm birth (PTB) remains a global health concern, impacting neonatal mortality and lifelong health consequences. Traditional methods for estimating PTB rely on electronic health records or biomedical signals, limited to short-term assessments in clinical settings. Recent studies have leveraged wearable technologies for in-home maternal health monitoring, offering continuous assessment of maternal autonomic nervous system (ANS) activity and facilitating the exploration of PTB risk. In this paper, we conduct a longitudinal study to assess the risk of PTB by examining maternal ANS activity through heart rate (HR) and heart rate variability (HRV). To achieve this, we collect long-term raw photoplethysmogram (PPG) signals from 58 pregnant women (including seven preterm cases) from gestational weeks 12-15 to three months post-delivery using smartwatches in daily life settings. We employ a PPG processing pipeline to accurately extract HR and HRV, and an autoencoder machine learning model with SHAP analysis to generate explainable abnormality scores indicative of PTB risk. Our results reveal distinctive patterns in PTB abnormality scores during the second pregnancy trimester, indicating the potential for early PTB risk estimation. Moreover, we find that HR, average of interbeat intervals (AVNN), SD1SD2 ratio, and standard deviation of interbeat intervals (SDNN) emerge as significant PTB indicators.

Machine learning (ML)-driven diagnosis systems are particularly relevant in pediatrics given the well-documented impact of early-life health conditions on later-life outcomes. Yet, early identification of diseases and their subsequent impact on length of hospital stay for this age group has so far remained uncharacterized, likely because access to relevant health data is severely limited. Thanks to a confidential data use agreement with the California Department of Health Care Access and Information, we introduce Ped-BERT: a state-of-the-art deep learning model that accurately predicts the likelihood of 100+ conditions and the length of stay in a pediatric patients next medical visit. We link mother-specific pre- and postnatal period health information to pediatric patient hospital discharge and emergency room visits. Our data set comprises 513.9K mother-baby pairs and contains medical diagnosis codes, length of stay, as well as temporal and spatial pediatric patient characteristics, such as age and residency zip code at the time of visit. Following the popular bidirectional encoder representations from the transformers (BERT) approach, we pre-train Ped-BERT via the masked language modeling objective to learn embedding features for the diagnosis codes contained in our data. We then continue to fine-tune our model to accurately predict primary diagnosis outcomes and length of stay for a pediatric patients next visit, given the history of previous visits and, optionally, the mothers pre- and postnatal health information. We find that Ped-BERT generally outperforms contemporary and state-of-the-art classifiers when trained with minimum features. We also find that incorporating mother health attributes leads to significant improvements in model performance overall and across all patient subgroups in our data. Our most successful Ped-BERT model configuration achieves an area under the receiver operator curve (ROC AUC) of 0.927 and an average precision score (APS) of 0.408 for the diagnosis prediction task, and a ROC AUC of 0.855 and APS of 0.815 for the length of hospital stay task. Further, we examine Ped-BERTs fairness by determining whether prediction errors are evenly distributed across various subgroups of mother-baby demographics and health characteristics, or if certain subgroups exhibit a higher susceptibility to prediction errors.

Continued spread of chronic wasting disease (CWD) through wild cervid herds negatively impacts populations, erodes wildlife conservation, drains resource dollars, and challenges wildlife management agencies. Risk factors for CWD have been investigated at state scales, but a regional model to predict locations of new infections can guide increasingly efficient surveillance efforts. We predicted CWD incidence by county using CWD surveillance data depicting white-tailed deer (Odocoileus virginianus) in 16 eastern and midwestern US states. We predicted the binary outcome of CWD-status using four machine learning models, utilized five-fold cross-validation and grid search to pinpoint the best model, then compared model predictions against the subsequent year of surveillance data. Cross validation revealed that the Light Boosting Gradient model was the most reliable predictor given the regional data. The predictive model could be helpful for surveillance planning. Predictions of false positives emphasize areas that warrant targeted CWD surveillance because of similar conditions with counties known to harbor CWD. However, disagreements in positives and negatives between the CWD Prediction Web App predictions and the on-the-ground surveillance data one year later underscore the need for state wildlife agency professionals to use a layered modeling approach to ensure robust surveillance planning. The CWD Prediction Web App is at https://cwd-predict.streamlit.app/ .

We construct non-linear machine learning (ML) prediction models for systolic and diastolic blood pressure (SBP, DBP) using demographic and clinical variables and polygenic risk scores (PRSs). We developed a two-model ensemble, consisting of a baseline model, where prediction is based on demographic and clinical variables only, and a genetic model, where we also include PRSs. We evaluate the use of a linear versus a non-linear model at both the baseline and the genetic model levels and assess the improvement in performance when incorporating multiple PRSs. We report the ensemble models performance as percentage variance explained (PVE) on a held-out test dataset. A non-linear baseline model improved the PVEs from 28.1 to 30.1% (SBP) and 14.3% to 17.4% (DBP) compared with a linear baseline model. Including seven PRSs in the genetic model computed based on the largest available GWAS of SBP/DBP improved the genetic model PVE from 4.8 to 5.1% (SBP) and 4.7 to 5% (DBP) compared to using a single PRS. Adding additional 14 PRSs computed based on two independent GWASs further increased the genetic model PVE to 6.3% (SBP) and 5.7% (DBP). PVE differed across self-reported race/ethnicity groups, with primarily all non-White groups benefitting from the inclusion of additional PRSs. In summary, non-linear ML models improves BP prediction in models incorporating diverse populations.

Managing patients with kidney allografts largely depends on biopsy diagnosis which is based on semiquantitative assessments of rejection features and extent of acute and chronic changes within the renal parenchyma. Current methods lack reproducibility while digital image data-driven computational models enable comprehensive and quantitative assays. In this study we aimed to develop a computational method for automated assessment of histopathology transformations within the tubulointerstitial compartment of the renal cortex. Whole slide images of modified Picrosirius red-stained biopsy slides were used for the training (n = 852) and both internal (n = 172) and external (n = 94) tests datasets. The pipeline utilizes deep learning segmentations of renal tubules, interstitium, and peritubular capillaries from which morphometry features were extracted. Seven indicators were selected for exploring the intrinsic spatial interactions within the tubulointerstitial compartment. A principal component analysis revealed two independent factors which can be interpreted as representing chronic and acute tubulointerstitial injury. A K-means clustering classified biopsies according to potential phenotypes of combined acute and chronic transformations of various degrees. We conclude that multivariate analyses of tubulointerstitial morphometry transformations enable extraction of and quantification of acute and chronic components of injury. The method is developed for renal allograft biopsies; however, the principle can be applied more broadly for kidney pathology assessment.

Understanding shared pathways and mechanisms involved in the pathogenesis of diseases like oral squamous cell carcinoma (OSCC) and COVID-19 could lead to the development of novel therapeutic strategies and diagnostic biomarkers. This study aims to predict the interactome of OSCC and COVID-19 based on salivary inflammatory proteins. Datasets for OSCC and COVID-19 were obtained from https://www.salivaryproteome.org/differential-expression and selected for differential gene expression analysis. Differential gene expression analysis was performed using log transformation and a fold change of two. Hub proteins were identified using Cytoscape and Cytohubba, and machine learning algorithms including naïve Bayes, neural networks, gradient boosting, and random forest were used to predict hub genes. Top hub genes identified included ALDH1A1, MT-CO2, SERPINC1, FGB, and TF. The random forest model achieved the highest accuracy (93%) and class accuracy (84%). The naive Bayes model had lower accuracy (63%) and class accuracy (66%), while the neural network model showed 55% accuracy and class accuracy, possibly due to data pre-processing issues. The gradient boosting model outperformed all models with an accuracy of 95% and class accuracy of 95%. Salivary proteomic interactome analysis revealed novel hub proteins as potential common biomarkers. [ABSTRACT FROM AUTHOR]

Accurate prediction of cardiovascular disease (CVD) mortality is essential for effective treatment decisions and risk management. Current models often lack comprehensive integration of key biomarkers, limiting their predictive power. This study aims to develop a predictive model for CVD-related mortality using a machine learning-based feature selection algorithm and assess its performance compared to existing models. We analyzed data from a cohort of 4,882 adults recruited between 1999 and 2004, followed for up to 20 years. After applying the Boruta algorithm for feature selection, key biomarkers including NT-proBNP, cardiac troponins, and homocysteine were identified as significant predictors of CVD mortality. Predictive models were built using these biomarkers alongside demographic and clinical variables. Model performance was evaluated using the concordance index (C-index), sensitivity, specificity, and accuracy, with internal validation conducted through bootstrap sampling. Additionally, decision curve analysis (DCA) was performed to assess clinical benefit. The combined model, incorporating both biomarkers and demographic variables, demonstrated superior predictive performance with a C-index of 0.9205 (95% CI: 0.9129–0.9319), outperforming models with demographic variables alone (C-index: 0.9030 (95% CI: 0.8938–0.9147)) or biomarkers alone (C-index: 0.8659 (95% CI: 0.8519–0.8826)). Cox regression analysis further identified key predictors of CVD mortality, including elevated AST/ALT, TyG, BUN, and systolic blood pressure, with protective factors such as higher chloride and iron levels. Nomogram construction and DCA confirmed that the combined model provided substantial net benefit across various time points. The integration of cardiac biomarkers, lipid profiles, and inflammatory markers significantly improves the accuracy of predictive models for CVD-related mortality. This novel approach offers enhanced prognostication, with potential for further optimization through the inclusion of additional clinical and lifestyle data. [ABSTRACT FROM AUTHOR]

One of the most significant, intricate, and little-discussed aspects of pressurized irrigation is cost estimation. This study attempts to model the early-stage cost of the drip irrigation system using a database of 515 projects divided into four sections the cost of the pumping station and central control system (TCP), the cost of on-farm equipment (TCF), the cost of installation and operation on-farm and pumping station (TCI), and the total cost (TCT). First, 39 environmental and management features affecting the cost of the listed sectors were extracted for each of the 515 projects previously mentioned. A database (a matrix of 515 × 43) was created, and the costs of all projects were updated for the baseline year of 2022. Then, several feature selection algorithms, such as WCC, LCA, GA, PSO, ACO, ICA, LA, HTS, FOA, DSOS, and CUK, were employed to choose the most significant features that had the biggest influence on the system cost. The selection of features was carried out for all features (a total of 39 features) as well as for easily available features (those features that existed before the irrigation system's design phase, 18 features). Then, different machine learning models such as Multivariate Linear Regression, Support Vector Regression, Artificial Neural Networks, Gene Expression Programming, Genetic Algorithms, Deep Learning, and Decision Trees, were used to estimate the costs of each of the of the aforementioned sections. Support vector machine (SVM) and optimization algorithms (Wrapper) were found to be the best learner and feature selection techniques, respectively, out of all the available feature selection algorithms. The two LCA and FOA algorithms produced the best estimation, according to the evaluation criteria results. Their RMSE for all features was 0.0020 and 0.0018, respectively, and their R2 was 0.94 and 0.94. For readily available features, these criteria were 0.0006 and 0.95 for both algorithms. In the part of the overall feature, the early-stage cost modeling with selected features revealed that the SVM model (with RBF Kernel) is the best model among the four cost sections discussed. Its evaluation criteria in the training stage are R2 = 0.923, RMSE = 0.008, and VE = 0.082; in the testing stage, they are R2 = 0.893, RMSE = 0.009, and VE = 0.102. The ANN model (MLP) was found to be the best model for a subset of features in the easily available feature part, with R2 = 0.912, RMSE = 0.008, and VE = 0.083 in the training stage and R2 = 0.882, RMSE = 0.009, and VE = 0.103 in the testing stage. The findings of this study can be utilized to highly accurately estimate the cost of local irrigation systems based on the recognized environmental and management parameters and by employing particular models. [ABSTRACT FROM AUTHOR]

Studying the propagation prediction of low-frequency (LF) radio waves is very significant for supporting applications in fixed and mobile long-distance communication, remote navigation, and timing service. Therefore, to enhance the predicting accuracy of LF sky wave propagation, we proposed an improved method based on the machine learning method. Firstly, we employed a machine learning method to create a prediction model for the critical frequency of the low ionospheric E layer (foE), which significantly affects LF sky wave propagation. Secondly, we enhanced the method for predicting LF sky wave propagation based on the model of low ionospheric parameters. By comparing the measured data from East Asia and predicted data based on the wave-hop theory, the proposed method achieved a 6.16% improvement in LF sky wave field strength. [ABSTRACT FROM AUTHOR]

Cardiovascular disease (CVD) remains a leading cause of global morbidity and mortality. Timely diagnosis is important in reducing both short and long-term health complications. Saliva has emerged as a potential source for biomarker discovery, offering a non-invasive tool for early detection of individuals at elevated risk for CVD, yet large-scale extensive proteomic analysis using saliva for a comprehensive biomarker discovery remains limited. In an effort to develop a diagnostic tool using saliva samples, our study aims to assess the salivary and plasma proteomes in subjects with high risk of developing CVD using a large-scale proteomic approach. Leveraging on the SOMAscan platform, we analyzed 1,317 proteins in saliva and plasma collected from subjects at a high risk of CVD (HR-CVD) and compared the profiles to subjects with low risk of CVD (LR-CVD). Our analysis revealed significant differences in the plasma and salivary proteins between the two groups. Pathway enrichment analysis of the differentially detected proteins revealed that the immune system activation and extracellular matrix remodeling are the most enriched pathways in the CVD-HR group. Comparing proteomic signatures between plasma and saliva, we found approximately 42 and 17 differentially expressed proteins associated with CVD-HR uniquely expressed in plasma and saliva respectively. Additionally, we identified eight common CVD-risk biomarkers shared between both plasma and saliva, demonstrating promising diagnostic tools for identifying individuals at high risk of developing CVD. In conclusion, saliva proteomics holds a significant promise to identify subjects with a high risk to develop CVD. Further studies are needed to validate our findings. [ABSTRACT FROM AUTHOR]

This study proposes a novel approach for generating dual-targeted adversarial examples in Graph Neural Networks (GNNs), significantly advancing the field of graph-based adversarial attacks. Unlike traditional methods that focus on inducing specific misclassifications in a single model, our approach creates adversarial samples that can simultaneously target multiple models, each inducing distinct misclassifications. This innovation addresses a critical gap in existing techniques by enabling adversarial attacks that are capable of affecting various models with different objectives. We provide a detailed explanation of the method’s principles and structure, rigorously evaluate its effectiveness across several GNN models, and visualize the impact using datasets such as Reddit and OGBN-Products. Our contributions highlight the potential for dual-targeted attacks to disrupt GNN performance and emphasize the need for enhanced defensive strategies in graph-based learning systems. [ABSTRACT FROM AUTHOR]

This study investigates the experimental assessment and mathematical modeling of the water sorption isotherms in dried specialty coffee beans processed by wet and semidry postharvest methods. The wet and semidry sorption isotherms were experimentally obtained over a range of water activities between 0.1 and 0.85 at temperatures of 25, 35, and 45 °C using the dynamic dew point method (DDI). Mathematical modeling was conducted to describe the influence of water activity, temperature, and postharvest method on the equilibrium moisture content. Twelve conventional sorption equations and four machine learning techniques were employed for modeling, using 75% of the experimental data for training and 25% for validation. The selection of the best model was carried out via multifactor Analysis of Variance (ANOVA). Experimental results showed that wet and semidry coffee beans exhibited a type II S-shaped isotherm (Brunauer–Emmett–Teller classification) and a significant (p < 0.05) influence of temperature on sorption curves. Additionally, the mucilaginous coating found in semidry coffee beans provided a protective role against water sorption. The Support Vector Machine (SVM) model provided the best fit for describing the sorption isotherms (mean relative error, MRE < 1% and adjusted coefficient of determination, R2adj > 99%), demonstrating its robustness in predicting the equilibrium moisture content as a function of water activity, temperature, and postharvest processing method. This mathematical model could serve as a virtual representation of the storage process, facilitating real-time decision-making to enhance coffee quality management during storage. [ABSTRACT FROM AUTHOR]

Plants are essential at all stages of living things. Plant pests, diseases, and symptoms are most regularly visible in plant leaves and fruits and sometimes within the roots. Yet, their diagnosis by experts in the laboratory is expensive, tedious, and time-consuming if the samples involve laboratory analysis. Failure to detect early plant symptoms and diseases is the core biotic cause of increased plant stresses, structure, health, reduced subsistence farming, and threats to global food security. To mitigate these problems at a social, economic, and environmental level, inappropriate herbicide application reduction and early plant disease detection and classification (PDDC) are significant solutions in this case. Advancements in transfer learning techniques have resulted in effective results in smart farming and have become extensively used in disease identification and classification research studies. This study presents a novel hybrid inception-xception (IX) using a convolution neural network (CNN). The presented model combines inception and depth-separable convolution layers to capture multiple-scale features while reducing model complexity and overfitting. In contrast to ordinary CNN architectures, it extends the network for better feature extraction, improving PDDC performance that demands diverse feature competencies. It further presents a real-time artificial intelligence (AI) application available in MATLAB, Android, and Servlet to automatically identify and classify diseases based on the leaf environment using improved CNN, machine learning (ML), and computer vision techniques. To assess the presented IX-CNN model performance, different classifiers, namely, support vector machine (SVM), decision tree (DT) and random forest (RF), were used. The experiments used six datasets, including PlantVillage, Turkey Disease, Plant Doc, Rice Disease, RoCole, and NLB datasets. Plant Doc, PlantVillage, and Turkey Disease datasets demonstrated an accuracy of 100%. Rice Disease, RoCole, and NLB attained an accuracy of 99.79%, 99.95%, and 98.64%, respectively. [ABSTRACT FROM AUTHOR]

Transcarotid artery revascularization (TCAR) is a relatively new and technically challenging procedure that carries a non-negligible risk of complications. Risk prediction tools may help guide clinical decision-making but remain limited. We developed machine learning (ML) algorithms that predict 1-year outcomes following TCAR. The Vascular Quality Initiative (VQI) database was used to identify patients who underwent TCAR between 2016 and 2023. We identified 115 features from the index hospitalization (82 pre-operative [demographic/clinical], 14 intra-operative [procedural], and 19 post-operative [in-hospital course/complications]). The primary outcome was 1-year post-procedural stroke or death. The data was divided into training (70%) and test (30%) sets. Six ML models were trained using pre-operative features with tenfold cross-validation. Overall, 38,325 patients were included (mean age 73.1 [SD 9.0] years, 14,248 [37.2%] female) and 2,672 (7.0%) developed 1-year stroke or death. The best pre-operative prediction model was XGBoost, achieving an AUROC of 0.91 (95% CI 0.90–0.92). In comparison, logistic regression had an AUROC of 0.68 (95% CI 0.66–0.70). The XGBoost model maintained excellent performance at the intra- and post-operative stages, with AUROC’s (95% CI’s) of 0.92 (0.91–0.93) and 0.94 (0.93–0.95), respectively. Our ML algorithm has potential for important utility in guiding peri-operative risk-mitigation strategies to prevent adverse outcomes following TCAR. [ABSTRACT FROM AUTHOR]

Prediction of isocitrate dehydrogenase (IDH) mutation status and epilepsy occurrence are important to glioma patients. Although machine learning models have been constructed for both issues, the correlation between them has not been explored. Our study aimed to exploit this correlation to improve the performance of both of the IDH mutation status identification and epilepsy diagnosis models in patients with glioma II-IV. 399 patients were retrospectively enrolled and divided into a training (n = 279) and an independent test (n = 120) cohort. Multi-center dataset (n = 228) from The Cancer Imaging Archive (TCIA) was used for external test for identification of IDH mutation status. Region of interests comprising the entire tumor and peritumoral edema were automatically segmented using a pre-trained deep learning model. Radiomic features were extracted from T1-weighted, T2-weighted, post-Gadolinium T1 weighted, and T2 fluid-attenuated inversion recovery images. We proposed an iterative approach derived from LASSO to select features shared by two tasks and features specific to each task, before using them to construct the final models. Receiver operating characteristic (ROC) analysis was employed to evaluate the model. The IDH mutation identification model achieved area under the ROC curve (AUC) values of 0.948, 0.946 and 0.860 on the training, internal test, and external test cohorts, respectively. The epilepsy diagnosis model achieved AUCs of 0.924 and 0.880 on the training and internal test cohorts, respectively. The proposed models can identify IDH status and epilepsy with fewer features, thus having better interpretability and lower risk of overfitting. This not only improves its chance of application in clinical settings, but also provides a new scheme to study multiple correlated clinical tasks. [ABSTRACT FROM AUTHOR]

Breast cancer, with its high incidence and mortality globally, necessitates early prediction of local and distant recurrence to improve treatment outcomes. This study develops and validates predictive models for breast cancer recurrence and metastasis using Recurrence-Free Survival Analysis and machine learning techniques. We merged datasets from the Molecular Taxonomy of Breast Cancer International Consortium, Memorial Sloan Kettering Cancer Center, Duke University, and the SEER program, creating a comprehensive dataset of 272, 252 rows and 23 columns. Our methodology utilized three predictive strategies: assessing recurrence risk, differentiating local from distant recurrences, and identifying potential metastatic sites. Key prognostic factors were identified through survival analysis. LightGBM, XGBoost, and Random Forest models were employed and validated against data from the Baheya Foundation. The models demonstrated strong performance; the survival analysis achieved a C-index of 0.837. The LightGBM model reached an AUC of 92% in predicting recurrences, while XGBoost and Random Forest models distinguished recurrence types with up to 86% accuracy, and they effectively differentiated between bone metastasis and all other locations combined (brain, liver, and lungs). This study highlights the significant potential of machine learning in advancing breast cancer management and sets a new benchmark for predictive analytics. Future research will integrate genetic data to further enhance these models. [ABSTRACT FROM AUTHOR]

This study developed a predictive model using deep learning (DL) and natural language processing (NLP) to identify emergency cases in pediatric emergency departments. It analyzed 87,759 pediatric cases from a South Korean tertiary hospital (2012–2021) using electronic medical records. Various NLP models, including four machine learning (ML) models with Term Frequency-Inverse Document Frequency (TF-IDF) and two DL models based on the KM-BERT framework, were trained to differentiate emergency cases using clinician transcripts. Gradient Boosting, among the ML models, performed best with an AUROC of 0.715, AUPRC of 0.778, and F1-score of 0.677. DL models, especially the fine-tuned KM-BERT model, showed superior performance, achieving an AUROC of 0.839, AUPRC of 0.879, and F1-score of 0.773. Shapley-based explanations provided insights into model predictions, underlining the potential of these technologies in medical decision-making. This study demonstrates the potential of advanced DL techniques for NLP in emergency medical settings, offering a more precise and efficient approach to managing healthcare resources and improving patient outcomes. [ABSTRACT FROM AUTHOR]

Indian agriculture is vital sector in the country's economy, providing employment and sustenance to millions of farmers. However, Plant diseases are a serious risk to crop yields and farmers' livelihoods. Traditional plant disease diagnosis methods rely heavily on human expertise, which can lead to inaccuracies due to the invisible nature of early disease symptoms and the labor-intensive process, making them inefficient for large-scale agricultural management. To recover from this and, address these challenges, this study explores deep learning, specifically Convolutional Neural Networks (CNN), as a means to enhance the accuracy and efficiency of plant disease detection. Deep learning architectures, like convolutional neural network, can autonomously learn and extract complicated characteristics and patterns from huge datasets. Our research, conducted on mango and groundnut leaves collected during field visits in western Maharashtra and supplemented by online datasets, demonstrates a CNN model that achieves an impressive 96% accuracy as compared to machine learning techniques that follow tedious feature extraction. Furthermore, image processing contributes to enhancing the dataset through normalization, resizing, and augmentation for better classification results. Overall, CNN can continuously improve and adapt its performance through iterative training, resulting in higher accuracy rates and reduced false positives in contrast to conventional machine learning methods. [ABSTRACT FROM AUTHOR]

Busy walking paths, like in a park, city centre, or shopping mall, frequently necessitate collision avoidance behaviour. Lab-based research has shown how different situation- and person-specific factors, typically studied independently, affect avoidance behaviour. What happens in the real world is unclear. Thus, we filmed unscripted pedestrian walking behaviours on a busy urban path. We leveraged deep learning algorithms to identify and extract pedestrian walking trajectories and had unbiased raters characterize situations where two pedestrians approached each other from opposite ends. We found that smaller medial-lateral distance between approaching pedestrians and smaller crowd size predicted an increased likelihood of a subsequent path deviation. Furthermore, we found that whether a pedestrian looked distracted or held, pushed, or pulled an object predicted medial-lateral distance between pedestrians at time of crossing. Our results highlight both similarities and differences with lab-based behaviour and offer insights relevant to developing accurate computational models for realistic pedestrian movement. [ABSTRACT FROM AUTHOR]

As a multivariate time series, the prediction of curling trajectories is crucial for athletes to devise game strategies. However, the wide prediction range and complex data correlations present significant challenges to this task. This paper puts forward an innovative deep learning approach, CasLSTM, by introducing integrated inter-layer memory, and establishes an encoder-predictor curling trajectory forecasting model accordingly. Additionally, tailored training techniques involving non-teacher-forcing, ExMSE loss and incremental multi-trajectory learning are devised to enhance model performance. Notably, the model demonstrates astounding accuracy, achieving sub-1cm average errors over 30m trajectories, outperforming vanilla LSTM by 41.8%. It also showcases robustness across various curling settings, with strict validation metrics on a static test set further verifying precision. Field test results reveal promising predictive capabilities for real-world scenarios as well, exhibiting applicability. The proposed technique liberates data-driven curling stone trajectory prediction from sole reliance on analytical models and tackles key challenges of long sequence forecasting. The presented technologies and insights could also generalize to prediction tasks in other remote trajectories and multivariate time series domains. [ABSTRACT FROM AUTHOR]

In this paper, the usage of a predictive surrogate model for the estimate of flow variables in the transient phase of coolant injection from the nose cone by combining the Long Short-Term Memory (LSTM) and Proper Orthogonal Decomposition (POD) technique. The velocity, pressure, and mass fraction of the counterflow jet is evaluated via this hybrid technique and the source of discrepancy of a predictive surrogate model with Full order model is explained in this study. The POD modes for the efficient prediction of the different flow variables are defined. The performance of the POD + LSTM for different ranges of training and test is evaluated and it is found that the performance of this hybrid technique is acceptable when 80% of the available data is training test. The predictive errors of coolant mass fraction and axial velocity is higher due to the complexity of the vortex in the recirculation region. [ABSTRACT FROM AUTHOR]

Water conveyance channels in cold and arid regions pass through several saline-alkali soil areas. Canal water leakage exacerbates the salt expansion traits of such soil, damaging canal slope lining structures. To investigate the mechanical properties of saline clay, this study conducted indoor tests, including direct shear, compression, and permeation tests, and scanning electron microscopy (SEM) analysis of soil samples from typical sites. This study aims to elucidate the impact of various factors on the mechanical properties of saline clay from a macro–micro perspective and reveal its physical mechanisms. A prediction model is formulated and validated. The findings indicate the following: (1) Cohesion in direct shear tests has a linear negative correlation with water content and a positive correlation with dry density and initially decreases with increasing salt content until 2%, after which it increases. The internal friction angle initially increases and then decreases with increasing water content, reaching a peak at the optimal water content, and then gradually increases with dry density while initially decreasing, followed by an increase in salt content, stabilizing thereafter. Water content, dry density, or salt content chiefly affect cohesion by influencing electrostatic attraction, van der Waals forces, particle cementation, and valence bonds at particle contact points. (2) Compression tests reveal a linear positive correlation between the compression coefficient and water content, a negative correlation with dry density, and a stepwise linear correlation with salt content, peaking at 2%. The compression index decreases with increasing water content and dry density, following a trend similar to that of the compression coefficient with increasing salt content. The rebound index shows a linear negative correlation with water content and dry density, transitioning from a negative to a positive correlation at 2% salt content. Scanning electron microscopy analysis revealed particle flattening and increased aggregation with increasing consolidation pressure, reducing compressibility. Large pores and three-dimensional porosity have the greatest influence on soil compressibility. (3) Permeability tests reveal an exponential negative correlation between the permeability coefficient and dry density. As the dry density increases, the particle arrangement becomes denser, decreasing the pore quantity, with micropores disproportionately impacting the permeability coefficient. An increase in salinity initially increases the permeability coefficient before it decreases. The boundary point of the 2% salt content divides the effect of salt ions from promoting free water flow to blocking seepage channels, with the proportion of micropores being the primary influencing factor. (4) Employing statistical theory and machine learning algorithms, dry density, water content, and salinity are used to predict mechanical index values. The improved particle swarm optimization-support vector regression (PSO-SVR) model has high accuracy and general applicability. These findings offer insights for the construction and upkeep of open channel projects in arid regions. [ABSTRACT FROM AUTHOR]

Glaucoma poses a growing health challenge projected to escalate in the coming decades. However, current automated diagnostic approaches on Glaucoma diagnosis solely rely on black-box deep learning models, lacking explainability and trustworthiness. To address the issue, this study uses optical coherence tomography (OCT) images to develop an explainable artificial intelligence (XAI) tool for diagnosing and staging glaucoma, with a focus on its clinical applicability. A total of 334 normal and 268 glaucomatous eyes (86 early, 72 moderate, 110 advanced) were included, signal processing theory was employed, and model interpretability was rigorously evaluated. Leveraging SHapley Additive exPlanations (SHAP)-based global feature ranking and partial dependency analysis (PDA) estimated decision boundary cut-offs on machine learning (ML) models, a novel algorithm was developed to implement an XAI tool. Using the selected features, ML models produce an AUC of 0.96 (95% CI: 0.95–0.98), 0.98 (95% CI: 0.96–1.00) and 1.00 (95% CI: 1.00–1.00) respectively on differentiating early, moderate and advanced glaucoma patients. Overall, machine outperformed clinicians in the early stage and overall glaucoma diagnosis with 10.4 –11.2% higher accuracy. The developed user-friendly XAI software tool shows potential as a valuable tool for eye care practitioners, offering transparent and interpretable insights to improve decision-making. [ABSTRACT FROM AUTHOR]

Electric load forecasting is crucial in the planning and operating electric power companies. It has evolved from statistical methods to artificial intelligence-based techniques that use machine learning models. In this study, we investigate short-term load forecasting (STLF) for large-scale electricity usage datasets. We propose a new prediction model for STLF that combines data clustering and dimensionality reduction schemes to handle large-scale electricity usage data effectively. Here, we adapt k-means clustering for data clustering, kernel principal component analysis (kernel PCA), universal manifold approximation and projection (UMAP), and t-stochastic nearest neighbor (t-SNE) for dimensionality reduction. To verify the effectiveness of the proposed model, we extensively apply it to neural network-based models. We compare and analyze the performance of the proposed model with the comparisons using actual electricity usage data for 4710 households. Experimental results demonstrate that data clustering with dimensionality reduction can improve the performance of baseline models. As a result, the prediction accuracy of the proposed method outperforms those of the existing methods by 1.01–1.76 times for summer data and by 1.03–1.36 times for winter data in terms of mean absolute percentage error (MAPE). [ABSTRACT FROM AUTHOR]

This study presents a comprehensive workflow to detect low seismic amplitude gas fields in hydrocarbon exploration projects, focusing on the West Delta Deep Marine (WDDM) concession, offshore Egypt. The workflow integrates seismic spectral decomposition and machine learning algorithms to identify subtle anomalies, including low seismic amplitude gas sand and background amplitude water sand. Spectral decomposition helps delineate the fairway boundaries and structural features, while Amplitude Versus Offset (AVO) analysis is used to validate gas sand anomalies. The entire seismic volume is classified into facies domains using machine learning, which isolates target features from seismic background data. The study area, covering 1850 km2, includes major structures such as the Rosetta fault and Nile Delta offshore anticline, with reservoirs consisting of layered sandstones and mudstones. Over 90 wells, including exploration and development wells, have been drilled in the area. Seismic amplitude data, including full and partial offset stacked, were analyzed to classify gas, water, and shale zones. The workflow's performance is demonstrated through the successful identification of the low-amplitude Swan-E Messinian anomaly, characterized as a high-risk gas prospect. Machine learning techniques, specifically neural network models, were trained to differentiate seismic features such as low-amplitude gas sand from background-amplitude water sand and shale. By iterating over multiple attributes and validating the models on blind test sets and on a blind section, which excluded a known shallow gas field, the workflow significantly improved the ability to detect potential hydrocarbon reservoirs characterized by low seismic amplitude. The results show that this integrated approach reduces exploration risk, quantifies the chance of success, and enhances decision-making in well placement and hydrocarbon exploration. This method is particularly useful for identifying low seismic amplitude anomalies, which are often challenging to detect with conventional seismic analysis. (1) This study developed a workflow to detect low seismic amplitude gas fields in near-field exploration. (2) It uses a machine learning algorithm to classify and explore low-seismic-amplitude gas sand reservoirs. (3) This approach helps estimate the likelihood of success and reduces the risk associated with hydrocarbon exploration wells. [ABSTRACT FROM AUTHOR]

Recent advancements of large language models (LLMs) like generative pre-trained transformer 4 (GPT-4) have generated significant interest among the scientific community. Yet, the potential of these models to be utilized in clinical settings remains largely unexplored. In this study, we investigated the abilities of multiple LLMs and traditional machine learning models to analyze emergency department (ED) reports and determine if the corresponding visits were due to symptomatic kidney stones. Leveraging a dataset of manually annotated ED reports, we developed strategies to enhance LLMs including prompt optimization, zero- and few-shot prompting, fine-tuning, and prompt augmentation. Further, we implemented fairness assessment and bias mitigation methods to investigate the potential disparities by LLMs with respect to race and gender. A clinical expert manually assessed the explanations generated by GPT-4 for its predictions to determine if they were sound, factually correct, unrelated to the input prompt, or potentially harmful. The best results were achieved by GPT-4 (macro-F1 = 0.833, 95% confidence interval [CI] 0.826–0.841) and GPT-3.5 (macro-F1 = 0.796, 95% CI 0.796–0.796). Ablation studies revealed that the initial pre-trained GPT-3.5 model benefits from fine-tuning. Adding demographic information and prior disease history to the prompts allows LLMs to make better decisions. Bias assessment found that GPT-4 exhibited no racial or gender disparities, in contrast to GPT-3.5, which failed to effectively model racial diversity. [ABSTRACT FROM AUTHOR]

Training a machine learning system to evaluate any type of facial deformity is impeded by the scarcity of large datasets of high-quality, ethics board-approved patient images. We have built a deep learning-based cleft lip generator called CleftGAN designed to produce an almost unlimited number of high-fidelity facsimiles of cleft lip facial images with wide variation. A transfer learning protocol testing different versions of StyleGAN as the base model was undertaken. Data augmentation maneuvers permitted input of merely 514 frontal photographs of cleft-affected faces adapted to a base model of 70,000 normal faces. The Frechet Inception Distance was used to measure the similarity of the newly generated facial images to the cleft training dataset. Perceptual Path Length and the novel Divergence Index of Normality measures also assessed the performance of the novel image generator. CleftGAN generates vast numbers of unique faces depicting a wide range of cleft lip deformity with variation of ethnic background. Performance metrics demonstrated a high similarity of the generated images to our training dataset and a smooth, semantically valid interpolation of images through the transfer learning process. The distribution of normality for the training and generated images were highly comparable. CleftGAN is a novel instrument that generates an almost boundless number of realistic facial images depicting cleft lip. This tool promises to become a valuable resource for the development of machine learning models to objectively evaluate facial form and the outcomes of surgical reconstruction. [ABSTRACT FROM AUTHOR]

Membrane incompatibility poses significant health risks, including severe complications and potential fatality. Surface modification of membranes has emerged as a pivotal technology in the membrane industry, aiming to improve the hemocompatibility and performance of dialysis membranes by mitigating undesired membrane-protein interactions, which can lead to fouling and subsequent protein adsorption. Affinity energy, defined as the strength of interaction between membranes and human serum proteins, plays a crucial role in assessing membrane-protein interactions. These interactions may trigger adverse reactions, potentially harmful to patients. Researchers often rely on trial-and-error approaches to enhance membrane hemocompatibility by reducing these interactions. This study focuses on developing machine learning algorithms that accurately and rapidly predict affinity energy between novel chemical structures of membrane materials and human serum proteins, based on a molecular docking dataset. Various membrane materials with distinct characteristics, chemistry, and orientation are considered in conjunction with different proteins. A comparative analysis of linear regression, K-nearest neighbors regression, decision tree regression, random forest regression, XGBoost regression, lasso regression, and support vector regression is conducted to predict affinity energy. The dataset, comprising 916 records for both training and test segments, incorporates 12 parameters extracted from data points and involves six different proteins. Results indicate that random forest (R² = 0.8987, MSE = 0.36, MAE = 0.45) and XGBoost (R² = 0.83, MSE = 0.49, MAE = 0.49) exhibit comparable predictive performance on the training dataset. However, random forest outperforms XGBoost on the testing dataset. Seven machine learning algorithms for predicting affinity energy are analyzed and compared, with random forest demonstrating superior predictive accuracy. The application of machine learning in predicting affinity energy holds significant promise for researchers and professionals in hemodialysis. These models, by enabling early interventions in hemodialysis membranes, could enhance patient safety and optimize the care of hemodialysis patients. [ABSTRACT FROM AUTHOR]

Metabolic syndrome (Mets) in adolescents is a growing public health issue linked to obesity, hypertension, and insulin resistance, increasing risks of cardiovascular disease and mental health problems. Early detection and intervention are crucial but often hindered by complex diagnostic requirements. This study aims to develop a predictive model using NHANES data, excluding biochemical indicators, to provide a simple, cost-effective tool for large-scale, non-medical screening and early prevention of adolescent MetS. After excluding adolescents with missing diagnostic variables, the dataset included 2,459 adolescents via NHANES data from 2007–2016. We used LASSO regression and 20-fold cross-validation to screen for the variables with the greatest predictive value. The dataset was divided into training and validation sets in a 7:3 ratio, and SMOTE was used to expand the training set with a ratio of 1:1. Based on the training set, we built eight machine learning models and a multifactor logistic regression model, evaluating nine predictive models in total. After evaluating all models using the confusion matrix, calibration curves and decision curves, the LGB model had the best predictive performance, with an AUC of 0.969, a Youden index of 0.923, accuracy of 0.978, F1 score of 0.989, and Kappa value of 0.800. We further interpreted the LGB model using SHAP, the SHAP hive plot showed that the predictor variables were, in descending order of importance, BMI age sex-specific percentage, weight, upper arm circumference, thigh length, and race. Finally, we deployed it online for broader accessibility. The predictive models we developed and validated demonstrated high performance, making them suitable for large-scale, non-medical primary screening and early warning of adolescent Metabolic syndrome. The online deployment of the model allows for practical use in community and school settings, promoting early intervention and public health improvement. [ABSTRACT FROM AUTHOR]

Exploring the potential of advanced artificial intelligence technology in predicting microsatellite instability (MSI) and Ki-67 expression of endometrial cancer (EC) is highly significant. This study aimed to develop a novel hybrid radiomics approach integrating multiparametric magnetic resonance imaging (MRI), deep learning, and multichannel image analysis for predicting MSI and Ki-67 status. A retrospective study included 156 EC patients who were subsequently categorized into MSI and Ki-67 groups. The hybrid radiomics model (HMRadSum) was developed by extracting quantitative imaging features and deep learning features from multiparametric MRI using emerging attention mechanism. Tumor markers were subsequently predicted utilizing an XGBoost classifier. Model performance and interpretability were evaluated using standard classification metrics, Gradient-weighted Class Activation Mapping (Grad-CAM), and SHapley Additive exPlanations (SHAP) techniques. For the MSI prediction task, the HMRadSum model achieved area-under-curve (AUC) value of 0.945 (95% CI 0.862-1.000) and accuracy of 0.889. For the Ki-67 prediction task, the AUC and accuracy of HMRadSum model was 0.888 (95% CI 0.743-1.000) and 0.810. This hybrid radiomics model effectively extracted features associated with EC gene expression, providing potential clinical implications for personalized diagnosis, treatment, and treatment strategy optimization. [ABSTRACT FROM AUTHOR]

Athlete engagement is influenced by several factors, including cohesion, passion and mental toughness. Machine learning methods are frequently employed to construct predictive models as a result of their high efficiency. In order to comprehend the effects of cohesion, passion and mental toughness on athlete engagement, this study utilizes the relevant methods of machine learning to construct a prediction model, so as to find the intrinsic connection between them. The construction and comparison methods of predictive models by machine learning algorithms are investigated to evaluate the level of predictive models in order to determine the optimal predictive model. The results show that the PSO-SVR model performs best in predicting athlete engagement, with a prediction accuracy of 0.9262, along with low RMSE (0.1227), MSE (0.0146) and MAE (0.0656). The prediction accuracy of the PSO-SVR model exhibits an obvious advantage. This advantage is mainly attributed to its strong generalization ability, nonlinear processing ability, and the ability to optimize and adapt to the feature space. Particularly noteworthy is that the PSO-SVR model reduces the RMSE (7.54%), MSE (17.05%), and MAE (3.53%) significantly, while improves the R2 (1.69%), when compared to advanced algorithms such as SWO. These results indicate that the PSO-SVR model not only improves the accuracy of prediction, but also enhances the reliability of the model, making it a powerful tool for predicting athlete engagement. In summary, this study not only provides a new perspective for understanding athlete engagement, but also provides important practical guidance for improving athlete engagement and overall performance. By adopting the PSO-SVR model, we can more accurately identify and optimise the key factors affecting athlete engagement, thus bringing far-reaching implications for research and practice in sport science and related fields. [ABSTRACT FROM AUTHOR]

The rapid advancements in additive manufacturing (AM) across different scales and material classes have enabled the creation of architected materials with highly tailored properties. Beyond geometric flexibility, multi-material AM further expands design possibilities by combining materials with distinct characteristics. While machine learning has recently shown great potential for the fast inverse design of lattice structures, its application has largely been limited to single-material systems. In this work, we propose a novel approach that incorporates material properties as edge features within the graph representation of multi-material truss lattices, utilizing graph neural networks (GNNs) to develop a fast and efficient inverse design framework. We validate this framework by designing lattices with tunable thermal expansion and stiffness properties, showcasing its ability to explore a broad and flexible design space. We showcase the framework's inverse design capabilities for both single and multi-objective optimization tasks and assess its limitations. Additionally, we demonstrate the superior capacity of GNNs in capturing structure-property relationships for multi-material systems. We anticipate that the continued advancement of GNN-assisted inverse design will play a key role in unlocking the full potential of multi-material truss lattices. [ABSTRACT FROM AUTHOR]

Angio-based microvascular resistance (AMR) as a potential alternative to the index of microcirculatory resistance (IMR) and its relationship with microvascular obstruction (MVO) and other cardiac magnetic resonance (CMR) parameters still lacks comprehensive validation. This study aimed to validate the correlation between AMR and CMR-derived parameters and to construct an interpretable machine learning (ML) model, incorporating AMR and clinical data, to forecast MVO in ST-segment elevation myocardial infarction (STEMI) patients undergoing primary percutaneous coronary intervention (PPCI). We enrolled 452 STEMI patients from Nanjing Drum Tower Hospital between 2018 and 2022, who received both PPCI and CMR. After PPCI, AMR measurements and CMR-derived parameters were recorded, and clinical data were gathered. The ML workflow comprised feature selection using the Boruta algorithm, model construction with seven classifiers, hyperparameter optimization via ten-fold cross-validation, model comparison based on the area under the curve (AUC), and a Shapley additive explanations (SHAP) analysis to analyze the significance of different features. 32.29% of patients showed inconsistency between AMR and MVO, but we successfully constructed a predictive model for MVO. Among the classifiers, Extreme gradient boosting (XGBoost) post hyperparameter optimization displayed superior performance, achieving an AUC of 0.911 and 0.846 in the training and validation sets, respectively. SHAP analysis identified AMR as a pivotal predictor of MVO. Although we observed the inconsistency between AMR and MVO but the ML-based construction of MVO prediction model is feasible, which brings the possibility of timely prediction of patients with MVO and timely imposition of interventions during PPCI. [ABSTRACT FROM AUTHOR]

Refractory High-Entropy Alloys (RHEAs), such as NbMoTaW, MoNbTaVW, HfNbTaZr, Re0.1Hf0.25NbTaW0.4, Nb40Ti25Al15V10Ta5Hf3W2, TixNbMoTaW (x = 0, 0.25, 0.5, 0.75 and 1), and 3d transition metal HEAs such as Al10.3Co17Cr7.5Fe9Ni48.6Ti5.8Ta0.6Mo0.8W0.4 have demonstrated superior performance compared to traditional superalloys, particularly in high-temperature applications for engine components. However, the development of these alloys often depends on critical raw materials (CRMs) such as Ta, W, Nb, Hf, and others. The reliance on critical raw materials (CRMs) not only generates substantial emissions during recycling processes but also imposes considerable risks across global supply chains, hindering the pursuit of Net-zero ambitions. In this pioneering work, we unveil an inventive approach to inversely predict novel multicomponent alloy compositions, meticulously crafted to eliminate CRMs while achieving hardness levels comparable to those of CRM-containing multi-principal element alloys (MPEAs). A robust machine learning (ML) model was developed using a computational database of 3,608 entries, covering unary and binary materials from the Thermo-Calc 2024a software. Among various ML models, the Extra Trees Regressor (ETR) exhibited superior performance and was integrated with metaheuristic optimization techniques to identify novel MPEA compositions. The Cuckoo Search Optimization (CSO) method produced reduced-CRM MPEAs that closely matched Thermo-Calc predictions, with an error margin below ± 20%. To assess the efficacy of these reduced-CRM MPEAs, we compared the hardness of newly synthesized MPEA with CRM-containing counterparts reported in the literature, particularly those with high-risk critical raw materials like Niobium (Nb) and Tantalum (Ta). For example, the CoCrFeNb0.309Ni alloy, which includes CRMs Nb and Co exhibits a Vickers hardness of 480 HV. In contrast, our proposed composition, Ti0.01111NiFe0.4Cu0.4 achieves a comparable hardness of 488 HV without using a CRM. Our objective was not to develop high hardness alloy but to facilitate the development of reduced-CRM multi-principal element alloys (R-CRM-MPEAs). We validated our computational approach through the experimental synthesis of an FCC-phase alloy, Al6.25Cu18.75Fe25Co25Ni25. Thermo-Calc evaluation and ML model predictions of the Vickers hardness showed excellent agreement with the experimental hardness values, which lends credence to our approach. In conclusion, this study provides a robust framework for accelerating the discovery of novel R-CRM-MPEAs, effectively addressing challenges related to supply chain vulnerabilities, import dependence, and related environmental concerns. [ABSTRACT FROM AUTHOR]

A major histologic feature of cirrhosis is the loss of liver architecture with collapse of tissue and vascular changes per unit. We developed qVessel to quantify the arterial density (AD) in liver biopsies with chronic disease of varied etiology and stage. 46 needle liver biopsy samples with chronic hepatitis B (CHB), 48 with primary biliary cholangitis (PBC) and 43 with metabolic dysfunction-associated steatotic liver disease (MASLD) were collected at the Shuguang Hospital. The METAVIR system was used to assess stage. The second harmonic generation (SHG)/two-photon images were generated from unstained slides. Collagen proportionate area (CPA) using SHG. AD was counted using qVessel (previously trained on manually labeled vessels by stained slides (CD34/a-SMA/CK19) and developed by a decision tree algorithm). As liver fibrosis progressed from F1 to F4, we observed that both AD and CPA gradually increases among the three etiologies (P < 0.05). However, at each stage of liver fibrosis, there was no significant difference in AD or CPA between CHB and PBC compared to MASLD (P > 0.05). AD and CPA performed similar diagnostic efficacy in liver cirrhosis (P > 0.05). Using the qVessel algorithm, we discovered a significant correlation between AD, CPA and METAVIR stages in all three etiologies. This suggests that AD could underpin a novel staging system. [ABSTRACT FROM AUTHOR]

To integrate machine learning and multiomic data on lactylation-related genes (LRGs) for molecular typing and prognosis prediction in lung adenocarcinoma (LUAD). LRG mRNA and long non-coding RNA transcriptomes, epigenetic methylation data, and somatic mutation data from The Cancer Genome Atlas LUAD cohort were analyzed to identify lactylation cancer subtypes (CSs) using 10 multiomics ensemble clustering techniques. The findings were then validated using the GSE31210 and GSE13213 LUAD cohorts. A prognosis model for LUAD was developed using the identified hub LRGs to divide patients into high- and low-risk groups. The effectiveness of this model was validated. We identified two lactylation CSs, which were validated in the GSE31210 and GSE13213 LUAD cohorts. Nine hub LRGs, namely HNRNPC, PPIA, BZW1, GAPDH, H2AFZ, RAN, KIF2C, RACGAP1, and WBP11, were used to construct the prognosis model. In the subsequent prognosis validation, the high-risk group included more patients with stage T3 + 4, N1 + 2 + 3, M1, and III + IV cancer; higher recurrence/metastasis rates; and lower 1, 3, and 5 year overall survival rates. In the oncogenic pathway analysis, most of the oncogenic mutations were detected in the high-risk group. The tumor microenvironment analysis illustrated that immune activity was notably elevated in low-risk patients, indicating they might more strongly respond to immunotherapy than high-risk patients. Further, oncoPredict analysis revealed that low-risk patients have increased sensitivity to chemotherapeutics. Overall, we developed a model that combines multiomic analysis and machine learning for LUAD prognosis. Our findings represent a valuable reference for further understanding the important function of lactylation modification pathways in LUAD progression. [ABSTRACT FROM AUTHOR]

Foam ceramics are widely used in industrial applications due to their unique properties, including high porosity, lightweight, and high-temperature resistance. However, their complex microstructure presents significant challenges for image analysis. Traditional machine learning methods often fall short in capturing both global feature dependencies and detailed representations. To address this, a novel artificial intelligence recognition model, FD-Conv, is proposed, which combines the global information processing capabilities of Transformers with the local feature extraction strengths of convolutional neural networks. Additionally, a frequency domain block detail enhancement mechanism is introduced to improve recognition accuracy. Experimental results demonstrate that the FD-Conv model enhances recognition accuracy by at least 7.6% compared to state-of-the-art methods. Furthermore, the model effectively identifies foam ceramics with varying compositions and formulations and quantifies their microstructural phase characteristics. This research aims to advance the application of foam ceramic microstructure image analysis by improving recognition accuracy, particularly in multi-source microscopic image feature learning and pattern recognition. [ABSTRACT FROM AUTHOR]

The increasing trend in land surface temperature (LST) and the formation of urban heat islands (UHIs) has emerged as a persistent challenge for urban planners and decision-makers. The current research was carried out to study the land use and land cover (LULC) changes and associated LST patterns in the planned city (Kabul) and the unplanned city (Jalalabad), Afghanistan, using Support Vector Machine (SVM) and Landsat data from 1998 to 2018. Future changes in LULC and LST were predicted for 2028 and 2038 using Cellular Automata-Markov (CA-Markov) and Artificial Neural Network (ANN) models. The results clearly emphasize different LULC changes and LST patterns between Kabul and Jalalabad. Between 1998 and 2018, the built-up areas in Kabul and Jalalabad increased by 16% and 30%, respectively, while bare soil and vegetation decreased by 15% and 1% in Kabul and 4% and 30% in Jalalabad. The built-up areas showed the highest seasonal and annual LST, followed by bare soil and vegetation. The maximum seasonal LST occurred during the summer for both cities from 1998 to 2018. Future predictions showed that the built-up areas (48% and 55% in 2018) will increase to approximately 59% and 68% by 2028 and to 68% and 79% by 2038 in Kabul and Jalalabad, respectively. Similarly, LST simulations showed that the percentage of areas with higher LST (> 35°C) would increase from (0% and 5% in 2018) to 4% and 5% and 22% and 43% in Kabul and Jalalabad by 2028 and 2038, respectively. Kabul's planned city shows lower LST than Jalalabad's unplanned city, primarily due to planned urbanization and greater vegetation cover in the city center. Urban planners should limit unplanned development and increase vegetation in Jalalabad to reduce the potential impacts of high temperatures. [ABSTRACT FROM AUTHOR]

The experiments were conducted at different levels of infrared power, airflow, and temperature. The relationships between the input process factors and response factors' physicochemical properties of dried garlic were optimized by a self-organizing map (SOM), and the model was developed using machine learning. Artificial Neural Network (ANN) with 99% predicting accuracy and Self-Organizing Maps (SOM) with 97% clustering accuracy were used to determine the quality characteristics of garlic. Specifically, five key areas were identified, and valuable insights were offered for optimizing garlic production and improving its overall quality. The (aw) values for the sample ranged from 0.43 to 0.48. The maximum vitamin C content was 0.112 mg/g, followed by an air temperature of 40 °C and 0.7 m/s air velocity under 1500 W/m². The total color change values increased with IR and higher air temperature but declined with higher air velocity. Also, the garlic's flavor strength, allicin content, water activity, and vitamin C levels decreased as the IR and air temperature increased. The results demonstrated a significant impact of the independent parameters on the response parameters (P < 0.01). Interestingly, the machine learning predictions closely matched the test data sets, providing valuable insights for understanding and controlling the factors affecting garlic drying performances. [ABSTRACT FROM AUTHOR]

This study develops predictive models for Chinese female patients with VL utilizing machine learning techniques. The aim is to create an effective model that can assist in clinical diagnosis and treatment of vaginal relaxation, thereby enhancing women's pelvic floor health. In total, 1184 women with VL have been randomly selected and categorized into groups using the finger measurement method. Among them, there are 383 cases of mild VL, 405 cases of moderate VL, and 396 cases of severe VL. Concurrently, 396 healthy women without VL who underwent routine health examinations have been chosen at random and assigned to the non-VL group. Based on 1580 cases, we have established LightGBM, Random Forest, XGBoost, and AdaBoost models based on training dataset using 5-fold cross-validation and GridSearch, and analyzed the performance of the models on the hold-out test dataset. The confusion matrix, precision, recall, F1 score, overall accuracy, and ROC curve of the models on the hold-out test dataset are compared. The overall accuracy of LightGBM model, RF model, XGBoost model, and AdaBoost model are 0.8987, 0.8987, 0.8987, and 0.8457, respectively. The average AUC of LightGBM model is 0.976, the one of RF model is 0.9763, the one of XGBoost model is 0.9775, and the one of AdaBoost model is 0.928. The XGBoost model has the more comprehensive and reasonable performance among the four prediction models, which can accurately distinguish between healthy, mild VL, as well as moderate VL and severe VL, which can assist doctors in diagnosing persons' conditions more accurately, devising personalized treatment plans, avoiding unnecessary surgeries, reducing persons' psychological stress, improving patient compliance and treatment outcomes, thus enhancing overall treatment results. [ABSTRACT FROM AUTHOR]

This study employed some machine learning (ML) techniques with Python programming to forecast the adsorption capacity of MOF adsorbents for thiophenic compounds namely benzothiophene (BT), dibenzothiophene (DBT), and 4,6-dimethyl dibenzothiophene (4,6-DMDBT). Five ML models were developed with the help of a dataset containing 676 rows to correlate the adsorbent features, adsorption conditions, and adsorbate characteristics to the MOF sample's sulfur adsorption capability. Among the ML approaches, MLP model achieved the best performance with a low mean squared error (MSE) of 0.0032 on the test set and 0.0021 on the training set and mean relative error (MRE) of 15.26% on the test set. Also, Random Forest model yielded a higher test MSE of 0.0045 and MRE of 17.83%. Feature importance analysis was performed by utilizing MLP model and shapely additive plan (SHAP) method, and the findings revealed that "initial concentration of sulfur" (SHAP value 0.51) and "contact time" (SHAP value 0.37) were the crucial factors influenced desulfurization process efficiency. Additionally, a comparative analysis of the features utilizing the MLP network classified the factors into three primary categories: process conditions, adsorbent characteristics, and adsorbate characteristics. Consequently, the process condition was identified as the most significant group compared to others. Finally, the desulfurization process optimization indicated the maximum DBT adsorption of 161.6 mg/g for Zr-based MOF could be achieved when the features including BET, TPV, pore size, oil/adsorbent ration, and temperature were tuned around 756 m2/g, 0.955 cm3/g, 5.96 nm, 449.85 g/g, 20.1 °C, respectively. [ABSTRACT FROM AUTHOR]

Concrete-filled steel tube (CFST) columns are widely employed in high-rise buildings, long-span bridges, and seismic-resistant structures due to their superior load-bearing capacity, structural efficiency, and resilience under extreme loading conditions. This study uses symbolic regression with structural design code provisions to predict the eccentric strength of concrete filled-steel tubular columns with circular shape (CCFST) and rectangular shape (RCFST). Previous studies have used two distinct approaches for estimating eccentric strength: explainable models based on theoretical derivations and black-box models derived from machine learning (ML) methods. This study proposes a hybrid model derived from the design code standards, with performance enhanced by the symbolic regression technique. This model is based on a comprehensive experimental database of 464 tests for CCFST columns and 313 tests for RCFST columns under eccentric loading from various research papers. The developed code-based symbolic regression (C-SR) displays both robust and interpretable, demonstrating high prediction accuracy with mean values of the prediction-to-actual ratios of 1.006 and 0.997 and coefficient of variation (CoV) values of 0.117 and 0.098 for CCFSTs and RCFSTs, respectively, while providing explainable mathematical expressions that align with the mechanical principles of code provisions. The developed C-SR model is benchmarked against EC4 and AISC360 standards and evaluated against the various ML techniques, demonstrating acceptable performance. The results highlight the C-SR model's effectiveness in providing reliable predictions and valuable insights for practical engineering applications. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has burdened healthcare systems globally. To curb high hospital admission rates, only patients with genuine medical needs are admitted. However, machine learning (ML) models to predict COVID-19 hospitalization in Asian children are lacking. This study aimed to develop and validate ML models to predict pediatric COVID-19 hospitalization. We collected secondary data with 2200 patients and 65 variables from Malaysian aged 0 to 12 with COVID-19 between 1st February 2020 and 31st March 2022. The sample was partitioned into training, internal, and external validation groups. Recursive Feature Elimination (RFE) was employed for feature selection, and we trained seven supervised classifiers. Grid Search was used to optimize the hyperparameters of each algorithm. The study analyzed 1988 children and 30 study variables after data were processed. The RFE algorithm selected 12 highly predicted variables for COVID-19 hospitalization, including age, male sex, fever, cough, rhinorrhea, shortness of breath, vomiting, diarrhea, seizures, body temperature, chest indrawing, and abnormal breath sounds. With external validation, Adaptive Boosting was the highest-performing classifier (AUROC = 0.95) to predict COVID-19 hospital admission in children. We validated AdaBoost as the best to predict COVID-19 hospitalization among children. This model may assist front-line clinicians in making medical disposition decisions. [ABSTRACT FROM AUTHOR]

Physical Layer Security (PLS) in Cognitive Radio Networks (CRN) improves the confidentiality, availability, and integrity of the external communication between the devices/ users. The security models for sensing and beamforming reduce the impact of adversaries such as eavesdroppers in the signal processing layer. To such an extent, this article introduces a Secure Channel Estimation Model (SCEM) using Channel State Information (CSI) and Deep Learning (DL) to improve the PLS. In this proposed model, the CSI is exploited to evaluate the channel utilization and actual capacity availability throughout the allocation intervals. The change in channel capacity and utilization augments the need for security through 2-level key shared authentication. The deep learning algorithm verifies the authentication completeness for maximum channel capacity utilization irrespective of adversary interference. This verification follows mutual authentication between the primary and secondary users sharing the maximum capacity channel with high secrecy. The learning monitors the outage secrecy rates to verify failed allocations such that the replacement for allocation is pursued. Thus, the physical layer security between different user categories is administered through maximum CSI exploitation with high beamforming abilities. The proposed model leverages the secrecy rate by 10.77% and the probability of detection by 15.01% and reduces the interference rate by 11.07% for the varying transmit powers. [ABSTRACT FROM AUTHOR]

Enhancement of security, personalization, and safety in advanced transportation systems depends on driver identification. In this context, this work suggests a new method to find drivers by means of a Random Forest model optimized using the osprey optimization algorithm (OOA) for feature selection and the salp swarm optimization (SSO) for hyperparameter tuning based on driving behavior. The proposed model achieves an accuracy of 92%, a precision of 91%, a recall of 93%, and an F1-score of 92%, significantly outperforming traditional machine learning models such as XGBoost, CatBoost, and Support Vector Machines. These findings show how strong and successful our improved method is in precisely spotting drivers, thereby providing a useful instrument for safe and quick transportation systems. [ABSTRACT FROM AUTHOR]

Ureteropelvic junction obstruction (UPJO) is a common pediatric condition often treated with pyeloplasty. Despite the surgical intervention, postoperative urinary tract infections (UTIs) occur in over 30% of cases within six months, adversely affecting recovery and increasing both clinical and economic burdens. Current prediction methods for postoperative UTIs rely on empirical judgment and limited clinical parameters, underscoring the need for a robust, multifactorial predictive model. We retrospectively analyzed data from 764 pediatric patients who underwent unilateral pyeloplasty at the Children's Hospital affiliated with the Capital Institute of Pediatrics between January 2012 and January 2023. A total of 25 clinical features were extracted, including patient demographics, medical history, surgical details, and various postoperative indicators. Feature engineering was initially performed, followed by a comparative analysis of five machine learning algorithms (Logistic Regression, SVM, Random Forest, XGBoost, and LightGBM) and the deep learning TabNet model. This comparison highlighted the respective strengths and limitations of traditional machine learning versus deep learning approaches. Building on these findings, we developed an ensemble learning model, meta-learner, that effectively integrates both methodologies, and utilized SHAP(Shapley Additive Explanation, SHAP) to complete the visualization of the integrated black-box model. Among the 764 pediatric pyeloplasty cases analyzed, 265 (34.7%) developed postoperative UTIs, predominantly within the first three months. Early UTIs significantly increased the likelihood of re-obstruction (P < 0.01), underscoring the critical impact of infection on surgical outcomes. In evaluating the performance of six algorithms, TabNet outperformed traditional models, with the order from lowest to highest as follows: Logistic Regression, SVM, Random Forest, XGBoost, LightGBM, and TabNet. Feature engineering markedly improved the predictive accuracy of traditional models, as evidenced by the enhanced performance of LightGBM (Accuracy: 0.71, AUC: 0.78 post-engineering). The proposed ensemble approach, combining LightGBM and TabNet with a Logistic Regression meta-learner, achieved superior predictive accuracy (Accuracy: 0.80, AUC: 0.80) while reducing dependence on feature engineering. SHAP analysis further revealed eGFR and ALB as significant predictors of UTIs post-pyeloplasty, providing new clinical insights into risk factors. In summary, we have introduced the first ensemble prediction model, incorporating both machine learning and deep learning (meta-learner), to predict urinary tract infections following pediatric pyeloplasty. This ensemble approach mitigates the dependency of machine learning models on feature engineering while addressing the issue of overfitting in deep learning-based models like TabNet, particularly in the context of small medical datasets. By improving prediction accuracy, this model supports proactive interventions, reduces postoperative infections and re-obstruction rates, enhances pyeloplasty outcomes, and alleviates health and economic burdens. Level of evidence IV Case series with no comparison group. [ABSTRACT FROM AUTHOR]