ABSTRACT This study presents a state-of-the-art ensemble machine learning (ML) model-based comprehensive evaluation of chloride ion penetration resistance of Self-Compacting Concrete (SCC) durability, simulated on an extensive dataset of 396 Rapid Chloride Penetration Test (RCPT), incorporating variations in cement, fly ash (FA), silica fume (SF), aggregate content, and temperature. Descriptive statistical analysis and visualizations revealed that moderate cement content, higher SF levels, and optimized FA replacement significantly enhance durability by reducing RCPT values. Elevated temperatures lead to increased permeability, although their impact can be mitigated in the mixes with higher SCM content. Four ensemble-based ML algorithms, Random Forest (RF), Gradient Boosting Machine (GBM), XGBoost, and CATBoost, were developed and compared for predictive modelling of RCPT values. The comparative analysis reveals XGBoost as the top performer with an R 2 of 0.9989, followed by RF (R 2 = 0.987), with CATBoost (R 2 = 0.963) and GBM (R 2 = 0.956) also performing satisfactorily. XGBoost excels in the validation with an R 2 of 0.9799. The study also presents a GUI framework for user-friendly, rapid, and cost-effective preliminary assessments. The study highlights the combined role of SCMs and ML-driven predictive modelling in developing durable, sustainable, and cost-effective SCCs for modern infrastructure.

ABSTRACT With the rapid growth of artificial intelligence, machine learning has emerged as a useful tool in the construction industry for optimizing materials and predicting concrete properties. This study explores the use of steel fiber and recycled demolition waste (DW) to improve concrete’s mechanical properties. The main objective is to predict the compressive strength (CS) of DW-modified fiber-reinforced concrete (FRC) by assessing the effects of cement, RHA, fiber content, natural and demolished waste aggregates, water, and superplasticizer dosages. Seven mix designs with varying fiber and DW levels were experimentally tested. Machine learning methods, including adaptive boosting (ADB), extreme gradient boosting (XGB), random forest (RF), and stacking models (XGB-ADB, XGB-RF), were applied to analyze these variables’ impact on CS. A dataset of 405 points was compiled from literature via a systematic review. The hybrid XGB-RF and XGB models showed the best performance with R2 values of 0.849 and 0.845, respectively. SHAP analysis identified cement, water, and superplasticizer as key factors affecting CS. Experimental validation supported the modeling results and the development of a graphical user interface. The novelty lies in integrating hybrid ML, explainable analysis, and experimental validation to predict the CS of the modified concrete and support mix design through a user-friendly GUI.

CT-based radiomics for predicting PD-L1 expression status in non-small cell lung cancer using a hybrid machine learning model.

Tunable triazole-based cholera toxin inhibitors: A QSAR-guided design and evaluation approach.

Interpretable machine learning-based prediction of liver metastasis risk in elderly patients with small cell lung Cancer: A study based on the SEER database and external validation in a Chinese cohort.

Periodontitis (PD) and oral squamous cell carcinoma (OSCC) frequently co‑occur in clinical populations. However, shared molecular determinants that could support risk assessment across diseases remain insufficiently defined. This study aimed to identify shared diagnostic biomarkers and potential mechanisms linking PD and OSCC using integrated bioinformatics and machine learning. PD and OSCC transcriptomic datasets were analysed to identify shared programmed cell death-related biomarkers using differentially expressed genes (DEGs), WGCNA and multiple machine learning algorithms (LASSO, Random Forest, GLM). Diagnostic performance was validated in external cohorts via ROC analysis, while immune landscapes and cellular interactions were characterised using CIBERSORT and single-cell analysis. At the cellular level, we evaluated SERPINA1 expression in an LPS-induced periodontal inflammation model and conducted siRNA-mediated knockdown in CAL27 oral cancer cells to examine its effects on proliferation, migration and invasion. Transcriptomic analyses of PD and OSCC revealed a shared immune-inflammatory signature with enrichment of programmed cell death pathways. SERPINA1 was consistently prioritised by LASSO, random forest and GLM and showed strong diagnostic performance across training and external validation cohorts (AUC > 0.8). CIBERSORT indicated remodelling of the immune microenvironment associated with SERPINA1 expression, with positive correlations to macrophage and neutrophil abundance. Single-cell analyses localised SERPINA1 to myeloid populations and suggested putative crosstalk with the pro-inflammatory mediator interleukin-1β (IL-1β). In the periodontal inflammation model, SERPINA1 expression was significantly upregulated upon LPS stimulation alongside increased inflammatory mediators such as IL-8 and IL-1β. In contrast, SERPINA1 knockdown in oral cancer cells led to reduced proliferative, migratory and invasive capacities. SERPINA1 emerges as a cross‑disease candidate biomarker linking chronic periodontal inflammation and oral cancer biology. The integrative framework provides a transparent roadmap for identifying clinically meaningful, reproducible biomarkers across oral diseases, going forwards.

Machine learning classification of PAH exposure in the Antarctic limpet Nacella concinna.

Non-sampling estimation of waste composition based on incineration flue gas pollutant fingerprints and machine learning approach.

A diagnosis tool for early detection and classification of heart disease in individuals using transformer mechanisms.

Response strategies of bacterial and micro-eukaryotic communities to environmental changes: Evidence from alpine lakes sedimentary DNA in Southwest China.

Development and validation of machine learning models to predict PTSD at multiple time points in hospitalized trauma patients.

High-selectivity phenol detection in cumene process wastewater via bromination and dynamic optical path.

Revolutionizing hepatic fibrosis staging: A machine learning approach combining clinical, biochemical, and microbiome insights.

Integrative bioinformatics identifies NSCLC biomarkers associated with LPS metabolism and circadian disruption.

As a crucial component of wind power generation systems, wind turbines must operate safely to prevent sudden failures. This requires effective identification of abnormal operational states. In this study, we propose a novel approach for categorizing abnormal states based on operational data from wind turbines, leveraging an improved density-based spatial clustering of applications with noise (DBSCAN) algorithm in conjunction with random forests. We establish a wind turbine performance model as a benchmark, employing a DBSCAN–bidirectional long short-term memory network (DBSCAN-BiLSTM) framework for anomaly detection. The random forest algorithm is then applied to accurately identify abnormal data points. Furthermore, we implement real-time adjustments to operational state thresholds based on identified anomalies, facilitating precise classification of abnormal operational data. A case study is presented, detailing steps including abnormal data cleaning, performance model construction, and abnormal state classification to validate our approach. Our results demonstrate that the DBSCAN-BiLSTM method significantly reduces the root mean square error (RMSE) by 41.5% and 49.7% compared to traditional LSTM and convolutional neural network (CNN) algorithms, respectively. This research provides an effective solution for identifying abnormal states in supervisory control and data acquisition (SCADA) data of wind turbines, which is vital for fault detection and maintenance in the wind power industry.

Machine learning-based prediction of three-year mortality in elderly inpatients with coronary artery disease combined with heart failure.

The complexity of many AI models hinders their clinical adoption because the clinicians using them do not regard them as transparent. This study addresses the lack of clinician-centered explainable AI (XAI) interfaces by designing and evaluating intuitive visual explanations for intubation prediction, testing the hypothesis that workflow-compatible designs enhance acceptance. This study compares three, time-aware, visual explanations for XAI-based intubation prediction and evaluate their acceptance, comprehension, and perceived utility among clinicians. We developed machine learning models to estimate the near-term risk of deterioration in the patient's condition which may lead to mechanical intubation using ICU time-series data. We generated global and local explanations using SHAP and designed three customized visual formats-a temporal force plot, a temporal bar chart, and a dual-encoded SHAP heatmap. Clinicians (n=206) evaluated comprehension and usability using objective questions and a Likert-based survey. Based on 4608 critically ill patients with 10 medical variables over 7hours of data for each patient, the Random Forest (RF) model achieved the highest area under the curve (AUC): 0.94. Furthermore, the local explanations were customized and evaluated by 206 clinicians through a survey conducted on the Prolific platform. A customized heatmap representation was selected as the visualization with the highest perceived clinical utility and alignment with clinical workflows. The reported findings support the need for explanation formats to be tailored to clinical reasoning and task context, supporting the concept of cognitive fit. The heatmap's close alignment with clinicians' mental models and its graphical integrity enhances interpretability and trust. This study demonstrates that explanation effectiveness depends on contextual relevance, rather than a universal standard, and that the presentation format itself significantly shapes clinicians' trust in XAI systems. This study advances clinical XAI by introducing a time-aware explanation framework for ICU intubation decisions. By integrating temporal trends with model reasoning, our visualizations closely align with clinicians' cognitive workflows. Rigorous clinician-centered evaluation identified the dual-encoded SHAP heatmap as the most useful and workflow-compatible visualization, highlighting the importance of explanation design alongside predictive accuracy for clinical adoption.

Who engages? Machine learning insights into digital mindfulness-based intervention for generalized anxiety disorder.

Ensuring reliable feature importance in food chemistry AI.

Long-term observations uncover sustained carbon dioxide emissions from lakes following aquaculture retreat.