Browse Articles
Discover research articles across all indexed journals
A robust machine learning approach to predicting remission and stratifying risk in rheumatoid arthritis patients treated with bDMARDs
Abstract Rheumatoid arthritis (RA) is a chronic autoimmune disease affecting millions worldwide, leading to inflammation, joint damage, and reduced quality of life. Although biological disease-modifying antirheumatic drugs (bDMARDs) are effective, they are costly, and up to 40% of patients do not achieve remission within six months. Accurate prediction of treatment response is crucial for optimizing care, minimizing side effects, and enhancing cost efficiency. This study proposes a robust machine learning framework for predicting six-month remission in RA patients using baseline routine clinical data. The framework also integrates risk stratification and explainability to enhance its clinical applicability. We evaluated multiple machine learning models, AdaBoost, Random Forest, XGBoost, and Support Vector Machines, using data from Austrian RA patients. We externally validated the results on an independent dataset from the Erlangen Hospital. To improve the reliability of probability estimates for actionable risk stratification, we employed calibration techniques, including Platt scaling, Isotonic regression, Beta calibration, and Spline calibration. We generated calibration curves to assess and visualize the alignment between predicted probabilities and observed outcomes. In addition, we used SHapley Additive exPlanations (SHAP) to analyze the contributions of different patient characteristics to the prediction of RA remission. AdaBoost demonstrated stronger performance than the other models, achieving an accuracy of 85.71% and a Brier score of 0.13 with isotonic regression calibration. SHAP identified DAS28, visual analog scales (VAS), age, and swollen joint count (SJC) as important characteristics for the prediction of RA remission. We also stratified patients into low-, medium-, and high-risk categories based on model predictions to support follow-up scheduling and treatment prioritization. Our framework predicts RA remission before the initiation of bDMARD therapy. It enables personalized care, actionable risk stratification, and optimized resource allocation. Its robustness was validated on two different individual cohort datasets, which highlights its potential for integration into routine clinical workflows.
Expression and diagnostic value evaluation of urinary exosomal miR-142-3p in diabetic nephropathy
Modeling Pseudomonas aeruginosa-Staphylococcus aureus interactions in zebrafish to assess the host inflammatory response upon co-infection
Thread design optimization of a dental implant using explicit dynamics finite element analysis
Fine-tuning of language models for automated structuring of medical exam reports to improve patient screening and analysis
Abstract The analysis of medical imaging reports is labour-intensive but crucial for accurate diagnosis and effective patient screening. Often presented as unstructured text, these reports require systematic organisation for efficient interpretation. This study applies Natural Language Processing (NLP) techniques tailored for European Portuguese to automate the analysis of cardiology reports, streamlining patient screening. Using a methodology involving tokenization, part-of-speech tagging and manual annotation, the MediAlbertina PT-PT language model was fine-tuned, achieving 96.13% accuracy in entity recognition. The system enables rapid identification of conditions such as aortic stenosis through an interactive interface, substantially reducing the time and effort required for manual review. It also facilitates patient monitoring and disease quantification, optimising healthcare resource allocation. This research highlights the potential of NLP tools in Portuguese healthcare contexts, demonstrating their applicability to medical report analysis and their broader relevance in improving efficiency and decision-making in diverse clinical environments.
Anatomical reduction of marginal impaction in posterior wall acetabular fractures ensures mid-term satisfactory outcomes
Reconfigurable model clusters for scalable modelling of feed drive dynamics
Identification and single-cell analysis of prognostic genes related to mitochondrial and neutrophil extracellular traps in bladder cancer
Multi-omics analysis identifies SNP-associated immune-related signatures by integrating Mendelian randomization and machine learning in hepatocellular carcinoma
Bioinsecticidal activity of aspergillus-derived endophytes from Olea europaea against Culex pipiens: toxicity, histology, and GC-MS profiling
Abstract Endophytic fungi offer a sustainable and eco-friendly alternative to synthetic insecticides, providing targeted pest control with reduced environmental toxicity and lower risk of resistance development. In this study, four endophytic fungi were isolated from Olea europaea and their ethyl acetate extracts’ larvicidal efficacy of Culex pipiens. Boxplot distributions revealed a positive correlation between metabolite concentration and larval mortality, with higher doses inducing greater lethality. Extracts 1 and 2 exhibited the highest larvicidal activity, with median mortality rates exceeding 60%, while Extracts 3 and 4 demonstrated significantly weaker effects. Dose-response curves further confirmed the potency of Extracts 1 and 2, with lower LC₅₀ values (148.36 and 153.36 µg/mL) compared to Extracts 3 and 4. The histological analysis revealed a dose-dependent impact, with minimal tissue alterations at 25–50 µg/mL, while 150–200 µg/mL caused severe disorganization, apoptosis, and inflammatory responses. The most potent isolates, Aspergillus niger(Extract 1) and Aspergillus flavus(Extract 2), were identified morphologically and molecularly, with NCBI accession numbers PQ269689 and PQ269690, respectively. GC-MS analysis identified key bioactive compounds, including oleic acid and hexadecanoic acid, supporting their insecticidal potential. These findings highlight A. niger and A. flavus ethyl acetate extracts as promising sources of bioinsecticides for pest control.
An international multi-cohort investigation of self-reported sleep and future depressive symptoms in older adults
A lightweight and secure authentication and privacy protection scheme for internet of medical things
Analysis of conservation practices for black soil based on organic matter and nitrogen contents in the black soil region of Northeast China
Analyzing the effects of pickling sludge and fly ash valorized cement sand bricks
Abstract The disposal of Stainless-Steel Pickling Sludge (SSPS) in landfills remains an important issue. Utilizing SSPS as construction material mitigates the negative environmental effects associated with its disposal, providing a sustainable solution. This study investigates co-utilization of SSPS and fly ash as partial substitution of river sand on cement sand bricks properties. Nine cement sand bricks compositions, including control mix, were prepared with varying composition of SSPS, fly ash and river sand. Four compositions were developed with SSPS varied from 2.5 to 10% with fixed fly ash content of 50%. Four additional compositions with varying fly ash content from 40 to 47.5% and varying SSPS 2.5–10% content as partial substitution of river sand were prepared. The developed bricks demonstrated that gradual increment of SSPS (2.5–10%) and reduction of fly ash (47.5–40%) proved incremental to the compressive strength up to 28 MPa. In addition, the morphological analysis using Scanning Electron Microscopy (SEM), X-ray Diffraction (XRD) and X-ray Fluorescence (XRF) were conducted for the compositions. The microstructure analysis showed that with inclusion of fly ash, Mix 2 (M2) compositions revealed a dense microstructure validating the sorptivity results as compared to Mix 1 (M1) compositions. Finally, the cost estimation of the waste valorized bricks as compared to the control bricks was observed to be significantly low. The experiment outcomes concluded adoption of SSPS-fly ash waste valorized bricks as a greener alternative to disposal.
Genetic analysis of silique and seed traits in Brassica juncea (L.) Czern. under differential doses of nitrogen application
Abstract Mustard, a major source of edible and industrial grade oils in the Indian subcontinent and various regions of Australia, Eastern Europe, and Canada, is also a protein resource for the animal feed industry. Silique and seed size are key traits for mustard improvement, but their inheritance mechanisms are not fully understood. We evaluated 92 inbred lines for silique length, seeds per silique, seed size, and rupture energy required to shatter a pod at three levels of nitrogen application in two crop seasons. Genotypes showed large phenotypic variations and a continuous distribution for all silique-related traits, suggesting a quantitative inheritance. Genotype × nitrogen interactions were significant for silique length, seeds per silique, and rupture energy. Association analysis identified 59 significant SNPs, whose annotations facilitated the prediction of 16 important genes underlying observed phenotypic variations. These genes are associated with silique formation (SHP2), grain formation (BG1, BG4), cell elongation (BRI1), grain filling (AT1G12500, AT1G77610, SWEET1, and AT3G14410), and silique shattering (UBP15, CO, INDEHISCENT, AGAMOUS1, FRUITFULL, and SHP2). RNA-seq data from 18 mustard genotypes revealed expression-level variations for identified candidate genes. Upregulation of SHP2 for SPS and resistance to silique shattering was observed, while downregulation of BG4 was observed in two genotypes with smallest seeds. This study provides valuable insight that may facilitate the marker-assisted selection (MAS) breeding for silique traits.
Physical activity and internalization problems in middle school students: the chain mediating role of rumination thinking and peer acceptance
Abstract This study explores methods to improve internalizing problems in middle school students through physical exercise, aiming to construct a mediational model that includes physical exercise, Rumination, peer acceptance, and internalizing problems. The model examines the interactions and influence pathways among these variables to provide a more comprehensive understanding of the mechanism through which physical exercise impacts internalizing problems in middle school students. 671 junior high school students were surveyed by physical exercise scale, Rumination scale, peer acceptance scale and internalization question scale. SPSS26.0 was used to analyze the data, followed by descriptive analysis, reliability analysis, Pearson correlation analysis, Harman single factor test and Bootstrap analysis. (1) Physical exercise was negatively correlated with internalizing problems (r = − 0.286), but the direct path between physical exercise and internalizing problems was not statistically significant (β = − 0.060, p > 0.05, CI [− 0.137, 0.018]); (2) Physical exercise was significantly negatively correlated with Rumination (β = − 0.174, p < 0.001, CI [− 0.255, − 0.094]) and significantly positively correlated with peer acceptance (β = 0.357, p < 0.001, CI [0.286, 0.429]). Rumination was significantly negatively correlated with peer acceptance (β = − 0.128, p < 0.001, CI [− 0.199, − 0.056]) and significantly positively correlated with internalizing problems (β = 0.289, p < 0.001, CI [0.222, 0.356]); peer acceptance was significantly negatively correlated with internalizing problems (β = − 0.372, p < 0.001, CI [− 0.448, − 0.296]). (3) The total effect of the chain mediation model was − 0.251, and the direct effect of physical exercise and internalized problems was − 0.059,The total effect accounted for 23.51%, Rumination and Peer acceptance played a mediating role between physical exercise and internalized behavior problems, and the total indirect effect size was − 0.192, The total effect accounted for 76.49%. The mediating effect values of Rumination and Peer acceptance between physical exercise and internalized behavior problems were − 0.050 and − 0.133, accounting for 19.92% and 52.99%, respectively. The chain-mediated effect between Rumination and Peer acceptance between physical exercise and internalized behavior problems was − 0.008, and the chain-mediated effect accounted for 3.19%. Physical exercise is negatively correlated with internalization, Rumination is negatively correlated with peer acceptance; Rumination and peer acceptance, as mediating variables, not only play an independent mediating role between physical exercise and junior high school students’ internalization problems, but also further strengthen their role in junior high school students’ internalization problems through chain mediation. Although the effect value of chain mediation is small, it is worthy of attention, suggesting that there is a synergistic effect among the variables. The results of this study provide an important reference for the design of school physical exercise curriculum. Through reasonable design of physical education curriculum, the interest and attraction of physical exercise can be increased, and the participation of students can be improved indirectly, thus promoting the improvement of internalization problems. At the same time, it can be considered to combine physical exercise with mental health education, and cultivate students’ methods and abilities to cope with pressure and negative emotions through physical exercise courses, cultivate positive and optimistic attitude, and improve junior middle school students’ mental toughness and social adaptability.