To identify factors associated with need for pharmacotherapy among patients with glucose intolerance and one elevation on 3-hour oral glucose tolerance test (OGTT). Retrospective cohort study of singleton gestations 24 weeks or greater with at least one elevation on OGTT. The primary outcome was the need for pharmacotherapy for glycemic control. To evaluate timing of OGTT elevation with need for pharmacotherapy, a logistic regression model controlled for maternal race/ethnicity, body mass index (BMI), gestational age at time of OGTT, age, parity, history of gestational diabetes and history of large neonate. The need for pharmacotherapy was assessed based on number and extent of testing elevations. A predictive model based on linear discriminant analysis was developed. 480 patients had 1 OGTT elevation; of these, 19.2% required pharmacotherapy. Fasting elevations were most associated with development of medication-requiring diabetes. A predictive model for risk of pharmacotherapy in patients with an abnormal OGTT based on BMI and extent of elevation at each OGTT time point increased the identification of patients requiring pharmacotherapy by 15.4%. 376 patients had at least 2 elevations on OGTT; these patients were more likely to require pharmacotherapy for glycemic control compared to those with 1 elevation. Patients with increasing BMI values had increased need for pharmacotherapy regardless of the number of abnormal values. Among patients with one OGTT elevation, fasting elevation and BMI are predictive of need for anti-glycemic medications. Predictive models may be useful in assessing need for pharmacotherapy for patients with abnormal OGTT not otherwise meeting criteria for gestational diabetes.

Bacterial biofilms around the apex are crucial in disease progression and persistence of apical periodontitis. While intracanal biofilms initiate infection, extraradicular biofilms contribute to treatment resistance and persistence. Thus, a comprehensive understanding of these biofilms may help elucidate mechanisms underlying persistent apical periodontitis. Therefore, in this study, we aimed to compare the microbiome and predicted functional profiles in matched apical root canals with those of extraradicular biofilms associated with persistent apical periodontitis. Seventeen root apices from patients with persistent apical periodontitis were collected via surgery. After extraradicular biofilm was collected, intracanal biofilm was obtained by cryopulverisation. Bacterial communities were detected by amplicon sequencing of the V1-V2 region of the 16S rRNA gene. Diversity, microbial composition and predicted bacterial functions were compared between matched intracanal and extraradicular biofilms. Alpha diversity analysis of the microbiome revealed no significant differences between the two sampling sites. In contrast, the beta diversity of the microbiota of the same root (matched samples) was significantly lower than that of the microbiota of unpaired samples. There were no statistically significant differences in permutational multivariate analysis of variance for the microbiome between paired extraradicular and intracanal biofilms, regardless of the presence of the sinus tract. The abundances of the predominant genera, namely Fusobacterium, Treponema, Prevotella, Porphyromonas and Bacteroides as well as gram-positive bacteria, including Actinomyces, were similar between extraradicular and intraradicular biofilms. Linear discriminant analysis effect size analysis identified bacterial taxa significantly enriched in extraradicular biofilms, whereas no taxa were significantly enriched in intraradicular biofilms. Phylogenetic Investigation of Communities by Reconstruction of Unobserved States analysis revealed several differences in Kyoto Encyclopaedia of Genes and Genomes pathways between these biofilms. While comparison of the microbiome between extraradicular and intracanal biofilms of the same root apices revealed differences in bacterial composition, certain similarities were noted, particularly in dominant bacterial species abundance, indicating a close microbial relationship between intracanal and extraradicular biofilms, with some exceptions. Additionally, some differences in predicted functional profiles were observed between the two biofilm types. Thus, the characterisation of bacterial communities around the apical foramen may guide the development of appropriate antimicrobial strategies.

Nail diseases, including such common conditions as fungus, and more serious issues like melanoma, may be important clues to the overall health and require a clear diagnosis to be treated. The purpose of the paper is to create a nail disease detection system based on the advanced machine learning methods, including transfer learning and federated learning. The research seeks to show how machine learning and federated learning can be combined to detect nail disease performance with high accuracy without having to share data. The data include pictures of diverse nail conditions including Acral Lentiginous Melanoma, Onychogryphosis, and Pitting among others that are checked to maintain the quality of data in a uniform manner to facilitate the effective training of the models. The most common feature extraction models are ResNet152V2, DenseNet201, MobileNetV2, and InceptionResNetV2 that produce between 1,280 and 2,048 features per image. These characteristics are then pooled to create a unified feature space of 6,784 dimensions which is further narrowed to five major characteristics with Linear Discriminant Analysis (LDA) to create an efficient form of classification. A range of classification models, including Deep Neural Networks (DNN) and Long Short-Term Memory (LSTM) and Bidirectional LSTM (Bi-LSTM) are compared, with the last one reaching the highest classification accuracy of 91.8%. The federated learning strategy enables the joint training of DL models by different clients to ensure data-privacy and has validation-accuracy rates exceeding 99-percent in both uniformly random and structured data distributions. The proposed federated learning-based models resulted high in both uniformly random and structured data distributions.

The gut microbiome (GM) is increasingly recognized for its role in atherosclerosis development. However, its potential as a biomarker for risk-stratification in patients with atherosclerotic cardiovascular (CV) comorbidities remains under-explored. This study aimed to identify distinct GM clusters associated with elevated CV risk. In this prospective observational cohort, patients with coronary artery disease, hypertension, hyperlipidemia, or diabetes mellitus referring to Mayo Clinic from 2013 to 2018 were enrolled. Bacterial DNA was analyzed in the V3-V5 region of 16S rDNA. Beta-diversity was plotted using Principal Coordinates Analysis. Unsupervised hierarchical clustering of the GM classified participants into two clusters. Cox regression evaluated the association between clusters and Major Adverse Cardiac Events (MACE), defined as a composite of cardiac events, heart failure, and all-cause mortality. Permutational Multivariate Analysis of Variance identified clinical factors contributing to cluster assignment. Linear Discriminant analysis identified GM taxa with differential abundance among clusters and their effect sizes. Among 211 participants (median age 60 [IQR: 50-70] years; 57.3% male), two distinct GM profiles emerged (Cluster H: N = 104; Cluster L: N = 107, P < 0.001). Cluster L participants were younger (P < 0.001), more likely female (P = 0.009), and had healthier CV profiles, including lower BMI (P = 0.007), hypertension (P = 0.010), hyperlipidemia (P = 0.005), and lower coronary artery disease prevalence (P = 0.003). Over a median follow-up of 7.4 years, Cluster L had a significantly lower incidence of MACE compared to Cluster H (HR = 0.48, 95% CI: 0.26-0.91, P = 0.024). Cluster L had higher operational taxonomic units (P < 0.001) and lower Bacillota-to-Bacteroidetes ratio (P < 0.001) compared to Cluster H. The predominant taxa in Cluster L included Bacteroides, Alistipes, and Parabacteroides, whereas Blautia, Agathobacter, and Clostridium sensu stricto-1 were more abundant in Cluster H. Distinct GM profiles are associated with varying CV risk, highlighting the potential of unsupervised GM profiling as a novel tool for risk stratification and individualized therapy.

Background/Objectives: Fetal health is essential in prenatal care, influencing both maternal and fetal outcomes. Cardiotocography (CTG) monitors uterine contractions and fetal heart rate, yet manual interpretation exhibits significant inter-examiner variability. Machine learning offers automated alternatives; however, class imbalance in CTG datasets where pathological cases constitute less than 10% leads to poor detection of minority classes. This study aims to provide the first systematic benchmark comparing five resampling strategies across seven classifier families for multi-class CTG classification, evaluated using imbalance-aware metrics rather than overall accuracy alone. Methods: Seven machine learning models were employed: Naïve Bayes (NB), Random Forest (RF), Linear Discriminant Analysis (LDA), k-Nearest Neighbors (KNN), Linear Support Vector Machine (SVM), Multinomial Logistic Regression (MLR), and Multi-Layer Perceptron (MLP). To address class imbalance, we evaluated the original unbalanced dataset (base) and five resampling methods: SMOTE, BSMOTE, ADASYN, NearMiss, and SCUT. Performance was evaluated on a held-out test set using Balanced Accuracy (BACC), Macro-F1, the Macro-Matthews Correlation Coefficient (Macro-MCC), and Macro-Averaged ROC-AUC. We also report per-class ROC curves. Results: Among all models, RF proved most reliable. Training on the original distribution (base) yielded the highest BACC (0.9118), whereas RF combined with BSMOTE provided the strongest class-balanced performance (Macro-MCC = 0.8533, Macro-F1 = 0.9073) with a near-perfect ROC-AUC (approximately 0.986–0.989). Overall, resampling effects proved model dependent. While some classifiers achieved optimal performance on the natural class distribution, oversampling techniques, particularly SMOTE and BSMOTE, demonstrated significant improvements in minority class discrimination and class-balanced metrics across multiple model families. Notably, certain models benefited substantially from resampling, exhibiting enhanced Macro-F1, BACC, and minority class recall without sacrificing overall accuracy. Conclusions: These findings establish robust, model-agnostic baselines for CTG-based fetal health screening. They highlight that strategic oversampling can translate improved minority class discrimination into clinically meaningful performance gains, supporting deployment in cost-sensitive and threshold-aware clinical settings.

Storage-induced spectral changes of human platelet-rich plasma revealed by SERS.

Fecal microbiota transplantation (FMT) is increasingly recognized as an alternative to antibiotics for treating recurrent Clostridioides difficile infection. The success of FMT heavily depends on the appropriate selection of donors, encompassing factors such as diet patterns, lifestyle, environmental exposures, and intestinal microbiota diversity. A potential super donor was identified from 5 healthy adults and provided stool samples periodically over 2 years (2021-2022). The samples underwent 16S rRNA sequencing via the Illumina MiSeq platform, and microbial diversity was analyzed using QIIME 2 in comparison with 152 healthy individuals. The stool microbiome composition of the potential super donor remained stable without significant changes over a 2-year period. Both alpha and beta diversity analyses revealed significant differences between the super donor and the 152 healthy individuals. The super donor exhibited significantly higher microbial diversity based on alpha diversity metrics (P< 0.0001) and distinct compositional profiles as shown by beta diversity. Linear discriminant analysis effect size (LEfSe) analysis indicated that Faecalibacterium and Prevotella strains comprised a significant proportion, with notable differences in relative abundance patterns (P< 0.05). Furthermore, 7 bacterial species were isolated from the super donor, all of which demonstrated inhibitory effects on the growth of C. difficile in vitro. These findings suggest that selecting donors with specific microbiota profiles, particularly those exhibiting higher microbial diversity, may potentially contribute to the inhibition of C. difficile, and further clinical studies are warranted to validate these findings.

The analysis of multivariate data is a central issue in biomedical research, where the accurate classification of patients and the extraction of reliable conclusions are of critical importance. Linear Discriminant Analysis (LDA) remains one of the most established methods for both dimensionality reduction and classification of data. In this paper, we examine in detail the theoretical foundations, assumptions, and statistical properties of LDA, and apply the method step by step to real data from the Breast Cancer Wisconsin (Diagnostic) database, which includes cellular features from breast biopsy samples with the aim of distinguishing benign from malignant tumors. Emphasis is placed on the importance of the method’s assumptions, such as multivariate normality, equality of covariance matrices, and absence of multicollinearity, demonstrating that their fulfillment leads to significant improvements in model performance. Specifically, careful preprocessing and strict adherence to these assumptions increase classification accuracy from 95.6% (94.7% cross-validated) to 97.8% (97.4% cross-validated). To our knowledge, this study is the first to demonstrate the dual use of LDA as both a dimensionality-reduction tool and a predictive classification model for this medical database within the same biomedical analysis framework. Moreover, we provide, for the first time, a systematic comparison between our assumption-aware LDA model and related studies employing the most accurate machine-learning classifiers reported in the literature for this dataset, showing that classical LDA achieves accuracy comparable to these more complex methods. The resulting discriminant model, which uses 13 variables out of the original 30, can be applied easily by clinical researchers to classify new cases as benign or malignant, while simultaneously providing interpretable coefficients that reveal the underlying relationships among variables. The implementation is carried out in the SPSS environment, following the theoretical steps described in the paper, thus offering a user-friendly and reproducible framework for reliable application. In addition, the study establishes a structured and transparent workflow for the proper application of LDA in biomedical research by explicitly linking assumption verification, preprocessing, dimensionality reduction, and classification.

To describe the oral microbiota in bearded dragons (Pogona vitticeps) with or without dental disease and evaluate the impact of fruit consumption. 42 total client-owned bearded dragons were categorized into groups: healthy (absent or mild dental disease [n = 21]) and diseased (moderate to severe dental disease [21]). An additional analysis compared fruit-eating (n = 17) and non-fruit-eating bearded dragons (25). Following dentition assessment, all oral quadrants were sampled at 1 time point for DNA extraction via next-generation sequencing targeting bacterial 16S rRNA and fungal internal transcribed spacer 2 regions. The α- and β-diversity, taxonomic abundance, core microbiota analysis, and linear discriminant effect size analyses were compared between groups. The oral microbiota comprised 1,317 and 163 fungal species. Although there were no significant differences in bacterial or fungal α-diversity between healthy and diseased groups, bacterial β-diversity differed significantly. Certain taxa were more abundant in the dental disease group, including Pseudomonas aeruginosa, Devriesea agamarum, Serratia marcescens, and the Aspergillus genus. Additionally, the microbiota of bearded dragons that consumed fruit was significantly altered. There were distinct organisms in the oral microbiota attributed to dental disease, with specific organisms more abundant in diseased individuals, suggesting an association with disease. Bearded dragons fed fruit had more abundant microbial species, indicating fruit consumption may promote oral microbial overgrowth. Both differences in the oral microbiota and increased prevalence of specific species associated with dental health status and diet should be considered when making husbandry and therapeutic decisions for bearded dragons.

Several members of the Rutaceae (citrus) family are widely cultivated and processed. Tocopherol (T) synthesis and composition are well-documented, while tocotrienols (T3) in most plant families remain underreported. To amend this, mass screening of Rutaceae species’ seed tocochromanols were analysed. Of the 53 analysed species, seed tocochromanols were tocotrienol-dominated in 22 species, including a majority of species Zanthoxyloideae (Choisya, Dictamnus, Melicope, Ptelea, Skimmia, Tetradium, Zanthoxylum) and the Cneoroideae (Cneorum) subfamily. Total tocochromanol content ranged from 0.20–25.98 mg 100 g−1 dry weight (dw) seeds. The highest tocochromanol content was observed in Murraya paniculata, Ruta graveolens seeds, the highest tocotrienol (T3) content was observed in Skimmia anquetilia and Dictamnus albus—19.80 and 19.70 mg 100 g−1 dw, respectively. The major tocochromanols in the seeds were γ-T and γ-T3, while others were present in low concentration or absent. Linear discriminant analysis (LDA), principal component analysis (PCA) and non-hierarchal cluster analysis (N-HCA) identified similar tocochromanol content trends in the Rutoideae subfamily species and the Bergera and Murraya genus, while the Zanthoxyloideae subfamily species’ seed tocochromanol composition was highly variable. The efficient extractability of tocochromanols using sustainable solvent–ethanol is demonstrating suitability of this approach for daily samples screening and bioactive extraction.

Dual-channel surface-enhanced Raman spectroscopy integrated with machine learning for accurate classification of mixed dyes.

Pattern recognition-based dual-channel colorimetric platform for on-site and accurate identification of Chinese wolfberry origin.

Differentiation of ex-vivo intracranial tumors based on frequency band segmentation integrated in linear discriminant analysis (FBS-LDA) of impedance-derived dielectric spectra

A fluorescence sensor array based on triple-color emission S,N-co-doped graphene quantum dots for metal ions discrimination.

Using nonlinear dynamic analysis to differentiate fall status in older women.

Employing Fourier-transform infrared spectroscopy as dereplication strategy in foodborne outbreak investigation of cereulide-producing Bacillus cereus.

Integrative morphological and genomic analyses reveal diversity, reticulate evolution, and adaptation in diploid and tetraploid Rosa species from Xinjiang.

Tea in color: Single-probe nanozyme array for multichannel visual fingerprinting and portable quality evaluation.

Word classification across speech modes from low-density electrocorticography signals

Non-destructive identification of forensically relevant body fluid stains using a portable electronic nose: A pilot study.