Principal Component Analysis Research Articles

Biomolecular Adsorption on Nanomaterials: Combining Molecular Simulations with Machine Learning.

Adsorption free energies of 32 small biomolecules (amino acids side chains, fragments of lipids, and sugar molecules) on 33 different nanomaterials, computed by the molecular dynamics - metadynamics methodology, have been analyzed using statistical machine learning approaches. Multiple unsupervised learning algorithms (principal component analysis, agglomerative clustering, and K-means) as well as supervised linear and nonlinear regression algorithms (linear regression, AdaBoost ensemble learning, artificial neural network) have been applied. As a result, a small set of biomolecules has been identified, knowledge of adsorption free energies of which to a specific nanomaterial can be used to predict, within the developed machine learning model, adsorption free energies of other biomolecules. Furthermore, the methodology of grouping of nanomaterials according to their interactions with biomolecules has been presented.

RNA M6A modification shaping cutaneous melanoma tumor microenvironment and predicting immunotherapy response.

Recent years have seen rising mortality rates linked to cutaneous melanoma (SKCM), despite advances in immunotherapy. Understanding RNA N6-methyladenosine (M6A) significance in SKCM is crucial for prognosis, tumor microenvironment (TME), immune cell presence, and immunotherapy efficacy. We analyzed 23 M6A regulators using SKCM samples from TCGA and GEO databases, identifying three M6A modification patterns linked to TME cell infiltration. Principal component analysis (PCA) yielded an M6A score for individual tumors, utilizing patient gene expression profiles and CNV data from TCGA. M6A modification patterns play a crucial role in SKCM development and progression, influencing tumor attributes such as inflammatory stage, subtype, TME interstitial activity, and genetic mutations. The M6A score independently predicts patient outcomes and correlates with improved response to immunotherapy, validated across anti-PD-1 and anti-PD-L1 therapy cohorts. M6A modifications significantly impact the TME landscape, with the M6A score serving as a predictive marker for immunotherapy response. Integrating M6A-related information into clinical practice could revolutionize SKCM management and treatment strategies.

Archetype analysis and the PHATE algorithm as methods to describe and visualize pregnant women's levels of physical activity knowledge.

The knowledge of physical activity (PA) recommended for pregnant women and practical application of it has positive impact on the outcome. Nevertheless, it is estimated that in high-income countries over 40% of pregnant women are insufficiently physically active. One of the reasons is insufficient knowledge pregnant women have about allowed effort during pregnancy and both recommended and not recommended physical activities. Description of knowledge about physical activity the women have and distinguishing patterns of their knowledge is becoming an increasingly important issue. A common approach to handle survey data that reflect knowledge involves clustering methods or Principal Component Analysis (PCA). Nevertheless, new procedures of data analysis are still being sought. Using survey data collected by the Institute of Mother and Child Archetypal analysis has been applied to detect levels of knowledge reflected by answers given in a questionnaire and to derive patterns of knowledge contained in the data. Next, PHATE (Potential of Heat-diffusion for Affinity-based Trajectory Embedding) algorithm has been used to visualize the results and to get a deeper insight into the data structure. The results were compared with picture derived from PCA. Three archetypes representing three patterns of knowledge have been distinguished and described. The presentation of complex data in a low dimension was obtained with help of PHATE. The formations revealed by PHATE have been successfully described in terms of knowledge levels reflected by the survey. Finally, comparison of PHATE with PCA has been shown. Archetype analysis combined with PHATE provides novel opportunities in examining nonlinear structure of survey data and allows for visualization that captures complex relations in the data. PHATE has made it possible to distinguish sets of objects that have common features but were captured neither by Archetypal analysis nor PCA. Moreover, for our data, PHATE provides an image of data structure which is more detailed than interpretation of PCA.

A Dynamics-Weighted Principal Components Analysis of Dominant Atmospheric Drivers of Ocean Variability with an Application to the North Atlantic Subpolar Gyre

Abstract This paper describes a framework for identifying dominant atmospheric drivers of ocean variability. The method combines statistics of atmosphere–ocean fluxes with physics from an ocean general circulation model to derive atmospheric patterns optimized to excite variability in a specified ocean quantity of interest. We first derive the method as a weighted principal components analysis and illustrate its capabilities in a toy problem. Next, we apply our analysis to the problem of interannual upper ocean heat content (HC) variability in the North Atlantic Subpolar Gyre (SPG) using the adjoint of the MITgcm and atmosphere–ocean fluxes from the ECCOv4-r4 state estimate. An unweighted principal components analysis reveals that North Atlantic heat and momentum fluxes in ECCOv4-r4 have a range of spatiotemporal patterns. By contrast, dynamics-weighted principal components analysis collapses the space of these patterns onto a small subset—principally associated with the North Atlantic Oscillation—that dominates interannual SPG HC variance. By perturbing the ECCOv4-r4 state estimate, we illustrate the pathways along which variability propagates from the atmosphere to the ocean in a nonlinear ocean model. This technique is applicable across a range of problems across Earth system components, including in the absence of a model adjoint. Significance Statement While the oceans have absorbed 90% of the excess heat associated with human-forced climate change, the change in the ocean’s heat content is not steady, with peaks and troughs superimposed upon a general increase. These fluctuations come from chaotic changes in the atmosphere and ocean and can be hard to disentangle. We use this case of ocean heat content variability to introduce a new method for determining the patterns of weather and climate in the atmosphere that are most effective at generating fluctuations in the ocean. To do this, we combine the statistics of recent atmospheric activity with output from a state-of-the-art numerical ocean model that reveals physical processes driving changes in ocean quantities including ocean heat content. This approach suggests that the atmospheric patterns that stimulate the most energetic changes in ocean heat content in the northern North Atlantic are not necessarily the most energetic patterns present in the atmosphere. We test our findings by preventing these patterns from affecting the ocean in our numerical model and measure a strong reduction in ocean heat content fluctuations.

Impacts of Feature Selection on Predicting Machine Failures by Machine Learning Algorithms

In the context of Industry 4.0, managing large amounts of data is essential to ensure informed decision-making in intelligent production environments. It enables, for example, predictive maintenance, which is essential for anticipating and identifying causes of failures in machines and equipment, optimizing processes, and promoting proactive management of human, financial, and material resources. However, generating accurate information for decision-making requires adopting suitable data preprocessing and analysis techniques. This study explores the identification of machine failures based on synthetic industrial data. Initially, we applied the feature selection techniques Principal Component Analysis (PCA), Minimum Redundancy Maximum Relevance (mRMR), Neighborhood Component Analysis (NCA), and Denoising Autoencoder (DAE) to the collected data and compared their results. In the sequence, a comparison among three widely known machine learning classifiers, namely Random Forest (RF), Support Vector Machine (SVM), and Multilayer Perceptron neural network (MLP), was conducted, with and without considering feature selection. The results showed that PCA and RF were superior to the other techniques, allowing the classification of failures with rates of 0.98, 0.97, and 0.98 for the accuracy, precision, and recall metrics, respectively. Thus, this work contributes by solving an industrial problem and detailing techniques to identify the most relevant variables and machine learning algorithms for predicting machine failures that negatively impact production planning. The findings provided by this study can assist industries in giving preference to employing sensors and collecting data that can contribute more effectively to machine failure predictions.

Development of a Measurement System Using Infrared Spectroscopy-Attenuated Total Reflectance, Principal Component Analysis and Artificial Intelligence for the Safe Quantification of the Nucleating Agent Sorbitol in Food Packaging

Sorbitol derivatives and other additives are commonly used in various products, such as packaging or food packaging, to improve their mechanical, physical, and optical properties. To accurately and precisely evaluate the efficacy of adding sorbitol-type nucleating agents to these articles, their quantitative determination is essential. This study systematically investigated the quantification of sorbitol-type nucleating agents in food packaging made from impact copolymers of polypropylene (PP) and polyethylene (PE) using attenuated total reflectance infrared spectroscopy (ATR-FTIR) together with analysis of principal components (PCA) and machine learning algorithms. The absorption spectra revealed characteristic bands corresponding to the C–O–C bond and hydroxyl groups attached to the cyclohexane ring of the molecular structure of sorbitol, providing crucial information for identifying and quantifying sorbitol derivatives. PCA analysis showed that with the selected FTIR spectrum range and only the first two components, 99.5% of the variance could be explained. The resulting score plot showed a clear pattern distinguishing different concentrations of the nucleating agent, affirming the predictability of concentrations based on an impact copolymer. The study then employed machine learning algorithms (NN, SVR) to establish prediction models, evaluating their quality using metrics such as RMSE, R2, and RMSECV. Hyperparameter optimization was performed, and SVR showed superior performance, achieving near-perfect predictions (R2 = 0.9999) with an RMSE of 0.100 for both calibration and prediction. The chosen SVR model features two hidden layers with 15 neurons each and uses the Adam algorithm, balanced precision, and computational efficiency. The innovative ATR-FTIR coupled SVR model presented a novel and rapid approach to accurately quantify sorbitol-type nucleating agents in polymer production processes for polymer research and in the analysis of nucleating agent derivatives. The analytical performance of this method surpassed traditional methods (PCR, NN).

Large-scale genome-wide SNP analysis reveals the rugged (and ragged) landscape of global ancestry, phylogeny, and demographic history in chicken breeds.

The worldwide chicken gene pool encompasses a remarkable, but shrinking, number of divergently selected breeds of diverse origin. This study was a large-scale genome-wide analysis of the landscape of the complex molecular architecture, genetic variability, and detailed structure among 49 populations. These populations represent a significant sample of the world's chicken breeds from Europe (Russia, Czech Republic, France, Spain, UK, etc.), Asia (China), North America (USA), and Oceania (Australia). Based on the results of breed genotyping using the Illumina 60K single nucleotide polymorphism (SNP) chip, a bioinformatic analysis was carried out. This included the calculation of heterozygosity/homozygosity statistics, inbreeding coefficients, and effective population size. It also included assessment of linkage disequilibrium and construction of phylogenetic trees. Using multidimensional scaling, principal component analysis, and ADMIXTURE-assisted global ancestry analysis, we explored the genetic structure of populations and subpopulations in each breed. An overall 49-population phylogeny analysis was also performed, and a refined evolutionary model of chicken breed formation was proposed, which included egg, meat, dual-purpose types, and ambiguous breeds. Such a large-scale survey of genetic resources in poultry farming using modern genomic methods is of great interest both from the viewpoint of a general understanding of the genetics of the domestic chicken and for the further development of genomic technologies and approaches in poultry breeding. In general, whole genome SNP genotyping of promising chicken breeds from the worldwide gene pool will promote the further development of modern genomic science as applied to poultry.

Genetic structure analyses reveal multiple origins of Elymus sensu stricto (Poaceae)

AbstractElymus sensu stricto includes tetraploid species in the tribe Triticeae with a StH genome combination derived from Hordeum (H) and Pseudoroegneria (St). The group is related to many important cereals and forage grasses and is considered an important wild genetic resource for crop breeding. The relationships within Elymus s.s. have been difficult to resolve due to the large number of widely distributed species, a high degree of morphological diversity, introgression between species, and potential multiple origins. Using genome-wide DArTseqLD data from 57 taxa, including the StH group and associated species, it is possible to gain representative information about the genetic structure of the genus. SplitsTree, STRUCTURE and principal component analysis (PCoA) show that Elymus s.s. is divided into two major clades highly correlating with geographical origin separating American species from Eurasians. These findings, together with the contribution of different Pseudoroegneria and Hordeum species to the two clades, support the conclusion that Elymus s.s. has at least two independent origins. The Eurasian and American clades are further divided into three and two subclades, respectively. The results show that Elymus species have migrated multiple times from North to South America and between America and Eurasia. The new findings could be used to develop a new classification of Elymus s.s. with sections based on phylogenetic data.

Classification of sugarcane areas in Landsat images using machine learning algorithms

Monitoring sugarcane areas through remote sensing is essential for planning and management of the national sugarcane industry. The use of machine learning algorithms has provided many benefits to remote sensing. This article aims to compare the prediction quality of three important machine learning methods in identifying sugarcane areas using Landsat images: Logistic Regression (LR), Decision Tree (DT) and RandomForest (RF). LR was applied in three versions: LR without penalization, LR with Ridge penalization (LR-R) and LR with Lasso penalization (LR-L). Data obtained in this study refer to a region of approximately 306,000 ha located in the state of São Paulo, Brazil, which was segmented into approximately 46,000 polygons (observations). Six spectral bands and vegetation indices observed along 17 months resulted in 102 covariates, which were reduced via Principal Component Analysis (PCA). In total, 19 Principal Components were chosen to account for 94.61% of the cumulative explained variance ratio and were used in machine learning methods to classify each polygon as sugarcane or other land covers. The method with the highest accuracy considering testing sample of 20% of data was RF (78.51%), followed by DT (72.30%), LR-L (69.64%), LR-R (69.64%), and LR (69.52%).

Exploiting a variational auto-encoder to represent the evolution of sudden stratospheric warmings

Sudden stratospheric warmings (SSWs) are the most dramatic events in the wintertime stratosphere. Such extreme events are characterized by substantial disruption to the stratospheric polar vortex, which can be categorized into displacement and splitting types depending on the morphology of the disrupted vortex. Moreover, SSWs are usually followed by anomalous tropospheric circulation regimes that are important for subseasonal-to-seasonal prediction. Thus, monitoring the genesis and evolution of SSWs is crucial and deserves further advancement. Despite several analysis methods that have been used to study the evolution of SSWs, the ability of deep learning methods has not yet been explored, mainly due to the relative scarcity of observed events. To overcome the limited observational sample size, we use data from historical simulations of the Whole Atmosphere Community Climate Model version 6 to identify thousands of simulated SSWs, and use their spatial patterns to train the deep learning model. We utilize a convolutional neural network combined with a variational auto-encoder (VAE)—a generative deep learning model—to construct a phase diagram that characterizes the SSW evolution. This approach not only allows us to create a latent space that encapsulates the essential features of the vortex structure during SSWs, but also offers new insights into its spatiotemporal evolution mapping onto the phase diagram. The constructed phase diagram depicts a continuous transition of the vortex pattern during SSWs. Notably, it provides a new perspective for discussing the evolutionary paths of SSWs: the VAE gives a better-reconstructed vortex morphology and more clearly organized vortex regimes for both displacement-type and split-type events than those obtained from principal component analysis. Our results provide an innovative phase diagram to portray the evolution of SSWs, in which particularly the splitting SSWs are better characterized. Our findings support the future use of deep learning techniques to study the underlying dynamics of extreme stratospheric vortex phenomena, and to establish a benchmark to evaluate model performance in simulating SSWs.

Convective drying of golden delicious apple enhancement: drying characteristics, artificial neural network modeling, chemical and ATR-FTIR analysis of quality parameters

AbstractIn recent years, many innovative methods have been investigated to provide alternative approaches to the food drying industry, but currently the most widely used method is convective drying. There are difficulties in integrating innovative methods into the food industry due to cost, inapplicability to every food material, or product quality, etc. In addition, it is possible to improve the convective drying method by pre-treating of foods. Thanks to the convective drying method with increased efficiency, shorter drying processes can be achieved. This study investigates the effects of ethanol and citric acid pretreatments on the convective drying process of apple slices and the drying rate, diameter and thickness shrinkage, color properties, total phenolic content (TPC), antioxidant activity (AA), ATR-FTIR spectra, and principal component analysis (PCA) of the dried samples. The results indicate that both ethanol and citric acid pretreatments significantly enhance the drying rate and decrease drying time, with the most favorable outcomes observed for apple slices immersed in an ethanol solution for 20 min. The study employs thin-layer and artificial neural network (ANN) modeling, revealing that ANN modeling outperforms thin-layer models in predicting moisture ratio. Shrinkage ratios in diameter and thickness were observed, but no significant statistical differences are found among the sample groups. The color properties of dried apple slices are influenced by pretreatments. L* values decreased in the ethanol-pretreated samples, whereas a* and b* values increased in all samples. On the other hand, drying process leads to a decrease in TPC and AA. Ethanol pretreatments caused higher losses; lower losses were observed in the citric acid–pretreated and untreated apples slices. ATR-FTIR analysis suggests distinct spectral changes in dried samples, particularly influenced by ethanol and citric acid pretreatments. The ATR-FTIR spectra highlighted shifts in water and carbohydrate levels, proteins, fibers, organic acids, and the occurrence of Maillard reactions throughout the drying process. PCA reveals that samples dried with ethanol and citric acid share a similar plane, while fresh samples and those dried at 60 °C exhibit different arrangements.

Impact of Nature of Medical Data on Machine and Deep Learning for Imbalanced Datasets: Clinical Validity of SMOTE Is Questionable

Dataset imbalances pose a significant challenge to predictive modeling in both medical and financial domains, where conventional strategies, including resampling and algorithmic modifications, often fail to adequately address minority class underrepresentation. This study theoretically and practically investigates how the inherent nature of medical data affects the classification of minority classes. It employs ten machine and deep learning classifiers, ranging from ensemble learners to cost-sensitive algorithms, across comparably sized medical and financial datasets. Despite these efforts, none of the classifiers achieved effective classification of the minority class in the medical dataset, with sensitivity below 5.0% and area under the curve (AUC) below 57.0%. In contrast, the similar classifiers applied to the financial dataset demonstrated strong discriminative power, with overall accuracy exceeding 95.0%, sensitivity over 73.0%, and AUC above 96.0%. This disparity underscores the unpredictable variability inherent in the nature of medical data, as exemplified by the dispersed and homogeneous distribution of the minority class among other classes in principal component analysis (PCA) graphs. The application of the synthetic minority oversampling technique (SMOTE) introduced 62 synthetic patients based on merely 20 original cases, casting doubt on its clinical validity and the representation of real-world patient variability. Furthermore, post-SMOTE feature importance analysis, utilizing SHapley Additive exPlanations (SHAP) and tree-based methods, contradicted established cerebral stroke parameters, further questioning the clinical coherence of synthetic dataset augmentation. These findings call into question the clinical validity of the SMOTE technique and underscore the urgent need for advanced modeling techniques and algorithmic innovations for predicting minority-class outcomes in medical datasets without depending on resampling strategies. This approach underscores the importance of developing methods that are not only theoretically robust but also clinically relevant and applicable to real-world clinical scenarios. Consequently, this study underscores the importance of future research efforts to bridge the gap between theoretical advancements and the practical, clinical applications of models like SMOTE in healthcare.

Predicting Progression of Intracranial Hemorrhage in the Prehospital TXA for TBI Trial.

Progression of intracranial hemorrhage is a common, potentially devastating complication after moderate/severe traumatic brain injury (TBI). Clinicians have few tools to predict which patients with traumatic intracranial hemorrhage on their initial head computed tomographic scan (hCT) scan will progress. The objective of this investigation was to identify clinical, imaging, and/or protein biomarkers associated with progression of intracranial hemorrhage (PICH) after moderate/severe TBI and to create an accurate predictive model of PICH based on clinical features available at presentation. We analyzed a subset of subjects from the phase II double-blind, multi-center randomized "Prehospital Tranexamic Acid Use for TBI" trial. This subset was limited to the placebo arm of the parent trial with evidence of hemorrhage on the initial head CT and a follow-up CT 6 hours after. PICH was defined as an increase in hemorrhage size by 30% or more, or the development of new hemorrhage in the intra- and extra-axial intracranial vault between the initial and the follow-up head CT. Two independent radiologists evaluated each head CT, and conflicts were adjudicated by a third. Clinical and radiographic characteristics were collected, along with plasma protein biomarkers at admission. Principal component analysis (PCA) was performed, and each principal component (PC) was interrogated for its association with PICH. Finally, expert opinion and recursive feature extraction (RFE) were used to select input features for the construction of several super- vised classification models. Their ability to predict PICH were quantified and compared. In this subset of subjects (n=104), 46% (n=48) demonstrated PICH. Univariate analyses showed no association between PICH and age, sex, admission Glasgow Coma Scale (GCS), GCS motor subscore, presence of midline shift, admission platelet count or admission INR. Radiographic severity scores (Marshall score [P=0.007], Rotterdam score [P=0.004]) and initial hematoma volume [P=0.005] were associated with PICH. Higher levels of admission GFAP (P<0.001) and MAP (P=0.011) were also associated with PICH. Of the PCs, PC1 was significantly associated with PICH (P=0.0125). Using multimodal data input, machine learning classifiers successfully discriminated patients with or without PICH. Models composed of machine selected features performed better than models composed of expert selected variables (reaching an average of 77% accuracy, AUC = 0.78 versus AUC = 0.68 for the expert selected variables). Predictive models utilizing variables measured at admission can accurately predict PICH, confirmed by the 6-hour follow-up head CT. Our best performing models must now be externally validated in a separate cohort of TBI patients with low GCS and initial head CT positive for hemorrhage.

Infant feeding practices and body mass index up to 7.5 years in the French nationwide ELFE study.

The infant diet represents one of the main modifiable determinants of early growth. This study aimed to investigate the associations of infant feeding practices with body mass index (BMI) until 7.5 years. Analyses were based on data from the French nationwide ELFE birth cohort. Data on breastfeeding (BF) and complementary feeding (CF) were collected monthly from 2 to 10 months. Infant feeding practices were characterized using principal component analyses (PCA) and hierarchical ascendant classification. BMI z-score was computed at 1, 2, 3, 5 and 7.5 years, from data collected in the child's health booklet; 7.5-year overweight was defined according to IOTF references. Associations between infant feeding practices and BMI were investigated by linear regression models adjusted for main confounders. Ever breastfeeding was not associated with BMI up to 7.5 years. Compared to intermediate breastfeeding duration (1 to <3 months), longer breastfeeding duration (≥6 months) was related to lower 1-year BMI, but not at older ages. Compared to the recommended age at CF introduction (4-6 months), early CF (<4 months) was related to higher BMI up to 5 years with a similar trend at 7.5 years, but not to the risk of overweight. The PCA patterns characterized by early baby cereal introduction and late food pieces introduction or by frequent intake of main food groups were related to a lower BMI up to 7.5 years. Breastfeeding was related with a lower BMI in infancy but not thereafter, whereas an early CF initiation (<4 months) was associated with a higher BMI in childhood.

Metabolomic Approach to Identify Potential Biomarkers in KRAS-Mutant Pancreatic Cancer Cells

Pancreatic cancer is characterized by its high mortality rate and limited treatment options, often driven by oncogenic RAS mutations. In this study, we investigated the metabolomic profiles of pancreatic cancer cells based on their KRAS genetic status. Utilizing both KRAS-wildtype BxPC3 and KRAS-mutant PANC1 cell lines, we identified 195 metabolites differentially altered by KRAS status through untargeted metabolomics. Principal component analysis and hierarchical condition trees revealed distinct separation between KRAS-wildtype and KRAS-mutant cells. Metabolite set enrichment analysis highlighted significant pathways such as homocysteine degradation and taurine and hypotaurine metabolism. Additionally, lipid enrichment analysis identified pathways including fatty acyl glycosides and sphingoid bases. Mapping of identified metabolites to KEGG pathways identified nine significant metabolic pathways associated with KRAS status, indicating diverse metabolic alterations in pancreatic cancer cells. Furthermore, we explored the impact of TRPML1 inhibition on the metabolomic profile of KRAS-mutant pancreatic cancer cells. TRPML1 inhibition using ML-SI1 significantly altered the metabolomic profile, leading to distinct separation between vehicle-treated and ML-SI1-treated PANC1 cells. Metabolite set enrichment analysis revealed enriched pathways such as arginine and proline metabolism, and mapping to KEGG pathways identified 17 significant metabolic pathways associated with TRPML1 inhibition. Interestingly, some metabolites identified in PANC1 compared to BxPC3 were oppositely regulated by TRPML1 inhibition, suggesting their potential as biomarkers for KRAS-mutant cancer cells. Overall, our findings shed light on the distinct metabolite changes induced by both KRAS status and TRPML1 inhibition in pancreatic cancer cells, providing insights into potential therapeutic targets and biomarkers for this deadly disease.

Application of texture signatures based on multiparameter-magnetic resonance imaging for predicting microvascular invasion in hepatocellular carcinoma: Retrospective study

BACKGROUND Despite continuous changes in treatment methods, the survival rate for advanced hepatocellular carcinoma (HCC) patients remains low, highlighting the importance of diagnostic methods for HCC. AIM To explore the efficacy of texture analysis based on multi-parametric magnetic resonance (MR) imaging (MRI) in predicting microvascular invasion (MVI) in preoperative HCC. METHODS This study included 105 patients with pathologically confirmed HCC, categorized into MVI-positive and MVI-negative groups. We employed Original Data Analysis, Principal Component Analysis, Linear Discriminant Analysis (LDA), and Non-LDA (NDA) for texture analysis using multi-parametric MR images to predict preoperative MVI. The effectiveness of texture analysis was determined using the B11 program of the MaZda4.6 software, with results expressed as the misjudgment rate (MCR). RESULTS Texture analysis using multi-parametric MRI, particularly the MI + PA + F dimensionality reduction method combined with NDA discrimination, demonstrated the most effective prediction of MVI in HCC. Prediction accuracy in the pulse and equilibrium phases was 83.81%. MCRs for the combination of T2-weighted imaging (T2WI), arterial phase, portal venous phase, and equilibrium phase were 22.86%, 16.19%, 20.95%, and 20.95%, respectively. The area under the curve for predicting MVI positivity was 0.844, with a sensitivity of 77.19% and specificity of 91.67%. CONCLUSION Texture analysis of arterial phase images demonstrated superior predictive efficacy for MVI in HCC compared to T2WI, portal venous, and equilibrium phases. This study provides an objective, non-invasive method for preoperative prediction of MVI, offering a theoretical foundation for the selection of clinical therapy.

The impact of uranium contamination in groundwater on human health: a toxicological risk assessment

ABSTRACT All living beings are exposed to naturally occurring radionuclides. Uranium, a primordial radionuclide forms compounds that are water soluble and enter the human body through food and drinking water. This study conducted a health risk assessment to evaluate the potential dangers associated with consuming uranium through groundwater in the Mansa district of Punjab, India. The uranium concentration varied from 43.43 µg/L to 449.79 µg/L. The biokinetic model of uranium including hair compartment as excretion pathway is used to calculate uranium retention in different organs of the human body. The Pearson correlation and principal component analysis are used for assessing the correlation of uranium with TDS, salinity, conductivity and pH. The high correlation is observed between TDS and uranium. The hazard quotient is 17.42 and excess cancer risk is higher than the permissible limit of 10−4. Owing to the high uranium concentration in Mansa district, periodic monitoring is advised along with clinical investigation of the population.

Improving prediction of maternal health risks using PCA features and TreeNet model

Maternal healthcare is a critical aspect of public health that focuses on the well-being of pregnant women before, during, and after childbirth. It encompasses a range of services aimed at ensuring the optimal health of both the mother and the developing fetus. During pregnancy and in the postpartum period, the mother’s health is susceptible to several complications and risks, and timely detection of such risks can play a vital role in women’s safety. This study proposes an approach to predict risks associated with maternal health. The first step of the approach involves utilizing principal component analysis (PCA) to extract significant features from the dataset. Following that, this study employs a stacked ensemble voting classifier which combines one machine learning and one deep learning model to achieve high performance. The performance of the proposed approach is compared to six machine learning algorithms and one deep learning algorithm. Two scenarios are considered for the experiments: one utilizing all features and the other using PCA features. By utilizing PCA-based features, the proposed model achieves an accuracy of 98.25%, precision of 99.17%, recall of 99.16%, and an F1 score of 99.16%. The effectiveness of the proposed model is further confirmed by comparing it to existing state of-the-art approaches.

Characterization and Discrimination of Marigold Oleoresin from Different Origins Based on UPLC-QTOF-MS Combined Molecular Networking and Multivariate Statistical Analysis

Marigold oleoresin is an oil-soluble natural colorant mainly extracted from marigold flowers. Xinjiang of China, India, and Zambia of Africa are the three main production areas of marigold flowers. Therefore, this study utilized ultra-performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-QTOF-MS/MS) technology, combined with Global Natural Products Social Molecular Networking (GNPS) and multivariate statistical analysis, for the qualitative and discriminant analysis of marigold oleoresin obtained from three different regions. Firstly, 83 compounds were identified in these marigold oleoresin samples. Furthermore, the results of a principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) indicated significant differences in the chemical compositions of the marigold oleoresin samples from different regions. Finally, 12, 23, and 38 differential metabolites were, respectively, identified by comparing the marigold oleoresin from Africa with Xinjiang, Africa with India, and Xinjiang with India. In summary, these results can be used to distinguish marigold oleoresin samples from different regions, laying a solid foundation for further quality control and providing a theoretical basis for assessing its safety and nutritional aspects.

Multiple faces of stress in the zebrafish (Danio rerio) brain

The changing expressions of certain genes as a consequence of exposure to stressors has not been studied in detail in the fish brain. Therefore, a stress trial with zebrafish was conducted, aiming at identifying relevant gene regulation pathways in different regions of the brain. As acute stressors within this trial, feed rewarding, feed restriction, and air exposure have been used. The gene expression data from the experimental fish brains have been analyzed by means of principal component analyses (PCAs), whereby the individual genes have been compiled according to the regulation pathways in the brain. The results did not indicate a mutual response across the treatment and gender groups. To evaluate whether a similar sample structure belonging to a large sample size would have allowed the classification of the gene expression patterns according to the treatments, the data have been bootstrapped and used for building random forest models. These revealed a high accuracy of the classifications, but different genes in the female and male zebrafish were found to have contributed to the classification algorithms the most. These analyses showed that less than eight genes are, in most cases, sufficient for an accurate classification. Moreover, mainly genes belonging to the stress axis, to the isotocin regulation pathways, or to the serotonergic pathways had the strongest influence on the outcome of the classification models.