PLoS ONE最新文献

{"title":"Protocol for a systematic review and meta-analysis of the combination of ezetimibe and statins for hyperlipidemia.","authors":"Tianfu Yang, Weijuan Li, Weiwei Chen, Donghong Zhu, Yuxi Ren, Xiongfeng Huang","doi":"10.1371/journal.pone.0312588","DOIUrl":"https://doi.org/10.1371/journal.pone.0312588","url":null,"abstract":"<p><strong>Introduction: </strong>Hyperlipidemia is increasingly recognized as a significant global health issue, often associated with conditions such as hypertension, diabetes, and obesity. While statins are frequently prescribed to manage lipid levels, recent studies indicate that reliance solely on statin therapy may present certain disadvantages, including prolonged treatment durations, the potential for drug resistance, and various adverse effects. Research indicates that the combination of ezetimibe and statins demonstrates a favorable therapeutic effect in the management of hyperlipidemia. However, existing studies have not consistently confirmed these benefits, and there is no current meta-analysis available. As a result, we will perform a meta-analysis to assess the effectiveness and safety of the combination of ezetimibe and statins in managing hyperlipidemia, aiming to offer evidence-based medical guidance for clinical practice.</p><p><strong>Methods and analysis: </strong>The systematic review and meta-analysis will adhere to the PRISMA guidelines for systematic reviews and meta-analyses. We will search for randomized controlled trials that investigate the efficacy and safety of the combination of ezetimibe and statins in treating hyperlipidemia, based on specific criteria. The following electronic databases will be searched by two researchers for relevant records published up to October 1, 2024: Cochrane Central Register of Controlled Trials (CENTRAL) in Cochrane Library, Embase.com, Web of Science, MEDLINE (via PubMed), Wanfang China Database, China National Knowledge Infrastructure (CNKI), Chinese Biomedical Literature Database (CBM) and Chinese Scientific Journal Database (VIP). They will also check references and relevant journals manually. Two independent reviewers will handle screening, data extraction, and quality assessment. Subgroup analysis, sensitivity analysis, and publication bias analysis will be performed to assess consistency and reliability. Review Manager 5.4 will be used for data analysis and synthesis, while the GRADE approach will be employed to evaluate the overall study's evidence quality.</p><p><strong>Expected results: </strong>The findings of this systematic review will be shared with various stakeholders who are interested in the combination of ezetimibe and statins for hyperlipidemia. This will offer valuable insights for researchers undertaking future investigations and for clinical practitioners specializing in the treatment of hyperlipidemia.</p><p><strong>Ethics and dissemination: </strong>This study is based on a secondary analysis of the literature, so ethical review approval is not required. The final report will be published in a peer-reviewed journal.</p><p><strong>Study registration: </strong>The protocol of the systematic review has been registered on Open Science Framework, with a registration DOI https://doi.org/10.17605/OSF.IO/TEVUY.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0312588"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of a sequential laser microdissection tissue cuts workflow for the spatial and quantitative analysis of drugs in fresh frozen tissue sections.","authors":"Elias Maris, Farid Jahouh, Kathleen Allaerts, Rob Vreeken","doi":"10.1371/journal.pone.0312542","DOIUrl":"https://doi.org/10.1371/journal.pone.0312542","url":null,"abstract":"<p><p>Mass spectrometry imaging (MSI) is a well-established technique that allows to determine the distribution of small molecules, such as lipids, metabolites, and drugs, as well as large molecules in tissue sections. Because of the tissue heterogeneity, resulting in different matrix effects, and to the fact that the measured compounds are not entirely \"extracted\" from the tissue during the measurement, the absolute quantitative aspect of MSI is limited. To combine compound quantification with spatial information on fresh frozen unstained tissue sections, laser (capture) microdissection has been used to isolate tissue sections for compound extraction and LC-MS/MS quantification. Although this method relying on manual ROIs selection is rather sensitive compared to traditional MSI methods, it lacks the throughput needed to screen entire tissue sections. To apply a higher throughput tissue screening approach, we propose herein a workflow for performing indiscriminate and sequential LMD tissue section cuts that can cover up to 96 cuts collected in a 96 well plate on Leica LMD systems, for further extractions and LC-MS/MS analysis. Our workflow relies on the creation and implementation of 96 squares microgrid templates for the LMD cut of different area sizes (30x30 μm2, 50x50 μm2, 100x100 μm2, 200x200 μm2, 270x270 μm2 and 500x500 μm2) using 5 different magnifications (5x, 10x, 20x, 40x and 63x), on fresh frozen tissue sections. The method was applied on 20μm mouse brain and liver tissue sections. The tissue cut collection yields were evaluated visually and by the detection of the sprayed standards on the tissue sections, and found to be ranging from 78% to 91%, and the throughput of the LMD cuts and collection in a 96 well format, was measured to be from 19 to 37 minutes per tissue section, depending on the 96 squares microgrid template and the corresponding magnification lens used. Further extraction and LC-MS/MS analysis of 3 different compounds previously sprayed on a mouse liver tissue section allowed to determine the LLOQ the workflow allows to achieve when using the different templates.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0312542"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Association between cardiac autonomic regulation, visceral adipose tissue, cardiorespiratory fitness and ambient air pollution: 4HAIE study (Program-4).","authors":"Tomas Dostal, Martina Dankova, Tomas Barot, Dominik Sindler, Petr Kutac, Vera Jandackova, Daniel Jandacka, Peter Hofmann, Lukas Cipryan","doi":"10.1371/journal.pone.0315767","DOIUrl":"https://doi.org/10.1371/journal.pone.0315767","url":null,"abstract":"<p><strong>Background: </strong>The main objective of the present cross-sectional cohort study was to determine whether there is an association between cardiac autonomic regulation, as expressed through heart rate variability (HRV), and cardiorespiratory fitness (CRF), visceral adipose tissue (VAT), and over the long-term living in areas with low or high air pollution.</p><p><strong>Methods: </strong>The study sample included 1036 (487 females) healthy runners (603) and inactive participants (age 18-65 years) who had lived for at least 5 years in an area with high (Moravian-Silesian; MS) or low (South Bohemian; SB) air pollution in the Czech Republic. A multivariable regression analysis was used to evaluate the associations between multiple independent variables (CRF (peak oxygen consumption), VAT, sex, socioeconomic status (education level), and region (MS region vs. SB region) with dependent variable HRV. The root mean square of successive RR interval differences (rMSSD) was employed for the evaluation of HRV.</p><p><strong>Results: </strong>The multivariable linear regression model revealed that cardiac autonomic regulation (rMSSD) was significantly associated with CRF level (p < .001) and age (p < .001). There were no associations between rMSSD and region (high or low air-pollution), sex, education level or VAT (p > 0.050).</p><p><strong>Conclusions: </strong>We showed that living in an area with low or high air pollution is not associated with cardiac autonomic modulation in healthy runners and inactive individuals. CRF and age significantly directly and inversely, respectively, associated with HRV. There were no other significant associations.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0315767"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Area under the ROC Curve has the most consistent evaluation for binary classification.","authors":"Jing Li","doi":"10.1371/journal.pone.0316019","DOIUrl":"https://doi.org/10.1371/journal.pone.0316019","url":null,"abstract":"<p><p>The proper use of model evaluation metrics is important for model evaluation and model selection in binary classification tasks. This study investigates how consistent different metrics are at evaluating models across data of different prevalence while the relationships between different variables and the sample size are kept constant. Analyzing 156 data scenarios, 18 model evaluation metrics and five commonly used machine learning models as well as a naive random guess model, I find that evaluation metrics that are less influenced by prevalence offer more consistent evaluation of individual models and more consistent ranking of a set of models. In particular, Area Under the ROC Curve (AUC) which takes all decision thresholds into account when evaluating models has the smallest variance in evaluating individual models and smallest variance in ranking of a set of models. A close threshold analysis using all possible thresholds for all metrics further supports the hypothesis that considering all decision thresholds helps reduce the variance in model evaluation with respect to prevalence change in data. The results have significant implications for model evaluation and model selection in binary classification tasks.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0316019"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882718","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatiotemporal dynamic and regional differences of public attention to vaccination: An empirical study in China.","authors":"Yaming Zhang, Xiaoyu Guo, Yanyuan Su","doi":"10.1371/journal.pone.0312488","DOIUrl":"https://doi.org/10.1371/journal.pone.0312488","url":null,"abstract":"<p><strong>Background: </strong>Internet searches offer an indicator of public attention and possible demand for certain things. Studying the spatiotemporal characteristics of the public's concern for vaccination can determine the spatiotemporal distribution of demand for vaccines in China, and capture the changes in the health awareness of the Chinese population, thus informing future vaccination strategies.</p><p><strong>Methods: </strong>Based on the collection of Baidu search indices for vaccination-related keywords in 363 cities in China, This paper seeks to explore the spatiotemporal changes and regional differences in public attention toward vaccination in China by using the seasonal index, seasonal concentration index, Herfindahl index, Moran index, and Dagum Gini coefficient.</p><p><strong>Results: </strong>The following findings are presented. First, there are significant seasonal fluctuations and unbalanced monthly distributions of vaccination-related public attention in China. Second, the public attention in Chinese cities shows the spatial characteristics of \"leading in the east, followed by the central, western and northeastern regions\". The spatial correlation of attention has been strengthened, and the high-high clusters are mainly distributed in the Beijing-Tianjin-Hebei (BTH), Yangtze River Delta (YRD), and Greater Bay Area (GBA) urban agglomerations. Third, regional differences in overall public attention narrowed in China, with intra-regional differences narrowing in seven regions (Northwest China, Central China, and so on), and intra-regional differences increasing in East China. The dominant role in the Gini coefficient changes from transvariation intensity to inter-regional differences.</p><p><strong>Conclusion: </strong>Major public health emergencies stimulate the public's attention to health topics. Although the short-term increase in vaccination-related public attention was not observed to translate into a long-term increase in public vaccine literacy, the seasonal and regional differences in vaccination-related public attention in China have significantly narrowed before and after COVID-19, suggesting that the imbalance between public health literacy levels has improved.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0312488"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive characterization of high-risk coding and non-coding single nucleotide polymorphisms of human CXCR4 gene.","authors":"Bonoshree Sarkar, Muhammad Safiul Alam Mondal, Taibur Rahman, Md Ismail Hosen, Atiqur Rahman","doi":"10.1371/journal.pone.0312733","DOIUrl":"https://doi.org/10.1371/journal.pone.0312733","url":null,"abstract":"<p><p>CXCR4, a chemokine receptor known as Fusin or CD184, spans the outer membrane of various human cells, including leukocytes. This receptor is essential for HIV infection as well as for many vital cellular processes and is implicated to be associated with multiple pathologies, including cancers. This study employs various computational tools to investigate the molecular effects of disease-vulnerable germ-line missense and non-coding SNPs of the CXCR4 gene. In this investigation, the tools SIFT, PROVEAN, PolyPhen-2, PANTHER, SNAP 2.0, PhD-SNP, and SNPs&GO were used to predict potentially harmful and disease-causing nsSNPs in CXCR4. Additionally, their impact on protein stability was examined by I-mutant 3.0, MUpro, Consurf, and Netsurf 2.0, combined with conservation and solvent accessibility analyses. Structural analysis with normal and mutant residues of the protein harboring these disease-associated functional SNPs was conducted using TM-align and SWIS MODEL, with visualization aided by PyMOL and the BIOVINA Discovery Studio Visualizer. The functional impact of wild-type and mutated CXCR4 variants was evaluated through molecular docking with its natural ligand CXCR4-modulator 1, using the PyRx tool. Non-coding SNPs in the 3' -UTR were investigated for their regulatory effects on miRNA binding sites using PolymiRTS. Five non-coding SNPs were identified in the 3'-UTR that can disrupt existing miRNA binding sites or create new ones. Non-coding SNPs in the 5' and 3'-UTRs, as well as in intronic regions, were also examined for their potential roles in gene expression regulation. Furthermore, RegulomeDB databases were employed to assess the regulatory potential of these non-coding SNPs based on chromatin state and protein binding regulation. In the mostly annotated variant (ENSP00000241393) of the CXCR4 gene, we found 23 highly deleterious and pathogenic nsSNPs and these were selected for in-depth analysis. Among the 23 nsSNPs, five (G55V, H79P, L80P, H113P, and P299L) displayed notable structural alternation, with elevated RMSD values and reduced TM (TM-score) values. A molecular docking study revealed the significant impact of the H113P variant on the protein-ligand binding affinity, supported by MD simulation over 100 nanoseconds, which highlighted substantial stability differences between wild-type and H113P mutated proteins during ligand binding. This comprehensive analysis shed light on the potential functional consequences of genetic variation in the CXCR genes, offering valuable insights into the implications of disease susceptibility and may pave the way for future therapeutic interventions.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0312733"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Influences of aging and mating history in males on paternity success in the red flour beetle Tribolium castaneum.","authors":"Renya Kawakami, Kentarou Matsumura","doi":"10.1371/journal.pone.0316008","DOIUrl":"https://doi.org/10.1371/journal.pone.0316008","url":null,"abstract":"<p><p>The investment of males in reproductive traits is often associated with their age. For example, several empirical and theoretical studies have demonstrated that older males make greater investment in reproduction compared with younger males. However, with regards to post-copulatory sexual selection, male reproductive success might be influenced by decreasing sperm quality with male age and the interaction between aging and mating experience in males. However, only a few studies that investigated influences of male aging as well as male mating experience on their post-copulatory sexual selection. In this study, we investigated paternity success influenced by the post-copulatory sexual selection of males at different ages in the red flour beetle Tribolium castaneum. To investigate the effects of the mating experience, the paternity success of older males who had experienced multiple matings (mated male) was compared with older males who had not experienced mating (naive male). The results of this study revealed that paternity success was not affected by male aging. In fact, naive old males exhibited significantly higher paternity success compared with old males who had previously mated. These results suggest that an interaction between male aging and their mating experience affected their paternity success, but not male aging. Our study has demonstrated that male aging affects their reproductive success in a complex interaction of multiple factors in T. castaneum.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0316008"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882801","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Inherent stochasticity, noise and limits of detection in continuous and time-gated fluorescence systems.","authors":"Nicholas H Vitale, Arjang Hassibi, Hyongsok Tom Soh, Boris Murmann, Thomas H Lee","doi":"10.1371/journal.pone.0313949","DOIUrl":"https://doi.org/10.1371/journal.pone.0313949","url":null,"abstract":"<p><p>We present a model for the noise and inherent stochasticity of fluorescence signals in both continuous wave (CW) and time-gated (TG) conditions. When the fluorophores are subjected to an arbitrary excitation photon flux, we apply the model and compute the evolution of the probability mass function (pmf) for each quantum state comprising a fluorophore's electronic structure, and hence the dynamics of the resulting emission photon flux. Both the ensemble and stochastic models presented in this work have been verified using Monte Carlo molecular dynamic simulations that utilize the Gillespie algorithm. The implications of the model on the design of biomolecular fluorescence detection systems are explored in three relevant numerical examples. For a given system, the quantum-limited signal-to-noise ratio (QSNR) and limits of detection are computed to demonstrate how key design tradeoffs are quantified. We find that as systems scale down to micro- and nano- dimensions, the interplay between the fluorophore's photophysical qualities and use of CW or TG has ramifications on optimal design strategies when considering optical component selection, measurement speed, and system energy requirements. While CW systems remain a gold standard, TG systems can be leveraged to overcome cost and system complexity hurdles when paired with the appropriate fluorophore.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0313949"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Perceived benefits and challenges one year after receiving brief therapy in a district psychiatric centre. An exploration of patients' and GPs' experiences: A qualitative study.","authors":"Hilde V Markussen, Lene Aasdahl, Petter Viksveen, Marit B Rise","doi":"10.1371/journal.pone.0312513","DOIUrl":"https://doi.org/10.1371/journal.pone.0312513","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Scarce treatment resources put pressure on mental health services prompting innovations in service provision. Various innovative strategies have been introduced to provide patients with improved and effective treatment due to increased demands for mental health services. In 2016 a district psychiatric centre (DPC) started a brief treatment program to provide early and effective help for moderate depression and anxiety. There is little evidence regarding the potential benefits that different mental health patients may experience after receiving brief therapy treatment. Moreover, few studies have explored the experiences of referring general practitioners (GPs) with different patient outcomes after brief therapy. The aim of this study was to investigate the long-term experiences of patients who received brief therapy at a DPC at least one year ago, as well as the experiences of general practitioners (GPs) who referred patients for such treatment since 2016. Specifically, the study sought to determine patterns in the patients' stories and GPs' experiences to see if it could provide new insight for further studies.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We conducted individual interviews with a total of seventeen participants, consisting of eleven patients and six GPs. Using an exploratory approach, we analyzed patients' narratives as they described them in the interviews, employing inductive and thematic analysis techniques. The GPs' experiences of referring several patients to brief therapy were also subjected to thematic analysis. In addition, we synthesized the patients' experiences into condensed stories for comparison. The experiences of GPs, who had referred patients to this brief treatment program over several years, were compared with the patients' experiences and reflections one year after receiving brief therapy. This comparison aimed to challenge and deepen the understanding of condensed patient stories.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The main results are presented as three condensed patient stories: A) Coping with mental problems; B) A path to another treatment; and C) Confusion and lost hope. The GPs' experiences are included in each condensed patient story to challenge and elaborate on relevant aspects.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;Time-limited brief therapy was experienced as beneficial for patients with moderate affective and anxiety disorders but was experienced as unsuitable for those with more severe conditions. This raises important questions about the appropriateness of offering brief therapy to a diverse patient population and the efficient use of healthcare resources. We recommend further research to enhance understanding and develop tailored treatment services for different ailments. Identifying which patients benefit most from specific therapies can improve outcomes and resource allocation. An important improvement measure might be to enhance early communication between patients, General Pract","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0312513"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing hypertension prediction using ensemble learning approaches.","authors":"Isteaq Kabir Sifat, Md Kaderi Kibria","doi":"10.1371/journal.pone.0315865","DOIUrl":"https://doi.org/10.1371/journal.pone.0315865","url":null,"abstract":"<p><p>Hypertension (HTN) prediction is critical for effective preventive healthcare strategies. This study investigates how well ensemble learning techniques work to increase the accuracy of HTN prediction models. Utilizing a dataset of 612 participants from Ethiopia, which includes 27 features potentially associated with HTN risk, we aimed to enhance predictive performance over traditional single-model methods. A multi-faceted feature selection approach was employed, incorporating Boruta, Lasso Regression, Forward and Backward Selection, and Random Forest feature importance, and found 13 common features that were considered for prediction. Five machine learning (ML) models such as logistic regression (LR), artificial neural network (ANN), random forest (RF), extreme gradient boosting (XGB), light gradient boosting machine (LGBM), and a stacking ensemble model were trained using selected features to predict HTN. The models' performance on the testing set was evaluated using accuracy, precision, recall, F1-score, and area under the curve (AUC). Additionally, SHapley Additive exPlanations (SHAP) was utilized to examine the impact of individual features on the models' predictions and identify the most important risk factors for HTN. The stacking ensemble model emerged as the most effective approach for predicting HTN risk, achieving an accuracy of 96.32%, precision of 95.48%, recall of 97.51%, F1-score of 96.48%, and an AUC of 0.971. SHAP analysis of the stacking model identified weight, drinking habits, history of hypertension, salt intake, age, diabetes, BMI, and fat intake as the most significant and interpretable risk factors for HTN. Our results demonstrate significant advancements in predictive accuracy and robustness, highlighting the potential of ensemble learning as a pivotal tool in healthcare analytics. This research contributes to ongoing efforts to optimize HTN prediction models, ultimately supporting early intervention and personalized healthcare management.</p>","PeriodicalId":20189,"journal":{"name":"PLoS ONE","volume":"19 12","pages":"e0315865"},"PeriodicalIF":2.9,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142882075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}