Suppr超能文献

利用计数回归探索血小板与其他成分之间的关系:孟加拉国的一项横断面研究。

Exploring the relationship between blood platelet and other components utilizing count regression: A cross-sectional study in Bangladesh.

作者信息

Honey Umme, Saleh Sm Arif Bin, Salan Md Sifat Ar, Kabir Mohammad Alamgir, Ali Akher

机构信息

Department of Statistics and Data Science Jahangirnagar University Dhaka Savar Bangladesh.

NYU Langone Health Long Island New York USA.

出版信息

Health Sci Rep. 2024 Aug 20;7(8):e70007. doi: 10.1002/hsr2.70007. eCollection 2024 Aug.

Abstract

BACKGROUND AND AIMS

Blood, vital for transporting nutrients and maintaining balance, comprises red blood cells, white blood cells, and platelets, each pivotal. Imbalances lead to issues-low red cells cause fatigue (anemia), high white cells hint at infection, low counts raise infection risks. Using trendy statistical approaches, investigating the complex link between platelet counts and numerous blood components. Our investigation, leveraging count regression approaches, revealed deep insights into the interaction between platelet counts and other important hematological markers.

METHODS

A cross-sectional study utilized data from 3120 individuals, including both male and female participants, who visited these hospitals between June 16, 2022 and December 17, 2022, to assess their blood samples through testing by using convenience non-parametric sampling framework. Platelet count was taken into account as a measure of outcome in this research. This specific study region was chosen for its easy accessibility, which helped the seamless execution of the data-gathering technique. Count regression, negative binomial regression, and quasi-Poisson regression techniques have been employed for examining relationship of the data sets.

RESULTS

Three different count regression models were utilized to assess the proper association between the response and the relevant covariates and we found negative binomial count regression model (Akaike information criterion = 76.55, Bayesian information criterion = 76.59, and deviance = 3.14) was providing comparatively better performance than others. Based on the chosen model we found white blood cell, erythrocyte sedimentation rate, and eosinophils are significant but neutrophil, monocyte, and lymphocyte are not significant. We have also gone through proper model adequacy checking for our selected model and we found enough evidence to justify our model.

CONCLUSION

From the result, we found insightful remarks into the mechanisms involved in platelet production and regulation, which can aid in developing increased effective treatments and interventions to maintain optimal platelet levels and prevent health problems related to abnormal platelet counts.

摘要

背景与目的

血液对于运输营养物质和维持平衡至关重要,它由红细胞、白细胞和血小板组成,每种成分都至关重要。血液成分失衡会引发各种问题——红细胞数量过低会导致疲劳(贫血),白细胞数量过高提示感染,血小板计数过低会增加感染风险。本研究采用前沿统计方法,探究血小板计数与多种血液成分之间的复杂联系。我们的研究利用计数回归方法,深入揭示了血小板计数与其他重要血液学指标之间的相互作用。

方法

一项横断面研究使用了2022年6月16日至2022年12月17日期间到这些医院就诊的3120名个体(包括男性和女性参与者)的数据,通过便利非参数抽样框架对他们的血液样本进行检测评估。本研究将血小板计数作为结果指标。选择该特定研究区域是因其交通便利,有助于数据收集技术的顺利实施。采用计数回归、负二项回归和拟泊松回归技术来检验数据集之间的关系。

结果

使用三种不同的计数回归模型来评估响应变量与相关协变量之间的恰当关联,我们发现负二项计数回归模型(赤池信息准则=76.55,贝叶斯信息准则=76.59,偏差=3.14)的表现相对优于其他模型。基于所选模型,我们发现白细胞、红细胞沉降率和嗜酸性粒细胞具有显著性,但中性粒细胞、单核细胞和淋巴细胞不具有显著性。我们还对所选模型进行了恰当的模型拟合优度检验,发现有足够证据证明我们的模型合理。

结论

从结果中,我们对血小板生成和调节机制有了深刻认识,这有助于开发更有效的治疗方法和干预措施,以维持最佳血小板水平,预防与异常血小板计数相关的健康问题。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/2944/11335575/1890cf07e2e7/HSR2-7-e70007-g002.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验