通讯作者: 周巍, zhouweishmily@163.com

DOI：10.12201/bmr.202503.00046

Construction and Analysis of a Prediction Model for Hypertension Combined with Left Ventricular Diastolic Dysfunction Based on Random Forest AlgorithmWANG Tingting1 ,ZHOU Wei1*

• 摘要：背景 高血压对心脏的损害进程中，左心室舒张功能障碍的发生和发展是一个重要而关键的过渡阶段，左心室舒张功能减退往往比收缩功能的减退更早发生，分析和探讨高血压合并左室舒张功能障碍的危险因素，对于及早预防和干预进一步心脏损伤有重要意义。目的 通过构建随机森林模型进行危险因素预测验证，探讨EH合并LVDD的影响因素，为临床防治提供依据。方法 选取2023年10月至2024年7月于我院住院治疗的原发性高血压患者300例作为研究对象。收集患者入院时基本资料、实验室检查结果等资料，通过是否合并LVDD将其分为单纯EH组（n=60）和合并LVDD组（LVDD，n=240），比较两组一般资料及生化指标，Logistic回归分析EH合并LVDD的影响因素，构建随机森林模型。结果 与EH组相比，LVDD组年龄、高血压分级、高血压危险分层、病程、合并冠心病占比、WBC、PLT、IL-6、TC、LDL-C、ALP、CK、Pro-BNP升高（p＜0.05）；HDL-C、ALB降低（p＜0.05）。Logistic 回归分析显示，高血压分级、高血压危险分层、病程、合并冠心病占比、WBC、IL-6、TC、LDL-C、ALB、ALT、Pro-BNP，是EH合并LVDD的影响因素，其中ALB是保护因素。随机森林模型显示，当决策树数量为300时，模型袋外数据(OOB)错误率达到最低，为2.86%。根据重要性排序，影响EH合并LVDD的影响因素从高到低依次为：IL-6、ALB、ALP、LDL-C、Pro-BNP、TC、高血压分级、高血压危险分层、WBC。结论 IL-6、ALB、ALP、LDL-C、Pro-BNP、TC、高血压分级、高血压危险分层、WBC是EH合并LVDD的影响因素，有助于临床早期防治。

   

  Abstract: Background The occurrence and development of left ventricular diastolic dysfunction is an important and critical transitional stage in the damage process of hypertension to the heart. Left ventricular diastolic dysfunction often occurs earlier than systolic dysfunction. Analyzing and exploring the risk factors of hypertension combined with left ventricular diastolic dysfunction is of great significance for early prevention and intervention of further heart damage. Objective To construct a random forest model for risk factor prediction and validation, explore the influencing factors of EH combined with LVDD, and provide a basis for clinical prevention and treatment. Method 300 patients with primary hypertension who were hospitalized in our hospital from October 2023 to July 2024 were selected as the research subjects. Collect basic information and laboratory test results of patients upon admission, and divide them into a simple EH group (n=60) and a combined LVDD group (LVDD, n=240) based on whether LVDD is present or not. Compare the general information and biochemical indicators of the two groups, perform logistic regression analysis on the influencing factors of EH combined with LVDD, and construct a random forest model. Results Compared with the EH group, the LVDD group showed an increase in age, hypertension grading, hypertension risk stratification, disease duration, proportion of concomitant coronary heart disease, WBC, PLT, IL-6, TC, LDL-C, ALP, CK, and Pro BNP (p<0.05); HDL-C and ALB decreased (p<0.05). Logistic regression analysis showed that hypertension grading, hypertension risk stratification, disease duration, proportion of concomitant coronary heart disease WBC、IL-6、TC、LDL-C、ALB、ALT、Pro-BNP, It is the influencing factor of EH merging with LVDD, among which ALB is the protective factor. The random forest model shows that when the number of decision trees is 300, the out of bag data (OOB) error rate of the model reaches its lowest point, which is 2.86%. According to the ranking of importance, the factors that affect the merging of EH and LVDD, from high to low, are as follows: IL-6、ALB、ALP、LDL-C、Pro-BNP、TC、 Hypertension classification, hypertension risk stratification WBC。Conclusion IL-6 ALB、ALP、LDL-C、Pro-BNP、TC、Hypertension grading, hypertension risk stratification, and WBC are influencing factors of EH combined with LVDD, which are helpful for early clinical prevention and treatment.

  提交时间：2025-03-15

  版权声明：作者本人独立拥有该论文的版权，预印本系统仅拥有论文的永久保存权利。任何人未经允许不得重复使用。
• html

• 图表

• 阮旭凌, 刘琦, 郭志恒, 晏峻峰. 基于LDA和XGBoost算法的乳腺癌预测模型构建研究. 2022. doi: 10.12201/bmr.202106.00007

  韦云师, 莫伟, 向娅, 廖秋姣, 何柳, 凌朝灵, 陆启祥, 刘芳印. 老年髋部骨折患者术后谵妄风险预测模型的研究进展. 2024. doi: 10.12201/bmr.202409.00029

  毕方平, 刘蕊, 刘小丽, 金泽宏. 左室整体长轴应变在糖尿病患者早期心功能损伤中的应用价值及相关因素分析. 2024. doi: 10.12201/bmr.202411.00042

  杨道玲, 周一波, 王戏丹, 陈静婉, 陈傅华, 陶肖樱. 胎儿心脏定量分析技术对妊娠期高血压患者胎儿早期心功能损害的评估. 2024. doi: 10.12201/bmr.202410.00076

  陈佳琦, 向红英, 凌迎春. 脑电图定量参数、血清miR-146a-5p对阿尔茨海默病患者认知功能障碍的预测价值分析. 2024. doi: 10.12201/bmr.202412.00027

  岳晓波, 赵兴胜. 血清MHR、NLR及PLR与原发性高血压的相关性研究. 2024. doi: 10.12201/bmr.202411.00052

  庞震, 顾继昱, 吴宇飞, 颜仕星, 李汪洋, 孙越. 一种基于三元组抽取策略解决中医诊治高血压医疗实体提取问题的研究. 2021. doi: 10.12201/bmr.202107.00015

  冯利, 岳小飞. 9种结合式机器学习算法在基于高维数据的肿瘤早期诊断准确性的比较研究. 2021. doi: 10.12201/bmr.202108.00016

  王丽丽, 刘贺. 基于5W1H分析法的医院图书馆阅读推广模式构建. 2022. doi: 10.12201/bmr.202208.00003

  胡欣. 剖宫产术中寒战风险预测模型及列线图的构建. 2025. doi: 10.12201/bmr.202501.00053

• 序号	提交日期	编号	操作
  1	2025-02-20	bmr.202503.00046V1	下载

• 公开评论  匿名评论  仅发给作者 提交评论