基于不同治疗方法对脑卒中患者水肿体积进展研究  

Research on the progression of edema volumein stroke patients based on different treatment methods

作　　者：李腾 朱孟坤 杨柳 朱奥宇 李金红 LI Teng;ZHU Mengkun;YANG Liu;ZHU Aoyu;LI Jinhong(Faculty of Mathematics and Artificial Intelligence,Qilu University of Technology(Shandong Academy of Science),Jinan 250353,China)

机构地区：[1]齐鲁工业大学(山东省科学院)数学与人工智能学部,山东济南250353

出　　处：《齐鲁工业大学学报》2025年第2期72-80,共9页Journal of Qilu University of Technology

基　　金：国家自然科学基金(12201333);山东省科技型中小企业创新能力提升工程(2021TSGC1017)。

摘　　要：出血性脑卒中后,血肿周围的水肿作为脑出血后继发性损伤的标志,在近年来引起了广泛关注。研究患者血肿周围水肿发生及演进规律,进行早期识别和预测对于改善患者预后、提升其生活质量具有重要意义。通过患者个人信息、影像信息、治疗方案和预后等真实临床数据,一是采用多项式回归模型和岭回归模型对数据进行拟合,二是采用高斯函数对数据拟合,根据计算残差结果选择最佳拟合方法,构建全体患者水肿体积随时间进展曲线。再根据年龄把患者分为4个亚组,采用所得最佳拟合模型分析患者水肿体积随时间进展模式的年龄差异。最终根据已知数据计算水肿体积的变化,利用随机森林和多层感知器模型进行训练,使用测试集对模型进行预测,分析基于不同治疗方法对水肿体积进展模式的影响,得出最佳治疗方案。The edema around the hematoma after hemorrhagic stroke has attracted widespread attention in recent years as a marker of secondary injury after cerebral hemorrhage.Studying the occurrence and progression of edema around hematoma in patients,early identification and prediction,is of great significance for improving patient prognosis and enhancing their quality of life.Based on real clinical data such as patient personal information,imaging information,treatment plans,and prognosis,polynomial regression models and ridge regression models are used to fit the data,and Gaussian functions are used to fit the data.The best fitting method is selected based on the calculated residual results to construct a curve of edema volume progression over time for all patients.Divide the patients into four subgroups based on age,and use the obtained best fit model to analyze the age differences in the progression pattern of edema volume over time.Finally,the change in edema volume was calculated based on known data,and a random forest and multi-layer perceptron model were used for training.The model was predicted using a test set,and the impact of different treatment methods on the progression pattern of edema volume was analyzed to determine the optimal treatment plan.

关 键 词：水肿体积 多项式回归 岭回归 随机森林 多层感知器 

分 类 号：O29[理学—应用数学]