基于医疗大数据的老年骨质疏松性骨折院前急救分类优化研究  

Optimization of pre-hospital emergency classification for elderly patients with osteoporotic fracture based on medical big data

作　　者：沈蓝 浦同青 周志聪 陆春梅 邓学林 谢好 陈朝明 SHEN Lan;PU Tongqing;ZHOU Zhicong;LU Chunmei;DENG Xuelin;XIE Hao;CHEN Chaoming(Department of Emergency Medicine,Southern Central Hospital of Yunnan Province,the First People′s Hospital of Honghe State,Mengzi,Yunnan,661199;Intensive Care Unit,Southern Central Hospital of Yunnan Province,the First People′s Hospital of Honghe State,Mengzi,Yunnan,661199)

机构地区：[1]云南省滇南中心医院/红河州第一人民医院急诊医学部,云南蒙自661199 [2]云南省滇南中心医院/红河州第一人民医院重症医学科,云南蒙自661199

出　　处：《实用临床医药杂志》2023年第15期7-13,共7页Journal of Clinical Medicine in Practice

基　　金：2023年度云南省科学技术厅昆明医科大学应用基础研究联合专项面上项目(202201AC070006);青岛大学医疗集团科研项目(YLJT20222013)。

摘　　要：目的提取基于医疗大数据的老年骨质疏松性骨折院前急救的结构、过程、结果指标分类要素,并应用现代数理模型进行多学科交叉分类优化研究,分析其相关特征。方法运用文献回顾法检索20篇老年骨质疏松性骨折院前急救的相关文献,采用迭代求解的聚类分析(K-MEANS算法)算法对院前急救指标进行针对性的分类优化分析。结果经矩阵分析证实,结构指标、过程指标、结果指标的一级指标中各维度的权重分别为0.3324、0.1395、0.5276;矩阵一致性检验结果(CR)值均<0.1,其中结构、过程、结果指标的CR值分别为0.0345、0.0394和0.0395,符合检验要求。结论将K-MEANS算法与DeepFM预测模型用于医疗大数据下的老年骨质疏松性骨折院前急救分类指标的复杂关系的预测、定性、定量有益,可为该病的多学科交叉分类优化处理与各项临床基础工作的贯彻落实提供必要条件。Objective To extract the classification elements of indexes such as structure,process and outcome of pre-hospital emergency treatment for elderly patients with osteoporotic fracture based on medical big data,and to analyze their related characteristics by applying modern mathematical models for interdisciplinary classification optimization research.Methods The literature review method was used to search for 20 relevant literatures on pre-hospital emergency treatment for elderly patients with osteoporosis fracture,and the iterative clustering analysis(K-MEANS algorithm)algorithm was used to classify and optimize the pre-hospital emergency treatment indicators.Results Through matrix analysis,it was confirmed that the weights of the dimensions in the primary indicators of structural indicators,process indicators and outcome indicators were 0.3324,0.1395 and 0.5276,respectively;the consistency results(CR)of the matrix were all smaller than 0.1.Among them,the CR values of structure,process and outcome indicators were 0.0345,0.0394 and 0.0395,which met the requirements of the test.Conclusion Applications of the K-MEANS algorithm and the DeepFM predictive model are beneficial for predication,qualification and quantification of the complex relationship between classification indicators for pre-hospital emergency care in elderly patients with osteoporotic fracture based on medical big data,which can provide necessary conditions for optimizing the interdisciplinary classification and implementation of various clinical foundational work for this disease.

关 键 词：医疗大数据 老年骨质疏松性骨折 院前急救 分类优化 迭代求解的聚类分析 DeepFM预测理论 

分 类 号：R683[医药卫生—骨科学] R311[医药卫生—外科学]