基于贝叶斯网络的心血管疾病与其他慢性病因果关系分析
|
摘要
目的:利用贝叶斯网络研究心血管疾病以及一些常见慢性病的因果关系。方法:收集2 752例心血管疾病患者(男1 382例,女1 370例)的病历数据,使用基于分数-搜索学习方法构建贝叶斯网络,采用爬山算法,并使用Bootstrap模型平均增加模型的鲁棒性。构造疾病、二次住院和死亡之间的有向无环图,进而发现变量之间的因果关系。结果与结论:绘制出心血管疾病因果网络图,发现对心血管疾病患者死亡影响最大的路径为冠心病、高血压→脑梗死→肺部感染→死亡。
Aim: To study the causal relationship between cardiovascular diseases and common chronic diseases using Bayesian network. Methods: The data of 2 752 patients( 1 382 males,1 370 females) with cardiovascular diseases were collected. Bayesian network was constructed based on scoring-search learning method,and hill-climbing algorithm and Bootstrap model were used to increase the robustness of the model. A directed acyclic graph containing the diseases,second hospitalization and death,were constructed,and then the causal relationship between these variables were found. Results and Conclusion: Bayesian network about the relationship between cardiovascular diseases and common chronic diseases was constructed,and the causal routine which had the most serious impact on death was coronary artery disease and hypertension→cerebral infarction→pulmonary infection→death.
Aim: To study the causal relationship between cardiovascular diseases and common chronic diseases using Bayesian network. Methods: The data of 2 752 patients( 1 382 males,1 370 females) with cardiovascular diseases were collected. Bayesian network was constructed based on scoring-search learning method,and hill-climbing algorithm and Bootstrap model were used to increase the robustness of the model. A directed acyclic graph containing the diseases,second hospitalization and death,were constructed,and then the causal relationship between these variables were found. Results and Conclusion: Bayesian network about the relationship between cardiovascular diseases and common chronic diseases was constructed,and the causal routine which had the most serious impact on death was coronary artery disease and hypertension→cerebral infarction→pulmonary infection→death.
引文
[1]陈伟伟,高润霖,刘力生,等.《中国心血管病报告2015》概要[J].中国循环杂志,2016,31(6):521
[2] HECKERMAN D,GEIGER D,CHICKERING DM. Learning Bayesian networks:the combination of knowledge and statistical data[J]. Machine Learning,1995,20(3):197
[3] MCNALLY RJ,MAIR P,MUGNO BL,et al. Co-morbid obsessive-compulsive disorder and depression:a Bayesian network approach[J]. Psychol Med,2017,47(7):1204
[4] DJEBBARI A,QUACKENBUSH J. Seeded Bayesian networks:constructing genetic networks from microarray data[J]. BMC Syst Biol,2008,2(1):57
[5] BENJAMIN EJ,LEVY D,VAZIRI SM,et al. Independent risk factors for atrial fibrillation in a population-based cohort:the framingham heart study[J]. JAMA,1994,271(11):840
[6] HEDELAND H,OSTBERG G,HOKFELT B. On the prevalence of adrenocortical adenomas in an autopsy material in relation to hypertension and diabetes[J]. Acta Med Scand,1968,184(3):211
[7] SACHS K,PEREZ O,PE'ER D,et al. Causal protein-signaling networks derived from multiparameter single-cell data[J]. Science,2005,308(5721):523
[8]钟女娟,宋咏梅,刘更生,等.中药经验要素贝叶斯网络模型构建及应用[J].山东大学学报(医学版),2012,50(2):157
[9] BERARTI MA,GOSHU AT. Learning Bayesian networks using heart failure data[J]. Cardiol Angiol,2015,4(2):43
[10]PALMERINI T,BENEDETTO U,BACCHI-REGGIANI L,et al. Mortality in patients treated with extended duration dual antiplatelet therapy after drug-eluting stent implantation:a pairwise and Bayesian network meta-analysis of randomised trials[J]. Lancet,2015,385(9985):2371
[11]ONISKO A,DRUZDZEL MJ. Impact of precision of Bayesian network parameters on accuracy of medical diagnostic systems[J]. Artif Intell Med,2013,57(3):197
[12]ONG IM,GLASNER JD,PAGE D. Modelling regulatory pathways in E. coli from time series expression profiles[J].Bioinformatics,2002,18(Suppl 1):S241
[2] HECKERMAN D,GEIGER D,CHICKERING DM. Learning Bayesian networks:the combination of knowledge and statistical data[J]. Machine Learning,1995,20(3):197
[3] MCNALLY RJ,MAIR P,MUGNO BL,et al. Co-morbid obsessive-compulsive disorder and depression:a Bayesian network approach[J]. Psychol Med,2017,47(7):1204
[4] DJEBBARI A,QUACKENBUSH J. Seeded Bayesian networks:constructing genetic networks from microarray data[J]. BMC Syst Biol,2008,2(1):57
[5] BENJAMIN EJ,LEVY D,VAZIRI SM,et al. Independent risk factors for atrial fibrillation in a population-based cohort:the framingham heart study[J]. JAMA,1994,271(11):840
[6] HEDELAND H,OSTBERG G,HOKFELT B. On the prevalence of adrenocortical adenomas in an autopsy material in relation to hypertension and diabetes[J]. Acta Med Scand,1968,184(3):211
[7] SACHS K,PEREZ O,PE'ER D,et al. Causal protein-signaling networks derived from multiparameter single-cell data[J]. Science,2005,308(5721):523
[8]钟女娟,宋咏梅,刘更生,等.中药经验要素贝叶斯网络模型构建及应用[J].山东大学学报(医学版),2012,50(2):157
[9] BERARTI MA,GOSHU AT. Learning Bayesian networks using heart failure data[J]. Cardiol Angiol,2015,4(2):43
[10]PALMERINI T,BENEDETTO U,BACCHI-REGGIANI L,et al. Mortality in patients treated with extended duration dual antiplatelet therapy after drug-eluting stent implantation:a pairwise and Bayesian network meta-analysis of randomised trials[J]. Lancet,2015,385(9985):2371
[11]ONISKO A,DRUZDZEL MJ. Impact of precision of Bayesian network parameters on accuracy of medical diagnostic systems[J]. Artif Intell Med,2013,57(3):197
[12]ONG IM,GLASNER JD,PAGE D. Modelling regulatory pathways in E. coli from time series expression profiles[J].Bioinformatics,2002,18(Suppl 1):S241