基于电子病历的临床医疗大数据挖掘流程与方法

doi:10.11959/j.issn.2096-0271.2017054

Abstract

Abstract:

Electronic medical records from hospitals record the patient's disease,diagnosis and treatment information.It forms the basis of clinical data.Mining such data can assist doctors in clinical research and clinical diagnosis and treatment.Firstly,challenges encountered in the process of big data mining on EMR were raised,then the core process was summarized.The process includes tasks such as clinical data integration,the construction of clinical specialist disease database based on knowledge graph,the quality assessment methods on EMR,and comparative effectiveness and risk prediction of diseases as the core of clinical big data applications.A solution for each task was proposed,and the experimental results were given.Finally,the future directions of technologies and applications of big data mining on healthcare were presented.

Key words: medical knowledge graph, clinical specialist disease database, evaluation of data quality, electronic medical record, risk prediction of diseases, comparative effectireness

CLC Number:

TP311.13

Tong RUAN, Ju GAO, Donglei FENG, Xiyuan QIAN, Ting WANG, Chenglin SUN. Process and methods of clinical big data mining based on electronic medical records[J]. Big Data Research, 2017, 3(5): 83-98.

Figures/Tables 8

References 33

[1]	王茜 . 英国大数据战略分析[J]. 全球科技经济瞭望, 2013(8): 24-27.
	WANG X . British state strategy of developing big data[J]. Global Science,Technology and Economy Outlook, 2013(8): 24-27.
[2]	王忠 . 美国推动大数据技术发展的战略价值及启示[J]. 中国发展观察, 2012(6): 44-45.
	WANG Z . The strategic value and enlightenment of promoting big data technology development in America[J]. China Development Observation, 2012(6): 44-45.
[3]	BROWN J S , HOLMES J H , SHAH K ,et al. Distributed health data networks:a practical and preferred approach to multiinstitutional evaluations of comparative effectiveness,safety,and quality of care[J]. Med Care, 2010,48(6): 45-51.
[4]	SHERMAN R E , ANDERSON S A , DALPAN G J ,et al. Real-world evidencewhat is it and what can it tell us[J]. New England Journal of Medicine, 2016,375(23):2293.
[5]	董景五 . 疾病和有关健康问题的国际统计分类第十次修订本（ICD-10）[M]. 北京: 人民卫生出版社, 1996.
	DONG J W . The international statistical classification of diseases and related health problems 10th revision[M]. Beijing: People’s Medical Publishing HousePress, 1996.
[6]	BODENREIDER O . The unified medical language system (UMLS):integrating biomedical terminology[J]. Nucleic Acids Research, 2004,32(suppl 1): D267-D270.
[7]	WEINSTEIN J N , COLLISSON E A , MILLS G B ,et al. The cancer genome atlas pan-cancer analysis project[J]. Nature Genetics, 2013,45(10): 1113-1120.
[8]	SAMWALD M , JENTZSCH A , BOUTON C ,et al. Linked open drug data for pharmaceutical research and development[J]. Journal of Cheminformatics, 2011,3(1):19.
[9]	BELLEAU F , NOLIN M A , TOURIGNY N ,et al. Bio2RDF:towards a mashup to build bioinformatics knowledge systems[J]. Journal of Biomedical Informatics, 2008,41(5): 706-716.
[10]	HEARST M A , . Automatic acquisition of hyponyms from large text corpora[C]// The 14th Conference on Computational Linguistics,August 23-28,1992,Nantes,France. New York:ACM Press, 1992: 539-545.
[11]	DALKEY N C , ROURKE D L . Experimental assessment of Delphi procedures with group value judgements:advanced research projects agency[J]. Cluster Analysis, 1971:58.
[12]	CHENG Y , WANG F , ZHANG P ,et al. Risk prediction with electronic health records:a deep learning approach[C]// The 2016 SIAM International Conference on Data Mining,May 5-7,2016,Miami,USA.[S.l.:s.n.], 2016: 432-440.
[13]	SUTHERLAND S M , CHAWLA L S,KANE-GILL S L ,et al. Utilizing electronic health records to predict acute kidney injury risk and outcomes:workgroup statements from the 15th,ADQI consensus conference[J]. Canadian Journal of Kidney Health ＆ Disease, 2016,3(1): 1-14.
[14]	WOLFSON J , BANDYOPADHYAY S , ELIDRISI M ,et al. A naive Bayes machine learning approach to risk prediction using censored,time-to-event data[J]. Statistics in Medicine, 2014,34(21): 2941-2957.
[15]	马宗帅 . 基于深度学习的心脑血管疾病预测方法研究[D]. 西安:西安建筑科技大学, 2015.
	MA Z S . Research on cardiovascular disease prediction based on deep learning technical[D]. Xi’an:Xi’an University of Architecture and Technology, 2015.
[16]	AULI M , GALLEY M , QUIRK C ,et al. Joint language and translation modeling with recurrent neural networks[J]. American Journal of Psychoanalysis, 2013,74(2): 212-213.
[17]	RUFFINI G , IBA?EZ D , CASTELLANO M ,et al. EEG-driven RNN classification for prognosis of neurodegeneration in at-risk patients[C]// International Conference on Artificial Neural Networks,September 6-9,2016,Barcelona,Spain. Berlin:Springer, 2016: 306-313.
[18]	MIOTTO R , LI L , DUDLEY J T . Deep learning to predict patient future diseases from the electronic health records[M]. Berlin: Springer International PublishingPress, 2016.
[19]	LIU L , TANG J , CHENG Y ,et al. Mining diabetes complication and treatment patterns for clinical decision support[C]// The 22nd ACM international conference on Information ＆ Knowledge Management,October 27-November 1,2013,San Francisco,USA. New York:ACM Press, 2013: 279-288.
[20]	HUANG Z , DONG W , BATH P ,et al. On mining latent treatment patterns from electronic medical records[J]. Data Mining＆ Knowledge Discovery, 2015,29(4): 1-36.
[21]	CHENG Y , WANG F , ZHANG P ,et al. Risk prediction with electronic health records:a deep learning approach[C]// The 2016 SIAM International Conference on Data Mining,May 5-7,2016,Miami,USA.[S.l.:s.n. ], 2016: 432-440.
[22]	SUTHERLAND S M , CHAWLA L S , KANEGILL S L ,et al. Utilizing electronichealth records to predict acute kidney injury risk and outcomes:workgroup statements from the 15th ADQI consensus conference[J]. Canadian Journal of Kidney Health and Disease, 2016,3(1): 1-14.
[23]	WOLFSON J , BANDYOPADHYAY S , ELIDRISI M ,et al. A naive Bayes machine learning approach to risk prediction using censored,time-to-event data[J]. Statistics in Medicine, 2011,34(21): 2941-2957.
[24]	SUTSKEVER I , VINYALS O , LE Q V . Sequence to sequence learning with neural networks[C]// The 27th International Conference on Neural Information Processing Systems,December 8-13,2014,Montreal,Canada. New York:ACM Press, 2014: 3104-3112.
[25]	ZHANG J , MANI I . kNN approach to unbalanced data distributions:a case study involving information extraction[C]// The ICML 2003 Workshop on Learning from Imbalanced Datasets,December 3-8,2003,Piscataway,USA.[S.l.:s.n.], 2003.
[26]	TOMEK I . Two modifications of CNN[J]. IEEE Transactions on Systems Man and Communications, 1976,SMC-6(11): 769-772.
[27]	KUBAT M , MATWIN S . Addressing the course of imbalanced training sets:onesided selection[C]// The 14th International Conference on Machine Learning (ICML 1997),July 8-12,1997,Nashville,USA.[S.l.:s.n.],. 1997: 179-186.
[28]	WILSOND L . Asymptotic properties of nearest neighbor rules using edited data[J]. IEEE Transactions on Systems,Man,and Communications, 2007,SMC-2(3): 408-421.
[29]	LAURIKKALA J , . Improving identification of difficult small classes by balancing class distribution[C]// Conference on Artificial Intelligence in Medicine in Europe,July 1-4,2001,Cascais,Portugal. Berlin:Springer Berlin Heidelberg, 2001: 63-66.
[30]	CHAWLAN V , BOWYER K W , HALL L O ,et al. SMOTE:synthetic minority oversampling technique[J]. Journal of Artificial Intelligence Research, 2002(16): 321-357.
[31]	ROSENBAUM P R , RUBIN D B . The central role of the propensity score in observational studies for causal effects[J]. Biometrika, 1983,70(1): 41-55.
[32]	SOBEL M E , . Causal inference in the social and behavioral sciences[M]// Handbook of Statistical Modeling for the Social and Behavioral Sciences. New York:Springer US, 1995: 1-38.
[33]	HOLLAND P W . Statistics and causal inference[J]. Journal of the American Statistical Association, 1986,81(396): 945-960.

Metrics

Recommended 0

No Suggested Reading articles found!

名称	描述	示例（粗体表示）
原子症状	不可拆分的症状	水肿、麻木、抽筋、疼、痛
连词	表示并列或选择关系的词语	尿频伴尿急和尿痛、三角肌反射减弱或消失
否定词	表示不存在某种事物的词语	产后无乳汁分泌、双肺未闻及明显啰音
存在词	表示存在某种事物的词语	流清涕、胸腔内出现积液、抽搐发作
程度词	形容症状严重程度或出现频率的词语	轻微创伤、剧烈疼痛、少量腹水、偶有自言自语
发展词	形容症状发展状况的词语	气喘加重、失眠缓解
能够词	表示具有某种能力的词语	上肢可抬举、颈部能够前屈
不能词	表示不具有某种能力的词语	难以入睡、呼吸困难、颈部无法仰伸
动作词	表示特定动作的词语	上肢可抬举、颈部无法仰伸
情景限定词	表示某种特定情景的词语	产后腹痛、进食后心绞痛、行走时踩棉花感
方位词	表示方位的词语	左心室肥厚、小腿后侧感觉障碍、双肺水泡音
部位词	表示身体部位的词语	手部肿块、脐带过长、足部畸形
中心词	除身体部位外症状描述主体	血压升高、呼吸音减弱、左上腹肿块
感觉词	描述感觉的词语	髌前空虚感、肛门坠胀感、喉部灼热感
特征词	表示事物特征的词语	米汤样大便、蚯蚓状肿物、压迫性头痛
修饰词	其他修饰词	呼吸急促、面色泛黄、体重下降

信息类别	字段	评估度量
人口学信息	性别	完整性
		一致性
	年龄	完整性
诊断信息	疾病名称	完整性
		准确性
	诊断编码	完整性
		正确性
	诊断编码与疾病名称	不一致
	诊断	完整性
		精确性
	诊断日期	完整性
	整体	正确性
体征信息	心律	完整性
	血压	完整性
基础信息	病史	文本抽取复杂度
	家族史	文本抽取复杂度
检验信息	至少包含一项检验指标	完整性
自定义数据表	病人记录密度	1次住院信息
		2～3次住院信息
		4～5次住院信息
		5次以上信息
	临床事件	完整性
用药信息	西药	完整性
		治疗心衰药物
	中成药	完整性
疗效数据	入院时间	完整性
	出院时间	完整性
	出/入院时间	准确性
	再入院率	30天再入院率

疾病名	deep	PCA	K-means	GMM	hand	count	percent	sample
高血压	0.770 8	0.545 2	0.536 5	0.652	0.653 7	7 097	69.43%	N
糖尿病	0.660 2	0.618 5	0.631 1	0.627	0.626 8	3 674	35.94%	N
冠心病	0.745 8	0.600 7	0.617 1	0.741	0.740 7	5 072	49.62%	N
房颤	0.521 8	0.535	0.403 9	0.645	0.644 4	3 053	29.87%	Y
慢性肾功能不全	0.726 7	0.372 9	0.565 5	0.699 1	0.698 3	896	8.77%	Y
心脏瓣膜病	0.841 9	0.258	0.5	0.882	0.902 2	80	0.78%	Y
扩张性心肌病	0.776 7	0.415 7	0.439 5	0.674 5	0.674 4	321	3.14%	Y
肥厚性心肌病	0.814 2	0.221 7	0.396 1	0.437 8	0.437 4	146	1.43%	Y
慢性阻塞性肺疾病	0.546 6	0.576 6	0.45 7	0.522 2	0.522 1	818	8.00%	Y
脑梗塞/一过性脑缺血	0.739 2	0.697 2	0.762 3	0.871 7	0.873 4	2 579	25.23%	Y

变量名称	逻辑回归系数	P值	是否具有显著性
高血压	-0.431 987	0.000 009 6	是
糖尿病	0.328 216	0.000 007 42	是
冠心病	0.248 723	0.000 744	是
房颤	-0.243 055	0.001 509	是
慢性肾功能不全	0.420 103	0.000 586	是
心脏瓣膜疾病	-0.09 113	0.807 312	否
扩张性心肌病	-0.465 246	0.018 189	是
肥厚性心肌病	0.446 251	0.060 175	否
慢性阻塞性肺疾病	0.005 168	0.970 185	否
脑梗塞或一过性脑缺血	0.256 523	0.002 834	是

Process and methods of clinical big data mining based on electronic medical records

RichHTML

PDF下载

Knowledge

Cited

Abstract

Cite this article

share this article

Figures/Tables 8

References 33

Related Articles 3

Metrics

Recommended 0

[1]	Ling TAN, Haihong E, Zemin KUANG, Meina SONG, Yu LIU, Zhengyu CHEN, Xiaoxuan XIE, Jundi LI, Jiawei FAN, Qingchuan WANG, Xiaoyang KANG. Key technologies and research progress of medical knowledge graph construction [J]. Big Data Research, 2021, 7(4): 80-104.
[2]	Liangchen XU, Chonghui GUO. Analysis and utilization framework of electronic medical records under the background of smart hospital construction [J]. Big Data Research, 2021, 7(4): 141-156.
[3]	Xiaoyuan BAO, Kai ZHANG, Meng JIN, Shuanglian XIE, Kai SONG. A data-space based platform for the integration and application of electronic health records [J]. Big Data Research, 2019, 5(6): 47-61.