文献详情
The application of unsupervised deep learning in predictive models using electronic health records
文献类型:期刊
通讯作者:Tong, LP (reprint author), Advocate Aurora Hlth, 3075 Highland Pkwy, Downers Grove, IL 60515 USA.
期刊名称:BMC MEDICAL RESEARCH METHODOLOGY影响因子和分区
年:2020
卷:20
期:1
关键词:Autoencoder; LASSO; Enhanced Reg; Predictive model; Predictive performance; Important response-specific predictors
所属部门:统计学院;信息学院
摘要:Background The main goal of this study is to explore the use of features representing patient-level electronic health record (EHR) data, generated by the unsupervised deep learning algorithm autoencoder, in predictive modeling. Since autoencoder features are unsupervised, this paper focuses on their general lower-dimensional representation of EHR information in a wide variety of predictive tasks. Methods We compare the model with autoencoder features to traditional models: logistic model with le ...More
Background The main goal of this study is to explore the use of features representing patient-level electronic health record (EHR) data, generated by the unsupervised deep learning algorithm autoencoder, in predictive modeling. Since autoencoder features are unsupervised, this paper focuses on their general lower-dimensional representation of EHR information in a wide variety of predictive tasks. Methods We compare the model with autoencoder features to traditional models: logistic model with least absolute shrinkage and selection operator (LASSO) and Random Forest algorithm. In addition, we include a predictive model using a small subset of response-specific variables (Simple Reg) and a model combining these variables with features from autoencoder (Enhanced Reg). We performed the study first on simulated data that mimics real world EHR data and then on actual EHR data from eight Advocate hospitals. Results On simulated data with incorrect categories and missing data, the precision for autoencoder is 24.16% when fixing recall at 0.7, which is higher than Random Forest (23.61%) and lower than LASSO (25.32%). The precision is 20.92% in Simple Reg and improves to 24.89% in Enhanced Reg. When using real EHR data to predict the 30-day readmission rate, the precision of autoencoder is 19.04%, which again is higher than Random Forest (18.48%) and lower than LASSO (19.70%). The precisions for Simple Reg and Enhanced Reg are 18.70 and 19.69% respectively. That is, Enhanced Reg can have competitive prediction performance compared to LASSO. In addition, results show that Enhanced Reg usually relies on fewer features under the setting of simulations of this paper. Conclusions We conclude that autoencoder can create useful features representing the entire space of EHR data and which are applicable to a wide array of predictive tasks. Together with important response-specific predictors, we can derive efficient and robust predictive models with less labor in data extraction and model training. ...Hide
DOI:10.1186/s12874-020-00923-1
百度学术:The application of unsupervised deep learning in predictive models using electronic health records
语言:外文
人气指数:1
浏览次数:1
基金:China Scholarship CouncilChina Scholarship Council
作者其他论文
Enhanced Phosphate Removal from Water by Honeycomb-Like Microporous Lanthanum-Chitosan Magnetic Spheres.Cheng, Rong, Shen, Liang-Jie, Zhang, Ying-Ying, et al. .WATER. 2018, 10(11).
Microcystis aeruginosa Synergistically Facilitate the Photocatalytic Degradation of Tetracycline Hydrochloride and Cr(VI) on PAN/TiO2/Ag Nanofiber Mats.Wang, Lei, Zhang, Changbo, Cheng, Rong, et al. .CATALYSTS. 2018, 8(12).
Machine Learning to Instruct Single Crystal Growth by Flux Method.Yao, Tang-Shi, Tang, Cen-Yao, Yang, Meng, et al. .CHINESE PHYSICS LETTERS. 2019, 36(6).
Seeing Is Believing: Experimental Evidence on the Impact of Eyeglasses on Academic Performance, Aspirations, and Dropout among Junior High School Students in Rural China.Nie, Jingchun, Pang, Xiaopeng, Wang, Lei, et al. .ECONOMIC DEVELOPMENT AND CULTURAL CHANGE. 2020, 68(2), 335-355.
Feature screening for ultrahigh-dimensional additive logistic models.Wang, Lei, Ma, Xuejun, Zhang, Jingxiao,.JOURNAL OF STATISTICAL PLANNING AND INFERENCE. 2020, 205, 306-317.
删除或更新信息,请邮件至freekaoyan#163.com(#换成@)
The application of unsupervised deep learning in predictive models using electronic health records
本站小编 Free考研/2020-04-17
相关话题/文献 信息学院
《玉函山房辑佚书·连山》佚文还原及文献价值举隅
文献详情《玉函山房辑佚书·连山》佚文还原及文献价值举隅外文标题:RestorationofLostScriptsandExamplesofDocumentationValueonLianShanofTheEditingLostBookofYuhanshanfang文献类型:期刊期刊名称:中州学刊年: ...中国人民大学科研学术 本站小编 Free考研 2020-04-17增值税与公司财务行为:文献综述
文献详情增值税与公司财务行为:文献综述文献类型:期刊期刊名称:财会月刊年:2019期:4页码:148-155ISSN:1004-0994关键词:增值税;企业财务行为;治理角色;激励角色;信息角色所属部门:商学院链接地址:http://d.oldg.wanfangdata.com.cn/Periodi ...中国人民大学科研学术 本站小编 Free考研 2020-04-17我国珍贵少数民族档案文献遗产保护需求研究——基于对《中国档案文献遗产名录》的统计分析
文献详情我国珍贵少数民族档案文献遗产保护需求研究——基于对《中国档案文献遗产名录》的统计分析外文标题:ResearchonNeedsforProtectionofPreciousMinorityArchivesDocumentaryHeritageinChina—BasedontheStatisti ...中国人民大学科研学术 本站小编 Free考研 2020-04-17消费者碎片化行为:文献回顾与研究展望
文献详情消费者碎片化行为:文献回顾与研究展望外文标题:ActivityFragmentationofConsumers:LiteratureReview文献类型:期刊期刊名称:科学决策年:2019期:3页码:73-98ISSN:1006-4885关键词:消费者;碎片化行为;碎片化阅读;移动营销所属部 ...中国人民大学科研学术 本站小编 Free考研 2020-04-17日本中世汉文文献对汉语辞书编纂的价值——以《运步色叶集》为例(一)
文献详情日本中世汉文文献对汉语辞书编纂的价值——以《运步色叶集》为例(一)文献类型:期刊期刊名称:宁夏大学学报(人文社会科学版)年:2019卷:41期:1页码:16-18ISSN:1001-5744关键词:《运步色叶集》;《汉语大词典》;日本中世;辞书编纂所属部门:文学院链接地址:http://d. ...中国人民大学科研学术 本站小编 Free考研 2020-04-17互联网金融信用风险管理的文献综述
文献详情互联网金融信用风险管理的文献综述文献类型:期刊期刊名称:现代管理科学年:2019期:4页码:49-51ISSN:1007-368X关键词:互联网金融;信息识别;投资行为;信用风险管理;监管所属部门:财政金融学院链接地址:http://d.oldg.wanfangdata.com.cn/Per ...中国人民大学科研学术 本站小编 Free考研 2020-04-17股票回购信号假说的文献综述
文献详情股票回购信号假说的文献综述文献类型:期刊期刊名称:现代管理科学年:2019期:4页码:79-81ISSN:1007-368X关键词:信号假说;要约回购;公开市场回购所属部门:财政金融学院链接地址:http://d.oldg.wanfangdata.com.cn/Periodical_xdgl ...中国人民大学科研学术 本站小编 Free考研 2020-04-17农地确权的资源禀赋效应研究——一个文献综述
文献详情农地确权的资源禀赋效应研究——一个文献综述文献类型:期刊期刊名称:现代管理科学年:2019期:4页码:76-78ISSN:1007-368X关键词:农地确权;资源禀赋效应所属部门:财政金融学院链接地址:http://d.oldg.wanfangdata.com.cn/Periodical_x ...中国人民大学科研学术 本站小编 Free考研 2020-04-17房价对人力资本存量水平的影响:一个文献综述
文献详情房价对人力资本存量水平的影响:一个文献综述文献类型:期刊期刊名称:现代管理科学年:2019期:5页码:115-117ISSN:1007-368X关键词:房价;人力资本存量;生育水平;人力资本投资;劳动力流动所属部门:劳动人事学院链接地址:http://d.oldg.wanfangdata.c ...中国人民大学科研学术 本站小编 Free考研 2020-04-17董事多样性与公司治理国内外文献综述
文献详情董事多样性与公司治理国内外文献综述文献类型:期刊期刊名称:现代管理科学年:2019期:3页码:112-114ISSN:1007-368X关键词:董事;多样性;人口统计特征;人力资本;社会资本所属部门:财政金融学院链接地址:http://d.oldg.wanfangdata.com.cn/Pe ...中国人民大学科研学术 本站小编 Free考研 2020-04-17