{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T18:05:17Z","timestamp":1784829917156,"version":"3.55.0"},"reference-count":36,"publisher":"Oxford University Press (OUP)","issue":"6","license":[{"start":{"date-parts":[[2017,11,30]],"date-time":"2017-11-30T00:00:00Z","timestamp":1512000000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/academic.oup.com\/journals\/pages\/about_us\/legal\/notices"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2018,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Objective<\/jats:title><jats:p>A key challenge in clinical data mining is that most clinical datasets contain missing data. Since many commonly used machine learning algorithms require complete datasets (no missing data), clinical analytic approaches often entail an imputation procedure to \u201cfill in\u201d missing data. However, although most clinical datasets contain a temporal component, most commonly used imputation methods do not adequately accommodate longitudinal time-based data. We sought to develop a new imputation algorithm, 3-dimensional multiple imputation with chained equations (3D-MICE), that can perform accurate imputation of missing clinical time series data.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We extracted clinical laboratory test results for 13 commonly measured analytes (clinical laboratory tests). We imputed missing test results for the 13 analytes using 3 imputation methods: multiple imputation with chained equations (MICE), Gaussian process (GP), and 3D-MICE. 3D-MICE utilizes both MICE and GP imputation to integrate cross-sectional and longitudinal information. To evaluate imputation method performance, we randomly masked selected test results and imputed these masked results alongside results missing from our original data. We compared predicted results to measured results for masked data points.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>3D-MICE performed significantly better than MICE and GP-based imputation in a composite of all 13 analytes, predicting missing results with a normalized root-mean-square error of 0.342, compared to 0.373 for MICE alone and 0.358 for GP alone.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>3D-MICE offers a novel and practical approach to imputing clinical laboratory time series data. 3D-MICE may provide an additional tool for use as a foundation in clinical predictive analytics and intelligent clinical decision support.<\/jats:p><\/jats:sec>","DOI":"10.1093\/jamia\/ocx133","type":"journal-article","created":{"date-parts":[[2017,10,27]],"date-time":"2017-10-27T19:11:17Z","timestamp":1509131477000},"page":"645-653","source":"Crossref","is-referenced-by-count":87,"title":["3D-MICE: integration of cross-sectional and longitudinal imputation for multi-analyte longitudinal clinical data"],"prefix":"10.1093","volume":"25","author":[{"given":"Yuan","family":"Luo","sequence":"first","affiliation":[{"name":"Department of Preventive Medicine, Northwestern University, Chicago, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peter","family":"Szolovits","sequence":"additional","affiliation":[{"name":"Computer Science and Artificial Intelligence Lab, Massachusetts Institute of Technology, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anand S","family":"Dighe","sequence":"additional","affiliation":[{"name":"Department of Pathology, Massachusetts General Hospital, Boston, MA, USA"},{"name":"Harvard Medical School, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jason M","family":"Baron","sequence":"additional","affiliation":[{"name":"Department of Pathology, Massachusetts General Hospital, Boston, MA, USA"},{"name":"Harvard Medical School, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2017,11,30]]},"reference":[{"issue":"158","key":"2020110612380719100_ocx133-B1","doi-asserted-by":"crossref","first-page":"158rv11","DOI":"10.1126\/scitranslmed.3003528","article-title":"Computational medicine: translating models to clinical care","volume":"4","author":"Winslow","year":"2012","journal-title":"Sci Translational Med."},{"issue":"6243","key":"2020110612380719100_ocx133-B2","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1126\/science.aab1328","article-title":"Ten things we have to do to achieve precision medicine","volume":"349","author":"Kohane","year":"2015","journal-title":"Science."},{"issue":"8","key":"2020110612380719100_ocx133-B3","doi-asserted-by":"crossref","first-page":"e002847","DOI":"10.1136\/bmjopen-2013-002847","article-title":"Comparison of imputation methods for missing laboratory data in medicine","volume":"3","author":"Waljee","year":"2013","journal-title":"BMJ Open."},{"issue":"6","key":"2020110612380719100_ocx133-B4","doi-asserted-by":"crossref","first-page":"1134","DOI":"10.1093\/jamia\/ocx071","article-title":"Biases introduced by filtering electronic health records for patients with \u201ccomplete data.\u201d","volume":"24","author":"Weber","year":"2017","journal-title":"J Am Med Inform Assoc."},{"issue":"11","key":"2020110612380719100_ocx133-B5","doi-asserted-by":"crossref","first-page":"2370","DOI":"10.1002\/sim.2715","article-title":"Multiple imputation for the comparison of two screening tests in two-phase Alzheimer studies","volume":"26","author":"Harel","year":"2007","journal-title":"Stat Med."},{"issue":"25","key":"2020110612380719100_ocx133-B6","doi-asserted-by":"crossref","first-page":"2592","DOI":"10.1002\/sim.4016","article-title":"A comparison of multiple imputation and fully augmented weighted estimators for Cox regression with missing covariates","volume":"29","author":"Qi","year":"2010","journal-title":"Stat Med."},{"issue":"3","key":"2020110612380719100_ocx133-B7","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v045.i03","article-title":"mice: Multivariate imputation by chained equations in R","volume":"45","author":"Buuren","year":"2011","journal-title":"J Stat Software."},{"issue":"1","key":"2020110612380719100_ocx133-B8","doi-asserted-by":"crossref","first-page":"112","DOI":"10.1093\/bioinformatics\/btr597","article-title":"MissForest: non-parametric missing value imputation for mixed-type data","volume":"28","author":"Stekhoven","year":"2012","journal-title":"Bioinformatics."},{"key":"2020110612380719100_ocx133-B9","volume-title":"Imputing Missing Data for Gene Expression Arrays","author":"Hastie","year":"1999"},{"issue":"1","key":"2020110612380719100_ocx133-B10","first-page":"85","article-title":"A multivariate technique for multiply imputing missing values using a sequence of regression models","volume":"27","author":"Raghunathan","year":"2001","journal-title":"Survey Methodol."},{"issue":"2","key":"2020110612380719100_ocx133-B11","doi-asserted-by":"crossref","first-page":"1","DOI":"10.18637\/jss.v045.i02","article-title":"Multiple imputation with diagnostics (mi) in R: opening windows into the black box","volume":"45","author":"Su","year":"2011","journal-title":"J Stat Software."},{"issue":"20","key":"2020110612380719100_ocx133-B12","doi-asserted-by":"crossref","first-page":"3503","DOI":"10.1002\/sim.2452","article-title":"Survival analysis using auxiliary variables via non-parametric multiple imputation","volume":"25","author":"Hsu","year":"2006","journal-title":"Stat Med."},{"key":"2020110612380719100_ocx133-B13","first-page":"949","article-title":"Robust likelihood-based analysis of multivariate data with missing values","volume":"14","author":"Little","year":"2004","journal-title":"Statistica Sinica."},{"key":"2020110612380719100_ocx133-B14","doi-asserted-by":"crossref","first-page":"149","DOI":"10.5705\/ss.2010.069","article-title":"Doubly robust nonparametric multiple imputation for ignorable missing data","volume":"22","author":"Long","year":"2012","journal-title":"Statistica Sinica."},{"issue":"6","key":"2020110612380719100_ocx133-B15","doi-asserted-by":"crossref","first-page":"778","DOI":"10.1093\/ajcp\/aqw064","article-title":"Using machine learning to predict laboratory test results","volume":"145","author":"Luo","year":"2016","journal-title":"Am J Clin Pathol."},{"issue":"3","key":"2020110612380719100_ocx133-B16","doi-asserted-by":"crossref","first-page":"911","DOI":"10.1111\/j.1541-0420.2008.01155.x","article-title":"Extensions of the penalized spline of propensity prediction method of imputation","volume":"65","author":"Zhang","year":"2009","journal-title":"Biometrics."},{"issue":"6","key":"2020110612380719100_ocx133-B17","doi-asserted-by":"crossref","first-page":"681","DOI":"10.1002\/(SICI)1097-0258(19990330)18:6<681::AID-SIM71>3.0.CO;2-R","article-title":"Multiple imputation of missing blood pressure covariates in survival analysis","volume":"18","author":"Van Buuren","year":"1999","journal-title":"Stats Med."},{"issue":"6","key":"2020110612380719100_ocx133-B18","doi-asserted-by":"crossref","first-page":"520","DOI":"10.1093\/bioinformatics\/17.6.520","article-title":"Missing value estimation methods for DNA microarrays","volume":"17","author":"Troyanskaya","year":"2001","journal-title":"Bioinformatics."},{"key":"2020110612380719100_ocx133-B19","doi-asserted-by":"crossref","first-page":"21689","DOI":"10.1038\/srep21689","article-title":"Multiple imputation for general missing data patterns in the presence of high-dimensional data","volume":"6","author":"Deng","year":"2016","journal-title":"Sci Rep."},{"issue":"1","key":"2020110612380719100_ocx133-B20","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1198\/000313007X172556","article-title":"Much ado about nothing: a comparison of missing data methods and software to fit incomplete data regression models","volume":"61","author":"Horton","year":"2007","journal-title":"Am Stat."},{"issue":"10","key":"2020110612380719100_ocx133-B21","doi-asserted-by":"crossref","first-page":"1137","DOI":"10.1002\/sim.4201","article-title":"A functional multiple imputation approach to incomplete longitudinal data","volume":"30","author":"He","year":"2011","journal-title":"Stats Med."},{"key":"2020110612380719100_ocx133-B22","article-title":"Fast multivariate spatio-temporal analysis via low rank tensor learning","year":"2014","journal-title":"Adv Neural Inf Process Syst."},{"key":"2020110612380719100_ocx133-B23","year":"2015"},{"key":"2020110612380719100_ocx133-B24","first-page":"1493","article-title":"Uncovering the spatio-temporal dynamics of memes in the presence of incomplete information","volume-title":"Proceedings of the 25th ACM International Conference on Information and Knowledge Management","author":"Ge","year":"2016"},{"issue":"2","key":"2020110612380719100_ocx133-B25","first-page":"106","article-title":"A functional data approach to missing value imputation and outlier detection for traffic flow data","volume":"2","author":"Chiou","year":"2014","journal-title":"Transportmetrica B."},{"issue":"18","key":"2020110612380719100_ocx133-B26","doi-asserted-by":"crossref","first-page":"3130","DOI":"10.1002\/sim.6166","article-title":"A Bayesian approach to functional mixed-effects modeling for longitudinal data with binomial outcomes","volume":"33","author":"Kliethermes","year":"2014","journal-title":"Stats Med."},{"issue":"4","key":"2020110612380719100_ocx133-B27","doi-asserted-by":"crossref","first-page":"1157","DOI":"10.1111\/j.0006-341X.2000.01157.x","article-title":"Multiple imputation and posterior simulation for multivariate missing data in longitudinal studies","volume":"56","author":"Liu","year":"2000","journal-title":"Biometrics."},{"issue":"2","key":"2020110612380719100_ocx133-B28","doi-asserted-by":"crossref","first-page":"437","DOI":"10.1198\/106186002760180608","article-title":"Computational strategies for multivariate linear mixed-effects models with missing values","volume":"11","author":"Schafer","year":"2002","journal-title":"J Comput Graph Stat."},{"issue":"11","key":"2020110612380719100_ocx133-B29","doi-asserted-by":"crossref","first-page":"2101","DOI":"10.1007\/s00122-016-2760-9","article-title":"Multi-task Gaussian process for imputing missing data in multi-trait and multi-environment trials","volume":"129","author":"Hori","year":"2016","journal-title":"Theor Appl Genet."},{"issue":"3","key":"2020110612380719100_ocx133-B30","doi-asserted-by":"crossref","first-page":"455","DOI":"10.1137\/07070111X","article-title":"Tensor decompositions and applications","volume":"51","author":"Kolda","year":"2009","journal-title":"SIAM Rev."},{"key":"2020110612380719100_ocx133-B31","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1007\/978-3-540-28650-9_4","article-title":"Gaussian processes in machine learning","volume-title":"Advanced Lectures on Machine Learning","author":"Rasmussen","year":"2004"},{"key":"2020110612380719100_ocx133-B32","author":"Bonilla","year":"2007"},{"key":"2020110612380719100_ocx133-B33","doi-asserted-by":"crossref","first-page":"1012","DOI":"10.1145\/1102351.1102479","volume-title":"Proceedings of the 22nd International Conference on Machine Learning","author":"Yu","year":"2005"},{"key":"2020110612380719100_ocx133-B34","article-title":"GPfit: an R package for Gaussian process model fitting using a new optimization algorithm","author":"MacDonald","year":"2013","journal-title":"arXiv preprint arXiv:1305.0759."},{"issue":"1","key":"2020110612380719100_ocx133-B35","doi-asserted-by":"crossref","first-page":"40","DOI":"10.1002\/mpr.329","article-title":"Multiple imputation by chained equations: what is it and how does it work?","volume":"20","author":"Azur","year":"2011","journal-title":"Int J Methods Psychiatr Res."},{"issue":"4","key":"2020110612380719100_ocx133-B36","first-page":"1","article-title":"gplots: various R programming tools for plotting data","volume":"2","author":"Warnes","year":"2009","journal-title":"R Package Version."}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/25\/6\/645\/34150454\/ocx133.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/academic.oup.com\/jamia\/article-pdf\/25\/6\/645\/34150454\/ocx133.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,8,5]],"date-time":"2022-08-05T18:55:49Z","timestamp":1659725749000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/25\/6\/645\/4675256"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,11,30]]},"references-count":36,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2017,11,30]]},"published-print":{"date-parts":[[2018,6,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocx133","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"value":"1067-5027","type":"print"},{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2018,6]]},"published":{"date-parts":[[2017,11,30]]}}}