Literature DB >> 29202205

3D-MICE: integration of cross-sectional and longitudinal imputation for multi-analyte longitudinal clinical data.

Yuan Luo1, Peter Szolovits2, Anand S Dighe3,4, Jason M Baron3,4.   

Abstract

Objective: 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 "fill in" 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.
Methods: 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.
Results: 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. Conclusions: 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.

Entities:  

Mesh:

Year:  2018        PMID: 29202205      PMCID: PMC7646951          DOI: 10.1093/jamia/ocx133

Source DB:  PubMed          Journal:  J Am Med Inform Assoc        ISSN: 1067-5027            Impact factor:   4.497


  20 in total

1.  Multiple imputation of missing blood pressure covariates in survival analysis.

Authors:  S van Buuren; H C Boshuizen; D L Knook
Journal:  Stat Med       Date:  1999-03-30       Impact factor: 2.373

2.  Missing value estimation methods for DNA microarrays.

Authors:  O Troyanskaya; M Cantor; G Sherlock; P Brown; T Hastie; R Tibshirani; D Botstein; R B Altman
Journal:  Bioinformatics       Date:  2001-06       Impact factor: 6.937

3.  Much ado about nothing: A comparison of missing data methods and software to fit incomplete data regression models.

Authors:  Nicholas J Horton; Ken P Kleinman
Journal:  Am Stat       Date:  2007-02       Impact factor: 8.710

4.  Multiple imputation by chained equations: what is it and how does it work?

Authors:  Melissa J Azur; Elizabeth A Stuart; Constantine Frangakis; Philip J Leaf
Journal:  Int J Methods Psychiatr Res       Date:  2011-03       Impact factor: 4.035

5.  Extensions of the penalized spline of propensity prediction method of imputation.

Authors:  Guangyu Zhang; Roderick Little
Journal:  Biometrics       Date:  2008-11-13       Impact factor: 2.571

6.  Using Machine Learning to Predict Laboratory Test Results.

Authors:  Yuan Luo; Peter Szolovits; Anand S Dighe; Jason M Baron
Journal:  Am J Clin Pathol       Date:  2016-06-21       Impact factor: 2.493

7.  Multi-task Gaussian process for imputing missing data in multi-trait and multi-environment trials.

Authors:  Tomoaki Hori; David Montcho; Clement Agbangla; Kaworu Ebana; Koichi Futakuchi; Hiroyoshi Iwata
Journal:  Theor Appl Genet       Date:  2016-08-19       Impact factor: 5.699

8.  A functional multiple imputation approach to incomplete longitudinal data.

Authors:  Yulei He; Recai Yucel; Trivellore E Raghunathan
Journal:  Stat Med       Date:  2011-02-22       Impact factor: 2.497

9.  Comparison of imputation methods for missing laboratory data in medicine.

Authors:  Akbar K Waljee; Ashin Mukherjee; Amit G Singal; Yiwei Zhang; Jeffrey Warren; Ulysses Balis; Jorge Marrero; Ji Zhu; Peter Dr Higgins
Journal:  BMJ Open       Date:  2013-08-01       Impact factor: 2.692

10.  Multiple Imputation for General Missing Data Patterns in the Presence of High-dimensional Data.

Authors:  Yi Deng; Changgee Chang; Moges Seyoum Ido; Qi Long
Journal:  Sci Rep       Date:  2016-02-12       Impact factor: 4.379

View more
  24 in total

1.  Supervised Nonnegative Matrix Factorization to Predict ICU Mortality Risk.

Authors:  Guoqing Chao; Chengsheng Mao; Fei Wang; Yuan Zhao; Yuan Luo
Journal:  Proceedings (IEEE Int Conf Bioinformatics Biomed)       Date:  2019-01-24

2.  Privacy-Preserving Methods for Vertically Partitioned Incomplete Data.

Authors:  Yi Deng; Xiaoqian Jiang; Qi Long
Journal:  AMIA Annu Symp Proc       Date:  2021-01-25

3.  Evaluating the state of the art in missing data imputation for clinical data.

Authors:  Yuan Luo
Journal:  Brief Bioinform       Date:  2022-01-17       Impact factor: 11.622

Review 4.  Advances in Machine Learning Approaches to Heart Failure with Preserved Ejection Fraction.

Authors:  Faraz S Ahmad; Yuan Luo; Ramsey M Wehbe; James D Thomas; Sanjiv J Shah
Journal:  Heart Fail Clin       Date:  2022-03-04       Impact factor: 3.179

5.  A Multi-directional Approach for Missing Value Estimation in Multivariate Time Series Clinical Data.

Authors:  Xiao Xu; Xiaoshuang Liu; Yanni Kang; Xian Xu; Junmei Wang; Yuyao Sun; Quanhe Chen; Xiaoyu Jia; Xinyue Ma; Xiaoyan Meng; Xiang Li; Guotong Xie
Journal:  J Healthc Inform Res       Date:  2020-06-04

6.  Predictive modeling of bacterial infections and antibiotic therapy needs in critically ill adults.

Authors:  Garrett Eickelberg; L Nelson Sanchez-Pinto; Yuan Luo
Journal:  J Biomed Inform       Date:  2020-08-16       Impact factor: 6.317

Review 7.  Natural Language Processing for EHR-Based Computational Phenotyping.

Authors:  Zexian Zeng; Yu Deng; Xiaoyu Li; Tristan Naumann; Yuan Luo
Journal:  IEEE/ACM Trans Comput Biol Bioinform       Date:  2018-06-25       Impact factor: 3.710

8.  Are My EHRs Private Enough? Event-Level Privacy Protection.

Authors:  Chengsheng Mao; Yuan Zhao; Mengxin Sun; Yuan Luo
Journal:  IEEE/ACM Trans Comput Biol Bioinform       Date:  2018-06-25       Impact factor: 3.710

9.  Predicting Missing Values in Medical Data via XGBoost Regression.

Authors:  Xinmeng Zhang; Chao Yan; Cheng Gao; Bradley A Malin; You Chen
Journal:  J Healthc Inform Res       Date:  2020-08-03

10.  MedGCN: Medication recommendation and lab test imputation via graph convolutional networks.

Authors:  Chengsheng Mao; Liang Yao; Yuan Luo
Journal:  J Biomed Inform       Date:  2022-01-29       Impact factor: 6.317

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.