PMID- 31610264
OWN - NLM
STAT- MEDLINE
DCOM- 20201014
LR  - 20201014
IS  - 1532-0480 (Electronic)
IS  - 1532-0464 (Linking)
VI  - 100
DP  - 2019 Dec
TI  - Deep representation learning for individualized treatment effect estimation using 
      electronic health records.
PG  - 103303
LID - S1532-0464(19)30222-9 [pii]
LID - 10.1016/j.jbi.2019.103303 [doi]
AB  - Utilizing clinical observational data to estimate individualized treatment 
      effects (ITE) is a challenging task, as confounding inevitably exists in clinical 
      data. Most of the existing models for ITE estimation tackle this problem by 
      creating unbiased estimators of the treatment effects. Although valuable, 
      learning a balanced representation is sometimes directly opposed to the objective 
      of learning an effective and discriminative model for ITE estimation. We propose 
      a novel hybrid model bridging multi-task deep learning and K-nearest neighbors 
      (KNN) for ITE estimation. In detail, the proposed model firstly adopts multi-task 
      deep learning to extract both outcome-predictive and treatment-specific latent 
      representations from Electronic Health Records (EHR), by jointly performing the 
      outcome prediction and treatment category classification. Thereafter, we estimate 
      counterfactual outcomes by KNN based on the learned hidden representations. We 
      validate the proposed model on a widely used semi-simulated dataset, i.e. IHDP, 
      and a real-world clinical dataset consisting of 736 heart failure (HF) patients. 
      The performance of our model remains robust and reaches 1.7 and 0.23 in terms of 
      Precision in the estimation of heterogeneous effect (PEHE) and average treatment 
      effect (ATE), respectively, on IHDP dataset, and 0.703 and 0.796 in terms of 
      accuracy and F1 score respectively, on HF dataset. The results demonstrate that 
      the proposed model achieves competitive performance over state-of-the-art models. 
      In addition, the results reveal several findings which are consistent with 
      existing medical domain knowledge, and discover certain suggestive hypotheses 
      that could be validated through further investigations in the clinical domain.
CI  - Copyright (c) 2019 Elsevier Inc. All rights reserved.
FAU - Chen, Peipei
AU  - Chen P
AD  - College of Biomedical Engineering and Instrumental Science, Zhejiang University, 
      310008 Hangzhou, China; School of Industrial Engineering, Eindhoven University of 
      Technology, Eindhoven, the Netherlands.
FAU - Dong, Wei
AU  - Dong W
AD  - Department of Cardiology, Chinese PLA General Hospital, 100853 Beijing, China.
FAU - Lu, Xudong
AU  - Lu X
AD  - College of Biomedical Engineering and Instrumental Science, Zhejiang University, 
      310008 Hangzhou, China; School of Industrial Engineering, Eindhoven University of 
      Technology, Eindhoven, the Netherlands.
FAU - Kaymak, Uzay
AU  - Kaymak U
AD  - School of Industrial Engineering, Eindhoven University of Technology, Eindhoven, 
      the Netherlands; College of Biomedical Engineering and Instrumental Science, 
      Zhejiang University, 310008 Hangzhou, China.
FAU - He, Kunlun
AU  - He K
AD  - Department of Cardiology, Chinese PLA General Hospital, 100853 Beijing, China. 
      Electronic address: kunlunhe@plagh.org.
FAU - Huang, Zhengxing
AU  - Huang Z
AD  - College of Biomedical Engineering and Instrumental Science, Zhejiang University, 
      310008 Hangzhou, China. Electronic address: zhengxing.h@gmail.com.
LA  - eng
PT  - Journal Article
PT  - Research Support, Non-U.S. Gov't
DEP - 20191011
PL  - United States
TA  - J Biomed Inform
JT  - Journal of biomedical informatics
JID - 100970413
SB  - IM
MH  - Algorithms
MH  - Datasets as Topic
MH  - *Deep Learning
MH  - *Electronic Health Records
MH  - Heart Failure/therapy
MH  - Humans
MH  - Prognosis
OTO - NOTNLM
OT  - Counterfactual inference
OT  - Deep representation learning
OT  - Individualized treatment effect estimation
OT  - K-Nearest neighbors
OT  - Multi-task learning
EDAT- 2019/10/15 06:00
MHDA- 2020/10/21 06:00
CRDT- 2019/10/15 06:00
PHST- 2019/02/17 00:00 [received]
PHST- 2019/09/22 00:00 [revised]
PHST- 2019/10/07 00:00 [accepted]
PHST- 2019/10/15 06:00 [pubmed]
PHST- 2020/10/21 06:00 [medline]
PHST- 2019/10/15 06:00 [entrez]
AID - S1532-0464(19)30222-9 [pii]
AID - 10.1016/j.jbi.2019.103303 [doi]
PST - ppublish
SO  - J Biomed Inform. 2019 Dec;100:103303. doi: 10.1016/j.jbi.2019.103303. Epub 2019 
      Oct 11.