Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Apr 9;13(4):e0195024.
doi: 10.1371/journal.pone.0195024. eCollection 2018.

Readmission prediction via deep contextual embedding of clinical concepts

Cao Xiao  1 , Tengfei Ma  2 , Adji B Dieng  3 , David M Blei  3 , Fei Wang  4
Affiliations

Readmission prediction via deep contextual embedding of clinical concepts

Cao Xiao et al. PLoS One. .

Abstract

Objective: Hospital readmission costs a lot of money every year. Many hospital readmissions are avoidable, and excessive hospital readmissions could also be harmful to the patients. Accurate prediction of hospital readmission can effectively help reduce the readmission risk. However, the complex relationship between readmission and potential risk factors makes readmission prediction a difficult task. The main goal of this paper is to explore deep learning models to distill such complex relationships and make accurate predictions.

Materials and methods: We propose CONTENT, a deep model that predicts hospital readmissions via learning interpretable patient representations by capturing both local and global contexts from patient Electronic Health Records (EHR) through a hybrid Topic Recurrent Neural Network (TopicRNN) model. The experiment was conducted using the EHR of a real world Congestive Heart Failure (CHF) cohort of 5,393 patients.

Results: The proposed model outperforms state-of-the-art methods in readmission prediction (e.g. 0.6103 ± 0.0130 vs. second best 0.5998 ± 0.0124 in terms of ROC-AUC). The derived patient representations were further utilized for patient phenotyping. The learned phenotypes provide more precise understanding of readmission risks.

Discussion: Embedding both local and global context in patient representation not only improves prediction performance, but also brings interpretable insights of understanding readmission risks for heterogeneous chronic clinical conditions.

Conclusion: This is the first of its kind model that integrates the power of both conventional deep neural network and the probabilistic generative models for highly interpretable deep patient representation learning. Experimental results and case studies demonstrate the improved performance and interpretability of the model.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. An example segment of EHR records, where visits could occur to different locations.
Patients who are re-admitted to “inpatient hospital” within 30 days of their releases from “inpatient hospital” are considered readmissions.
Fig 2
Fig 2. The CONTENT model.
Fig 3
Fig 3. Clustering of patient representations.
Fig 4
Fig 4. Top clinical events for selected clusters.
See this image and copyright information in PMC

References

    1. Basu Roy S, Teredesai A, Zolfaghar K, Liu R, Hazel D, Newman S, et al. Dynamic Hierarchical Classification for Patient Risk-of-Readmission. In: Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. KDD’15. New York, NY, USA: ACM; 2015. p. 1691–1700. Available from: http://doi.acm.org/10.1145/2783258.2788585. - DOI
    1. McIlvennan C, Eapen Z, Allen L. Hospital Readmissions Reduction Program. Circulation. 2015;131(20). doi: 10.1161/CIRCULATIONAHA.114.010270 - DOI - PMC - PubMed
    1. K Z. Predicting Risk-of-Readmission for Congestive Heart Failure Patients: A Multi-Layer Approach. IEEE Trans on Big Data. 2013;.
    1. Mathias J, Agrawal A, Feinglass J, Cooper A, Baker D, Choudhary A. Development of a 5 year life expectancy index in older adults using predictive mining of electronic health record data. Journal of American Medical Informatics Association. 2013;20(e1). doi: 10.1136/amiajnl-2012-001360 - DOI - PMC - PubMed
    1. Tran T, Luo W, Phung D, Gupta S, Rana S, Kennedy RL, et al. A framework for feature extraction from hospital medical data with applications in risk prediction. BMC Bioinformatics. 2014;15(1):425 doi: 10.1186/s12859-014-0425-8 - DOI - PMC - PubMed

Publication types