Demonstrating the consequences of learning missingness patterns in early warning systems for preventative health care: A novel simulation and solution

[Display omitted] •Early warning system performance worsens if missingness pattern changes in EHR data.•Generated synthetic EHR data with variational autoencoder and custom loss function.•Randomized and Bayesian regression imputation appropriate for tree-based methods.•Using proper imputation, we de...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Journal of biomedical informatics 2020-10, Vol.110, p.103528-103528, Article 103528
Hauptverfasser:	Gillies, Christopher E., Taylor, Daniel F., Cummings, Brandon C., Ansari, Sardar, Islim, Fadi, Kronick, Steven L., Medlin, Richard P., Ward, Kevin R.
Format:	Artikel
Sprache:	eng
Schlagworte:	Computer Science Computer Science, Interdisciplinary Applications Early warning systems Life Sciences & Biomedicine Machine learning Medical Informatics Patient deterioration Science & Technology Simulation Technology Tree-based methods Variational autoencoder
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!