Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation

Abstract Extensive monitoring in intensive care units (ICUs) generates large quantities of data which contain numerous trends that are difficult for clinicians to systematically evaluate. Current approaches to such heterogeneity in electronic health records (EHRs) discard pertinent information. We p...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Jacob Deasy, Pietro Liò, Ari Ercole
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2020
Materias:
R
Q
Acceso en línea:https://doaj.org/article/967a5528348445e399c446557e5db6f7
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:967a5528348445e399c446557e5db6f7
record_format dspace
spelling oai:doaj.org-article:967a5528348445e399c446557e5db6f72021-12-02T13:58:15ZDynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation10.1038/s41598-020-79142-z2045-2322https://doaj.org/article/967a5528348445e399c446557e5db6f72020-12-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-79142-zhttps://doaj.org/toc/2045-2322Abstract Extensive monitoring in intensive care units (ICUs) generates large quantities of data which contain numerous trends that are difficult for clinicians to systematically evaluate. Current approaches to such heterogeneity in electronic health records (EHRs) discard pertinent information. We present a deep learning pipeline that uses all uncurated chart, lab, and output events for prediction of in-hospital mortality without variable selection. Over 21,000 ICU patients and tens of thousands of variables derived from the MIMIC-III database were used to train and validate our model. Recordings in the first few hours of a patient’s stay were found to be strongly predictive of mortality, outperforming models using SAPS II and OASIS scores, AUROC 0.72 and 0.76 at 24 h respectively, within just 12 h of ICU admission. Our model achieves a very strong predictive performance of AUROC 0.85 (95% CI 0.83–0.86) after 48 h. Predictive performance increases over the first 48 h, but suffers from diminishing returns, providing rationale for time-limited trials of critical care and suggesting that the timing of decision making can be optimised and individualised.Jacob DeasyPietro LiòAri ErcoleNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 10, Iss 1, Pp 1-11 (2020)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Jacob Deasy
Pietro Liò
Ari Ercole
Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation
description Abstract Extensive monitoring in intensive care units (ICUs) generates large quantities of data which contain numerous trends that are difficult for clinicians to systematically evaluate. Current approaches to such heterogeneity in electronic health records (EHRs) discard pertinent information. We present a deep learning pipeline that uses all uncurated chart, lab, and output events for prediction of in-hospital mortality without variable selection. Over 21,000 ICU patients and tens of thousands of variables derived from the MIMIC-III database were used to train and validate our model. Recordings in the first few hours of a patient’s stay were found to be strongly predictive of mortality, outperforming models using SAPS II and OASIS scores, AUROC 0.72 and 0.76 at 24 h respectively, within just 12 h of ICU admission. Our model achieves a very strong predictive performance of AUROC 0.85 (95% CI 0.83–0.86) after 48 h. Predictive performance increases over the first 48 h, but suffers from diminishing returns, providing rationale for time-limited trials of critical care and suggesting that the timing of decision making can be optimised and individualised.
format article
author Jacob Deasy
Pietro Liò
Ari Ercole
author_facet Jacob Deasy
Pietro Liò
Ari Ercole
author_sort Jacob Deasy
title Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation
title_short Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation
title_full Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation
title_fullStr Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation
title_full_unstemmed Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation
title_sort dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation
publisher Nature Portfolio
publishDate 2020
url https://doaj.org/article/967a5528348445e399c446557e5db6f7
work_keys_str_mv AT jacobdeasy dynamicsurvivalpredictioninintensivecareunitsfromheterogeneoustimeserieswithouttheneedforvariableselectionorcuration
AT pietrolio dynamicsurvivalpredictioninintensivecareunitsfromheterogeneoustimeserieswithouttheneedforvariableselectionorcuration
AT ariercole dynamicsurvivalpredictioninintensivecareunitsfromheterogeneoustimeserieswithouttheneedforvariableselectionorcuration
_version_ 1718392214165389312