Early prediction of hemodynamic interventions in the intensive care unit using machine learning
Abstract Background Timely recognition of hemodynamic instability in critically ill patients enables increased vigilance and early treatment opportunities. We develop the Hemodynamic Stability Index (HSI), which highlights situational awareness of po...
Main Authors: | , , , , , , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
BioMed Central
2021
|
Online Access: | https://hdl.handle.net/1721.1/138177 |
_version_ | 1811085833129689088 |
---|---|
author | Rahman, Asif Chang, Yale Dong, Junzi Conroy, Bryan Natarajan, Annamalai Kinoshita, Takahiro Vicario, Francesco Frassica, Joseph Xu-Wilson, Minnan |
author2 | Massachusetts Institute of Technology. Institute for Medical Engineering & Science |
author_facet | Massachusetts Institute of Technology. Institute for Medical Engineering & Science Rahman, Asif Chang, Yale Dong, Junzi Conroy, Bryan Natarajan, Annamalai Kinoshita, Takahiro Vicario, Francesco Frassica, Joseph Xu-Wilson, Minnan |
author_sort | Rahman, Asif |
collection | MIT |
description | Abstract
Background
Timely recognition of hemodynamic instability in critically ill patients enables increased vigilance and early treatment opportunities. We develop the Hemodynamic Stability Index (HSI), which highlights situational awareness of possible hemodynamic instability occurring at the bedside and to prompt assessment for potential hemodynamic interventions.
Methods
We used an ensemble of decision trees to obtain a real-time risk score that predicts the initiation of hemodynamic interventions an hour into the future. We developed the model using the eICU Research Institute (eRI) database, based on adult ICU admissions from 2012 to 2016. A total of 208,375 ICU stays met the inclusion criteria, with 32,896 patients (prevalence = 18%) experiencing at least one instability event where they received one of the interventions during their stay. Predictors included vital signs, laboratory measurements, and ventilation settings.
Results
HSI showed significantly better performance compared to single parameters like systolic blood pressure and shock index (heart rate/systolic blood pressure) and showed good generalization across patient subgroups. HSI AUC was 0.82 and predicted 52% of all hemodynamic interventions with a lead time of 1-h with a specificity of 92%. In addition to predicting future hemodynamic interventions, our model provides confidence intervals and a ranked list of clinical features that contribute to each prediction. Importantly, HSI can use a sparse set of physiologic variables and abstains from making a prediction when the confidence is below an acceptable threshold.
Conclusions
The HSI algorithm provides a single score that summarizes hemodynamic status in real time using multiple physiologic parameters in patient monitors and electronic medical records (EMR). Importantly, HSI is designed for real-world deployment, demonstrating generalizability, strong performance under different data availability conditions, and providing model explanation in the form of feature importance and prediction confidence. |
first_indexed | 2024-09-23T13:16:21Z |
format | Article |
id | mit-1721.1/138177 |
institution | Massachusetts Institute of Technology |
language | English |
last_indexed | 2024-09-23T13:16:21Z |
publishDate | 2021 |
publisher | BioMed Central |
record_format | dspace |
spelling | mit-1721.1/1381772024-03-19T17:40:19Z Early prediction of hemodynamic interventions in the intensive care unit using machine learning Rahman, Asif Chang, Yale Dong, Junzi Conroy, Bryan Natarajan, Annamalai Kinoshita, Takahiro Vicario, Francesco Frassica, Joseph Xu-Wilson, Minnan Massachusetts Institute of Technology. Institute for Medical Engineering & Science Abstract Background Timely recognition of hemodynamic instability in critically ill patients enables increased vigilance and early treatment opportunities. We develop the Hemodynamic Stability Index (HSI), which highlights situational awareness of possible hemodynamic instability occurring at the bedside and to prompt assessment for potential hemodynamic interventions. Methods We used an ensemble of decision trees to obtain a real-time risk score that predicts the initiation of hemodynamic interventions an hour into the future. We developed the model using the eICU Research Institute (eRI) database, based on adult ICU admissions from 2012 to 2016. A total of 208,375 ICU stays met the inclusion criteria, with 32,896 patients (prevalence = 18%) experiencing at least one instability event where they received one of the interventions during their stay. Predictors included vital signs, laboratory measurements, and ventilation settings. Results HSI showed significantly better performance compared to single parameters like systolic blood pressure and shock index (heart rate/systolic blood pressure) and showed good generalization across patient subgroups. HSI AUC was 0.82 and predicted 52% of all hemodynamic interventions with a lead time of 1-h with a specificity of 92%. In addition to predicting future hemodynamic interventions, our model provides confidence intervals and a ranked list of clinical features that contribute to each prediction. Importantly, HSI can use a sparse set of physiologic variables and abstains from making a prediction when the confidence is below an acceptable threshold. Conclusions The HSI algorithm provides a single score that summarizes hemodynamic status in real time using multiple physiologic parameters in patient monitors and electronic medical records (EMR). Importantly, HSI is designed for real-world deployment, demonstrating generalizability, strong performance under different data availability conditions, and providing model explanation in the form of feature importance and prediction confidence. 2021-11-22T13:45:51Z 2021-11-22T13:45:51Z 2021-11-14 2021-11-21T04:22:26Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/138177 Critical Care. 2021 Nov 14;25(1):388 PUBLISHER_CC en https://doi.org/10.1186/s13054-021-03808-x Creative Commons Attribution https://creativecommons.org/licenses/by/4.0/ The Author(s) application/pdf BioMed Central BioMed Central |
spellingShingle | Rahman, Asif Chang, Yale Dong, Junzi Conroy, Bryan Natarajan, Annamalai Kinoshita, Takahiro Vicario, Francesco Frassica, Joseph Xu-Wilson, Minnan Early prediction of hemodynamic interventions in the intensive care unit using machine learning |
title | Early prediction of hemodynamic interventions in the intensive care unit using machine learning |
title_full | Early prediction of hemodynamic interventions in the intensive care unit using machine learning |
title_fullStr | Early prediction of hemodynamic interventions in the intensive care unit using machine learning |
title_full_unstemmed | Early prediction of hemodynamic interventions in the intensive care unit using machine learning |
title_short | Early prediction of hemodynamic interventions in the intensive care unit using machine learning |
title_sort | early prediction of hemodynamic interventions in the intensive care unit using machine learning |
url | https://hdl.handle.net/1721.1/138177 |
work_keys_str_mv | AT rahmanasif earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT changyale earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT dongjunzi earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT conroybryan earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT natarajanannamalai earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT kinoshitatakahiro earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT vicariofrancesco earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT frassicajoseph earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning AT xuwilsonminnan earlypredictionofhemodynamicinterventionsintheintensivecareunitusingmachinelearning |