Early prediction of hemodynamic interventions in the intensive care unit using machine learning
© 2021. The Author(s)..
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.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2021 |
|---|---|
| Erschienen: |
2021 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:25 |
|---|---|
| Enthalten in: |
Critical care (London, England) - 25(2021), 1 vom: 14. Nov., Seite 388 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Rahman, Asif [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Clinical decision support |
|---|
| Anmerkungen: |
Date Completed 30.12.2021 Date Revised 30.12.2021 published: Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1186/s13054-021-03808-x |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM333137272 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM333137272 | ||
| 003 | DE-627 | ||
| 005 | 20231225221027.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231225s2021 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1186/s13054-021-03808-x |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1110.xml |
| 035 | |a (DE-627)NLM333137272 | ||
| 035 | |a (NLM)34775971 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Rahman, Asif |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Early prediction of hemodynamic interventions in the intensive care unit using machine learning |
| 264 | 1 | |c 2021 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 30.12.2021 | ||
| 500 | |a Date Revised 30.12.2021 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © 2021. The Author(s). | ||
| 520 | |a 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 | ||
| 520 | |a 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 | ||
| 520 | |a 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 | ||
| 520 | |a 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 | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Clinical decision support | |
| 650 | 4 | |a Hemodynamics | |
| 650 | 4 | |a Machine learning | |
| 650 | 4 | |a Vasoactive therapy | |
| 700 | 1 | |a Chang, Yale |e verfasserin |4 aut | |
| 700 | 1 | |a Dong, Junzi |e verfasserin |4 aut | |
| 700 | 1 | |a Conroy, Bryan |e verfasserin |4 aut | |
| 700 | 1 | |a Natarajan, Annamalai |e verfasserin |4 aut | |
| 700 | 1 | |a Kinoshita, Takahiro |e verfasserin |4 aut | |
| 700 | 1 | |a Vicario, Francesco |e verfasserin |4 aut | |
| 700 | 1 | |a Frassica, Joseph |e verfasserin |4 aut | |
| 700 | 1 | |a Xu-Wilson, Minnan |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Critical care (London, England) |d 1997 |g 25(2021), 1 vom: 14. Nov., Seite 388 |w (DE-627)NLM097399787 |x 1466-609X |7 nnns |
| 773 | 1 | 8 | |g volume:25 |g year:2021 |g number:1 |g day:14 |g month:11 |g pages:388 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1186/s13054-021-03808-x |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 25 |j 2021 |e 1 |b 14 |c 11 |h 388 | ||