An AI-Enabled Dynamic Risk Stratification for Emergency Department Patients with ECG and CXR Integration
© 2023. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature..
Emergency department (ED) triage scale determines the priority of patient care and foretells the prognosis. However, the information retrieved from the initial assessment is limited, hindering the risk identification accuracy of triage. Therefore, we sought to develop a 'dynamic' triage system as secondary screening, using artificial intelligence (AI) techniques to integrate information from initial assessment data and subsequent examinations. This retrospective cohort study included 134,112 ED visits with at least one electrocardiography (ECG) and chest X-ray (CXR) in a medical center from 2012 to 2022. Additionally, an independent community hospital provided 45,614 ED visits as an external validation set. We trained an eXtreme gradient boosting (XGB) model using initial assessment data to predict all-cause mortality in 7 days. Two deep learning models (DLMs) using ECG and CXR were trained to stratify mortality risks. The dynamic triage levels were based on output from the XGB-triage and DLMs from ECG and CXR. During the internal and external validation, the area under the receiver operating characteristic curve (AUC) of the XGB-triage model was >0.866; furthermore, the AUCs of DLMs using ECG and CXR were >0.862 and >0.886, respectively. The dynamic triage scale provided a higher C-index (0.914-0.920 vs. 0.827-0.843) than the original one and demonstrated better predictive ability for 5-year mortality, 30-day ED revisit, and 30-day discharge. The AI-based risk scale provides a more accurate and dynamic stratification of mortality risk in ED patients, particularly in identifying patients who tend to be overlooked due to atypical symptoms.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:47 |
|---|---|
| Enthalten in: |
Journal of medical systems - 47(2023), 1 vom: 31. Juli, Seite 81 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Chen, Yu-Hsuan Jamie [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
Artificial intelligence |
|---|
| Anmerkungen: |
Date Completed 01.01.2024 Date Revised 01.01.2024 published: Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1007/s10916-023-01980-x |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM360229522 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM360229522 | ||
| 003 | DE-627 | ||
| 005 | 20240108140727.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s10916-023-01980-x |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1246.xml |
| 035 | |a (DE-627)NLM360229522 | ||
| 035 | |a (NLM)37523102 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Chen, Yu-Hsuan Jamie |e verfasserin |4 aut | |
| 245 | 1 | 3 | |a An AI-Enabled Dynamic Risk Stratification for Emergency Department Patients with ECG and CXR Integration |
| 264 | 1 | |c 2023 | |
| 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 01.01.2024 | ||
| 500 | |a Date Revised 01.01.2024 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © 2023. The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature. | ||
| 520 | |a Emergency department (ED) triage scale determines the priority of patient care and foretells the prognosis. However, the information retrieved from the initial assessment is limited, hindering the risk identification accuracy of triage. Therefore, we sought to develop a 'dynamic' triage system as secondary screening, using artificial intelligence (AI) techniques to integrate information from initial assessment data and subsequent examinations. This retrospective cohort study included 134,112 ED visits with at least one electrocardiography (ECG) and chest X-ray (CXR) in a medical center from 2012 to 2022. Additionally, an independent community hospital provided 45,614 ED visits as an external validation set. We trained an eXtreme gradient boosting (XGB) model using initial assessment data to predict all-cause mortality in 7 days. Two deep learning models (DLMs) using ECG and CXR were trained to stratify mortality risks. The dynamic triage levels were based on output from the XGB-triage and DLMs from ECG and CXR. During the internal and external validation, the area under the receiver operating characteristic curve (AUC) of the XGB-triage model was >0.866; furthermore, the AUCs of DLMs using ECG and CXR were >0.862 and >0.886, respectively. The dynamic triage scale provided a higher C-index (0.914-0.920 vs. 0.827-0.843) than the original one and demonstrated better predictive ability for 5-year mortality, 30-day ED revisit, and 30-day discharge. The AI-based risk scale provides a more accurate and dynamic stratification of mortality risk in ED patients, particularly in identifying patients who tend to be overlooked due to atypical symptoms | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Artificial intelligence | |
| 650 | 4 | |a Chest X-ray | |
| 650 | 4 | |a Deep learning | |
| 650 | 4 | |a Electrocardiogram | |
| 650 | 4 | |a Emergency department | |
| 650 | 4 | |a Triage scale | |
| 700 | 1 | |a Lin, Chin-Sheng |e verfasserin |4 aut | |
| 700 | 1 | |a Lin, Chin |e verfasserin |4 aut | |
| 700 | 1 | |a Tsai, Dung-Jang |e verfasserin |4 aut | |
| 700 | 1 | |a Fang, Wen-Hui |e verfasserin |4 aut | |
| 700 | 1 | |a Lee, Chia-Cheng |e verfasserin |4 aut | |
| 700 | 1 | |a Wang, Chih-Hung |e verfasserin |4 aut | |
| 700 | 1 | |a Chen, Sy-Jou |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Journal of medical systems |d 1977 |g 47(2023), 1 vom: 31. Juli, Seite 81 |w (DE-627)NLM000343242 |x 1573-689X |7 nnns |
| 773 | 1 | 8 | |g volume:47 |g year:2023 |g number:1 |g day:31 |g month:07 |g pages:81 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1007/s10916-023-01980-x |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 47 |j 2023 |e 1 |b 31 |c 07 |h 81 | ||