Predicting patient decompensation from continuous physiologic monitoring in the emergency department
Abstract Anticipation of clinical decompensation is essential for effective emergency and critical care. In this study, we develop a multimodal machine learning approach to predict the onset of new vital sign abnormalities (tachycardia, hypotension, hypoxia) in ED patients with normal initial vital...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-04-01
|
| Series: | npj Digital Medicine |
| Online Access: | https://doi.org/10.1038/s41746-023-00803-0 |
| _version_ | 1797426693402525696 |
|---|---|
| author | Sameer Sundrani Julie Chen Boyang Tom Jin Zahra Shakeri Hossein Abad Pranav Rajpurkar David Kim |
| author_facet | Sameer Sundrani Julie Chen Boyang Tom Jin Zahra Shakeri Hossein Abad Pranav Rajpurkar David Kim |
| author_sort | Sameer Sundrani |
| collection | DOAJ |
| description | Abstract Anticipation of clinical decompensation is essential for effective emergency and critical care. In this study, we develop a multimodal machine learning approach to predict the onset of new vital sign abnormalities (tachycardia, hypotension, hypoxia) in ED patients with normal initial vital signs. Our method combines standard triage data (vital signs, demographics, chief complaint) with features derived from a brief period of continuous physiologic monitoring, extracted via both conventional signal processing and transformer-based deep learning on ECG and PPG waveforms. We study 19,847 adult ED visits, divided into training (75%), validation (12.5%), and a chronologically sequential held-out test set (12.5%). The best-performing models use a combination of engineered and transformer-derived features, predicting in a 90-minute window new tachycardia with AUROC of 0.836 (95% CI, 0.800-0.870), new hypotension with AUROC 0.802 (95% CI, 0.747–0.856), and new hypoxia with AUROC 0.713 (95% CI, 0.680-0.745), in all cases significantly outperforming models using only standard triage data. Salient features include vital sign trends, PPG perfusion index, and ECG waveforms. This approach could improve the triage of apparently stable patients and be applied continuously for the prediction of near-term clinical deterioration. |
| first_indexed | 2024-03-09T08:33:30Z |
| format | Article |
| id | doaj.art-0d3c52e43e9548e7880f31067aa6300c |
| institution | Directory Open Access Journal |
| issn | 2398-6352 |
| language | English |
| last_indexed | 2024-03-09T08:33:30Z |
| publishDate | 2023-04-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | npj Digital Medicine |
| spelling | doaj.art-0d3c52e43e9548e7880f31067aa6300c2023-12-02T18:50:56ZengNature Portfolionpj Digital Medicine2398-63522023-04-016111010.1038/s41746-023-00803-0Predicting patient decompensation from continuous physiologic monitoring in the emergency departmentSameer Sundrani0Julie Chen1Boyang Tom Jin2Zahra Shakeri Hossein Abad3Pranav Rajpurkar4David Kim5School of Medicine, Vanderbilt UniversityDepartment of Computer Science, Stanford UniversityDepartment of Computer Science, Stanford UniversityDalla Lana School of Public Health, University of TorontoDepartment of Biomedical Informatics, Harvard Medical SchoolDepartment of Emergency Medicine, Stanford UniversityAbstract Anticipation of clinical decompensation is essential for effective emergency and critical care. In this study, we develop a multimodal machine learning approach to predict the onset of new vital sign abnormalities (tachycardia, hypotension, hypoxia) in ED patients with normal initial vital signs. Our method combines standard triage data (vital signs, demographics, chief complaint) with features derived from a brief period of continuous physiologic monitoring, extracted via both conventional signal processing and transformer-based deep learning on ECG and PPG waveforms. We study 19,847 adult ED visits, divided into training (75%), validation (12.5%), and a chronologically sequential held-out test set (12.5%). The best-performing models use a combination of engineered and transformer-derived features, predicting in a 90-minute window new tachycardia with AUROC of 0.836 (95% CI, 0.800-0.870), new hypotension with AUROC 0.802 (95% CI, 0.747–0.856), and new hypoxia with AUROC 0.713 (95% CI, 0.680-0.745), in all cases significantly outperforming models using only standard triage data. Salient features include vital sign trends, PPG perfusion index, and ECG waveforms. This approach could improve the triage of apparently stable patients and be applied continuously for the prediction of near-term clinical deterioration.https://doi.org/10.1038/s41746-023-00803-0 |
| spellingShingle | Sameer Sundrani Julie Chen Boyang Tom Jin Zahra Shakeri Hossein Abad Pranav Rajpurkar David Kim Predicting patient decompensation from continuous physiologic monitoring in the emergency department npj Digital Medicine |
| title | Predicting patient decompensation from continuous physiologic monitoring in the emergency department |
| title_full | Predicting patient decompensation from continuous physiologic monitoring in the emergency department |
| title_fullStr | Predicting patient decompensation from continuous physiologic monitoring in the emergency department |
| title_full_unstemmed | Predicting patient decompensation from continuous physiologic monitoring in the emergency department |
| title_short | Predicting patient decompensation from continuous physiologic monitoring in the emergency department |
| title_sort | predicting patient decompensation from continuous physiologic monitoring in the emergency department |
| url | https://doi.org/10.1038/s41746-023-00803-0 |
| work_keys_str_mv | AT sameersundrani predictingpatientdecompensationfromcontinuousphysiologicmonitoringintheemergencydepartment AT juliechen predictingpatientdecompensationfromcontinuousphysiologicmonitoringintheemergencydepartment AT boyangtomjin predictingpatientdecompensationfromcontinuousphysiologicmonitoringintheemergencydepartment AT zahrashakerihosseinabad predictingpatientdecompensationfromcontinuousphysiologicmonitoringintheemergencydepartment AT pranavrajpurkar predictingpatientdecompensationfromcontinuousphysiologicmonitoringintheemergencydepartment AT davidkim predictingpatientdecompensationfromcontinuousphysiologicmonitoringintheemergencydepartment |