Abstract
Continuous monitoring of high-risk patients and early prediction of severe outcomes is crucial to prevent avoidable deaths. Current clinical monitoring is primarily based on intermittent observation of vital signs and the early warning scores (EWS). The drawback is lack of time series dynamics and correlations among vital signs. This study presents an approach to real-time outcome prediction based on machine learning from continuous recording of vital signs. Systolic blood pressure, diastolic blood pressure, heart rate, pulse rate, respiration rate and peripheral blood oxygen saturation were continuously acquired by wearable devices from 292 post-operative high-risk patients. The outcomes from serious complications were evaluated based on review of patients’ medical record. The descriptive statistics of vital signs and patient demographic information were used as features. Four machine learning models K-Nearest-Neighbors (KNN), Decision Trees (DT), Random Forest (RF), and Boosted Ensemble (BE) were trained and tested. In static evaluation, all four models had comparable prediction performance to that of the state of the art. In dynamic evaluation, the models trained from the static evaluation were tested with continuous data. RF and BE obtained the lower false positive rate (FPR) of 0.073 and 0.055 on no-outcome patients respectively. The four models KNN, DT, RF and BE had area under receiver operating characteristic curve (AUROC) of 0.62, 0.64, 0.65 and 0.64 respectively on outcome patients. RF was found to be optimal model with lower FPR on no-outcome patients and a higher AUROC on outcome patients. These findings are encouraging and indicate that additional investigations must focus on validating performance in a clinical setting before deployment of the real-time outcome prediction.
Originalsprog | Engelsk |
---|---|
Artikelnummer | 105559 |
Tidsskrift | Computers in Biology and Medicine |
Vol/bind | 147 |
Antal sider | 8 |
ISSN | 0010-4825 |
DOI | |
Status | Udgivet - 2022 |
Bibliografisk note
Funding Information:The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Eske Kvanner Aasvang, Christian Sylvest Meyhoff and Helge B.D. Sorensen report financial support was provided by The Novo Nordic Foundation, the Danish Cancer Society , Radiometer, A.P. Møller Foundation, WARD247 ApS. Eske Kvanner Aasvang, Christian Sylvest Meyhoff and Helge B.D. Sorensen report a relationship with Ward247 ApS that includes: equity or stocks.
Funding Information:
The WARD-project received support from the Innovation Fund Denmark ( 8056-00055B ); the Novo Nordic Foundation , the Danish Cancer Society (R150-A9865-16-S48); Copenhagen Center for Health Technology (CACHET) ; Radiometer Medical Aps; Isansys Ltd; A.P. Møller Foundation, Bispebjerg Hospital, Rigshospitalet and the Technical University of Denmark. No sponsor had any role in the study design, data collection, analysis, access to and interpretation of data, writing the report, or the decision to submit the article.
Publisher Copyright:
© 2022 The Authors