Abstract
A significant proportion of motor vehicle crash fatalities are potentially preventable with improved acute care. By increasing the accuracy of triage more victims could be transported directly to the best suited care facility and be provided optimal care. We hypothesize that On Scene Injury Severity Prediction (OSISP) algorithms, developed utilizing machine learning methods, have potential to improve triage by complementing the field triage protocol. In this study, the accuracy of OSISP algorithms based on the “National Automotive Sampling System - Crashworthiness Data System” (NASS-CDS) of crashes involving adult occupants for calendar years 2010–2015 was evaluated. Severe injury was the dependent variable, defined as Injury Severity Score (ISS) > 15. The dataset contained 37873 subjects, whereof 21589 included injury data and were further analyzed. Selection of model predictors was based on potential for injury severity prediction and perceived feasibility of assessment by first responders. We excluded vehicle telemetry data due to the limited availability of these systems in the contemporary vehicle fleet, and because this data is not yet being utilized in prehospital care. The machine learning algorithms Logistic Regression, Ridge Regression, Bernoulli Naïve Bayes, Stochastic Gradient Descent and Artificial Neural Networks were evaluated. Best performance with small margin was achieved with Logistic Regression, achieving area under the receiver operator characteristic curve (AUC) of 0.86 (95% confidence interval 0.82–0.90), as estimated by 10-fold stratified cross-validation. Ejection, Entrapment, Belt use, Airbag deployment and Crash type were good predictors. Using only a subset of the 5–7 best predictors approached the prediction accuracy achieved when using the full set (14 predictors). A simplified benefit analysis indicated that nationwide implementation of OSISP in the US could bring improved care for 3100 severely injured patients, and reduce unnecessary use of trauma center resources for 94000 non-severely injured patients, every year.
Highlights
Motor vehicle crashes (MVC) in US produce around 30000 fatalities and 4 million injured people every year, adding up to a societal economic burden of totally $240 billion or $800 per citizen (Blincoe et al, 2015)
We evaluate if methods employing machine learning and variables that can be assessed on the scene of accident has potential to amend field triage
The main finding is that an On Scene Injury Severity Prediction (OSISP) algorithm is capable of predicting severe injury in a US population of MVC occupants with an AUC of 0.86, based only on variables deemed to be feasible to assess on the scene of crash by first re sponders
Summary
Motor vehicle crashes (MVC) in US produce around 30000 fatalities and 4 million injured people every year, adding up to a societal economic burden of totally $240 billion or $800 per citizen (Blincoe et al, 2015). A significant proportion of the fatalities are potentially preventable (Ray et al, 2016; Berwick et al, 2016). Military trauma care has achieved remarkably high survival rates, 98%, for patients reaching a treatment facility (Berwick et al, 2016). Patients with severe injury have a higher probability of surviving if they are directly transported to a trauma center, to provide specialized care with minimal delay (Hu et al, 2017; Haas et al, 2010; MacKenzie et al, 2006; Candefjord et al, 2020). A key to provide adequate care to a larger proportion of patients is to attain a high triage accuracy (Ray et al, 2016; Sasser et al, 2012), so that the rate of appropriate decisions on where to transport the patient can be increased
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
