Occupant transportation decision algorithm:​ an advanced automatic crash notification algorithm developed and evaluated for motor vehicle crash occupant triage

Stitzel, Joel; Weaver, Ashley; Schoell, Samantha; Barnard, Ryan; Talton, Jennifer; Martin, R. Shayn; Meredith, J. Wayne

Affiliations

¹Virginia Tech-Wake Forest Center for Injury Biomechanics, United States

²Wake Forest School of Medicine, United States

Abstract

The objective was to develop an Advanced Automatic Crash Notification (AACN) algorithm and evaluate its performance in making optimal occupant triage decisions. The developed AACN algorithm known as the Occupant Transportation Decision Algorithm (OTDA) uses measurements obtainable from vehicle telemetry to predict risk of overall occupant injury and recommend a transportation decision for the occupant following a motor vehicle crash (MVC), particularly whether transport to a Level I/II trauma center is recommended. A list of injuries necessitating treatment at a Level I/II trauma center (TC) was determined using an injury-based approach based on three facets (severity, time sensitivity, and predictability). These three facets were quantified for each injury from expert physician and emergency medical services (EMS) professional opinion and database analyses of the National Trauma Data Bank and National Inpatient Sample. Severity, Time Sensitivity, and Predictability Scores were summed for each injury to compute an Injury Score. Injuries with an Injury Score exceeding a particular threshold were included on the Master Target Injury List, which is a list of injuries more likely to require Level I/II TC treatment. OTDA inputs for development include the Master Target Injury List and 38,970 National Automotive Sampling System-Crashworthiness Data System (NASS-CDS) 2000-2011 occupants. The OTDA uses multivariate logistic regression to predict an occupant's risk of sustaining an injury on the Master Target Injury List from the following model variables: longitudinal/lateral delta-v, number of quarter turns (in rollover only), belt status, multiple impacts, and airbag deployment.

A parametric OTDA was developed with five tunable parameters allowing for extensive optimization. The OTDA was optimized with a genetic algorithm that compared the OTDA transportation decision for each NASS-CDS occupant to a dichotomous representation of their Injury Severity Score (ISS). Occupants with ISS 16+ should be transported to a Level I/II TC. OTDA optimization minimized under triage (UT) and over triage (OT) rates with the goal of producing UT rates < 5% and OT rates < 50% as recommended by the American College of Surgeons (ACS). For the optimized OTDA, UT rates by crash mode were 5.9% (frontal), 4.6% (near side), 2.9% (far side), 7.0% (rear), and 16.0% (rollover). OT rates by crash mode for the optimized OTDA were 49.7% (frontal), 47.9% (near side), 49.7% (far side), 44.0% (rear), and 49.7% (rollover).

The OTDA was developed with an injury-based approach that examined three injury facets to identify injuries necessitating treatment at a Level I/II TC. Large hospital and survey datasets containing information on injuries, mortality risk, treatment urgency, and hospital transfers were used in conjunction with large crash datasets with crash, vehicle, occupant, and injury data. The OTDA has been rigorously optimized and has demonstrated improved UT rates compared to other AACN algorithms in the literature and OT rates meeting ACS recommendations. Since the OTDA uses only vehicle telemetry measurements specified in Part 563 regulation, this AACN algorithm could be readily incorporated into new vehicles to inform emergency personnel of recommended triage decisions for MVC occupants. The overall societal purpose of this AACN algorithm is to reduce response times, increase triage efficiency, and improve overall patient outcome.

Links

URL: http://www-esv.nhtsa.dot.gov/Proceedings/24/files/24ESV-000190.PDF

Cited Works (14)

Year	Entry
2011	Kononen DW, Flannagan CAC, Wang SC. Identification and validation of a logistic regression model for predicting serious injuries associated with motor vehicle crashes. Accid Anal Prev. January 2011;43(1):112-122.
2009	Gabauer DJ, Gabler HC. Differential rollover risk in vehicle-to-traffic barrier collisions. In: 53rd Annual Proceedings, Association for the Advancement of Automotive Medicine (AAAM). October 5-7, 2009; Baltimore, MD.131-140.
2004	Bahouth GT, Digges KH, Bedewi NE, Kuznetsov A, Augenstein JS, Perdeck E. Development of URGENCY 2.1 for the prediction of crash injury severity. Top Emerg Med. April–June 2004;26(2):157-165.
2010	Hampton CE, Gabler HC. Evaluation of the accuracy of NASS/CDS delta-V estimates from the enhanced winsmash algorithm. In: 54th Annual Proceedings, Association for the Advancement of Automotive Medicine (AAAM). October 17-20, 2010; Las Vegas, NV.241-251.
2003	Augenstein J, Perdeck E, Stratton J, Digges K, Bahouth G. Characteristics of crashes that increase the risk of serious injuries. In: 47th Annual Proceedings, Association for the Advancement of Automotive Medicine (AAAM). September 22-24, 2003; Lisbon, Portugal.561-576.
2001	Augenstein J, Digges K, Ogata S, Perdeck E, Stratton J. Development and validation of the urgency algorithm to predict compelling injuries. In: Proceedings of the 17th International Technical Conference on the Enhanced Safety of Vehicles (ESV). June 4-7, 2001; Amsterdam, The Netherlands.
2012	Ridella SA, Rupp JD, Poland K. Age‐related differences in AIS 3+ crash injury risk, types, causation and mechanisms. In: Proceedings of the 2012 International IRCOBI Conference on the Biomechanics of Injury. September 12-14, 2012; Dublin, Ireland.43-60.
1997	Malliaris AC, Digges KH, DeBlois JH. Relationships between crash casualties and crash attributes. In: Proceedings of the SAE International Congress & Exposition. February 24-27, 1997; Detroit, MI. Warrendale, PA: Society of Automotive Engineers. SAE 970393.
2013	Weaver AA, Barnard RT, Kilgo PD, Martin RS, Stitzel JD. Mortality-based quantification of injury severity for frequently occurring motor vehicle crash injuries. In: 57th Annual Proceedings, Association for the Advancement of Automotive Medicine (AAAM). September 22-25, 2013; Quebec City, QC.235-245.
2007	Sharma D, Stern S, Brophy J, Choi E-H. An overview of NHTSA’s crash reconstruction software WinSMASH. In: Proceedings of the 20th International Technical Conference on the Enhanced Safety of Vehicles (ESV). June 18-21, 2007; Lyon, France.
2009	Rauscher S, Messner G, Baur P, Augenstein J, Digges K, Perdeck E, Bahouth G, Pieske O. Enhanced automatic collision notification system: improved rescue care due to injury prediction: first field experience. In: Proceedings of the 21st International Technical Conference on the Enhanced Safety of Vehicles (ESV). June 15-18, 2009; Stuttgart, Germany.
2012	Bahouth G, Digges K, Schulman C. Influence of injury risk thresholds on the performance of an algorithm to predict crashes with serious injuries. In: 56th Annual Proceedings, Association for the Advancement of Automotive Medicine (AAAM). October 14-17, 2012; Seattle, WA.223-230.
2005	Augenstein J, Perdeck E, Bahouth GT, Digges KH, Borchers N, Baur P. Injury identification: priorities for data transmitted. In: Proceedings of the 19th International Technical Conference on the Enhanced Safety of Vehicles (ESV). June 6-9, 2005; Washington, DC.
1989	Cohen D, Digges K, Nichols RH. Rollover crashworthiness classification and severity indices. In: Proceedings of the 12th International Technical Conference on Experimental Safety Vehicles (ESV). May 29–June 1, 1989; Göteborg, Sweden.477-488.

Cited By (1)

Year	Entry
2023	Baker C. AutoTriage: A Feasibility Assessment of Using in-Vehicle Sensors to Predict Traumatic Brain Injury in Road Traffic Collisions [PhD thesis]. Imperial College London; March 23, 2023.