The URGENCY algorithm uses vehicle crash sensor data in Automatic Crash Notification (ACN) systems to assist in instantly identifying crashes that are most likely to have time critical injuries. The algorithm also provides the capability of improving injury identification, using data obtained from the scene. The prime purpose of the algorithm is to automatically provide emergency medical responders with objective information on crash severity to assist in detecting the approximately 1% of crashes with serious injuries needing the most urgent medical care. The algorithm calculates the risk of a MAIS 3+ injury being present in the crashed vehicle, instantly at the time of the crash. The prediction can be subsequently updated as more information becomes available. The algorithm was based on a multiple regression analysis using data from the National Accident Sampling System/Crashworthiness Data System, (NASS/CDS) years 1988-95.
In this paper, the accuracy of the algorithm was evaluated for near side crashes by applying it retrospectively to the population of injured occupants in NASS 1997-2000. URGENCY was applied to the population of injured occupants in near side crashes. Using an injury risk criterion of 50%, URGENCY identified 69% of the crashes with MAIS 3+ injuries. By lowering injury risk criterion to 40%, URGENCY identified 78% of the crashes with MAIS 3+ injuries.
Vehicle side intrusion was found to be a highly influential variable. By changing side intrusion from a binary to a continuous variable, the correctly identified crashes increased from 69% to 81%.
Examination of the consequence of missing variables found that unknown values of occupant height and weight had a negligible effect on the ability to capture the MAIS 3+ injured. However, lack of knowledge of these variables did increase the magnitude of the false positives.
|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.|
|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.|
|2015||Gabler HC, Weaver AA, Stitzel JD. Automotive field data in injury biomechanics. In: Yoganandan N, Nahum AM, Melvin JW, eds. Accidental Injury: Biomechanics and Prevention. 3rd ed. New York: Springer; 2015:33-49.|
|2017||Stitzel J, Weaver A, Talton J, Barnard R, Schoell S, Martin RS, Meredith JW. Estimated triage improvement and societal cost savings with nationwide implementation of an advanced automatic crash notification algorithm in the U.S. In: Proceedings of the 25th International Technical Conference on the Enhanced Safety of Vehicles (ESV). June 5-8, 2017; Detroit, MI.|