An IT-solution for clinical evaluation of an on-scene injury severity prediction algorithm intended for road crash victims
In Sweden, as in most countries, the accuracy of triage of road crash patients appears to be low. There is a need for new complementary tools to improve the accuracy. An algorithm for On-Scene Injury Severity Prediction (OSISP) of car passengers involved in road crashes has been developed at the SAFER Vehicle and Traffic Safety Centre at Chalmers. The use of this algorithm in the prehospital stage has the potential to become a complementary tool to improve the triage accuracy. In this thesis, an implementation of the algorithm has been developed for use in Android smartphones. The App format has been designed to naturally fit into the normal workflow of ambulance personnel, via iterative refinements considering feedback from prehospital experts. The user is asked to provide Accident Characteristics for the OSISP algorithm to calculate the risk of severe injury, e.g. age, gender, airbag deployment, belt use, environment, type of accident and posted speed limit. According to the calculated risk, an example of how the clinical decision support may look like is presented. The App logs the data to a server via an File Transfer Protocol implementation. Data is sent through a mobile network or WiFi and automatically uploaded to the server when a network becomes available. Based on input from interviews, a possibility to triage several patients at the same time has been implemented. The final solution has been confirmed to appear to be usable in the field by several ambulance nurses. This solution is ready for implementation in a clinical study and evaluation of the OSISP algorithm. If the solution is successfully evaluated, the ambition is to integrate the algorithm in other ambulance ICT-platforms.