Accident; analysis and prevention
-
Review
Machine learning approaches to analysing textual injury surveillance data: a systematic review.
To synthesise recent research on the use of machine learning approaches to mining textual injury surveillance data. ⋯ The use of narrative text for injury surveillance has grown in popularity, complexity and quality over recent years. With advances in data mining techniques, increased capacity for analysis of large databases, and involvement of computer scientists in the injury prevention field, along with more comprehensive use and description of quality assurance methods in text mining approaches, it is likely that we will see a continued growth and advancement in knowledge of text mining in the injury field.
-
Traumatic brain injury (TBI) continues to be a leading cause of morbidity and mortality throughout the world. Research has been undertaken in order to better understand the characteristics of the injury event and measure the risk of injury to develop more effective environmental, technological, and clinical management strategies. This research used methods that have limited applications to predicting human responses. ⋯ The results of the methodology were consistent with current TBI research, describing TBI to occur in the range of 335-445g linear accelerations and 23.7-51.2krad/s(2) angular accelerations. More significantly, this research demonstrated that lower responses in the antero-posterior direction can cause TBI, with lateral impact responses requiring larger magnitudes for the same types of brain lesions. This suggests an increased likelihood of sustaining TBI for impacts to the front or back of the head, a result that has implications affecting current understanding of the mechanisms of TBI and associated threshold parameters.
-
Comparative Study
The effects of changes in the traffic scene during overtaking.
Overtaking maneuvers were studied in real traffic, by comparing cases where a change in the opposite traffic occurred during the overtaking maneuver i.e., appearance of an oncoming car, with cases where no change occurred during the maneuver i.e., either an already apparent oncoming car or no oncoming car. In total 45 naturally occurring cases of overtaking were analysed. By examining the time headways (TH) between the overtaking car and the other cars involved, at the end of the maneuver, a significant correlation was found between the TH to opposite traffic and the TH rear to the overtaken car. ⋯ It is suggested that drivers were probably expecting to be confronted with an oncoming car during the overtaking. However, the decreased available time to disambiguate this situation leads the overtaking driver to limit the rear safety margin of the vehicle being overtaken. The appropriateness of this practice, in terms of safety, remains questionable.
-
Several studies have shown that personality traits and attitudes toward traffic safety predict aberrant driving behaviors and crash involvement. However, this process has not been adequately investigated in professional drivers, such as bus drivers. The present study used a personality-attitudes model to assess whether personality traits predicted aberrant self-reported driving behaviors (driving violations, lapses, and errors) both directly and indirectly, through the effects of attitudes towards traffic safety in a large sample of bus drivers. ⋯ Personality traits relevant to emotionality directly predicted bus drivers' aberrant driving behaviors, without any mediation of attitudes. Finally, only self-reported violations were related to bus drivers' accident risk. The present findings suggest that the hypothesized personality-attitudes model accounts for aberrant driving behaviors in bus drivers, and provide the empirical basis for evidence-based road safety interventions in the context of public transport.