Search

The quality of vehicular collision data is crucial for studying the relationship between injury severity and collision factors. Misclassified injury severity data in the crash dataset, however, may cause inaccurate parameter estimates and consequently lead to biased conclusions and poorly designed countermeasures. This is particularly true for imbalanced data where the number of samples in one class far outnumber the other. To improve the classification performance of the injury severity, the paper presents a robust noise filtering technique to deal with the mislabels in the imbalanced crash dataset using the advanced machine learning algorithms. We examine the state-of-the-art filtering algorithms, including Iterative Noise Filtering based on the Fusion of Classifiers (INFFC), Iterative Partitioning Filter (IPF), and Saturation Filter (SatF). In the case study of Cairo (Egypt), the empirical results show that: (1) the mislabels in crash data significantly influence the injury severity predictions, and (2) the proposed M-IPF filter outperforms its counterparts in terms of the effectiveness and efficiency in eliminating the mislabels in crash data. The test results demonstrate the efficacy of the M-IPF in handling the data noise and mitigating the impacts thereof. [ABSTRACT FROM AUTHOR]

Considering the high burden of injuries, the Health Authority - Abu Dhabi developed a draft electronic and paper-based injury and poisoning notification system (IPNS) to generate better data on the nature and severity of injuries. The pilot testing and evaluation of IPNS was conducted with the specific objectives to (1) identify the characteristics of injury cases, (2) explore potential risk factors, (3) illustrate the nature and type of data, and (4) the working mechanism of data collection. Data were collected from selected hospitals on patient demographics, injury information and clinical assessment. Descriptive, bivariate and multivariate analyses were conducted. Of 4226 injury cases, nearly three-fourths were male, majority were non-UAE nationals, and the mean age was 21.9. Multivariate findings suggested that compared to UAE nationals, non-UAE nationals were 27% more likely to experience fatal, severe or moderate injuries (p= 0.01). Individuals with health insurance were 31% less likely to suffer a fatal, severe or moderate injury compared to those having no health insurance (p< 0.001). This is the first systematically standardised collection of injury data across three facilities in Abu Dhabi, and provides initial information on characteristics and injury risk factors that will help identify the need for evidence-based intervention for injury prevention and control. [ABSTRACT FROM PUBLISHER]