In Canada, poisoning deaths are classified as unintentional, suicide, homicide or undetermined. However, there is a significant proportion of deaths which are classified as undetermined. This work describes a methodology to classify cases of poisoning-related deaths that were misclassified as undetermined, using data from the Canadian Coroner and Medical Examiner Database (CCMED). The research work used Decision Trees (DTs), Case-Based Reasoning (CBR), and Artificial Neural Networks (ANNs), using a number of known cases of suicides and of unintentional deaths, to classify undetermined cases as a suicide or an unintentional poisoning. The performance achieved a sensitivity of 96% (true positives or suicides), a specificity of 84% (true negatives or unintentional), and an area under the Receiver Operating Characteristic Curve (ROC) of 0.96. The results will help to calculate better estimates of the rates of suicides and of unintentional deaths by poison.
CITATION STYLE
Frize, M., Martirosyan, H., Subaskaran, J., McFaull, S. R., Skinner, R., & Tiv, M. (2016). Classification of undetermined deaths by poisoning: Suicidal or unintentional. In IFMBE Proceedings (Vol. 57, pp. 814–817). Springer Verlag. https://doi.org/10.1007/978-3-319-32703-7_159
Mendeley helps you to discover research relevant for your work.