Using machine learning-based Natural Language Processing to quantify emergency department presentations related to suicide or self-harm in the Australian Capital Territory

Key Points

• The aim is to enhance the identification of emergency department presentations related to suicide and self-harm.
• Utilized machine learning and natural language processing for analysis
• Focused on emergency department presentations in the Australian Capital Territory
• Evaluated existing identification methods for effectiveness
• Less than 40% of suicide or self-harm cases were identified using current methods
• The new identification method significantly improves accuracy
• Increased potential for better understanding and addressing suicide and self-harm issues