Cross Disciplinary Consultancy to Bridge Public Health Technical Needs and Analytic Developers: Negation Detection...

Area of Science:

• Public Health
• Health Informatics
• Natural Language Processing

Background:

• The International Society for Disease Surveillance (ISDS) initiative aims to bridge public health practice and data analytics.
• Funding from the Defense Threat Reduction Agency supports consultancies on critical public health issues.
• Previous work included asthma exacerbation prediction and asyndromic surveillance tools.

Purpose of the Study:

• To define a roadmap for developing algorithms, tools, and datasets.
• To enhance the capabilities of text processing algorithms for negation detection.
• To improve the identification of negated terms in free-text clinical data.

Main Methods:

• A consultancy focused on negation detection in free-text chief complaints and triage reports.
• Brought together public health practitioners, academia, and industry analytics solution developers.
• Explored the development of algorithms, tools, and datasets for this specific text processing task.

Main Results:

• Focused on defining a roadmap for future development in negation detection.
• Identified key areas for improving the accuracy of identifying negated terms in clinical text.
• Highlighted the need for specialized algorithms and datasets.

Conclusions:

• Accurate negation detection is crucial for improving public health surveillance from clinical notes.
• Further development of text processing algorithms and resources is required.
• Collaboration between public health and data science is essential for advancing surveillance capabilities.