The UK Health Security Agency (UKHSA) is exploring the potential of artificial intelligence (AI) to detect and investigate foodborne illness outbreaks. Researchers at UKHSA have assessed various AI models to analyze online restaurant reviews for symptoms and food mentions that could indicate a foodborne illness outbreak.
The goal is to identify outbreaks earlier and prevent more people from getting sick. UKHSA's Chief Data Officer, Professor Steven Riley, believes that AI can help identify the likely source of more foodborne illness outbreaks when combined with traditional epidemiological methods.
The agency is using large language models to trawl thousands of online reviews for information about symptoms like diarrhea, vomiting, and abdominal pain, as well as different food types people report eating. However, there are challenges to overcome, such as access to real-time data, variations in spelling and slang, and people misattributing their illness to a given meal.
UKHSA's work on AI-powered outbreak detection is part of a broader effort to strengthen local and national surveillance and support prompt responses to infectious disease outbreaks. The agency is also expanding the list of pathogens that must be reported by registered medical professionals and laboratories in England.
