The UK Health Security Agency (UKHSA) is considering the possibility of using artificial intelligence (AI) to detect and investigate the outbreak of foodborne diseases.
UKHSA experts evaluated various types of AI for their ability to detect and classify texts in online restaurant reviews.
Food-borne gastrointestinal disease is a huge burden in the UK, causing millions of people to feel sick every year. However, it is estimated that most cases are not formally diagnosed, pose the challenges of traditional monitoring methods.
Estimates from the Food Standards Agency (FSA) show that there are 2.4 million diseases per year due to pathogens associated with foodborne diseases. Of these, 16,400 cases have been treated in hospitals, 180 have died, costing British society £10.4 billion.
Potential data issues
UKHSA technical experts and scientists look at various language models and assess their ability to trawl thousands of online reviews for information on symptoms that may be related to foodborne illnesses, such as diarrhea, vomiting, abdominal pain, and report dietary food types of food types of food types of food types, including meat and fish.
Scientists believe that gathering such information can be routine, providing more information about the rate of disease that is not captured by current systems, as well as clues as possible sources and causes of occurrence.
But a The research raised a challenge This needs to be overcome, especially with the focus on accessing real-time data. Experts said there are limitations to the use of restaurant review data, which highlights the need for careful interpretation of results.
While it is possible to use this approach to collect general information about the types of food people eat, it is possible to collect general information about the types of food that may be related to the disease, it is difficult to determine which specific ingredients or other linked factors. Spell variation and slang use were also identified as potential challenges, identifying people who misdirect illness into specific diets.
Supports existing monitoring
The researchers applied efforts to the review’s Yelp Open Dataset and included information about symptoms and food. They randomly sampled 3,000 reviews to manually annotate using a protocol designed by UKHSA epidemiologists. This left 1,148 gastrointestinal-related reviews.
Only a small portion of all foods consumed are catering foods that may generate reviews on websites like Yelp. Restaurant review surveillance may disproportionately capture the illness among high-income people. It is unclear whether income is the cause of people who post reviews online.
The duration of pathogen incubation may vary from hour to hour. This means that people can accidentally seduce symptoms into a certain food facility if they are not actually the source.
“Using AI in this way, combined with traditional epidemiological methods, can help identify the source of more food-borne outbreaks to prevent more people from tackling food disease outbreaks before more people get sick.
The UKHSA is also expanding the pathogens that have registered UK medical professionals and laboratories.
The aim is to strengthen local and national surveillance and support rapid response to infectious disease outbreaks.
Starting April 6, medical professionals should notify the UKHSA if they suspect that the patient has Middle East Respiratory Syndrome (MERS), influenza of costellate animal origin, Kretzfeld Jacob’s disease, and several other conditions.
Laboratories testing human samples in the UK should report 10 more drugs, including norovirus, tick-borne encephalitis virus, toxoplasma, tritinera and Yersinia.
(To sign up for a free food safety news subscription, click here))
