Dataset Information

Leveraging machine learning to assess market-level food safety and zoonotic disease risks in China.

ABSTRACT: While many have advocated for widespread closure of Chinese wet and wholesale markets due to numerous zoonotic disease outbreaks (e.g., SARS) and food safety risks, this is impractical due to their central role in China's food system. This first-of-its-kind work offers a data science enabled approach to identify market-level risks. Using a massive, self-constructed dataset of food safety tests, market-level adulteration risk scores are created through machine learning techniques. Analysis shows that provinces with more high-risk markets also have more human cases of zoonotic flu, and specific markets associated with zoonotic disease have higher risk scores. Furthermore, it is shown that high-risk markets have management deficiencies (e.g., illegal wild animal sales), potentially indicating that increased and integrated regulation targeting high-risk markets could mitigate these risks.

SUBMITTER: Gao Q

PROVIDER: S-EPMC9755119 | biostudies-literature | 2022 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Leveraging machine learning to assess market-level food safety and zoonotic disease risks in China.

Gao Qihua Q Levi Retsef R Renegar Nicholas N

Scientific reports 20221215 1

While many have advocated for widespread closure of Chinese wet and wholesale markets due to numerous zoonotic disease outbreaks (e.g., SARS) and food safety risks, this is impractical due to their central role in China's food system. This first-of-its-kind work offers a data science enabled approach to identify market-level risks. Using a massive, self-constructed dataset of food safety tests, market-level adulteration risk scores are created through machine learning techniques. Analysis shows ...[more]

PMID: 36522373

Similar Datasets

Project description:BackgroundFood safety in the food market is one of the key areas of focus in public health, because it affects people of every age, race, gender, and income level around the world. The local and international food marketing continues to have significant impacts on food safety and health of the public. Food supply chains now cross multiple national borders which increase the internationalization of health risks. This systematic review of literature was, therefore, conducted to identify common public health risks related to food safety issues in the food market.MethodsAll published and unpublished quantitative, qualitative, and mixed method studies were searched from electronic databases using a three step searching. Analytical framework was developed using the PICo (population, phenomena of interest, and context) method. The methodological quality of the included studies was assessed using mixed methods appraisal tool (MMAT) version 2018. The included full-text articles were qualitatively analyzed using emergent thematic analysis approach to identify key concepts and coded them into related non-mutually exclusive themes. We then synthesized each theme by comparing the discussion and conclusion of the included articles. Emergent themes were identified based on meticulous and systematic reading. Coding and interpreting the data were refined during analysis.ResultsThe analysis of 81 full-text articles resulted in seven common public health risks related with food safety in the food market. Microbial contamination of foods, chemical contamination of foods, food adulteration, misuse of food additives, mislabeling, genetically modified foods (GM foods), and outdated foods or foods past their use-by dates were the identified food safety-related public health risks in the food market.ConclusionThis systematic literature review identified common food safety-related public health risks in the food market. The results imply that the local and international food marketing continues to have significant impacts on health of the public. The food market increases internationalization of health risks as the food supply chains cross multiple national borders. Therefore, effective national risk-based food control systems are essential to protect the health and safety of the public. Countries need also assure the safety and quality of their foods entering international trade and ensure that imported foods conform to national requirements.

Project description:The rapid detection of pathogenic strains in food products is essential for the prevention of disease outbreaks. It has already been demonstrated that whole-metagenome shotgun sequencing can be used to detect pathogens in food but, until recently, strain-level detection of pathogens has relied on whole-metagenome assembly, which is a computationally demanding process. Here we demonstrated that three short-read-alignment-based methods, i.e., MetaMLST, PanPhlAn, and StrainPhlAn, could accurately and rapidly identify pathogenic strains in spinach metagenomes that had been intentionally spiked with Shiga toxin-producing Escherichia coli in a previous study. Subsequently, we employed the methods, in combination with other metagenomics approaches, to assess the safety of nunu, a traditional Ghanaian fermented milk product that is produced by the spontaneous fermentation of raw cow milk. We showed that nunu samples were frequently contaminated with bacteria associated with the bovine gut and, worryingly, we detected putatively pathogenic E. coli and Klebsiella pneumoniae strains in a subset of nunu samples. Ultimately, our work establishes that short-read-alignment-based bioinformatics approaches are suitable food safety tools, and we describe a real-life example of their utilization.IMPORTANCE Foodborne pathogens are responsible for millions of illnesses each year. Here we demonstrate that short-read-alignment-based bioinformatics tools can accurately and rapidly detect pathogenic strains in food products by using shotgun metagenomics data. The methods used here are considerably faster than both traditional culturing methods and alternative bioinformatics approaches that rely on metagenome assembly; therefore, they can potentially be used for more high-throughput food safety testing. Overall, our results suggest that whole-metagenome sequencing can be used as a practical food safety tool to prevent diseases or to link outbreaks to specific food products.

Project description:BackgroundAs one of the serious public health issues, vaccination refusal has been attracting more and more attention, especially for newly approved human papillomavirus (HPV) vaccines. Understanding public opinion towards HPV vaccines, especially concerns on social media, is of significant importance for HPV vaccination promotion.MethodsIn this study, we leveraged a hierarchical machine learning based sentiment analysis system to extract public opinions towards HPV vaccines from Twitter. English tweets containing HPV vaccines-related keywords were collected from November 2, 2015 to March 28, 2016. Manual annotation was done to evaluate the performance of the system on the unannotated tweets corpus. Followed time series analysis was applied to this corpus to track the trends of machine-deduced sentiments and their associations with different days of the week.ResultsThe evaluation of the unannotated tweets corpus showed that the micro-averaging F scores have reached 0.786. The learning system deduced the sentiment labels for 184,214 tweets in the collected unannotated tweets corpus. Time series analysis identified a coincidence between mainstream outcome and Twitter contents. A weak trend was found for "Negative" tweets that decreased firstly and began to increase later; an opposite trend was identified for "Positive" tweets. Tweets that contain the worries on efficacy for HPV vaccines showed a relative significant decreasing trend. Strong associations were found between some sentiments ("Positive", "Negative", "Negative-Safety" and "Negative-Others") with different days of the week.ConclusionsOur efforts on sentiment analysis for newly approved HPV vaccines provide us an automatic and instant way to extract public opinion and understand the concerns on Twitter. Our approaches can provide a feedback to public health professionals to monitor online public response, examine the effectiveness of their HPV vaccination promotion strategies and adjust their promotion plans.

Dataset Information

Leveraging machine learning to assess market-level food safety and zoonotic disease risks in China.

Publications

Leveraging machine learning to assess market-level food safety and zoonotic disease risks in China.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets