The mathematics are not simple, but IBM scientists have come up with methodology for analyzing retail scanner data from grocery stores against the locations of confirmed cases of foodborne illness to dramatically speed up foodborne illness investigations. Time is the enemy in foodborne illness investigations. In the paper announcing their research, the IBM scientists point to the 2011 outbreak of E. coli O104:H4 in Europe —where over 4,000 were sickened and 50 people from 16 counties died — and it took over 60 days to identify imported Egyptian fenugreek seeds as the source. By that time  all the sprouts produced by the seeds were consumed and economic damages to farmers not in anyway involved topped 150 million Euros. bigdata_406x250“Rapid identification of contaminated items is vital to minimize illness and loss in an outbreak,” say the IBM scientists. Their study, entitled “From Farm to Fork: How Spatial-Temporal Data can Accelerate Foodborne Illness Investigation in a Global Food Supply Chain” was published in the Association for Computing Machinery’s Sigspatial Journal. “Spatial information of each component in the food distribution and supply chain can be used to define a network relationship between sources of contaminated food, wholesalers, retailers and consumers, and subsequent public health case reports,” IBM scientists say. “In this study, we demonstrate a new approach to accelerate the foodborne illness outbreak investigation. It is a computational technique that can 1) help identify possible sources of contamination in the early stages of a disease outbreak, or 2) make pro-active predictions on likely contamination sources before the onset of a potential outbreak.” “Leveraging retail scanner data with spatial information already collected at any grocery store/supermarket along with the confirmed geo-coded cases reported from the public health agency makes it possible to quickly identify a small set of ‘suspect’ products that should be tested in the laboratory and investigated further,” the study explains. IBM demonstrates in the study that as few as ten medical-examination reports of foodborne illnesse can narrow down the investigation to 12 suspected food products in just a few hours. Fewer illnesses and less economic loss result from more rapid identification of the contaminated food source. In applying its work to an actual E. coli outbreak in Norway with 17 confirmed cases, the IBM team used its methodology to analyze grocery scanner data with more than 2,600 possible food products and was able to create a short list of 10 possible contaminants. Further lab analysis of the 10 narrowed the source of contamination down to batch and lot numbers of a specific sausage. “When there’s an outbreak of foodborne illnesses, the biggest challenge facing public health officials is the speed at which they can identify the contaminated food source and alert the public,” said Kun Hu, public health research scientist at IBM Research –Almaden in San Jose, CA. “While traditional methods  like interviews and surveys are still necessary, analyzing big data from grocery store scanners can significantly narrow down the list of contaminants in hours for further lab testing,” Hu said. “Our study shows that Big Data and analytics can profoundly reduce investigation time and human error and have an huge impact on public health.” About 3,000 researchers work for IBM Research in 12 labs located on six continents.  They include six Nobel Laureates,  and winners of numerous other science metals. (To sign up for a free subscription to Food Safety News, click here.)