FoodScan

About

Our company: PolyChord is a spin-out from Cambridge University Astrophysics department, with two Professors of Astrophysics on board in the company. Our core technology is uniquely able to cope with complex, messy, noisy and multivariate data - it was invented to analyse the microwave radiation present after the Big bang, effectively tackling a 10 billion year large dataset. FoodScan: In collaboration with Prof. Seamus Fanning (Queen's University Belfast) and Prof. Larry Goodrich (University of Guelph, Canada), we have brought together next-generation data science, with contemporary microbiology knowledge, in an Innovate UK/National Research Council of Canada £400K funded project (started Nov 2019). We currently have a working tool which is able to give more accurate predictions of pathogen growth/death than the WHO standard, with quantitative error bars on the estimates - not only does this provide a more accurate prediction, but also a clear idea of exactly how accurate that prediction is. Currently, the industry norm for predicting pathogen growth is based on old maths (Akaike), and lab strain microbes that simply do not behave the same as those grown from foodstuffs. Next steps: UK Consumers currently throw away 6.7 bn tonnes of food a year. Even though "best before" dates are about quality rather than safety, many consumers will throw a product away based on this date alone - exemplified by the 1.7 bn tonnes of food thrown away with its packaging intact, each year. At PolyChord, we want to unify the predictive tool we've created to assess a food's ongoing safety, with sensory analysis data which is able to tell a seller when a consumer would no longer want to eat their product. By knowing the level of microbes in a product, and the conditions under which it is maintained, we will be able to give an accurate answer, with quantitative error bars, to the question; "when does my product quality truly drop off?". Based on the information given to us by our industry expert partners, many foods currently reach their "microbial limit" before the sensory analysis score drops. The benefits of a predictive tool for sensory analysis scores would therefore be two-fold: 1. The best before date could be safely extended, reducing food waste by households and thus bringing down the CO2 emissions inherent in the UK's food wastage problem - helping the UK government achieve its net-zero target. 2. Ensuring a consistently high-quality product, right up to the Best Before date. We are now seeking an industry partner to bring this forward, one who has a wealth of data and expertise to grow a next-generation tool from the currently tightly focused FoodScan, to a revolutionary product bespoke to the needs of their company. We can also bring in Food Science partners we have already been working successfully alongside if this is thought desirable.