Novel use of Machine Learning to prevent consumer injuries and product recalls resulting from errors in food supply-chain data

当你的免疫系统对一种无害的食物蛋白质产生反应时，就会发生食物过敏，这通常是由于产生了针对这种特定蛋白质的免疫球蛋白E （IgE）抗体。当这种蛋白质被摄入时，抗体会识别这种蛋白质并与之结合，导致组胺等物质的释放，从而导致过敏症状。立即反应（进食后两小时内），通常涉及IgE抗体，并出现典型症状，如瘙痒皮疹、肿胀，在某些情况下还会出现呕吐和腹泻。症状的严重程度各不相同，在最坏的情况下，可能导致危及生命的问题，包括呼吸困难和晕倒，也被称为过敏反应。LiberEat提供了唯一的数字资源，提供产品扫描，产品过滤，菜单过滤，和一个食谱银行都在一个地方。我们允许多个配置文件，这可以帮助用户准确地看到谁可以通过我们的应用程序吃每一种产品，菜肴和食谱。我们确保我们的数据来自可靠的来源，我们从不筛选，我们只与直接提供信息的合作伙伴餐厅合作。LiberEat最近被评为Tech Nation明日之星3.0苏格兰和英国比赛的获胜者，并入选Tech Nation的应用人工智能项目。该项目现在将增加检测申报过敏原差异的能力，识别存在较高错误风险的情况，并允许我们的过敏原识别技术集成到供应商过敏原管理系统中。这项技术将成为一个至关重要的安全网，用来捕捉我们从经验中知道的由人为错误造成的错误。该提案已经得到了几家食品服务连锁店和食品生产商的验证，其中一些将参与项目期间的测试。

Food allergies occur when your immune system reacts to a harmless food protein, which are often due to the creation of Immunoglobulin E (IgE) antibodies which are specific for that particular protein. When the protein is consumed, the antibodies recognise the protein and link to it, causing the release of substances such as histamine, which will result in allergic symptoms. An immediate reaction (within two hours of eating the food), usually involves IgE antibodies, and the development of classic symptoms such as an itchy rash, swelling and in some cases vomiting and diarrhoea. Symptoms can vary in severity and can, in the worst case, lead to life threatening problems including difficulty in breathing and collapse, also known as anaphylaxis.LiberEat provides the only digital resource to offer product scanning, product filtering, menu filtering, and a recipe bank all in one place. We allow multiple profiles, which helps users to see exactly who can eat every product, dish and recipe via our app. We ensure our data comes from reliable sources and we never screen scrape, we only work with partner restaurants who give us the information directly. LiberEat was recently named as a winner of the Tech Nation Rising Stars 3.0 Scotland and UK Competitions and selected for Tech Nation's applied AI program.This project will now add the ability to detect discrepancies in declared allergens, identify situations where there is a higher risk of errors, and allow our allergen identification technology to be integrated into supplier allergen management systems. The technology will become a vital safety net to catch the mistakes we know from experience are being made through human error. The proposal has been validated by several food service chains and food producers, a number of which will be involved in testing during the project.