Are "bytes" and "blocks" the secret ingredients to transforming food safety?
Chances are you or someone you know has suffered from foodborne illness. According to the World Health Organization, contaminated food can cause more than 200 diseases.
Article information and share options
In the U.S. alone, one in six people are affected by foodborne diseases each year – resulting in 128 000 hospitalizations, 3 000 deaths, and USD 9 billion in medical costs. Add to the damage another USD 75 billion lost annually in contaminated food which must be recalled and discarded, and it is not difficult to see why ensuring a safe food supply is considered by many to be one of society’s great challenges.
Food Safety incidents have occurred since ancient times and despite much progress they continue today. Thanks to the increasingly global and complex food supply chains, new threats are expected. The U.S. Centers for Disease Control and Prevention has warned in recent years that a growing share of foodborne illness outbreaks are being caused by imported foods, including seafood, spices and fresh produce. The chain is only as safe as its weakest link.
The global supply chain gives us increased choices and lower costs, but also makes it nearly impossible to prevent foodborne illness. New scientific and technological methods are needed to mitigate the safety hazards within the system and to provide transparency. Can big data solutions help us gain insights to make proactive decisions rather than reactionary decisions? IBM, who has a long legacy in driving insights from big data, believes that big data can be a powerful tool for dealing with food safety issues around the globe. Two such tools on the immediate horizon are metagenomics and blockchain.
Food safety practices today
Today, food safety is about detecting pathogens - the bacteria, viruses, or other microorganisms that can cause illness, disease and death.
A recent article in Nature contributed by Mars, Incorporated explains the current situation. Even with industry’s best practices, today’s testing methods are inherently limited in their ability to detect risks. The Hazard Analysis and Critical Control Point (HACCP) approach introduced in the 1970s is a systematic method of examining every point of a production process to understand what hazards exist and how they could be controlled. This is currently the best available system for managing food safety. But HACCP relies on the experience of the practitioner; it does not incorporate numerical risk assessment methods. Factories can continually carry out microbiological testing to monitor samples, but an absence of pathogens and contaminants in the tests does not always tell the whole story. It would take 300 samples from every single delivery or batch to have a 95% probability that no more than 1% is contaminated. But this 1% still represents a high-risk level for high-volume manufacturers, and 300 samples is an unrealistic level of operational testing.
The situation is even worse if we consider that these statistics assume an underlying random distribution of organisms throughout a given product, however we know that bacterial contamination typically doesn’t behave this way; bacterial distributions tend to be clumped. What's more, current testing methods for pathogens can test for only one organism at a time, so critical information is easy to miss.
Looking at current approaches to food safety, it is clear that new tools are needed. Large data sets and big data analytics are beginning to be used in food safety and quality, but to date they primarily focus on providing improved root cause and retrospective analysis. In addition to the reactive approaches, we need proactive tools and predictive analytics. But how can we test for threats that aren’t yet known?
A new approach: Metagenomics
The motivation is a simple, but novel, conceptual twist - focus on health, rather than disease, using the emerging science of metagenomics.
Genomics is the study of the complete genome of a single organism, such as the human genome. In contrast, metagenomics is the study of genomes from environmental samples where multiple organisms naturally live in symbiotic relationships (as opposed to unnatural, artificial media of laboratory petri dishes).
Communities of living microorganisms exist in all foods (and everywhere else for that matter). Instead of isolating individual disease-causing pathogens, such as E. coli and salmonella, the idea is to understand what a healthy community looks like under normal conditions by using metagenomics to analyze its microbial genetic material from samples gathered along the food supply chain. It’s like using a canary in a coal mine as an early indicator of potential danger. The hypothesis is that if we can quantitatively understand what that community looks like under normal conditions, then changes in those conditions can alert us that something may be wrong.
It may be possible that the regular use of one microbiome test could take the place of many separate tests for distinct hazards and provide early detection of hazardous situations before they cause substantial health or economic impact. Safety practices that are now reactive could shift to pro-active prevention, and current practices for prevention could become more precise and data-driven.
Sequencing the Food Supply Chain Consortium
Recognizing that analyzing the metagenomics of the global food supply chain is an ambitious effort requiring the expertise of many, Mars, Incorporated and IBM co-founded the Sequencing the Food Supply Chain Consortium.
The Consortium is using a scientific approach that combines massive amounts of data, multi-stakeholder expertise and cutting-edge cloud-based analytic services to focus on surveillance, risk assessment and diagnosis of foodborne pathogens.
Researchers are harvesting and sequencing the DNA and RNA of simple food samples to determine where anomaly and selection occur when paired with common organisms, toxins, and heavy metals. The goal is to learn and understand how microorganisms in raw food materials react with each other, and then identify how these reactions stabilize or mutate in food chain environments. Establishing the baseline of stable microbiomes will help determine the interactions which cause mutations leading to foodborne illnesses, as well as identify the good bacteria (ie probiotics) that create disease-resistant communities. We believe that by using metagenomic analysis we will not only be able to detect the unknown issues creating food safety problems but will also be able to address a host of additional problems facing the food supply chain. For example:
- Food traceability becomes easier to manage.
- Food fraud can be better detected by automatic verification of a food’s identity and source, thus providing the means to quickly identify unsafe counterfeit ingredients which would cause health risks to consumers. This will also save billions of dollars.
- Antibiotic resistance and the potential devastating impact to human health can be better managed.
- Factory microbiome management will focus on how to keep factories healthy and productive.
- Spoilage (expiration) will be managed more proactively through understanding of spoilage mechanisms thus making a major impact on food sustainability through waste reduction.
- Probiotic applications will be more targeted and robust.
- Selection in food processing (eg kill steps) will be managed in a more scientific way, thus ensuring that food quality and nutritional content are maximised.
Metagenomic data and analysis will be a foundational platform for creating new food sciences that will significantly help manage food safety issues and will have a profound impact on food sustainability.
Another new technology addresses the challenge of creating a more transparent, authentic and trustworthy digital record of the journey that the physical products take across the multi-tiered supply chain.
A novel technology: Blockchain
Food authentication and supply chain tracking are critical steps to quickly finding and helping address sources of contamination around the world. Early indications are that blockchain presents an innovative new way to accomplish these goals.
Blockchain is a shared, immutable ledger for recording the history of transactions. It fosters a new generation of transactional applications that establish trust, accountability and transparency. It provides a permanent record of transactions which are then grouped in blocks that cannot be altered. It could serve as an alternative to traditional paper tracking and manual inspection systems, which can leave supply chains vulnerable to inaccuracies.
With blockchain, food products can be digitally tracked from an ecosystem of suppliers to store shelves and ultimately to consumers. When applied to the food supply chain, digital product information such as farm origination details, batch numbers, factory and processing data, expiration dates, storage temperatures and shipping detail are digitally connected to food items and the information is entered into the blockchain along every step of the process. Each piece of information provides critical data points that could potentially reveal food safety issues with the product. The information captured in each transaction is agreed upon by all members of the business network; once there is a consensus, it becomes a permanent record that can’t be altered. This helps assure that all information about the item is accurate.
Walmart and the pork supply chain in China
Walmart is collaborating with Tsinghua University and IBM to provide better food tracking and consumer safety throughout its supply chain. Scientists from IBM Research are among the leading-edge technologists in the forefront of the rapid evolution of blockchain. Working alongside top talent in transaction security and authentication technology from Tsinghua University and with Walmart’s expertise in supply chain, logistics and food safety, they are creating a new model for food traceability, supply chain transparency and auditability using IBM blockchain based on the open-source Linux Foundation Hyperledger Project fabric. This can help retailers like Walmart better supervise and manage the shelf life of products and reduce the risk of non-compliance in individual stores, ensuring that consumers are getting authenticated, safe products, which is exactly the answer required.
Metagenomics and blockchain convergence
These are two examples of the use of big data technologies to improve our food system. Disruptive technologies in metagenomics and blockchain offer new breakthrough approaches to the significant challenges in food safety. In addition, IBM Research is working on numerous other big data solutions to dramatically transform food safety and food sustainability practices throughout the world. These solutions will help to form a portfolio of offerings that is synergistic and will act as a powerful catalyst to help companies manage food supply chains in ways that were not even dreamed about in the past.