Flavour and Food Products Developed with AI technology

McCormick, a global leader in flavor, and IBM today publicly announced their ongoing research collaboration to pioneer the application of artificial intelligence (AI) for flavor and food product development.

Using IBM Research AI for Product Composition, McCormick is ushering in a new era of flavor innovation and changing the course of the industry. Product developers across McCormick’s global workforce will be able to explore flavor territories more quickly and efficiently using AI to learn and predict new flavor combinations from hundreds of millions of data points across the areas of sensory science, consumer preference and flavor palettes. This proprietary, cutting-edge technology sets McCormick apart in its ability to develop more creative, better tasting products and new flavor experiences across both its Consumer and Flavor Solutions business units.

McCormick expects to launch its first AI-enabled product platform, “ONE,” by mid-2019, with a set of initial one-dish Recipe Mix flavors including Tuscan Chicken, Bourbon Pork Tenderloin and New Orleans Sausage. The company’s flavor developers created this product platform by combining IBM’s expertise in AI and machine learning with McCormick’s 40+ years of proprietary sensory science and taste data, which includes decades of past product formulas and millions of data points related to consumer taste preferences and palettes. AI has enabled McCormick’s product developers access to an expanded portfolio of flavor profiles that enhances their creativity. The new ONE platform was specifically developed to deliver family-favorite flavors with the ability to season both the protein and vegetable. The new seasoning blends expect to be on U.S. retail shelves by late spring.

“McCormick’s use of artificial intelligence highlights our commitment to insight-driven innovation and the application of the most forward-looking technologies to continually enhance our products and bring new flavors to market,” said McCormick Chairman, President and CEO Lawrence Kurzius. “This is one of several projects in our pipeline where we’ve embraced new and emerging technologies.”

As a world leader in artificial intelligence software, services and technology for business, IBM is focused on working with clients and enterprises across many industry sectors to help advance data-driven technologies that push markets forward.

“IBM Research’s collaboration with McCormick illustrates our commitment to helping our clients and partners drive innovation across industries,” said Kathryn Guarini, VP, Industry Research, IBM. “By combining McCormick’s deep data and expertise in science and taste, with IBM’s AI capabilities, we are working together to unlock the bounds of creativity and transform the food and flavor development process.”

Pairing McCormick’s global expertise, particularly that of its research and product development teams, with leading AI research helped McCormick accelerate the speed of flavor innovation by up to three times and deliver highly effective, consumer-preferred formulas. Through the ONE platform as well as several other projects in the pipeline, McCormick’s product developers are now using AI to unlock creativity, access new insights and share data with their peers around the world. The company plans to scale this technology globally by 2021.

Source: IBM

Advertisements

Food Company Leverages Artificial Intelligence to Make Healthy Gummies

Catherine Lamb wrote . . . . . . . . .

One of the hard parts of being an adult — and there are many — is that gummies are no longer considered an acceptable afternoon snack. But today Journey Foods is unveiling a sort of reimagination of the fruit snack; one which is packed with nutrients and also promotes biodiversity.

The Chicago-based startup launched about 18 months ago, though the inspiration goes back much further. Founder and CEO Riana Lynn, who had previously started a food traceability company and served as an entrepreneur in residence at Google, got the idea for Journey Foods after she spent a stint traveling around the world. Inspired by the widespread biodiversity she saw (and tasted), Lynn decided to bring some of her most nutrient-rich findings — like baobab and seaweed — back to the U.S. and transform them into healthy snacks.

The first product is the aforementioned fruit snacks, which Lynn calls “Micro-Foods.” Not only does that sound more legit than “gummy candy,” the name also communicates the caliber of nutrition research and technology that goes into Journey Foods’ products. “We are more of a hybrid biotech/CPG company,” Lynn told me over the phone. “I guess you could just call it food tech.”

Tech indeed. The company has three patents pending on the nutrition biotech that powers their products. They’re also testing out different sugar technologies, so people with dietary restrictions can still eat the Micro Foods.

The fruit chews, which come in Strawberry Chia Seed and Mango and Cayenne Spice, cost less than $1.50 per single-serve pack, which is slightly more expensive than many natural fruit snacks — but not much. “We’re really focused on accessibility,” Lynn told me. Journey Foods will make the bulk of their money from B2B sales and custom product creation for big CPG companies.

As of now, the Micro-Foods are available on Amazon and the Journey Foods website, and are being tested in 80 retail locations around the country. They’re also in a variety of corporate offices, and Lynn told me they’ll debut the Micro-Foods in select hospitals later this year.

Journey Foods also has a B2B product development tool called JourneyAI. It’s essentially an AI-powered database that helps the startup identify and catalog ingredients which could be used to make nutrient-dense foods, speeding up the trial and error of product R&D with their third-party CPG partners.

Journey Foods’ chews take advantage of a few food trends in one fell swoop. First of all, they’re capitalizing off of growing consumer demand for healthier foods, specifically snack foods. The global healthy snacks market was valued at more than $23 billion in 2018, according to Grand View Research, and doesn’t show any signs of slowing. Consumers are also getting more adventurous in their snacking, looking for new, exotic flavors. Finally, adding the buzzword “biodiversity” to their marketing, legit as it is, could help Journey Foods capture more ethically-motived consumers.

The decision to target B2B partners is also a smart play, specifically in hospitals and tech company offices. At the former, the Micro-Foods could help patients, specifically kids, get the nutrients they need. And at tech offices, which are renowned for their gigantic snack walls, Journey Foods can offer a healthier alternative to, say, gummy bears.

Lynn told me that next, Journey Foods will add new flavors to its lineup of Micro Foods products, incorporating ingredients like marine greens, probiotics, and vegetables. Down the road, the company will launch new nutrient-dense products outside the gummy realm. Journey Foods is currently in the midst of a fundraising round, with participation from Backstage Capital and other VC firms. The startup was also the first chosen to join the Soylent Innovation Lab earlier this month and was awarded a $15,000 grant and office space in L.A. Hopefully that translates into a lot more gummy chews that we don’t have to feel guilty about.

Source: The Spoon

Appliance Maker Awarded Comprehensive Patent For Kitchen-Centric Computer Vision System

wrote . . . . . . . . .

Over the past couple years, there’s been what can only be described as an intellectual property land grab in the world of computer vision as Google, Amazon and Microsoft file more patents in an effort to establish foundations from which to launch an innumerable amount of AI-driven products and services over the next decade or more.

But as it turns out, big appliance isn’t quite ready to cede the entirety of this fast-moving space to big tech, especially when it comes to their home turf, err, the home. Last week appliance giant Whirlpool was awarded what appears to be a a fairly broad patent for a computer vision system to track behavior of people and objects within a home and enable all sorts of potential scenarios such as auto-replenishment, guided cooking and more.

The patent, entitled “Interaction recognition and analysis system”, starts with fairly broad language to describe a system that utilizes image capture technology to identify the contents of a person’s hand and trigger a range of reactions from appliances within the home. From there, the descriptions and related diagrams get more specific, illustrating how the system could enable a variety of specific scenarios such as suggesting a recipe or ordering more laundry detergent from an online retailer.

One such example is a description of how the system could be used in a fridge across a number of different “access regions” that correspond to the different storage areas within a fridge. The system would know the freezer drawer from the fresh food drawer and be able to monitor corresponding changes in frozen food or fresh food inventory.

In another example, the patent describes how the system could be used to monitor activity in the cooking cavity of a microwave or oven. It could also determine the level of doneness of the food. From there, it could initiate a specific timer or a series of cooking processes. In other words, the system could serve as a foundation for guided cooking.

Of course, we shouldn’t be all that surprised that Whirlpool would be delving into computer vision-powered services as suggested by this patent. At the last CES, the company showed off ingredient recognition capabilities of the new Yummly app and has been showing off concept demos utilizing computer vision-powered interfaces since CES 2014.

Still, this new patent is intriguing and I am very interested to see where Whirlpool takes this type of technology in the future. Because of the broadness of the patent, it could serve as the IP underpinning for what is essentially a kitchen operating system, where fridges, ovens and washing machines are not only in sync with the consumers that use them, but also with each other and all the various systems in the home. The end result could be a more sentient and anticipatory kitchen in place of the smart appliances operating independently of one another that occupy our homes today.

Sure, that’s a lot to infer from one patent, but that’s the direction Whirlpool, BSH and others have been heading in the last couple years. My guess is 2019 will see significant moves in this direction, starting in just a couple weeks at CES in Las Vegas.

Source: The Spoon


Read also:

Is Amazon Considering Making A Smart Fridge? Probably Not (But Maybe) . . . . .

New Computer Model Enables Researchers to Predict Course of Prostate Cancer

How does a normal cell turn into a deadly cancer? Seeking an answer to this question, and working alongside other international working groups, researchers from Charité – Universitätsmedizin Berlin examined the tumor genomes of nearly 300 prostate cancer patients. Their findings describe the ways in which changes in the prostate cells’ genetic information pave the way for cancer development. Using a newly-developed computer model, it is now possible to predict the course of the disease in individual patients. It is hoped this will enable clinicians to develop tailor-made treatments. On Monday, 10 December 2018, the results of this study were published in the latest issue of the journal Cancer Cell*.

In Germany, prostate cancer is the most common malignancy in men, with close to 60,000 new cases diagnosed every year. These Tumors are usually slow-growing, meaning that not all patients require immediate treatment. Until recently, physicians had been unable to distinguish between benign and aggressive forms of the disease, particularly when dealing with tumors diagnosed at an early stage in the disease process.

Working alongside a number of other research groups from within Germany and abroad, Charité-based researchers helped to develop criteria that would make this type of classification possible. To do so, they studied the molecular profiles of close to 300 prostate tumors. They sequenced the information encoded within the cells’ genetic material, recorded chemical changes to the genetic code, and measured the activity of specific genes within cancerous tissues. An analysis of their data has shed light upon the temporal order of mutational events involved in the development of prostate cancer. “We were able to identify tumor subtypes that progress at different rates and therefore require different types of treatment,” says one of the study’s lead authors, Prof. Dr. Thorsten Schlomm, Director of Charité’s Department of Urology.

He adds: “We now know which of these mutations occur first, initiating the process of change from prostate cells to tumor cells, and which of them are more likely to follow later.” The researchers then used these results to develop a computer-based model capable of predicting the likely course of the disease in individual patients. “When an individual patient’s tumor shows a specific mutation, we are now able to predict which mutation is likely to follow, and how good the patient’s prognosis is,” explains Prof. Schlomm. “Our team is currently busy incorporating our computer model into the treatment process at Charité. This will enable clinicians to model a particular treatment’s likelihood of success. As for the timescale involved, we expect it will take two to three years for this algorithm-based method to become clinical routine.”

In an effort to improve the reliability of prognoses, the research consortium is planning to spend the next few years collating additional data on thousands of patients, which they will then use to further develop and enhance their computer model. They will achieve this by working with Berlin’s newly established urology network (Hauptstadt-Urologie-Netzwerk), which brings together urology specialists from Charité and private practice. Their ultimate aim is to make it easier for physicians to decide on the most suitable treatments for individual patients.

Source: Charité – Universitätsmedizin Berlin

The Power of Advanced NGS Technology in Routine Pathogen Testing of Food

Stephanie Pollard wrote . . . . . . . . .

The food industry is beginning to transition into an era of big data and analytics unlike anything the industry has ever experienced. However, while the evolution of big data brings excitement and the buzz of new possibilities, it also comes coupled with an element of confusion due to the lack of tools for interpretation and lack of practical applications of the newly available information.

As we step into this new era and begin to embrace these changes, we need to invest time to educate ourselves on the possibilities before us, then make informed and action-oriented decisions on how to best use big data to move food safety and quality into the next generation.

Stephanie Pollard will be presenting “The Power of Advanced NGS Technology in Routine Pathogen Testing” at the 2018 Food Safety Consortium | November 13–15One of the big questions for big data and analytics in the food safety industry is the exact origins of this new data. Next Generation Sequencing (NGS) is one new and disruptive technology that will contribute significantly to a data explosion in our industry.

NGS-based platforms offer the ability to see what was previously impossible with PCR and other technologies. These technologies generate millions of sequences simultaneously, enabling greater resolution into the microbial ecology of food and environmental surfaces.

This represents a seismic shift in the food safety world. It changes the age-old food microbiology question from: “Is this specific microbe in my sample?” to “what is the microbial makeup of my sample?”

Traditionally, microbiologists have relied on culture-based technologies to measure the microbial composition of foods and inform risk management decisions. While these techniques have been well studied and are standard practices in food safety and quality measures, they only address a small piece of a much bigger microbial puzzle. NGS-based systems allow more complete visibility into this puzzle, enabling more informed risk management decisions.

With these advances, one practical application of NGS in existing food safety management systems is in routine pathogen testing. Routine pathogen testing is a form of risk assessment that typically gives a binary presence/absence result for a target pathogen.

NGS-based platforms can enhance this output by generating more than the standard binary result through a tunable resolution approach. NGS-based platforms can be designed to be as broad, or as specific, as desired to best fit the needs of the end user.

Imagine using an NGS-based platform for your routine pathogen testing needs, but instead of limiting the information you gather to yes/no answers for a target pathogen, you also obtain additional pertinent information, including: Serotype and/or strain identification, resident/transient designation, predictive shelf-life analysis, microbiome analysis, or predictive risk assessment.

By integrating an NGS-based platform into routine pathogen testing, one can begin to build a microbial database of the production facility, which can be used to distinguish resident pathogens and/or spoilage microbes from transient ones. This information can be used to monitor and improve existing or new sanitation practices as well as provide valuable information on ingredient quality and safety.

This data can also feed directly into supplier quality assurance programs and enable more informed decisions regarding building partnerships with suppliers who offer superior products.

Similarly, by analyzing the microbiome of a food matrix, food producers can identify the presence of food spoilage microbes to inform more accurate shelf-life predictions as well as evaluate the efficacy of interventions designed to reduce those microbes from proliferating in your product (e.g. modified packaging strategies, storage conditions, or processing parameters).

Envision a technology that enables all of the aforementioned possibilities while requiring minimal disruption to integrate into existing food safety management systems. NGS-based platforms offer answers to traditional pathogen testing needs for presence/absence information, all the while providing a vast amount of additional information. Envision a future in which we step outside of our age-old approach of assessing the safety of the food that we eat via testing for the presence of a specific pathogen. Envision a future in which we raise our standards for safety and focus on finding whatever is there, without having to know in advance what to look for.

Every year we learn of new advancements that challenge the previously limited view on the different pathogens that survive and proliferate on certain food products and have been overlooked (e.g., Listeria in melons). Advanced NGS technologies allow us to break free of those associations and focus more on truly assessing the safety and quality of our products by providing a deeper understanding of the molecular makeup of our food.

Source: Food Safety Tech