Researchers Envision Personalized AI Chefs
Hoping to be healthier, tastier, and more enjoyable at the table
Foods can let us enjoy the moment instantly — fulfill the hungry stomach and mind, and give us the energy to do things later. In the long run, it can potentially help us prevent ourselves from getting diseases such as cancer.
Combinations of foods using ML to introduce us Hyperfoods
Did you know that tea and garlic are a lot similar than we think? We probably never really think about it when we cook garlic and drink tea after lunch.
I read the article of Michael Bronstein, AI-designed “hyperfoods” can possibly help prevent cancer, and watched TEDTalk.
These were the dishes presented at their TEDTalk.
- Hypermuesli with green tea — grapes, apples, hazelnuts, oats, flaxseeds, soy milk, strawberries, raspberries.
- Hypergazpacho — tomatoes, red peppers, carrot, celery, cucumber, dill, garlic, red onion, wholewheat bread, cumin, black pepper, vinegar, olive oil, salt.
- Rye trofie (pasta) with hyperpesto — basil, garlic, walnuts, olive oil, salt.
- Hypersmoothie — pomegranate, orange, grapes, banana, ginger.
After reading the article and watching the talk, my late lunch was somewhat resonated with it — chopped cabbage, a small amount of meat, and sprinkled some garlic, black pepper, salt, cooked pasta, drizzled a good amount of olive oil, and topped it with three teaspoons of cream cheese in the bowl after cooking it in the microwave for about 2 minutes 50 seconds with high power.
The article talks about graph ML, graph neural networks (GNN). I found an article on this platform (How to get started with Graph Machine Learning), but the author also writes about it. Typical graph ML architecture is “ graph neural networks implement some form of message passing on the graph allowing different nodes to exchange information.”
What to consider to offer personalized dishes or recipes for the day?
It says, foods combination, personal preferences, and so on. It is inevitable to consider food allergies and other information.
Recipe design must be highly personalised taking into account one’s taste preferences — but also many other parameters such as dietary restrictions, genetics, disease history, and the gut microbiome.
Potentially healthier, tastier, and more effective ways of preparing and enjoying foods seem possible in the future.
I recently also read an article Healthy Aging and Food: It’s Personal by Gunnar De Winter on this platform, showing readers the longevity hotspots or the Blue Zone. The area covers five regions and countries (Ikaria of Greece, Okinawa of Japan, Sardinia of Italy, Loma Linda of CA USA, and Nicoya Peninsula of Costa Rica). People in these regions share some good mindsets of managing daily activities in general and their dietary habits.
Health, aging, and personalized dishes/nutrition for longevity in the future
Speaking of health conditions, foods, and Japan, there is an application that suggests a variety of healthy recipes using AI. The Japanese application provides the users healthy recipes of their choice that are selected particularly for their health conditions and preferences. It is from Oishi Kenko Inc., and a paid service for 1, 6, and 12 months. Users can sign up for a free trial. Currently, it seems they only offer it in Japanese. They officially released it in September 2021.
Why it was made
It was to rescue the Menu Refugees with health conditions who need healthy recipes the most.
Before their official release, they saw the users had become tired of thinking of menus. Users can customize the suggestion by choosing:
- I want to have something light because I have a stomach issue today
- I feel like Japanese style today
- I want to use foods available in the refrigerator
Other options include how strict you want the menu to be and the needs of other family members. So the company officially released the application, using AI to suggest the best possible balanced dishes for the users.
The menus
Menus in the application are from 1) about 10,000 menus suggested by a registered dietitian, 2) about 500,000 different menus specifically for the types of illnesses, and are all in alignment with 3) the guidelines of associations of the particular health conditions.
Health conditions include skin issues such as pimples, type 2 diabetes, heart disease, and more. They have increased 50 to 56 different health conditions. And they are still trying to improve the application.
Foods we take several times a day affect our health. Enormously. We can choose to change it or not to change it. One option, using an app like this can help users stay on track. Potentially using AI or machine learning can help them not fall off individuals’ sustainable healthy habits.
My Thoughts on These
1) Using AI and an ML technique has shown how it can help people’s food intake daily when they have specific conditions. Users can benefit from the application, reducing their burdens of coming up with a new menu daily and maintaining their motivation to keep eating healthily at the same time.
2) Among all other factors such as genes and learning activity, it is one of the crucial factors associated with healthy longevity. Foods are what people take daily and what people can change somehow within their preference.
It is one of the health-related applications I was hoping to see. And this application lets people who live alone or with family can use it regularly.
Thanks for reading!
I enjoy creating 3D images for my stories these days. And read articles on this platform almost every day. You may find these categories of my articles interesting: Stories of the Fish and Buddy, DL ML AI, Writing Tips.
My Ko-fi shop: Here
My Writing Prompts & Tips app (Web app): Here
My referral link (if you are considering becoming a Medium member): Here
