Food(lg)

Large-scale food intake data plays a significant role in revealing the dietary information and knowledge of patients, posing a profound impact on human health and well-being. Therefore, sufficient food data is essential for facilitating different healthcare analytic applications such as disease diagnosis, and chronic disease progression modeling. As a result, Food(lg) is designed to fulfill the increasing demand for food data and further well-being.

We build Food(lg) to record daily nutrient estimates with journal entries for achieving a well-balanced diet. Powered by a deep-learning framework, Apache Singa, and based on nutritional guidelines, Food(lg) has been used successfully in several scenarios, such as diabetes prevention in Ng Teng Fong General Hospital and record of dental health in Faculty of Dentistry of NUS, to help patients build a healthy diet.

The Food(lg) for diabetes prevention can be downloaded on the APP Store or Google Play. Alternatively, you can view it on Desktop. Watch this video or visit the main page if you want to know more details.

You could access Food(lg) for dental health by visiting the dental main page. In addition to the app, we have released a Localized Singaporean Food Dataset, named FoodSG-233, which can capture the unique characteristics of each country’s dish varieties, cooking styles, and food ingredients, for promoting future data management research in food computing. For more details, please visit the website.

Food(lg) could keep a daily record of users' eating diet and activities. In addition, Food(lg) provides a quick and easy way for reviewing dietary intake, eating plans, and smart suggestions based on the records. Food(lg) also supports social networks to browse, discover and share food.

This Food(lg) app helps users to develop healthy eating lifestyles. It has attractive functions, e.g., surfing food photos, reviewing eating habits, recording food and planning healthy meals.

The image-based food recognition part of our Food(lg) app is powered by deep learning. The model training phase is off-line while the food recognition and health analysis is on-line.

In addition to recording the food intake and indicating the eating trend for users, Food(lg) for Diabetes Prevention also enables users to chat with dietitians and receive personalized diet recommendations.

Food(lg) for Dental Health not only could record the amount of sugar consumed but also show the main nutrient information and extra information about the food. Besides, Food(lg) provides a large amount of common dental knowledge, such as tooth decay, suitable amount of sugar and frequency of sugar intake.