Yuru Huang is a third-year PhD Student and Gates Cambridge Scholar based at the MRC Epidemiology Unit. She uses large, publicly available datasets to explore the out-of-home food environment and has produced a novel database, MenuTracker, that provides nutritional information of menu items served by large food chains in the UK.
What is your research about?
My research focuses on characterising different aspects of the out-of-home food environment, for example the nutritional composition of food served by chain restaurants, using big data sources. While our food choices are often seen as a matter of personal preference, the food environment surrounding us plays a significant role in shaping our dietary behaviours.
Given the increasing popularity of eating out, improving the out-of-home food environment may be an effective and equitable strategy to improve population diet, and ultimately, health.
My research uses data science techniques to harness the utility of publicly available data and address research questions that have not been previously explored in food environment research.
What led you to your area of research?
Before coming to Cambridge, I worked as a data scientist at the Center for Food Safety and Applied Nutrition, US Food and Drug Administration (FDA). I saw first-hand the importance of data in understanding the food landscape and subsequent policymaking. Through such experiences, I developed a strong interest in leveraging publicly available data sources for dietary public health research.
What are you working on now?
As part of my PhD, I have established a novel database named MenuTracker. This database provides consumers, policymakers and researchers with a tool to assess the nutritional information of foods available from chain restaurants and food outlets in the UK. We will also use this database to evaluate the effectiveness of the calorie labelling policy in England.
This project is exciting as I am applying data science techniques and deep learning methods to harvest and automate food environment measures that would otherwise take a long time to audit. This method can be automated to understand the healthiness of out-of-home food outlets in large geographical areas.
Why is this topic important?
Unhealthy diets were the cause of 22% of adult deaths worldwide in 2017. This means that diet improvements could potentially prevent one out of every five deaths.
Many local authorities in the UK are actively striving to improve menu healthiness of food outlets to promote healthier eating, and I believe that the MenuTracker could provide valuable insights for identifying areas where interventions are needed. Ultimately, my hope is that this research will contribute to effective policy actions that focus on creating a healthy out-of-home food environment.
Where can we find out more about your research?
A publicly available codebase of MenuTracker is available here and can be used to harvest data for noncommercial use. A publication, Monitoring the Nutrient Composition of Food Prepared Out-of-Home in the United Kingdom: Database Development and Case Study describes the creation of MenuTracker, and an Evidence Brief discusses some of the difficulties we encountered in obtaining the data.
You can also read some of my previous work on using Google street view images and Twitter data to characterise the neighbourhood built environment here.
