Sina Masoud-Ansari1, Mark Gahegan2, Andrew Jull3, Cliona Ni Mhurchu4
1University of Auckland, Auckland, New Zealand, email@example.com
2University of Auckland, Auckland, New Zealand, firstname.lastname@example.org
3University of Auckland, Auckland, New Zealand, email@example.com
4University of Auckland, Auckland, New Zealand, firstname.lastname@example.org
Obesity is a complex system operating at many levels, containing a diverse set of actors, and operating via different mechanisms and operative pathways. These characteristics suggest the need for new and more dynamic methods to better understand determinants and identify solutions. Typically a “reductionist” approach has been taken in obesity research, which involves studying individual decontextualized risk factors that operate at one level only and don’t account for interrelatedness and reciprocity between exposures. In contrast, “systems thinking” suggests that complex, dynamic systems, which feature multiple interdependent components whose interactions may include feedback, non-linearity and lack of centralised control, are best understood holistically. A systems approach can thus complement other obesity research by adding dimensions that reductionist approaches cannot. The Kids’Cam  study, provides a unique source of data to build a simulation model of New Zealand children’s food and activity environments. The dataset contains four days of data from 169 ethnically and socioeconomically diverse NZ children on where they go (GPS data), what they see and who they interact with (1.3 million images collected using automated, wearable cameras).
The Kids’Cam data provide key insights into some of the exposures and interactions that are needed to build better simulation models. Information on children aged 11-13 years and their schools, homes and exposure to advertising and use of local food outlets can all be extracted from this data. Demographic data from the Kids’Cam participants can be used to assign characteristics to virtual children and GPS data collected on children’s movements during the study could be used to assign typical travel routes around a virtual neighbourhood. Our aim in this project was to develop, test and validate the recognition and extraction methods that are needed as a precursor to simulation modelling. The resulting datasets of GPS routes and exposure to food branding will provide the necessary starting point for building simulation models of exposure and using model experiments to assess the potential impact of new policies or interventions.
AUTOMATED CLASSIFICATION AND DATA INTEGRATION
The Centre for eResearch provided the expertise in automated image classification, data integration and spatial analysis. Manually annotating images in the Kids’Cam data is labour intensive due to the large number of images. We investigated the potential for automated classification of images to reduce the effort required to extract data for this and future data sets. The NVIDIA DIGITS deep learning framework was used to train classifiers to detect features of interest such as whether the image was taken at school, in a supermarket or at home and whether the image contains particular food or drink items. A sample of manually annotated data was used to train the classifier and classification worked well for detecting the image setting in homogenous environments such as public and private transport. Detection of heterogeneous environments such as a participant’s home or school proved more difficult requiring more training data to reach sufficient accuracy. By combining information from the annotated images and the associated GPS records of children’s activities, we were able to create maps of exposure to food/drink items of interest. Figure 1 shows an example of the distribution of ‘fast-food’ exposure in Porirua, New Zealand. These datasets will create the foundation for future projects which aim to create simulation models of children’s activities and the effect of policy decisions on exposure and obesity.
Figure 1: Distribution of fast-food exposure in Porirua, New Zealand
1. Signal LN, Smith MB, Barr M, Stanley J, Chambers TJ, Zhou J, Duane A, Jenkin GLS, Pearson AL, Gurrin C, Smeaton AF, Hoek J, Ni Mhurchu C. Kids’Cam: An objective methodology to study the world in which children live. American Journal of Preventive Medicine 2017; published online April 25, 2017: http://doi.org/10.1016/j.amepre.2017.02.016.
Research IT Specialist at the Centre for eResearch at the University of Auckland