Hostname: page-component-5db58dd55d-smskv Total loading time: 0 Render date: 2026-05-25T11:39:29.601Z Has data issue: false hasContentIssue false
  • English
  • Français

FoodFinder: developing a rapid low-cost crowdsourcing approach for obtaining data on meal size from meal photos

Published online by Cambridge University Press:  24 November 2016

L. Johnson ,
C.Y. England ,
P. Laskowski ,
P.R. Woznowski ,
L. Birch ,
J.P. Hamilton-Shield ,
D.A. Lawlor ,
I. Craddock  and
A. Skinner
L. Johnson
Affiliation:
Centre for Exercise, Nutrition and Health Science, University of Bristol, Bristol, BS8 1TZ NIHR Bristol Biomedical Research Unit in Nutrition, University of Bristol, Bristol, BS2 8AE
C.Y. England
Affiliation:
Centre for Exercise, Nutrition and Health Science, University of Bristol, Bristol, BS8 1TZ NIHR Bristol Biomedical Research Unit in Nutrition, University of Bristol, Bristol, BS2 8AE
P. Laskowski
Affiliation:
Faculty of Engineering, University of Bristol, Bristol, BS8 1UB
P.R. Woznowski
Affiliation:
Faculty of Engineering, University of Bristol, Bristol, BS8 1UB
L. Birch
Affiliation:
NIHR Bristol Biomedical Research Unit in Nutrition, University of Bristol, Bristol, BS2 8AE
J.P. Hamilton-Shield
Affiliation:
NIHR Bristol Biomedical Research Unit in Nutrition, University of Bristol, Bristol, BS2 8AE
D.A. Lawlor
Affiliation:
Medical Research Council Integrative Epidemiology Unit, University of Bristol, Bristol, BS8 2BN
I. Craddock
Affiliation:
Faculty of Engineering, University of Bristol, Bristol, BS8 1UB
A. Skinner
Affiliation:
Medical Research Council Integrative Epidemiology Unit, University of Bristol, Bristol, BS8 2BN
Rights & Permissions [Opens in a new window]

Abstract

An abstract is not available for this content. As you have access to this content, full HTML content is provided on this page. A PDF of this content is also available in through the ‘Save PDF’ action button.

Capturing meal images using mobile phone cameras is a promising alternative to traditional dietary assessment methods. Acquiring photos is reasonably simple but nutritional content analysis of images is a major challenge. Automated food identification and portion size assessment is computationally and participant intensive; relying on participant feedback for accuracy(Reference Boushey, Kerr and Schap1). Dietitian analysis of photos is accurate but time-consuming and expensive(Reference Martin, Correa and Han2). Crowdsourcing could offer a rapid low-cost alternative by utilising the life-long experience that all humans have in food identification. Previous crowdsourcing methods include the Eatery app, which produces a simple 11-point ‘healthiness’ scale for each meal(Reference Turner-McGrievy, Helander and Kaipainen3) and the PlateMate system, which creates a list of all individual foods with portion sizes, energy and macronutrient content(Reference Noronha, Hysen and Zhang4). While the Eatery produces limited and subjective data on meal content, PlateMate represents a complex integrated system of multiple tasks requiring on average 25 workers, costing £2·75 and taking 90 min per image. For feasible data-capture in a large-scale longitudinal studies, crowdsourcing data from meal photos needs to be cheaper and quicker. We aimed to develop a simpler task and tested it's feasibility for crowdsourcing dietary data.

FoodFinder, a single task for identifying food groups and portion sizes, developed using Qualtrics (www.qualtrics.com/), and linked to the Prolific Academic (https://prolific.ac/) crowdsourcing platform for recruitment and reimbursement of a UK crowd. Thirty meal photos with measured total meal weight (grams) were analysed by a dietitian and crowds ranging in size from 5 to 50 people. The difference between actual meal weight (the gold-standard) and total meal weight estimated by different sized crowds and ratings by a dietician were compared to each other. To establish group consensus crowd estimates were weighted by majority agreement(Reference Zhai, Hachen and Kijewski-Correa5). Bland-Altman analysis assessed agreement with actual meal weight.

Fig. 1

Fig. 2

A crowd of 5 people underestimated true meal weight by 63 g, equating to 15 % of actual meal weight with limits of agreement (LOA) from −299 to 174 g. In comparison experts overestimated by 28 g equating to 9 % of actual meal weight with LOA −158, 214 g. With a crowd of 5 people, crowdsourcing cost £3·35 and took a mean 2 mins 55 sec (SD 2 min 6 sec) per image. A crowd of 50 had similar accuracy and limits of agreement (−65 g LOA −278, 149 g) but was more expensive. Further development of FoodFinder is required to make rapid low-cost analysis of meal photos via crowdsourcing a feasible method for assessing diet.

This work was supported by a catalyst award from the Elizabeth Blackwell Institute for Health Research, which is co-funded by the University of Bristol and an Institutional Strategic Support Fund from the Wellcome Trust.

References

1.Boushey, CJ, Kerr, DA, Schap, TE, et al. (2012) Eur J Clin Nutr 6, 648.Google Scholar
2.Martin, CK, Correa, JB, Han, H, et al. (2012) Obesity (Silver Spiring) 20, 891–9.Google Scholar
3.Turner-McGrievy, GM, Helander, EE, Kaipainen, K, et al. (2014) J Am Med Inform Assoc. e1. e112–9.Google Scholar
4.Noronha, J, Hysen, E, Zhang, H, et al. (2011) Proc UIST 1–12.Google Scholar
5.Zhai, Z, Hachen, D, Kijewski-Correa, T, et al. (2012) 45th Int Conf Sys Sci, doi:10.1109/HICSS.2012.151Google Scholar
Figure 0

Fig. 1

Figure 1

Fig. 2