Use of Different Food Image Recognition Platforms in Dietary Assessment: Comparison Study.
accuracy
automated food recognition
dietary assessment
image recognition
Journal
JMIR formative research
ISSN: 2561-326X
Titre abrégé: JMIR Form Res
Pays: Canada
ID NLM: 101726394
Informations de publication
Date de publication:
07 Dec 2020
07 Dec 2020
Historique:
received:
23
07
2019
accepted:
02
10
2020
revised:
02
09
2020
entrez:
7
12
2020
pubmed:
8
12
2020
medline:
8
12
2020
Statut:
epublish
Résumé
In the domain of dietary assessment, there has been an increasing amount of criticism of memory-based techniques such as food frequency questionnaires or 24 hour recalls. One alternative is logging pictures of consumed food followed by an automatic image recognition analysis that provides information on type and amount of food in the picture. However, it is currently unknown how well commercial image recognition platforms perform and whether they could indeed be used for dietary assessment. This is a comparative performance study of commercial image recognition platforms. A variety of foods and beverages were photographed in a range of standardized settings. All pictures (n=185) were uploaded to selected recognition platforms (n=7), and estimates were saved. Accuracy was determined along with totality of the estimate in the case of multiple component dishes. Top 1 accuracies ranged from 63% for the application programming interface (API) of the Calorie Mama app to 9% for the Google Vision API. None of the platforms were capable of estimating the amount of food. These results demonstrate that certain platforms perform poorly while others perform decently. Important obstacles to the accurate estimation of food quantity need to be overcome before these commercial platforms can be used as a real alternative for traditional dietary assessment methods.
Sections du résumé
BACKGROUND
BACKGROUND
In the domain of dietary assessment, there has been an increasing amount of criticism of memory-based techniques such as food frequency questionnaires or 24 hour recalls. One alternative is logging pictures of consumed food followed by an automatic image recognition analysis that provides information on type and amount of food in the picture. However, it is currently unknown how well commercial image recognition platforms perform and whether they could indeed be used for dietary assessment.
OBJECTIVE
OBJECTIVE
This is a comparative performance study of commercial image recognition platforms.
METHODS
METHODS
A variety of foods and beverages were photographed in a range of standardized settings. All pictures (n=185) were uploaded to selected recognition platforms (n=7), and estimates were saved. Accuracy was determined along with totality of the estimate in the case of multiple component dishes.
RESULTS
RESULTS
Top 1 accuracies ranged from 63% for the application programming interface (API) of the Calorie Mama app to 9% for the Google Vision API. None of the platforms were capable of estimating the amount of food. These results demonstrate that certain platforms perform poorly while others perform decently.
CONCLUSIONS
CONCLUSIONS
Important obstacles to the accurate estimation of food quantity need to be overcome before these commercial platforms can be used as a real alternative for traditional dietary assessment methods.
Identifiants
pubmed: 33284118
pii: v4i12e15602
doi: 10.2196/15602
pmc: PMC7752530
doi:
Types de publication
Journal Article
Langues
eng
Pagination
e15602Informations de copyright
©Stephanie Van Asbroeck, Christophe Matthys. Originally published in JMIR Formative Research (http://formative.jmir.org), 07.12.2020.
Références
PLoS One. 2013 Oct 09;8(10):e76632
pubmed: 24130784
Epidemiol Health. 2014 Jul 22;36:e2014009
pubmed: 25078382
J Diabetes Sci Technol. 2015 May;9(3):525-33
pubmed: 25901024
Int J Epidemiol. 2012 Aug;41(4):1187-203
pubmed: 22933652
IEEE J Biomed Health Inform. 2014 Jul;18(4):1261-71
pubmed: 25014934
Nutrients. 2017 Jun 27;9(7):
pubmed: 28653995
J Med Internet Res. 2016 May 11;18(5):e101
pubmed: 27170498
J Nutr. 2015 Dec;145(12):2639-45
pubmed: 26468491
Ethn Dis. 2013 Autumn;23(4):445-51
pubmed: 24392607
JAMA. 2013 Nov 27;310(20):2137-8
pubmed: 24281443
Proc Int Conf Image Proc. 2010;:1853-1856
pubmed: 22025261
Am J Epidemiol. 2016 Apr 15;183(8):715-28
pubmed: 27022032
IEEE J Sel Top Signal Process. 2010 Aug;4(4):756-766
pubmed: 20862266
Public Health Nutr. 2016 Apr;19(5):893-905
pubmed: 26143683
BMJ. 2013 Nov 14;347:f6698
pubmed: 24231028
Nutrients. 2015 Jul 27;7(8):6128-38
pubmed: 26225994
Multimed Tools Appl. 2018 Aug;77(15):19769-19794
pubmed: 30202237
IEEE J Biomed Health Inform. 2016 May;20(3):848-855
pubmed: 25850095
Adv Nutr. 2014 Jul 14;5(4):447-55
pubmed: 25022993
J Nutr Sci Vitaminol (Tokyo). 2015;61 Suppl:S33-5
pubmed: 26598878
Nutrients. 2018 Dec 28;11(1):
pubmed: 30597864