How much are we exposed to alcohol in electronic media? Development of the Alcoholic Beverage Identification Deep Learning Algorithm (ABIDLA).
Alcohol exposure
Beverage recognition
Deep learning algorithms
Journal
Drug and alcohol dependence
ISSN: 1879-0046
Titre abrégé: Drug Alcohol Depend
Pays: Ireland
ID NLM: 7513587
Informations de publication
Date de publication:
01 03 2020
01 03 2020
Historique:
received:
27
09
2019
revised:
19
12
2019
accepted:
19
12
2019
pubmed:
20
1
2020
medline:
20
1
2021
entrez:
20
1
2020
Statut:
ppublish
Résumé
Evidence demonstrates that seeing alcoholic beverages in electronic media increases alcohol initiation and frequent and excessive drinking, particularly among young people. To efficiently assess this exposure, the aim was to develop the Alcoholic Beverage Identification Deep Learning Algorithm (ABIDLA) to automatically identify beer, wine and champagne/sparkling wine from images. Using a specifically developed software, three coders annotated 57,186 images downloaded from Google. Supplemented by 10,000 images from ImageNet, images were split randomly into training data (70 %), validation data (10 %) and testing data (20 %). For retest reliability, a fourth coder re-annotated a random subset of 2004 images. Algorithms were trained using two state-of-the-art convolutional neural networks, Resnet (with different depths) and Densenet-121. With a correct classification (accuracy) of 73.75 % when using six beverage categories (beer glass, beer bottle, beer can, wine, champagne, and other images), 84.09 % with three (beer, wine/champagne, others) and 85.22 % with two (beer/wine/champagne, others), Densenet-121 slightly outperformed all Resnet models. The highest accuracy was obtained for wine (78.91 %) followed by beer can (77.43 %) and beer cup (73.56 %). Interrater reliability was almost perfect between the coders and the expert (Kappa = .903) and substantial between Densenet-121 and the coders (Kappa = .681). Free from any response or coding burden and with a relatively high accuracy, the ABIDLA offers the possibility to screen all kinds of electronic media for images of alcohol. Providing more comprehensive evidence on exposure to alcoholic beverages is important because exposure instigates alcohol initiation and frequent and excessive drinking.
Sections du résumé
BACKGROUND
Evidence demonstrates that seeing alcoholic beverages in electronic media increases alcohol initiation and frequent and excessive drinking, particularly among young people. To efficiently assess this exposure, the aim was to develop the Alcoholic Beverage Identification Deep Learning Algorithm (ABIDLA) to automatically identify beer, wine and champagne/sparkling wine from images.
METHODS
Using a specifically developed software, three coders annotated 57,186 images downloaded from Google. Supplemented by 10,000 images from ImageNet, images were split randomly into training data (70 %), validation data (10 %) and testing data (20 %). For retest reliability, a fourth coder re-annotated a random subset of 2004 images. Algorithms were trained using two state-of-the-art convolutional neural networks, Resnet (with different depths) and Densenet-121.
RESULTS
With a correct classification (accuracy) of 73.75 % when using six beverage categories (beer glass, beer bottle, beer can, wine, champagne, and other images), 84.09 % with three (beer, wine/champagne, others) and 85.22 % with two (beer/wine/champagne, others), Densenet-121 slightly outperformed all Resnet models. The highest accuracy was obtained for wine (78.91 %) followed by beer can (77.43 %) and beer cup (73.56 %). Interrater reliability was almost perfect between the coders and the expert (Kappa = .903) and substantial between Densenet-121 and the coders (Kappa = .681).
CONCLUSIONS
Free from any response or coding burden and with a relatively high accuracy, the ABIDLA offers the possibility to screen all kinds of electronic media for images of alcohol. Providing more comprehensive evidence on exposure to alcoholic beverages is important because exposure instigates alcohol initiation and frequent and excessive drinking.
Identifiants
pubmed: 31954949
pii: S0376-8716(20)30006-5
doi: 10.1016/j.drugalcdep.2020.107841
pii:
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
107841Informations de copyright
Copyright © 2020 Elsevier B.V. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of Competing Interest There is no conflict of interest to declare.