Automatic food detection in egocentric images using artificial intelligence technology

OBJECTIVE To develop an artificial intelligence (AI)-based algorithm which can automatically detect food items from images acquired by an egocentric wearable camera for dietary assessment. DESIGN To study human diet and lifestyle, large sets of egocentric images were acquired using a wearable device, called eButton, from free-living individuals. Three thousand nine hundred images containing real-world activities, which formed eButton data set 1, were manually selected from thirty subjects. eButton data set 2 contained 29 515 images acquired from a research participant in a week-long unrestricted recording. They included both food- and non-food-related real-life activities, such as dining at both home and restaurants, cooking, shopping, gardening, housekeeping chores, taking classes, gym exercise, etc. All images in these data sets were classified as food/non-food images based on their tags generated by a convolutional neural network. RESULTS A cross data-set test was conducted on eButton data set 1. The overall accuracy of food detection was 91·5 and 86·4 %, respectively, when one-half of data set 1 was used for training and the other half for testing. For eButton data set 2, 74·0 % sensitivity and 87·0 % specificity were obtained if both 'food' and 'drink' were considered as food images. Alternatively, if only 'food' items were considered, the sensitivity and specificity reached 85·0 and 85·8 %, respectively. CONCLUSIONS The AI technology can automatically detect foods from low-quality, wearable camera-acquired real-world egocentric images with reasonable accuracy, reducing both the burden of data processing and privacy concerns.

[1]  Geoffrey E. Hinton,et al.  Acoustic Modeling Using Deep Belief Networks , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[2]  Guang-Zhong Yang,et al.  A pilot study to determine whether using a lightweight, wearable micro-camera improves dietary assessment accuracy and offers information on macronutrients and eating rate , 2015, British Journal of Nutrition.

[3]  Mingui Sun,et al.  Model-based measurement of food portion size for image-based dietary assessment using 3D/2D registration , 2013, Measurement science & technology.

[4]  Mingui Sun,et al.  A wearable electronic system for objective dietary assessment. , 2010, Journal of the American Dietetic Association.

[5]  Giovanni Maria Farinella,et al.  On the Exploitation of One Class Classification to Distinguish Food Vs Non-Food Images , 2015, ICIAP Workshops.

[6]  B. Koroušić Seljak,et al.  NutriNet: A Deep Learning Food and Drink Image Recognition System for Dietary Assessment , 2017, Nutrients.

[7]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[8]  Seyedmostafa Safavi,et al.  Conceptual Privacy Framework for Health Information on Wearable Device , 2014, PloS one.

[9]  P. Stumbo.  New technology in dietary assessment: a review of digital methods in improving food record accuracy , 2013, Proceedings of the Nutrition Society.

[10]  C. J. Boushey,et al.  New mobile methods for dietary assessment: review of image-assisted and image-based dietary assessment methods , 2016, Proceedings of the Nutrition Society.

[11]  P. Kelly,et al.  Feasibility of a SenseCam-assisted 24-h recall to reduce under-reporting of energy intake , 2013, European Journal of Clinical Nutrition.

[12]  Marios Anthimopoulos,et al.  Food Recognition for Dietary Assessment Using Deep Convolutional Neural Networks , 2015, ICIAP Workshops.

[13]  Keiji Yanai,et al.  Automatic Expansion of a Food Image Dataset Leveraging Existing Categories with Domain Adaptation , 2014, ECCV Workshops.

[14]  A Hirata,et al.  A novel automated detection system for swallowing sounds during eating and speech under everyday conditions. , 2015, Journal of oral rehabilitation.

[15]  Lenore Arab,et al.  Automated camera-phone experience with the frequency of imaging necessary to capture diet. , 2010, Journal of the American Dietetic Association.

[16]  G. O'loughlin,et al.  Using a wearable camera to increase the accuracy of dietary analysis. , 2013, American journal of preventive medicine.

[17]  E J Delp,et al.  Use of technology in children’s dietary assessment , 2009, European Journal of Clinical Nutrition.

[18]  Jennifer Utter,et al.  Image-assisted dietary assessment: a systematic review of the evidence. , 2015, Journal of the Academy of Nutrition and Dietetics.

[19]  Wolf-Joachim Fischer,et al.  Food Intake Monitoring: Automated Chew Event Detection in Chewing Sounds , 2014, IEEE Journal of Biomedical and Health Informatics.

[20]  Wenyao Xu,et al.  Wearable Food Intake Monitoring Technologies: A Comprehensive Review , 2017, Comput..

[21]  Nicholas Gant,et al.  Wearable cameras can reduce dietary under-reporting: doubly labelled water validation of a camera-assisted 24 h recall. , 2015, The British journal of nutrition.

[22]  Carl Lachat,et al.  Assessing food intake through a chest-worn camera device , 2014, Public Health Nutrition.

[23]  Giovanni Maria Farinella,et al.  Retrieval and classification of food images , 2016, Comput. Biol. Medicine.

[24]  Wenyan Jia,et al.  Minimizing Memory Errors in Child Dietary Assessment with a Wearable Camera: Formative Research , 2015 .

[25]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups , 2012, IEEE Signal Processing Magazine.

[26]  Michael McCarthy.  Federal privacy rules offer scant protection for users of health apps and wearable devices , 2016, British Medical Journal.

[27]  Majid Sarrafzadeh,et al.  Audio-based detection and evaluation of eating behavior using the smartwatch platform , 2015, Comput. Biol. Medicine.

[28]  Corby K. Martin,et al.  Measuring food intake with digital photography. , 2014, Journal of human nutrition and dietetics : the official journal of the British Dietetic Association.

[29]  Edward J. Delp,et al.  Image-based food volume estimation , 2013, CEA '13.

[30]  I. Huybrechts,et al.  Review and evaluation of innovative technologies for measuring diet in nutritional epidemiology. , 2012, International journal of epidemiology.

[31]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition , 2012 .

[32]  Zhen Li,et al.  An exploratory study on a chest-worn computer for evaluation of diet, physical activity and lifestyle. , 2015, Journal of healthcare engineering.

[33]  Steve E Hodges,et al.  Wearable cameras in health: the state of the art and future possibilities. , 2013, American journal of preventive medicine.

[34]  J. Burke,et al.  Feasibility Testing of an Automated Image-Capture Method to Aid Dietary Recall , 2011, European Journal of Clinical Nutrition.

[35]  Sen Zhang,et al.  Detection of Activities by Wireless Sensors for Daily Life Surveillance: Eating and Drinking , 2009, Sensors.

[36]  Yujie Dong,et al.  Detecting Periods of Eating During Free-Living by Tracking Wrist Motion , 2014, IEEE Journal of Biomedical and Health Informatics.

[37]  Camille Couprie,et al.  Learning Hierarchical Features for Scene Labeling , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[38]  Jennifer Utter,et al.  The use of a wearable camera to capture and categorise the environmental and social context of self-identified eating episodes , 2015, Appetite.

[39]  Yiran Chen,et al.  eButton: A wearable computer for health monitoring and personal assistance , 2014, 2014 51st ACM/EDAC/IEEE Design Automation Conference (DAC).

[40]  Kiyoharu Aizawa,et al.  Highly Accurate Food/Non-Food Image Classification Based on a Deep Convolutional Neural Network , 2015, ICIAP Workshops.

[41]  Mingui Sun,et al.  Accuracy of food portion size estimation from digital pictures acquired by a chest-worn camera , 2013, Public Health Nutrition.

[42]  Robert Steele,et al.  An Overview of the State of the Art of Automated Capture of Dietary Intake Information , 2015, Critical reviews in food science and nutrition.

[43]  Juan Cui,et al.  A Survey on Automated Food Monitoring and Dietary Management Systems. , 2017 .

[44]  S. Marshall,et al.  An ethical framework for automated, wearable cameras in health behavior research. , 2013, American journal of preventive medicine.