Current Project Highlight

NutriBonAR helps consumers to improve diets by simply looking at their receipts.

We 1 2 !

1: Retail, 2: Health

Receive our updates:

Overall Research

Our goal is to leverage existing retail data to develop digital solutions
that combat diet-related diseases.


Retail data

We leverage retail data assets, such as product ingredients & digital receipts, to overcome current limitations of mobile health apps.

Improve health

Our focus areas are increasing nutrition literacy, tailoring of health interventions & scalable, automatic monitoring of diet intake.

Combat diseases

Our work on scalable & inclusive health apps aims to contribute to mitigating diet-related diseases such as obesity, diabetes & hypertension.


Our research focus:

Assessing mHealth

Although thousands of mobile health (mHealth) apps exist, they suffer under low retention & acceptance. We review dieta mHealth's retention behavior & best practice.

Learn more!

Nutrition Literacy

Retail data can be integrated in mHealth & induce applicable nutrition literacy, especially relevant due to the growing influence of processed & convenience food.

Learn more!

Tailored Interventions

Different to printed static traffic labels, mHealth apps can leverage product ingredient data to tailor interventions to each consumer based on individual needs.

Learn more!

Scalable Monitoring

Monitoring diets is hard due to required manual logging of meals. We use machine learning to infer diets via automatically recorded purchase data from loyalty cards.

Learn more!

 Motivation


86% of Swiss citizens eat too much salt1

Excessive intake of sodium, saturated fat, sugar can lead to diet-related diseases, such as obesity, diabetes, hypertension, reduced life expectancy2,3


1: Chappuis, A., et. al. (2011). Centre Hospitalier Universitaire Vaudois (CHUV)

2: Bochud, M., et. al. (2012). Public Health Reviews

3: Forouzanfar, M.H. et al. (2015). The Lancet

 Current state of diet mHealth is dramatic

Although thousands of diet apps exist, the vast majority of society is still not using them. The design of diet apps has to improve dramatically & make use of late-breaking research to improve retention & effectivity.




mHealth myriad

Nowadays, users can choose from 2'814 diet mHealth apps1. With only little regulation by authorities2, users & dieticians suffer under a myriad of apps to choose from.

Extremely low retention

Diet apps suffer under self-selection of healthy users, short-lived retention & low adherence 3, such that at least 50% of patients with diet-related diseases do not actively use them 4.

Barriers to use

Barriers of usage among today's diet apps lie in the extreme effort involved in manual diet logging, underreporting of diet, non-personalized & abstract recommendations, non-pleasant user experience.

Checklist

We recommend mHealth designers to include automated data collection methods, just-in-time-adaptive interventions, gamification, & state-of-art implementations of features, design, behavior-change-technologies!

1: 42matters.com (2019), 2: FDA.gov (2019), 3: Nutritiontest (2018), 4: Langford (2019), JMIR mHealth

Retail data can increase literacy

Due to the recent EU1169 regulation, new retail data assets became public & can now be used in mHealth to train applicable health literacy.


Lacking Literacy

Literacy is a necessary prerequesite for healthy diets. Still, 50% of consumers lack important applicable skills1, required to mitigate today's challenge of identifying healthy alternatives among processed food items in convenience retail.

Risk Awareness

You can only improve, what you know. As most consumers are unaware about their own exposure to consumption-related risks , new tools are needed to self-assess individual risk potential.

Retail Data

With the EU1169 regulation, retailers & brands published their product data, including image and ingredients, which can can help to train literacy & assess risk exposure in mHealth-mediated empirical studies.


Recent results:

Skinprotect in Journal of Food & Toxicology

With our Skinprotect app, users become able to scan their household inventory of beauty and cleaning products to assess their exposure to toxicologic substances. The study demonstrated that retail data can support the collection of population-specific exposure factors for complex exposure models.

Swiss FoodQuiz published at ECIS

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

Swiss FoodTracker app handover to BLV finished

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

Swiss SaltTracker study finished

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.


1: Nutritiontest (2018)


Retail data enables interventions

Retail data can enhance Behavior Interventions Interventions help make right decisions. Todays Apps are too slow, often yield abstract, not directly actionable recommendations (e.g. eat less salt). Today’s retail plays an important role in diet / EU 1169 data Machine Learning, OCR


Lacking Literacy

Behavior > Link > Behavior Interventions ,* Behavior Interventions at Purchase Time , * Aggregating data instead of single products , * New interfaces

Risk Awareness

Do not know their intake levels or exposure. Do not know outcome expectancy of their actions.

Retail Data

Retail data, image and ingredients can help to train and assess exposure in empirical studies.


Recent results:

BetterChoice app launched

Together with SGE and Buero

BasketCheck study finished

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

AllergyScan app launched in the app stores

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

NutriBonAR started

We use OCR to leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

HoloSelecta

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

Nutrition Avatar

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

Nutrition Avatar 3D

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.


1: Author (Year), Outlet


Diet tracking via digital receipts

Due to the current diet apps fail to monitor, as more than 90% do not comply with logging their diets over more than 90 days. Long term habits can only be built after longer periods. / Categories , Shopping / Retail data / Digital Receipts / GDPR


Lacking Literacy

* high effort, low acceptance & retention, Do not know their intake levels or exposure. Do not know outcome expectancy of their actions.

Risk Awareness

Autonomous Passive ,* Scale > 4 M Households , * Relevant Data , * GDPR / Anonymous ,

Retail Data

Retail data, image and ingredients can help to train and assess exposure in empirical studies.


Recent results:

Receipt2Nutrition

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.

Image2Product

We leverage existing retail data assets, such as product ingredients & digital receipts, to overcome the limitations of state-of-the-art mobile health solutions.


1: Author (Year), Outlet

2: Author (Year), Outlet

3: Author (Year), Outlet


Partners

Our work is conducted together with our partners from academia and industry, in order to ensure rigor and relevance of our work.


Research


Auto-ID Labs ETH/HSG is a joint research institute at ETH Zurich and University St. Gallen with a focus on the internet-of-things in retail.

Swiss Society for Nutrition SGE-SSN disseminates information and ongoing research on nutrition and health within Switzerland.

The Center for Digital Health Interventions designs digital health interventions based on digital biomarker and digital coaching research.

Financial support


GS1 is the world's largest standardization organization and supports research on digitization in retail.

CSS is Switzerland's largest health insurance and supports digital health initiatives.

Emmi is Switzerland's largest dairy produce and wants to support consumers to follow a balanced diet as basis for an active lifestyle.

Data


Migros is Switzerland's largest retailer and published its ingredient data as part of their Open Data strategy.

Trustbox is Switzerland's largest manufacturer-mandated product composition database.

Openfood is the world's largest open-source product composition database and maintained by thousands of contributers.

Codecheck is Switzerland's largest product composition database and also features a lot of beauty and healthcare products.

FoodRepo is a product composition database maintained by EPFL.

Bitsaboutme is a fully encrypted GDPR-based service that allows consumers to retrieve and process their personal data themselves.

Holo-One is a Swiss startup that engages in augmented and mixed Reality applications.

Adnexo is a Swiss startup that works in the internet of things in the agriculture sector.

App Projects

We develop digital solutions to fight diet-related diseases, such as diabetes, obesity, hypertension.


Data

We leverage existing data such as product composition data, product images, printed & digital receipts.

Method

Our apps are designed with experts such as Swiss Society for Nutrition, & tested in realistic field-studies.





Publications

We frequently publish in relevant scientific outlets of digital epidemiology, mobile health, information systems, to share our findings with the public & scientific community.


Automation of Data Collection in diet-related mHealth: a Review of Publicly Available and Well-Adopted Apps
Fuchs, K., Vuckovac, D., Ilic, A., 2018,
in ICTC 2018 Proceedings, IEEE, Jeju, South Korea, DOI: 10.13140/RG.2.2.23332.48001
Assessing exposure factors in the smartphone generation: Design and Evaluation of a smartphone app that collects use patterns of cosmetics and household chemicals
Von Goetz, N., Garcia-Hidalgo, E., Balachandran, C., Fuchs, K. , Frey, R., Ilic, A., 2018,
in Journal Of Food and Chemical Toxicology, Elsevier, DOI: 10.1016/j.fct.2018.05.060
Universal Food Allergy Number
Frey, R., Ryder, B., Fuchs, K., and Ilic, A. 2016,
in IOT 2016 Proceedings (Best Poster Award), Stuttgart, Germany, DOI: 10.1145/2991561.2998462
Swiss FoodQuiz: Inducing Nutritional Knowledge via a Visual Learning based Serious Game
Fuchs, K., Huonder, V., Vuckovac, D., and Ilic, A. 2016,
in ECIS 2016 Proceedings, Istanbul, Turkey, DOI: 10.13140/RG.2.2.23332.48001
Swiss FoodQuiz: Poster
Fuchs, K., Huonder, V., Vuckovac, D., and Ilic, A. 2016,
in Prototype Poster, ECIS 2016
Proposing A New Standard For Food Allergies
Frey, R., Ryder, B., Fuchs, K., and Ilic, A. 2016,
in GS1 S+I Conference 2016, Brussels, Belgium
Leveraging Food Databases for Nutritional Behavior Change
Fuchs, K., Vuckovac, D., and Ilic, A. 2016,
in GS1 S+I Conference 2016, Brussels, Belgium

Collaboration: If you want to work together with us, please check our current projects and positions and get in touch

We new friends!

If you are interested in our work, we look forward to hearing from you: team@autoidlabs.ch




Receive our updates: