Speakers

 

[su_row]
[su_column size=”1/4″] [/su_column]
[su_column size=”3/4″]Prof. Jennifer Coates
Jennifer.Coates@tufts.edu
Friedman School of Nutrition Science and PolicyBoston, MA, USA

Jennifer Coates, PhD is an Assosiate Professor of Food Policy and Applied Nutrition at the Tufts Friedman School of Nutrition Science and Policy and a Senior Researcher at the Tufts Feinstein International Center.

Dr. Coates’s research focuses on the development of methods for improving the design, implementation, and evaluation of international nutrition and food security programs in both development and humanitarian emergency contexts. Methods-related initiatives include the development and validation of: methods for scaling up global dietary data collection and use (INDDEX); indicators of the affordability of quality diets in Africa (IMMANA-IANDA); a standardized approach to malnutrition causal analysis and response assessment (ACF); methods for evaluating the micronutrient impact and functional health outcomes of national fortification programs (GAIN); dietary diversity indicators in emergency-prone contexts (WFP); and global experiential food security measures (FANTA). She also conducts research to understand the implementation and impact of integrated food security programs, including identifying promising models for sustainable exit from Title II-funded food aid programs (FANTA III), and examining the implementation challenges of integrating agriculture and nutrition programming (Ethiopia/ENGINE). Dr. Coates serves on the Technical Working Group to improve the measurement of food consumption in household consumption and expenditure surveys, through the United Nations Inter-Agency Working Group On Agricultural And Rural Statistics; on the UN Expert Advisory Group on Food and Nutrition Security Measurement; and the Editorial Board of the Global Food Security Journal.

In addition to her research and policy engagement, Dr. Coates teaches a range of graduate courses at the Friedman School.

[/su_column]
[/su_row]
[su_accordion]
[su_spoiler icon=”caret” title=”Innovative Technology and Dietary Assessment in Low-Income Countries“]

Introduction

One in three people in the world are malnourished [1]. Approximately 1.9 billion adults are overweight or obese, 2 billion people have micronutrient deficiencies (“hidden hunger”), and over 200 million children under the age of five are stunted or wasted [1]. While undernutrition has been decreasing slowly [2], overweight and obesity has tripled in low- and middle-income countries in the past two decades [3]. Some estimates suggest that over 3 billion people will be overweight or obese by 2030 [4].

The nutrition transition, characterized by increased urbanization, sedentary lifestyles, and significant changes in food systems and dietary patterns, has contributed to these changes [3], [5]. Dietary patterns have shifted towards increased consumption of energy dense, nutrient-poor, processed foods resulting in increased overweight and obesity, often in countries that are still grappling with high rates of undernutrition [4]. The coexistence of underweight with overweight and obesity – frequently referred to as the double burden of malnutrition – is increasingly occurring in the same country, as well as in the same household, in both low and middle-income countries [3].

Given these challenges it is critical to draw from innovative technologies to better measure individual level dietary patterns in order to determine the drivers of the double burden of malnutrition and related chronic diseases such as diabetes and cardiovascular disease. In addition, high quality dietary data on the food and nutrient intake of individuals is also critical to develop policies and programs that effectively address micronutrient deficiencies and influence the effects of food systems on dietary outcomes.

 

The challenge

Despite the importance of individual level dietary data, large-scale dietary surveys are rare in most low-income countries (LICs) due to a number of factors including: high cost and time requirements; limited familiarity with dietary data collection, analysis, and use; and a lack of research infrastructure for dietary data collection. Computer-assisted approaches to data collection are rare: individual level surveys are still commonly collected with pen and paper in most LICs. Furthermore, the potential of alternative sources of food consumption data, such as household consumption and expenditure (HCES) surveys, to proxy for individual consumption has not been realized, in part due to lack of awareness of their utility in food and nutrition programming.

Solution

The International Dietary Data Expansion (INDDEX) Project [6], in collaboration with partner organizations and country-level actors, aims to increase the availability, access, and use of dietary data in order to inform evidence-based decisions for agriculture, food, and nutrition policies and programs. The INDDEX Project plans to achieve this goal through four interrelated initiatives: 1) Develop technologies to standardize and streamline collection and analysis of individual dietary data, 2) Improve the design and use of HCES for food and nutrition analyses, 3) Demonstrate the appropriate use of ‘fit-for-purpose’ indicators and analyses, and 4) Develop guidance to communicate these advancements to international stakeholders.

The presentation will describe the gaps and bottlenecks facing dietary assessment in LICs with a focus on innovative technologies and will highlight ways in which the INDDEX Project intends to support the scale-up of dietary data collection and use. In addition, this presentation will address the challenges and opportunities of self-monitoring dietary intake in LICS using existing technologies in contexts with limited connectivity, low literacy, and limited access to basic technology (e.g. smartphones). In order to highlight promising areas for use of innovative technologies in LICs, examples will be drawn from an in-depth landscape review of dietary assessment platforms [7] and from research on emerging innovative technologies [8], both recently conducted by the INDDEX Project.

References

[1] International Food Policy Research Institute, Global Nutrition Report 2015: Actions and Accountability to Advance Nutrition and Sustainable Development. 2015: Washington, DC.
[2] Black, R.E., et al., Maternal and Child Undernutrition and Overweight in Low-Income and Middle-Income Countries. The Lancet, 2013. 382: p. 427-451.
[3] Shrimpton, R. and C. Rokx, The Double Burden of Malnutrition: A Review of Global Evidence, W. Bank, Editor. 2012.
[4] Popkin, B.M., L.S. Adair, and S.W. Ng, The Global Nutrition Transition: The Pandemic of Obesity in Developing Countries. Nutrition Review, 2012. 70(1).
[5] Popkin, B.M. and P. Gordon-Larson, The nutrition transition: worldwide obesity dynamics and their determinants. International Journal of Obesity, 2004. 28.
[6] http://inddex.nutrition.tufts.edu/
[7] http://inddex.nutrition.tufts.edu/sites/inddex.nutrition.tufts.edu/files/publications/INDDEX_LA_Report_2016.pdf[8] Coates, J, et al. Scaling up Dietary Data for Decision-Making in Africa and Asia: New Technological Frontiers, The FASEB Journal, vol. 30, no. 1 Supplement 669.15.

[/su_spoiler]

[su_divider top=”no” size=”1″]

[su_row]
[su_column size=”1/4″] [/su_column]
[su_column size=”3/4″]Prof. Efstratios Gavves
egavves@uva.nl
University of Amsterdam, The Nethelands

Dr. Efstratios Gavves is an Assistant Professor with the University of Amsterdam in the Netherlands and Scientific Manager of the QUVA Deep Vision Lab. After Efstratios completed his PhD with Prof. Arnold Smeulders and Dr. Cees Snoek in 2014 at the University of Amsterdam, he worked as a Post-doctoral Researcher at the KU Leuven with Prof. Tinne Tuytelaars. He has authored several papers in major Computer Vision and Multimedia conferences and journals, including CVPR, ICCV, ECCV, PAMI, IJCV, CVIU. His research interests include, but are not limited to, statistical and deep learning with applications on computer vision tasks, like object recognition, image captioning, action recognition, memory networks and recurrent networks, tracking.

[/su_column]
[/su_row]
[su_spoiler icon=”caret” title=”A Novel Perspective of Image Search for Tracking and Actions“]

In this talk I will focus on how image retrieval and visual search can be re-purposed for tasks that traditionally are considered to be very different.More specifically, I will first discuss a new, retrieval-inspired tracker, which is radically different from state-of-the-art trackers: it requires no model updating, no occlusion detection, no combination of trackers, no geometric matching, and still deliver state-of-the-art tracking performance on state-of-the-art online tracking benchmarks (OTB) and other very challenging YouTube videos. Departing from tracking, I will next focus on the relation between image search and other types of modalities that are not strictly speaking images, such as motion. More specifically, I will discuss a novel method for converting motion, or other types of sequential, dynamical inputs into just standalone, single images, so called “dynamic images”. By encoding all the relevant dynamic, information into simple single images, dynamic images allow for the use of existing, off-the-shelf image convolutional neural networks or handcrafted machine learning algorithms. The works presented in the talk have been published in the latest CVPR 2016 conference.

[/su_spoiler]

[su_divider top=”no” size=”1″]

[su_row]
[su_column size=”1/4″] [/su_column]
[su_column size=”3/4″]Dr. Thomas Mensink
thomas.mensink@uva.nl
University of Amsterdam, The Nethelands

Dr. Thomas Mensink received the M.Sc. degree, with honors, in artificial intelligence, from the University of Amsterdam (UvA) in 2007. He worked towards a Ph.D. jointly with the LEAR team of INRIA Grenoble and the Computer Vision team of Xerox Research Centre Europe. His PhD has received the AFRIF Thesis award 2012 for the best Ph.D. thesis in pattern recognition in France. Currently, he is post-doctoral scholar at the University of Amsterdam and received the prestigious young talent VENI award from the Netherlands Organisation for Scientific Research (NWO) in 2015. He focuses on machine learning for image and video classification. More specifically, his current research addresses learning semantic representations for visual data, which is essential when learning from few, or even zero, examples. He is lecturer of a MSc course on Visual Search Engines and have taught several classes on zero-shot learning for computer vision for different audiences. Some milestones include: over 1200 citations for the ECCV’10 paper about the Fisher Vector for image classification, ACM Multimedia 2014 best paper award, and ACM ICMR 2016 best paper award.

[/su_column]
[/su_row]
[su_spoiler icon=”caret” title=”Learning to Reuse Visual Knowledge“]

The central question in my talk is how existing knowledge, in the form of available labeled datasets, can be (re-)used for solving a new (and possibly) unrelated image classification task. This brings together two of my recent research directions, which I’ll discuss both. First, I’ll present some recent works in zero-shot learning, where we use ImageNet objects and semantic embeddings for various classification tasks. Second, I’ll present our work on active-learning. To re-use existing knowledge we propose to use zero-shot classifiers as prior information to guide the learning process by linking the new task to the existing labels. The work discussed in this talk has been published at ACM MM, CVPR, ECCV, and ICCV.

[/su_spoiler]

[/su_accordion]