Precision global health in the digital age

Precision global health is an approach similar to precision medicine, which facilitates, through innovation and technology, better targeting of public health interventions on a global scale, for the purpose of maximising their effectiveness and relevance. Illustrative examples include: the use of remote sensing data to fight vector-borne diseases; large databases of genomic sequences of foodborne pathogens helping to identify origins of outbreaks; social networks and internet search engines for tracking communicable diseases; cell phone data in humanitarian actions; drones to deliver healthcare services in remote and secluded areas. Open science and data sharing platforms are proposed for fostering international research programmes under fair, ethical and respectful conditions. Innovative education, such as massive open online courses or serious games, can promote wider access to training in public health and improving health literacy. The world is moving towards learning healthcare systems. Professionals are equipped with data collection and decision support devices. They share information, which are complemented by external sources, and analysed in real time using machine learning techniques. They allow for the early detection of anomalies, and eventually guide appropriate public health interventions. This article shows how information- driven approaches, enabled by digital technologies, can help improving global health with greater equity.

Contact networks across ages: computational analysis of epidemiological dynamics

Gianrocco Lazzari

For decades mathematical modeling in epidemiology has helped understanding the dynamics of infectious diseases, as well as describe possible intervention scenarios to prevent and control them. However, such models were relying on several assumptions, such as the ones on the structure of the underlying contact networks. The robustness of their predictions was therefore limited by this lack of knowledge. About a decade ago, with the advent of digital epidemiology, scientist have finally started to try to corroborate those assumptions, for instance with the use of wearable sensors to measure indeed contact networks. In this thesis, together with collaborators, I try to combine the digital collection of public health data with computational tools, in order to have a more realistic understanding of the phenomena under consideration. In two projects it was possible to finalise such marriage, thanks also to fruitful collaborations with other researchers who provided the data. This is for instance the case for the two chapters respectively on the modeling of influenza and plague outbreaks. Although they involve different technologies for the data collection, historical epochs and data types, the traditional epidemiological modeling allowed us to derive interpretable conclusions, capable for instance to inform public health interventions. In other projects, either the relevant data collection is still ongoing in the lab (like for the FoodRepo project), or the data collection has not started yet (like for the project on measles), although our work provides insights on the importance of such data collection for future studies. In the first chapter, we explore different mechanistic interpretation compatible with our data on the 1630 plague outbreak in Venice, collected through the digitisation of parish books from the historical State Archives of Venice. The data shows a non trivial temporal structure, which led us to propose few different epidemiological explanations. Further data collection will be needed to better constrain such interpretations. In the second chapter, we use previously recorded contact data in a high-school to assess the relative effect of ventilation on the influenza spread, with respect to vaccination strategies. Our result suggest the usefulness of non pharmacological interventions such as indeed improved ventilation, which become even more meaningful in the context of vaccination hesitancy and low vaccine efficacy, due for instance to the high mutation rate of viruses like influenza virus. In the third chapter, we propose a simple network generation model to try to explain differences in the incidence of highly infectious diseases (such as measles), across countries with similar vaccination coverages. Such differences are indeed one of the main open questions in public health, which are not yet fully understood even considering social phenomena such as recent anti-vax movements. In the last chapter, we present our open database of barcoded food products, FoodRepo. This database represents on the one hand, the first piece of a large study ongoing in our lab, in the field of nutritional epidemiology, that aims to assess the variability of glycemic response in an healthy cohort. On the other hand, important features such as its openness and programmatic accessibility make it an important digital tool at the service of any private or public actors in the field of nutrition.

EPFL2020

Understanding Eating And Drinking In Context From Crowdsourced Data

Thanh Trung Phan

With the increasing rate of urbanization, understanding food and beverage consumption, including alcohol drinking behaviour with its consequences, is relevant in such a megacity in the future. Especially, investigation of alcohol drinking is necessary for uncovering more drinking activities of young people in context, and providing more scientific references to authorities to improve public policies, in particular in the domain of public health. In previous research, data were collected a posteriori in face-to-face interviews or by using questionnaires. Thanks to the development of feature phones, an evolution of more traditional methods in ubiquitous food and alcohol research could have collected survey data via SMS. However, these methods have limitations, including a low recall and an expensive scaling up. On the other hand, the adoption of smartphones and social media is opening new channels for investigating behaviours by collecting fine-grained in-situ data, following methodologies from social sciences, and using advanced technologies from computer sciences. Recently, crowdsourcing is a new paradigm that consists in using the inputs from a great number of people to facilitate and accelerate large scale data collection from broad samples, compared with traditional methods. In addition, mobile crowdsourcing, a form of crowdsourcing, has enormous potential in collecting in-situ data by taking advantage of embedded sensors, cameras, and being equipped with Internet connection. In this dissertation, we investigate the drinking and eating behaviour of young people in Switzerland, based on crowdsourcing data including records and metadata from mobile sensors (mobile crowdsensing) and data shared on social networks. Our contributions are three-fold. First, we conduct data analyses that uncover generic food and drink consumption on Instagram and reveal two types of drinking practices (casual and heavy drinking) on social media. This analysis provides an initial snapshot of the food consumption and alcohol drinking practices based on the way they appear online, and creates a preliminary alcohol consumption model that is developed in the rest of the dissertation. Second, we use mobile crowdsensing data, annotated a posteriori, both to identify heavy drinking and to understand the characteristics of private spaces (including ambiances) and activities (including drinking activities) of young people in the weekend nights in Switzerland. These results show how mobile crowdsensing data can be used to better understand and predict alcohol drinking practices and ambiances in private spaces. Third, we combine mobile crowdsensing data with social media to retrieve the multi-factorial characteristics of drinking events depending on the type of beverage (multiple alcoholic and non-alcoholic categories) based on images features and contextual cues from individual and joint data sources. This result indicates the feasibility of using, individually or combined, data from mobile crowdsensing and social networks in discriminating drinking behaviour. This is a promising sign towards the development of a system that uses machine learning for self-monitoring of alcohol consumption. This dissertation, by combining the advanced machine learning of computer science and literature of social science, demonstrates the relevance of using a multidisciplinary approach to investigate social behaviours in urban areas.

EPFL2020

Understanding Eating And Drinking In Context From Crowdsourced Data

Thanh Trung Phan

EPFL2020

Contact networks across ages: computational analysis of epidemiological dynamics

Gianrocco Lazzari

EPFL2020