Research

Data science research at Maastricht University aims to accelerate scientific discovery through the development of powerful Artificial Intelligence (AI) platforms coupled with FAIR data and services to systematically unlock knowledge about the world we live in. Furthermore the improvement of clinical care and well-being through the creation of intelligent systems to bring the science of medicine back into the practice of medicine. Also the empowering of communities to characterize, implement, and monitor data-driven solutions that optimize their investments to maximize their quality of life through data mining and machine learning.

 

Core research themes

data science - core research themes
Niet ge- definieerd

Accelerating scientific discovery

Accelerating scientific discovery through the development of powerful Artificial Intelligence (AI) platforms coupled with FAIR data and services to systematically unlock knowledge about the world we live in.

Most aspects of scientific inquiry remain entirely dependent on human expertise and effort. However, no one person can keep track of the vast amounts of data, tools, and knowledge being generated each and every day. Moreover, with recent reports claiming rates of non-reproducibility of 64% in psychological studies and up to 89% in pharmacological studies, there is an urgent need to find effective approaches to conduct scientific investigations that takes into account and makes sense of the vast amount of data in a more reliable manner. Recent developments in data-driven systems show that machines can effectively ingest complex information and outperform humans in many tasks.

Topics include, but are not limited to:

  • Summarizing what is known about a topic and contextualizing new findings for an individual
  • Incentive-friendly platforms for sharing digital research objects (e.g. data, software, publications) that are maximally FAIR – Findable, Accessible, Interoperable, Reusable
  • Uncovering evidence that supports or potentially disputes a scientific assertion
  • Designing, prioritizing, and orchestrating experiments to fill in knowledge gaps
  • Tracking scientific progress, evolution of scientific disciplines, and scientific impact.
  • Adaptive data-driven learning to maximize the accuracy and completeness of any human endeavour
Data science - Core research themes
Niet ge- definieerd

Improving clinical care and well-being

Improving clinical care and well-being through the creation of intelligent systems that bring the science of medicine back into the practice of medicine.

Despite ritualistic and time-consuming chronicling of patient encounters in electronic health care systems worldwide, these on their own cannot translate into improved patient outcomes. Meanwhile, new technology is filing the gap to monitor and advise patient health on a per second basis, but often have uncertain health benefits. Once again, vast amounts of data are becoming readily available, but we lack the infrastructure, methodology, and understanding of the social, legal, and ethical aspects of health information systems to maximally advantage of these developments. Our goal in this research theme is to exploit emerging sources of health-relevant data to improve clinical care and well being.

Topics include, but are not limited to:

  • Using distributed data in a privacy-preserving manner to identify environmental determinants of health at the level of an individual.
  • Platforms to undertake systematic comparative effectiveness research using millions of patients worldwide
  • Methods for continuous and personalized diagnosis, prognosis, and treatment.
  • Methods to improve the quality of the physician-patient interaction. e.g. constructing a brief, but accurate patient summary from patient’s social networking data.
  • Methods to aid care workers find and work with people who need help at the time they need it.
Data science - core research themes
Niet ge- definieerd

Empowering communities

Empowering communities to characterize, implement, and monitor data-driven solutions that optimize their investments to maximize their quality of life. It wasn’t that long ago that the tools for multivariate analysis, data mining and machine learning had limited utility and could only be used by specialists.

Today, data science is more accessible and easier to use than ever – open source frameworks are freely available, compute-hungry analyses can be readily executed on cloud infrastructure, and enormous online communities are sharing their recipes and their accomplishments. Indeed, people from all walks of life are applying da elieve that data science can be the bridge that brings youth and community leaders together to better understand the problems that they face and craft solutions to make life better for them and their neighbours. In this research theme, we will focus on bringing data science to a region nestled in the heart of Western Europe, which is making investments to overcome serious economic challenges, and must strive to include youth as part of a strategy for economic and social prosperity.

Topics of interest include, but are not limited to:

  • Digital infrastructure to enable collaboration on policy making
  • Infrastructure for data sharing while strengthening privacy controls and cybersecurity for citizens and industry
  • Data driven methods to foster job creation and reduce social inequality
  • Identification and removal of barriers to digital innovation and commercialization for young data scientists and entrepreneurs
  • Accelerating scientific discovery

    Dit is er niet
  • Improving clinical care and well-being

    Dit is er niet
  • Empowering communities

    Dit is er niet

Projects

Data science - projects
Niet ge- definieerd

Analyzing partitioned FAIR health data responsibly

The objective of the project is combining and learning from access-restricted FAIR health and socioeconomic data across entities in a manner that preserves privacy.

Project description

Health “Big Data” is extremely privacy sensitive. Using it responsibly is key to establish trust and unlock the potential of this data for the health challenges facing Dutch society now and in the future. One of the unique characteristics of Big Data in health is that it is extremely partitioned across different entities. Citizens, hospitals, insurers, municipalities, schools, etc. all have a partition of the data and nobody has the complete set. Sharing across these entities is not easy due to administrative, political, legal-ethical and technical challenges.

In this project, we will establish a scalable technical and governance framework which can combine access-restricted data from multiple entities in a privacy-preserving manner. From Maastricht UMC+, we will use clinical, imaging and genotyping data from the Maastricht study, an extensive (10.000 citizens) phenotyping study that focuses on the aetiology of type 2 diabetes. From Statistics Netherlands (CBS), which hosts some of the biggest and most sensitive datasets of the Netherlands, we will use data pertaining to morbidity, health care utilization, and mortality. Our driving scientific use case is to understand the relation between diabetes, lifestyle, socioeconomic factors and health care utilization, which will inform guidelines with major public health impact.

Principal investigators
Prof. Michel Dumontier, Institute of Data Science
Prof. Andre Dekker, chair of Clinical Data Science at Maastricht UMC+
Prof. David Townend, professor of Law and Legal Philosophy at the UM Faculty of Health, Medicine and Life Sciences
Assoc. Prof. Annemarie Koster, Maastricht Study
Bob van den Berg, MSc, Center for Big Statistics

Data science - Projects
Niet ge- definieerd

Student project

PREMIUM: Data Science Community

Vast amounts of data being generated across all segments of society. If taken advantage of, these data offer an unprecedented opportunity to accelerate scientific discovery, to improve healthcare and wellbeing, and to strengthen our communities. Data Science is an interdisciplinary field concerning the scientific methods, systems and workflows to obtain insights from data.

Data Science has the potential to affect all aspects of human activity. At the Institute of Data Science @UM, we embrace this development and are preparing a new generation of data scientists, scholars and entrepreneurs who work in a collaborative manner to tackle the world’s most pressing problems.

Our goal is to be the academic home for data scientist at Maastricht University and in the region. We will develop infrastructure to support this idea along with a series of activities to network disparate groups, whether they be, kids, students, researchers, developers, alumni, professionals, and entrepreneurs.

Read more
MSP - Student for a Day
Niet ge- definieerd

Student project

Honours+: Data Science Project

What makes a student happy with their studies at UM and why? If you could go back in time and study again, what course would you study? Why? What do you think about fields of study that are outside your own? How has rapidly advancing technology affected your choice of career and your perception of other fields of study? Would you consider being employed by UM?

To address the above questions, we would like to explore advanced techniques in data science such as machine learning, natural language processing and automated reasoning, to comprehend data not only from survey questionnaires, but also from corroborated psychological evaluations, as well as studies of cultural, social, legal and ethical influences on students study and employment choices.

This project will involve a deep study of student sentiment at UM to uncover:

  • Students' perception of their context in the broader learning landscape at UM and how perceptions may influence interdisciplinary activities at UM.
  • Students' perception of technology and attitudes about its penetration into fields that have historically been reluctant to embrace it.
  • Factors which affect the satisfaction of students’ career path and study choices.
  • Factors which influence a students’ positive perception of UM as a potential employer for them.
Read more
  • Analyzing partitioned FAIR health data responsibly

    Dit is er niet
  • Student project

    PREMIUM: Data Science Community

    Dit is er niet
  • Student project

    Honours+: Data Science Project

    Dit is er niet

Infrastructure

We will develop computing infrastructure to support the mission of the institute and the development of Data Science @UM. Consequently, we will do this in close conjunction with ICTS and via participation in and interactions with diverse IT workgroups within UM. This includes, but is not limited to:

  • State of the art hybrid-cloud infrastructure
  • State of the art data storage
  • State of the art data science applications
  • State of the art data governance and compliance
  • Sustainability
  • Software and Licenses
  • Physical Space