1 Department of Applied Mathematics and Computer Science, Technical University of Denmark2 Cognitive Systems, Department of Applied Mathematics and Computer Science, Technical University of Denmark3 Department of Informatics and Mathematical Modeling, Technical University of Denmark4 Copenhagen Center for Health Technology, Center, Technical University of Denmark
Driven by the ubiquitous availability of data and inexpensive data storage, our ability to sense human beings has increased dramatically. Big data has permeated the public discourse and led to surprising insights across the sciences and the humanities. This dissertation presents research on expanding our capabilities in collecting, handling, processing, and using data collected about human beings to create an integrated view of social systems. The goal of the thesis has been threefold. The first part of the thesis focuses on the need, design, and implementation of large-scale sensor-driven human data collection studies. Social networks can be measured with high resolution and on multiple channels, such as face-toface meetings, social networks, or phone calls, in order to generate a more comprehensive picture of social systems. The largest study to date measuring large-scale social system—the Copenhagen Networks Study—is described, together with motivation and challenges of the deployment. Preliminary results are presented, indicating how a possibly biased and incomplete picture can be generated when data are collected from a single channel and with a low resolution, thus emphasizing the importance of the proposed approach and deployed implementation. The second part of the thesis deals with expanding our capabilities to sense the cognitive and emotional state of the users through development of a system for mobile brain imaging—the Smartphone Brain Scanner. A developed framework allows for EEG data collection and processing. It also provides the ability to build end-user applications on top of raw data and extracted features using off-the-shelf and custom-built neuroheadsets and mobile devices, thereby potentially becoming another channel in integrated human sensing. The motivation for creating such system is presented, advanced data processing—3D source reconstruction—is explained, and applications and use-cases are discussed. In the third part, the privacy issues surrounding the handling of such sensitive behavioral and biomedical data are investigated. A comprehensive review of best privacy practices in sensor-driven human data collection is presented and recommendations for practitioners are made. Based on this review and experiences with the Copenhagen Networks Study and the Smartphone Brain Scanner, the concept of Living Informed Consent is presented, which postulates larger participant control over collected data for the benefit of users, researchers, and society at large. The same privacy principles are applied to a personal neuroinformatics context, resulting in a proposed new approach to sensitive EEG data handling.