Work

Identifying and addressing privacy challenges of health data

Year: 2025
Type: dissertation
Abstract: Health-related data (HRD) is more ubiquitous than ever. It is generated in the clinical setting through the use of healthcare information systems by practitioners, and in the personal setting, through the use of digital services designed for health purposes, or general-purpose services, not specifically designed for health. Although HRD has many benefits in the advancement of healthcare, they are considered highly sensitive. It is thus essential to address privacy and security issues pertaining to HRD. In this thesis, we explore multiple aspects of HRD privacy. First, we design, implement, and evaluate an efficient and privacy-preserving protocol for collaborative training of decision tree models on distributed biomedical datasets. Our protocol follows a federated learning approach, where raw data is not shared across providers nor centralized. Only updates to the trained model are shared and securely aggregated by data providers, using additive secret-sharing. Our protocol achieves a good balance between model accuracy, computational efficiency, and privacy, making it suitable for use in the healthcare setting. Second, we examine shadow HRD: HRD generated and/or processed by individuals by using general-purpose digital tools outside of a professional healthcare information system. These can be overlooked when enforcing data protection laws pertaining to HRD, as they are difficult to detect. Through focus-groups, we identify and categorize a broad variety of user behaviors that lead to the creation of shadow HRD, thus empowering policy makers, privacy researchers, and users to easily locate shadow HRD, and design or implement appropriate privacy protections. We also analyze the privacy policies of popular service providers, both manually, and programmatically using a large-language model (LLM). We show that although some service providers do treat shadow HRD as sensitive data, many do not even acknowledge their collection through their service. Third, using a mixed-methods approach of an online user survey and two sets of interviews, with technology users and clinicians, we assess the prevalence of shadow HRD-creating behaviors, users' awareness and concerns with respect to shadow HRD, and the relevant privacy-utility trade-offs. We find that users deem that the benefits of using general-purpose tools for managing their health outweigh the privacy risks to their health data, and hence will continue heavily engaging in shadow HRD-creating behaviors. Furthermore, users rarely resort to privacy protection measures when doing so. This highlights the need for better protections at the service providers' and legal levels.
Author: El Zein, Yamane
Language: English
Open Access status: green
OpenAlex ID: W7047873055
Cites: 320
Cited by:
Related to: 10