NEMO – Non-identifiability of electroencephalograms (EEG) and comparable sensor signals from medical care for Open Science
The goal of this project is to develop and test new methods for analyzing biosignals (using EEG data from sleep monitoring systems as an example) while also protecting individuals’ privacy.
Specifically, the project (called NEMO) will create a toolkit for quantifying the risks of re-identification (i.e. revealing an individual’s identity) based on different privacy metrics. The toolkit will also include methods for anonymizing the EEG data and related metadata to minimize these risks. To accomplish this, NEMO will use statistical analysis, differential privacy, and anonymization techniques for unstructured, time-dependent data.
The main challenge is to balance privacy requirements with data analysis needs. To address this, the toolkit will be modular and configurable so that users can evaluate the quality of anonymization and make trade-offs transparently. Additionally, the project will develop tools for exploring data and create an integrated, interoperable data platform that considers data protection and scalability requirements. The platform will be designed for maintainability to ensure sustainable data use.
