Synthetic Data and PETs for Privacy-Compliant mHealth Within the EHDS: A Viewpoint Analysis

Area of Science:

• Health Informatics
• Data Privacy
• Artificial Intelligence

Background:

• The European Health Data Space (EHDS) aims to standardize health data sharing across EU Member States.
• Ensuring compliance with the General Data Protection Regulation (GDPR) is a primary objective for the EHDS.
• Mobile health (mHealth) systems face challenges in data sharing and regulatory adherence.

Purpose of the Study:

• To investigate synthetic data generation using Variational Autoencoders (VAEs).
• To explore Privacy-Enhancing Technologies (PETs), including Federated Learning, for mHealth.
• To assess the suitability of these methods for privacy-preserving and interoperable mHealth systems within the EHDS.

Main Methods:

• Generation of synthetic health data utilizing Variational Autoencoders (VAEs).
• Application of Privacy-Enhancing Technologies (PETs) such as Federated Learning.
• Evaluation of these techniques against GDPR's privacy-by-design principles.

Main Results:

• Synthetic data and PETs demonstrate potential for enhancing data privacy in mHealth.
• These methods align with GDPR requirements for data protection.
• Interoperability of mHealth systems can be improved through these privacy-preserving solutions.

Conclusions:

• Synthetic data and PETs are viable solutions for secure mHealth data sharing under the EHDS.
• Adoption of VAEs and Federated Learning supports GDPR compliance and privacy-by-design.
• These technologies facilitate the development of interoperable and trustworthy mHealth ecosystems.