Framework to Audit the Compliance of Medical Images with Swiss Data Protection Law

State: Assigned to MK
Published: 2023-10-23

The importance of sensitive medical data in AI-based image recognition in the medical sector is closely tied to adherence to data protection regulations. In an era of heightened concern for patient privacy, following data protection regulations is not just a legal obligation but a moral imperative. Compliance with Swiss laws is critical. These regulations ensure that patient data is handled with the utmost care, minimizing the risk of data breaches and unauthorized access. By respecting these rules, healthcare organizations avoid costly penalties and maintain patient trust. 


Therefore, this thesis is dedicated to creating, implementing, and validating a software module responsible for auditing the compliance of medical data (images) utilized in AI-driven early diagnosis frameworks with Swiss data protection regulations. Specifically, the thesis begins by delving into Swiss data protection legislation to grasp the essential criteria for handling sensitive data stored in cloud environments. Subsequently, the work focuses on devising the necessary software components for scrutinizing whether medical images, as utilized by an existing framework, meet the identified compliance requirements. Then, these components will be implemented using Python as the programming language. Finally, the integrated software components will be integrated into an existing AI-based framework, where a series of experiments are conducted to assess the usability and performance of the proposed solution.


Additional info:

•The thesis will be done in the context of a research collaboration between the Balgrist University Hospital and the Communication System Group at UZH, where a part-time admin position under the Digital Society Initiative is possible to coordinate the collaboration.

•The thesis will be validated in a real AI-based framework with medical images from the radiology department of the Balgrist University Hospital.

•The thesis will be supervised by personnel from UZH/CSG and Balgrist University Hospital.



30% Design, 40% Implementation, 30% Documentation

Supervisors: Dr Alberto Huertas

back to the main page