Research Topics

​​We develop and evaluate digital infrastructures that support efficient patient care processes and enab​​le high-quality medical research. Our work focuses on the integration of clinical workflows, data interoperability, and secure information management to foster innovation and improve healthcare delivery.

Key research di​​​rections​​​ include:

Projects: PrivateAIM, SafeLink (PPRL), PATH, AMPEL​

Key research directions inclu​d​​e:

Our research group develops advanced artificial intelligence methods to enhance clinical decision support through the analysis of electronic health records (EHRs) and multimodal data integration. By combining structured clinical data, medical imaging, laboratory results, and patient narratives, we design machine learning and deep learning models that generate comprehensive, data-driven insights to assist clinicians in diagnosis, prognosis, and tr​eatment planning. Our work also emphasizes explainable and trustworthy AI, ensuring that the developed systems are transparent, ethically grounded, and reliable for real-world healthcare applications.

Advanced generative modeling techniques, such as Generative Adversarial Networks (GANs) and Variational Autoen​coders (VAEs), are employed to generate realistic, privacy-preserving medical data that address key challenges in clinical research, including limited data availability and strict privacy constraints. By generating synthetic datasets that accurately capture the statistical properties of real medical data without exposing sensitive information, this work enhances data accessibility and supports robust machine learning development and implementation. The research focuses on evaluating the fidelity and downstream utility of synthetic data across structured and unstructured modalities, with the goal of improving model generalizability, reproducibility, and performance in medical diagnosis, prognosis, and clinical decision support.