Phone
Research Interests
My research centers around the development of machine learning solutions for various medical image analysis applications. From a technical perspective, I am particularly interested in developing and advancing new generative machine learning methods that accurately model the complex dynamics and variations of normal or pathological processes in the human body. My methods usually rely on mathematically sound and verifiable concepts from fields such as computational anatomy and related areas. The developed models can then serve as computer-aided diagnosis support tools or as tools for systematic data exploration in research scenarios. While the sensitivity and specificity of the models is of paramount importance in a healthcare context, my work also explicitly focuses on the explainability/interpretability and fairness of their decisions to enhance acceptability and trust by clinicians and patients. Finally, I am also interested in developing methods that achieve good results even if trained with limited data, a common problem in medical settings.
Over the years, I have been involved in numerous applied, interdisciplinary research projects where the machine learning methods developed by me or by my mentees have been successfully applied. This includes work on respiratory motion modeling for radiation therapy, various neuroimaging-related tasks involving cross-sectional or longitudinal imaging data and ocular and non-ocular disease detection and imaging biomarker discovery from retinal imaging data.
Research Areas:
Machine learning, medical image analysis, decision support, generative AI, explainability/interpretability