Josefine Vilsbøll Sundgaard

Josefine Vilsbøll Sundgaard is a Postdoctoral Researcher in the Visual Computing section of the Department of Applied Mathematics and Computer Science (DTU Compute) at the Technical University of Denmark (DTU), Faculty of Engineering. Her work centers on medical image analysis using deep learning techniques for disease diagnosis and biomarker discovery in clinical settings.

She earned her PhD from DTU in 2022 through research on deep learning methods for pediatric middle ear diagnostics, supervised by R. R. Paulsen and A. N. Christensen. Her educational background includes specialized training in computational methods for medical applications.

Her research spans medical image analysis with emphasis on deep learning, reinforcement learning, and anomaly detection. Key applications include otitis media diagnosis using tympanometry and normative data, cardiovascular imaging for coronary artery segmentation and left ventricular remodeling analysis, and liver disease diagnosis through MRI biomarker identification for nonalcoholic fatty liver disease and fibrosis. She integrates computer vision with clinical diagnostics to address pediatric and adult conditions.

Analysis of her 2024-2025 publications reveals a strong trend toward deep learning applications in cardiovascular and hepatic imaging. Her work demonstrates expertise in active learning frameworks, segmentation optimization, and anomaly detection systems using clinical CT and MRI data, with significant contributions to coronary artery analysis, cardiac adipose tissue quantification, and liver fibrosis identification in large-scale biobanks.

Scientific awards: No awards or fellowships were documented in the provided materials.

She actively supervises three PhD candidates: Radutoiu, A.-T. (focusing on HFpEF disease manifestations), Aspe, A. W. (analyzing cardiometabolic image biomarkers in CT), and Belmpeisi, R. A. (identifying abdominal MRI biomarkers). Her research is supported by five major projects including Deep learning for identifying biomarkers in medical images and AI driven analysis of cardiometabolic image biomarkers, with funding extending through 2028.

As a core member of DTU Compute's Visual Computing research group, she collaborates extensively with medical institutions including Rigshospitalet, contributing to interdisciplinary teams that develop AI solutions for clinical diagnostics across otolaryngology, cardiology, and hepatology.