Machine Learning and Data Science Group
In our machine learning and data science group we aim to develop novel ways of handling and analyzing data in radiology. Recent advances in data science and computer vision provide numerous possibilities for radiology to improve patient care and provide added value in various clinical settings.
Our main interest lies in improving radiological reports through structured reporting and usage of quantitative imaging biomarkers. Furthermore, we develop and apply deep learning techniques to improve the imaging workflow, from image acquisition and reconstruction to image interpretation.
It is our belief that through openness and transparency we can strengthen radiology’s role as leader of innovation in medicine.
- Enhancement of current clinical infrastructure to allow for large-scale machine learning projects
- Development of a deep learning based image classification system for fully automated quality assurance in radiographs
- Development of quality assured templates for structured reporting
- Evaluation of reading patterns in radiological reports using eye-tracking
- Development and evaluation of methods for analysis of body composition in CT and MR
- Design of cross-department methods for data sharing
- Modeling radiation exposure of radiological staff members in interventional radiology
- Multiparametric Imaging and Radiomics Group, University of Cologne (PD Dr. B. Baeßler)
- Cancer & Immunometabolism Research Group, University of Cologne (PD Dr. S. Theurich)
- Department of Nuclear Medicine (Dr. J. Hammes)
- Institute of German Language and Literature I, University of Cologne (Dr. F. Kretzschmar)
- Visual Computing Group, Institute of Computer Science, University of Mainz (Prof. Dr. M. Wand / S. Brodehl MSc)
- Department of Radiology, University Medical Center Mainz (Prof. Dr. P. Mildenberger / PD Dr. R. Kloeckner / Dr. F. Jungmann)
- European Society of Medical Imaging Informatics
- Working Group on Information Technology, German Radiological Society