Research Topics
The aim of our research group is to develop, test and deploy artificial intelligence methods in radiology as well as to perform data curation, pre- and post-processing.
Our current projects include deep learning methods for image segmentation and classification and radiomics feature extraction and selection for image classification. However, with the rapid development of techniques, we are exploring other techniques such as unsupervised methods, generative adversarial networks, combining different sources of data among others.
The focus of the new techniques is to deal with the specificities of Medical Imaging: few amount of data and/or annotations, unbalanced data, lack of protocol harmonisation and biased datasets.
Research Projects
- Segmentation, Classification and Characterisation of Lymph Nodes
- Classification and Characterisation of Lung Nodules and Masses
- Segmentation, Classification and Characterisation of Brain tumours
- Characterisation of Prostate Lesions
- Radiomics reproducibility
- Pulmonary Hypertension Evaluation
- Kidney Stones Segmentation and Characterisation
- Kidney Evaluation for ADPKD patients
- Aneurysma Evaluation
Grants
- RISK – Risk Maps using T2 mapping and Diffusion MR Sequences of the Prostate (ESR Seed Grant Award from the European Society of Radiology (ESR) and the European Institute for Biomedical Imaging Research (EIBIR))
- GAIA – Group for Applied Imaging Application of Artifical Intelligence (Köln Fortune)
- DeepLesion - Bring Brain Tumour Segmentation into Clinical Practice
- RACOON - Radiological Cooperation Network