Joint reconstruction and segmentation from undersampled MRI data

Researchers: Martin Benning, Veronica Corona, Carola-Bibiane Schönlieb

Magnetic resonance imaging (MRI) is widely used in medical applications, providing excellent soft tissue contrast. It uses fast acquisition techniques which may yield incomplete measurements, posing challenges for reconstruction and further analysis of the data. One common imaging task is segmentation, which is typically done after acquisition. However, it has been shown for different imaging techniques (x-ray, SPECT, PET) that solving both problems simultaneously can improve performances (Storath et al, 2015).

In this project, we want to provide a new reconstruction tool to simultaneously recover the image data from limited and noisy samples and identify a partition of the image in its regions of interests (e.g. organs, tumours).

Who's involved