The relationships between the heart and the kidney occupy an important place in the clinic, but are still poorly understood at the fundamental level. For example, the majority of patients with heart failure die of heart-related, not kidney-related problems. The goal of this project is to use MRI to correlate different renal and cardiac parameters in patients, particularly those related to tissue fibrosis. As part of an FNS project, we have developed a RESOLVE-based protocol that allows us to significantly improve image quality and resolution, to robustly measure diffusion parameters in the renal medulla and cortex. In collaboration with the Department of Nephrology, we conducted a study of 130 patients with chronic kidney disease and we were able to show a significant correlation between the cortico-medullary difference of the diffusion coefficient measured by MRI and the level of fibrosis measured by biopsy. In order to extend the validity of these results, we will start a multicenter study as part of a European COST project. We are also working on an MRI protocol to simultaneously measure cardiac and renal function in a clinically acceptable time.

Currently, the diagnosis of pulmonary hypertension relies mainly on invasive measurement of pressures in the pulmonary artery by catheterization. In collaboration with the Pulmonology and Cardiology departments, we are working on a method of estimating pressure using MRI and 4D flow imaging.

3D printing of heart models is very useful for planning a surgical or cardiological procedures. However, such processes take a considerable amount of time to complete. We are working on the development of a fast and effective strategy for the segmentation of CT or cardiac MRI to facilitate 3D printing.

Cardiac MRI can benefit from hybrid techniques incorporating other image modalities. We have developed a platform and hardware for in bore MRI compatible ultrasound. This project has two main areas of focus. At high MRI field, in obese patients or foetus, the ECG is often unreliable or absent. Doppler ultrasound can be used to replace this ECG signal to synchronise cardiac MRI acquisitions. In a second part, motion tracking of an ultrasound image can be used to send real-time positional information of the anatomy of interest and adjust the MRI acquisition accordingly. This could be particularly useful for cardiac valve imaging, where the valve moves out of a conventional 2D plane with the cardiac cycle.