Magnetic Resonance Imaging (MRI) is a well known, non-invasive medical imaging technique which is mainly used for diagnosis of soft tissue pathologies. Currently, the importance of MRI, especially in the context of Cardiovascular Magnetic Resonance (CMR), has been growing as it provides significant advantages over other imaging modalities such as high spatial and temporal resolution, absence of ionizing radiation and non-invasive imaging technology. Therefore, diverse cardiovascular applications have been developed in MRI scanner.

In CMR imaging, there are several important factors to consider for capturing an image of the heart with a sufficient image quality. In order to perform CMR imaging properly, the prescans such as localization, slice planning, shim box positioning, valve tracking or resting phase detection are needed which are done by medical expert manually, thus very time-consuming. With the help of deep learning algorithms, the prescan protocols can be performed automatically.

E-mail: seung.su.yoon@fau.de